Forthcoming and Online First Articles

International Journal of Modelling, Identification and Control

International Journal of Modelling, Identification and Control (IJMIC)

Forthcoming articles have been peer-reviewed and accepted for publication but are pending final changes, are not yet published and may not appear here in their final order of publication until they are assigned to issues. Therefore, the content conforms to our standards but the presentation (e.g. typesetting and proof-reading) is not necessarily up to the Inderscience standard. Additionally, titles, authors, abstracts and keywords may change before publication. Articles will not be published until the final proofs are validated by their authors.

Forthcoming articles must be purchased for the purposes of research, teaching and private study only. These articles can be cited using the expression "in press". For example: Smith, J. (in press). Article Title. Journal Title.

Articles marked with this shopping trolley icon are available for purchase - click on the icon to send an email request to purchase.

Online First articles are published online here, before they appear in a journal issue. Online First articles are fully citeable, complete with a DOI. They can be cited, read, and downloaded. Online First articles are published as Open Access (OA) articles to make the latest research available as early as possible.

Open AccessArticles marked with this Open Access icon are Online First articles. They are freely available and openly accessible to all without any restriction except the ones stated in their respective CC licenses.

Register for our alerting service, which notifies you by email when new issues are published online.

We also offer which provide timely updates of tables of contents, newly published articles and calls for papers.

International Journal of Modelling, Identification and Control (135 papers in press)

Regular Issues

  • Self-tuning fuzzy logic PID controller with a practical view to PEM fuel cell air supply system   Order a copy of this article
    by Mehdi Rakhtala 
    Abstract: The polymer electrolyte membrane fuel cell (PEMFC) is an appropriate candidate in renewable energy resources to using in vehicular, industrial and other applications. This research paper concentrates on the nonlinear model of the PEMFC system. The load fluctuations in the fuel cell stack affect the lifetime and cause fuel cell stack damage and ageing. So, a closed-loop control system is suggested to regulate the oxygen excess ratio at the desired value. In this paper, a self-tuning fuzzy logic PID (FPID) controller is suggested for an PEMFC air supply system because the fuel cell is a very severely nonlinear system. The oxygen excess ratio is regulated to a desired value by adjusting the air flow-rate. The control scope is to adjust the oxygen excess ratio in its operating range by controlling the compressor. The proposed FPID controller is a nonlinear and robust controller that ensures good efficiency around each equilibrium point under model uncertainties and avoids oxygen starvation during load fluctuations.
    Keywords: PEMFC; self-tuning; fuzzy logic PID; oxygen excess ratio; durability.

  • An optimal control approach for hybrid motion/force control of coordinated multiple non-holonomic mobile manipulators using neural network   Order a copy of this article
    by Komal Rani, Naveen Kumar 
    Abstract: This paper presents an intelligent optimal control approach for motion/force control of cooperative multiple non-holonomic mobile robot manipulators carrying a single rigid object. Firstly, a combined model of multiple mobile manipulators and the object is obtained in terms of object coordinates. Using this model, a state-space form of error dynamics is derived for quadratic optimisation. Then, the explicit solution of Hamilton Jacobi Bellman (HJB) equation for optimal control is obtained using the Riccati equation. The linear optimal control, neural network and adaptive bound are used to design the proposed controller. It is shown that the uncertainties of the system are compensated using radial basis function neural network and adaptive compensator. The radial basis function neural network approximates the unknown dynamics and the adaptive compensator estimates the bounds on the neural network approximation error and the unstructured uncertainties of the system. The asymptotic stability of the closed-loop system is demonstrated using Lyapunov stability analysis and optimal control theory. Finally, the proposed control approach are verified in a comparative manner through simulation results with two identical mobile manipulators grasping the single rigid object.
    Keywords: multiple mobile manipulators; optimal control; Hamilton Jacobi Bellman optimisation; motion/force control; RBF neural network; adaptive compensator; asymptotic stability.

  • Differentially flat trajectory generation and controller design for a quadrotor UAV   Order a copy of this article
    by Arindam Singha, Anjan Kumar Ray, Arun Baran Samaddar 
    Abstract: A control strategy of differentially flat trajectory generation and a backstepping controller for tracking the desired trajectory are developed for a quadrotor Unmanned Aerial Vehicle (UAV). A globally smooth trajectory is generated through multiple waypoints, which are at different planes. To show the effectiveness of the trajectory generation method, four different shapes of trajectories are generated using different numbers of waypoints. Simulation studies have shown that the proposed paradigm is able to generate smooth trajectories using multiple waypoints. Along with that, the quadrotor UAV has successfully tracked the desired trajectories by using the proposed controller. The proposed controller is also validated with constant and variable desired yaw angles. The robustness of the proposed controller is validated in the presence of external disturbances in the system control input. The proposed controller is also compared with other controllers and shows satisfactory performances.
    Keywords: Andrews' curve; backstepping controller; differential flatness; trajectory generation; trajectory tracking; quadrotor UAV.

  • Modified augmented fractional order control schemes for cart inverted pendulum using constrained Luus-Jaakola optimisation   Order a copy of this article
    by Deep Mukherjee, G. Lloyed Raja, Palash Kundu, Apurba Ghosh 
    Abstract: Since the upright position of an inverted pendulum system is an unstable equilibrium, it is extremely challenging to control. Fractional order based control schemes are becoming increasingly popular in stabilising an unstable system and achieving satisfactory closed-loop performance. Therefore, a novel combination of fractional order Lyapunov (FOLyapunov) rule and fractional order proportional integral (FOPI)/two-degrees of freedom FOPI (2DOF-FOPI) controller is proposed to tackle this problem. Parameters of FOPI/2DOF-FOPI controllers are obtained using multi-objective constrained the Luus-Jaakola multipass optimisation method. Comparative simulation studies are carried out with a direct synthesis based PID control scheme, combined with fractional order Massachusetts Institute of Technology (FOMIT) rule augmented with FOPI/2DOF-FOPI controllers using a mathematical model of inverted pendulum. It is evident that the proposed combination of FOLyapunov method and FOPI/2DOF-FOPI controllers outperforms the other schemes.
    Keywords: inverted pendulum; MRAC; fractional calculus; MIT rule; FOMIT rule; FOPI; 2DOF-FOPI; Luus-Jaakola algorithm; FOLyapunov stability rule.

  • H model reduction of discrete-time 2D T-S fuzzy systems in finite frequency ranges   Order a copy of this article
    by Abderrahim El-Amrani 
    Abstract: This paper deals with the problem of H model reduction for two-dimensional (2D) discrete Takagi-Sugeno (T-S) fuzzy systems described by the Fornasini-Marchesini local state-space (FM LSS) model, over finite frequency (FF) domain. The problem to be solved in the paper is to find a reduced-order model such that the approximation error system is asymptotically stable, which is able to approximate the original T-S fuzzy system with comparatively small and minimised H performance when frequency ranges of noises are known beforehand. Via the use of the generalised Kalman Yakubovich Popov (gKYP) lemma, new design conditions guaranteeing the FF H model reduction are established in terms of linear matrix inequalities. To highlight the effectiveness of the proposed H model reduction design, a numerical example is given.
    Keywords: multidimensional systems; finite frequency H∞; model reduction; T-S fuzzy systems.

  • Social spider optimisation based identification and optimal control of fractional order system   Order a copy of this article
    by Sandip Mehta, Dipak Adhyaru 
    Abstract: Fractional order derivatives and integrals are infinite-dimensional operators and non-local in time. Currently, the researchers are working on the solution of the fractional optimal control problem using some approximation and numerical analysis. In this paper, a social spider-based constrained optimisation method is proposed to control the fractional order system. An effort has been made to translate the fractional optimal control problem to the standard unity feedback system. A multi Simpson method has been used to solve the integration of the performance function. The proposed method has not used any matrix computation and it has been demonstrated that it is easier to implement the FOCP (fractional order optimal control) method on the given hardware. Along with the optimal control, a simple identification technique is proposed for the fractional order system. The optimal controller has been designed using computational intelligence techniques. The error analysis and the performance analysis have been carried out for the proposed methods.
    Keywords: fractional order system; SSO-C; FOCP; metaheuristic algorithms.

  • Optimal fractional order control for nonlinear systems represented by the Euler-Lagrange formulation   Order a copy of this article
    by Ahmad Taher Azar, Fernando E. Serrano, Nashwa Ahmad Kamal 
    Abstract: In this paper, a novel control strategy is shown for the control of fractional order systems established in the Euler-Lagrange formulation. This strategy is based on the design of an optimal controller considering a fractional order system based in the Euler-Lagrange formulation because this allows more degrees of freedom in the system establishment and the optimal controller design. The design procedure consists of establishing a performance index and then, by finding the gradient of this index, an optimal control law is obtained with the initial and final conditions of the system. It is important to note that there are a limited number of studies related to this topic found in the literature. Finally, in order to test the theoretical results obtained in this work, a numerical example that consists of the stabilisation of a two-links robotic manipulator is shown.
    Keywords: optimal fractional order control; fractional order systems; optimal control; Lagrangian systems.

  • Feature pair pyramid detector for small product defect detection   Order a copy of this article
    by Zihao Huang, Hong Xiao, Tao Wang, Junhao Zhou 
    Abstract: There are many object detection algorithms that have performed well on public datasets, and they can be used in product defect detection. But there are still many details, which can affect the detection performance of actual product defect detection, that need to be optimised. In this paper, we design a defect detector, called feature pair pyramid (FPP) detector, and optimise it for specific industrial application using three methods. Then we use FPP detector to detect defects of metal can products of an enterprise. The experiment results show that our FPP detector is more effective in detecting small size defects. The performance (AP@0.5) of our detector is better than current state-of-the-art detectors.
    Keywords: defect detection; small object detection; feature pair pyramid; one-stage object detector; Resnet; k-means; anchor box.

  • Assessing the feasibility of underwater vehicle controllers in underactuated hovercraft via simulations   Order a copy of this article
    by Przemyslaw Herman 
    Abstract: The paper presents a comparison of several selected trajectory tracking controllers used for an underactuated hovercraft. The aim of the study is to check if algorithms that have proved to be effective for another class of vehicles can be adapted to control a hovercraft. In order to compare the effectiveness of the controllers, one that is exclusively designed to control the hovercraft and three others that are suitable for the control of underactuated underwater vehicles were examined. This study also proposes a methodology for simulation analysis of control algorithms. The initial simulation tests demonstrated on the 3-DOF hovercraft model show the results that can be obtained by using each control algorithm under the proposed assumptions.
    Keywords: underactuated hovercraft; nonlinear control; trajectory tracking; simulation.

  • Position tracking and balancing control of ball balancer system using intelligent controllers   Order a copy of this article
    by Ankita Varshney, Bharat Bhushan, Rupam Singh 
    Abstract: This paper develops an intelligent control approach for achieving the nonlinear control of a two degrees of freedom (2DoF) ball balancer system. The ball balancer system provides an example of an inherently unstable and underactuated electromechanical system that can be used to realise the problem of position tracking and balancing control in robotic systems. Besides, intelligent control takes into consideration the system characteristics and parameters and incorporates the heuristic knowledge and a human logical approach to best train the controller for achieving the desired control task. In this research, the robustness of the intelligent controller is realised by developing an adaptive neuro-fuzzy inference system (ANFIS), which is applied to the ball balancer system to achieve position and balancing control. The numerical simulations are carried out using the first-principles method to perform the system modelling, design, and development of the control strategy for a two-dimensional ball balancer system. The performance of the proposed controller is assessed in terms of the time domain characteristics, and a comparative study is done to demonstrate the superiority of intelligent control techniques over classical control techniques.
    Keywords: adaptive neuro-fuzzy inference system; artificial neural network controller; ball balancer system; intelligent controller.

  • Track planning of an multi-rotor unmanned aerial vehicle in a complex environment space   Order a copy of this article
    by Yue Chu, Zhong Hua Han, Liying Yang 
    Abstract: Aiming at the dynamic track planning problem of a multi-rotor UAV in a complex environment space, this paper proposes a high-dimensionality-reduced space environment modelling method. In this way, the complexity of the environment model will be reduced and the planning efficiency will be improved. In addition, the paper proposes an improved artificial potential field (APF) method. First, obtain overall environmental information through the A* algorithm, and optimise the global path nodes. Then, improve the potential function of the APF method, and add the attraction of the global path to the UAV, so that it can guide the UAV movement smoothly. By analysing the simulation results, it can be found that this method can make up for the shortcomings of the APF method in path guidance to a certain extent. The effective combination of the two algorithms improves the UAV's path planning ability in complex environments.
    Keywords: environment modelling; track planning; A* algorithm; APF; optimised path nodes.

  • Trajectory tracking for mobile manipulator based on nonlinear active disturbance rejection control   Order a copy of this article
    by Mhmed Algrnaodi, Maarouf Saad, Mohamad Saad, Raouf Fareh, Abdelkrim Brahmi 
    Abstract: This paper designs a nonlinear active disturbance rejection control (ADRC) to solve the trajectory tracking problem of a mobile manipulator (MM) in the presence of parameter uncertainties and nonlinear dynamics coupling effects of the MM system. The control scheme consists of a nonlinear extended state observer (NESO) and a nonlinear proportional derivative (PD) controller. Based on the Lagrange formulation, a dynamical model of the MM is formulated, where external disturbances and modelling uncertainties are assumed to be part of the total disturbance, which is estimated with an observer and rejected online in the control law. Since the proposed controller cannot be performed unless the full transformed state vector of the system model is available, an NESO is designed to estimate the transformed state vector as well as the uncertainties. The nonlinear PD controller uses the state estimated by the NESO, and the effect of uncertainties is cancelled online by the control input. Experimental results of the MM proposed tracking controller show its validity and efficiency.
    Keywords: active disturbance rejection control; mobile manipulator; modelling uncertainty; external disturbances; nonlinear extended state observer.

  • Real-time system identification of an unmanned quadcopter system using fully tuned radial basis function neural networks   Order a copy of this article
    by Mohammad Fahmi Pairan, Syariful Syafiq Shamsudin, Mohd Fauzi Yaakub, Mohd Shazlan Mohd Anwar 
    Abstract: In this paper, we present the performance analysis of a fully tuned neural network trained with the Extended Minimal Resource Allocating Network (EMRAN) algorithm for real-time identification of a quadcopter. Radial basis function (RBF) network based on system identification can be used as an alternative technique for quadcopter modelling. The number of neurons in the RBF are typically determined by trial and error approach. In order to prevent the neurons and network parameters selection dilemma, RBF with EMRAN recursive training algorithm is proposed. This automatic tuning algorithm will implement the network growing and pruning method to add or eliminate neurons in the current network. The EMRANs performance is compared with the Minimal Resource Allocating Network (MRAN) training for 1000 input-output pair untrained attitude data. EMRAN uses the winner neuron strategy to reduce training time for large neuron size as much as 88 hidden neurons. The results indicate that the EMRAN algorithm produces faster mean training time, around 4.16 ms compared with MRAN at 5.89 ms, with a slight reduction in prediction accuracy. The proposed model can predict the attitude of a quadcopter using untrained data.
    Keywords: quadcopter; system identification; neural network; fully tuned neural network; radial basis function.

  • An intelligent online detection approach based on big data for mechanical properties of hot-rolled strip   Order a copy of this article
    by Jinxiang Chen, Ziming Fan 
    Abstract: An LightGBM prediction model based on big data is presented in order to online detect the mechanical properties of hot-rolled strip, which can achieve greater accuracy than both the existing prediction approaches and hardware detection method for the local strips. A dataset of mechanical properties of hot-rolled strip is constructed firstly by collecting a steel plant's hot-rolled process control parameters, which includes 17,000 samples, and every sample contains 17 input characteristics and three output mechanical property parameters. Based on the dataset, an LightGBM intelligent prediction model is established and trained to predict the three mechanical properties of the hot-rolled strip steels. A 17,000 dataset of hot rolling mill is used to verify the effectiveness of the model. Results show that the prediction accuracies for tensile strength, compressive strength and elongation are 0.99971, 0.99835, and 0.99631, respectively. Especially, the prediction accuracy for elongation is higher than the existing methods.
    Keywords: intelligent prediction; LightGBM; hot-rolled strip; machine learning; big data analysis; steel mechanical properties.

  • Modelling and optimised gait planning of biped robots with different leg mechanisms   Order a copy of this article
    by Behnam Dadashzadeh, Akbar Allahverdizadeh, Mehdi Azhdarzadeh 
    Abstract: Biped robots with point feet demonstrate faster gaits and more natural dynamics, although planning optimal mechanisms and gaits and designing stable control strategies for them is difficult. This research focuses on modelling and gait generation optimisation of four different real biped models that include practical extended models of the theoretical SLIP and compass gait as a novelty of the work. All of the models have point feet, their torso angle is constrained, and they move in the sagittal plane. The first model is a kneed biped model without spring, which is a five-rigid-link robot with four actuators in its hips and knees. The second model, a kneed biped model with springs in shins, is very similar to the first model, but its shins have linear springs. These springs make the system underactuated and their passive vibration makes calculations and gait generation very difficult. The third model is a semi-telescopic springy biped model. For this robot in the single support phase of walking, the weight-bearing springy stance leg is straight and the other leg bends its knee, swings forward, and then becomes straight and hits the ground. In the double support phase, both legs are aligned with springy telescopic joints. In this model, the existence of leg springs increases the cost of transport and gait error. For the fourth model, the semi-compass gait with kneed swing leg, in the single support phase, the knee of the stance leg stays straight, and the swing leg bends its knee to clear the ground, then it becomes straight and hits the ground. Dynamic equations of the different phases are combined to create a dynamic model of a full walking gait. In the following step, optimisation parameters, objective functions and constraints are presented, and successive stages of optimisation are completed to find the optimal gaits. The efficiency of the gaits and required motor torques for the optimal gait of each model are illustrated.
    Keywords: biped robot; walking; modelling; gait optimisation.

  • A new super-twisting sliding mode control based direct instantaneous power control of PWM-rectifier connected to grid   Order a copy of this article
    by Arezki Fekik, Hakim Denoun, Mohamed Lamine Hamida, Sundarapandian Vaidyanathan, Nacera Yassa 
    Abstract: A new super-twisting sliding mode control (STSMC) based direct instantaneous power control (DPC) of PWM-recti er connected to grid is studied in this paper. The new STSMC-DPC scheme uses instantaneous power controllers. The new scheme does not use current controllers as with traditional vector control. The two regulators possess a design parameter that facilitates to adjust their behaviour between a linear operation of the PI type and a behaviour in conventional sliding mode with constant gain. Our tests demonstrate that STSMC-DPC shows a robust behaviour and that it functions without chattering in steady state as in a classical regulator. The results obtained are very relevant and satisfactory for the control of the PWM converter without sensor, in terms of the reduction of the chattering as well as an operation under a power factor and having a low THD of the short absorbed and uncoupled control of the instantaneous powers.
    Keywords: direct power control; sliding mode control; super-twisting SMC; rectifier; converter; regulator; robust control; power factor; PWM; SVM; THD; UPF.

  • Combined control strategies for performance enhancement of a wind energy conversion system based PMSG   Order a copy of this article
    by Youssef Errami, Abdellatif Obbadi, Smail Sahnoun 
    Abstract: In this paper, an analysis of a Wind Farm System (WFS) based on Permanent Magnet Synchronous Generator (PMSG) under Sliding Mode Control (SMC) and Vector Control Approach (VCA) gives the basis for combining outstanding features of the two control strategies into a reliable and efficient control system to provide enhanced WFS performance. The system contains two PMSGs connected to a common dc-bus system through rectifiers. The dc-bus is connected to the power grid through Five-Level Neutral Point Clamped Voltage Source Converter (5L-NPC-VSC). The main objective of this control is to regulate the velocities of the PMSGs with Maximum Power Point Tracking (MPPT) for the complete functioning regions of a wind turbine system. The proposed control techniques are based on Sliding Mode Approach (SMA) and Vector Control (VC) to achieve the MPPT, to keep the dc link voltage constant and to the control active and reactive powers. The stability of the regulators is obtained using Lyapunov analysis. The simulation results in Matlab/Simulink are presented to verify the effectiveness of the proposed combined control strategies.
    Keywords: wind power generation system; PMSG; NPC; sliding mode control; VCA; nonlinear control; electric network connection.

  • An offline predictive control with ellipsoid invariant set for time-variant system   Order a copy of this article
    by Rosana Rego, Marcus Costa 
    Abstract: A new formulation is proposed to address constrained robust model predictive control (MPC) offline scheme for uncertain systems based on a quasi-min-max algorithm with Linear Matrix Inequalities (LMI). The approach involves an offline design of a robust state observer. The proposed offline MPC is applied to a three-state switching cell (3SSC) boost converter. The converter analyses the ellipsoids of stability and its projections in the closed loop impulse response. A comparison is made between the online and offline performances of the algorithm to show the efficiency of the offline proposed approach. The results show that the offline formulation obtained good results and ensures the stability of the converter.
    Keywords: model predictive control; boost converter; linear matrix inequalities; uncertainties.

  • Adaptive cubature quadrature Kalman filter for nonlinear state estimation with one-step randomly delayed measurements   Order a copy of this article
    by Sri Mannarayana Poluri, Aritro Dey 
    Abstract: This paper proposes an adaptive Cubature Quadrature Kalman filter for one-step randomly delayed measurements (ACQKFRD), which is capable of handling critical situations when the system suffers from unknown parameter variations. The proposed filter automatically tunes the unknown element of process noise covariance related to the uncertain parameter by using an adaptation algorithm based on maximum-likelihood estimation. The proposed filter has been validated in simulation with the help of two significant nonlinear case studies. Monte Carlo (MC) simulation demonstrates the efficacy and consistency of the proposed filter.
    Keywords: adaptation; aircraft tracking; Bayesian filter; cubature quadrature Kalman filter; one-step randomly delayed measurement; nonlinear filtering; process noise covariance.

  • Brain MRI monitoring approach of lesion progress in multiple sclerosis using active contours   Order a copy of this article
    by Chaima Dachraoui, Aymen Mouelhi, Salam Labidi 
    Abstract: Magnetic Resonance Imaging (MRI) is considered as powerful tool for no-invasive diagnosis and description of brain pathologies. This is particularly the case of multiple sclerosis, for monitoring this disease and its treatment. Multiple sclerosis is an autoimmune inflammatory disease of the central nervous system in which clinical markers are used today for diagnosis and for therapeutic evaluation. In order to automate a long and hard process for the clinician, we propose a semi-automatic segmentation approach of multi sclerosis lesions in longitudinal MRI sequences. We use firstly a robust algorithm that allows spatiotemporal extraction of these lesions by the geodesic active contour model. Then, we recommend an original scheme based on an automated image registration technique for evaluating the evolution of the detected lesions. A quantitative study is presented in this paper to validate our results using the BrainWeb simulator, MICCAI2008 and MICCAI2016. Very promising results are obtained in the case of clinical data. Our research was tested on 10 typical and atypical synthetic motifs and 1000 MRI images from different centres which contain healthy brain and brain suffering from multiple sclerosis disease.
    Keywords: multiple sclerosis lesions; brain MRI; segmentation approach; evolution.

  • Life prediction method of automobile electromagnetic relay based on dual self-attention one-dimensional convolutional neural network   Order a copy of this article
    by Zhan Li Zhan, Lan Chao Lan, Zhao Hao Zhao, Guo Jifeng Guo 
    Abstract: The automobile electromagnetic relay is an electrical component that is widely used in the field of automobile manufacturing. It plays an important role in the control, regulation, and protection of low-voltage electrical systems of vehicles. Because of their frequent use and relatively poor working environment, they may fail and be damaged after being corroded by sand, dust, oil, and other pollutants over a long duration. Consequently, a method for predicting the life of automotive relays based on dual self-attention convolutional neural networks is proposed. The network is composed of global and local time series convolution. Two convolutional neural networks are connected in parallel to extract the global and local features of the automotive electromagnetic relay life data, and the extracted data features are determined to predict the life of the relay. Finally, the root mean square error was used to evaluate the prediction results.
    Keywords: convolutional neural network; automotive electromagnetic relay; life prediction; root mean square error; self-attention.

  • Modelling and intelligent optimisation for hot-rolling roll changes of high-speed tool steel rod and wire   Order a copy of this article
    by Jinxiang Chen, Yilan Yin, Can Li, Yanjin Chen, Jinjun Song, Yimin Shi 
    Abstract: The production of high-speed tool steel rod and wire is characterised by multiple varieties and small batches. The optimisation of roll change times of hot continuous rolling mill must be considered during production scheduling. Otherwise, frequent roll changes will lead to the decrease of work efficiency and the huge increase of production cost. In order to solve the above problems, an optimal model with multi-constraints is presented, and a HACA to optimise the number of roll changes is presented by considering various factors of the hot continuous rolling and roll changing process in this paper. By comparing and analysing the single model GA, ACA, and empirical method in a special steel factory, it can be found that the optimised roll change times are better than the original empirical roll change times, which improves the production efficiency and reduces the labour cost.
    Keywords: high-speed tool steel rod and wire; hot continuous rolling; roll changes optimisation; ant colony algorithm; genetic algorithm.

  • Modelling and analysis of aesthetic characteristics using digital technology in the artworks   Order a copy of this article
    by Yi Song 
    Abstract: In the 1980s, digital technology began to be fully integrated into the field of art and combined with art in various forms. As an active, cross-innovative and extensively permeable technical factor, it supports the diversity of artistic creation and gradually derives new artistic languages, forms as well as aesthetic styles. Meanwhile, as a key tool to be intervened in the artistic process, digital technology not only expands peoples customary aesthetic experience, but also provides rich media for the manifestation forms and means of art, and then the artistic aesthetics is updated accordingly. From the perspective of aesthetics, the digital-technology-based aesthetic process no longer focuses only on the aesthetic object of material attributes. The creative process reflects a relatively strong integrity and systematisation, and pays attention to peoples new relationships in artistic activities, such as process participation, behavior triggering, interactive experience and so on. The identity of aesthetic subject tends to be fuzzy. It focuses on the empathy and interaction in aesthetic process. The aesthetic distance becomes closer and produces a strong sense of immersion, triggering a practical reflection. Aesthetic behavior becomes more convenient. The aesthetic trend constantly evolves, and the continuity and uncertainty coexist. In this paper, through analysing the influence of digital technology on art and based on the typical art form of digital technology, case analysis and comparative study are used to understand the aesthetic process of digital aesthetics, summarise the basic characteristics of digital aesthetics while providing a cognitive framework, reflect on the future possibility of digital aesthetics, and provide the reference for people to treat the aesthetic value of digital aesthetics.
    Keywords: digital technology; digital aesthetics; characteristic analysis.

  • Sparse Gaussian process regression in real-time myoelectric control   Order a copy of this article
    by Myong Chol Jung, Rifai Chai, Jinchuan Zheng, Hung Nguyen 
    Abstract: In myoelectric control, nonlinear regression models, Gaussian process in specific, have shown promising accuracy in estimation, but no study has been conducted to evaluate the real-time performance of Gaussian process regression. In this work, the real-time performance of sparse Gaussian process regression is evaluated with 17 able-bodied subjects. Unlike the existing training methods, in which training protocols are strictly pre-determined, a novel training method is proposed. The subjects' real-time performance adjusts training time and the number of training samples. While the majority of subjects showed similar learning rates, there was a significant difference between a few subjects (p<0.05). As a result of real-time performance, the subjects completed 97% of the average tasks and achieved 80% path efficiency comparable to existing methods.
    Keywords: sparse Gaussian process regression; GP; regression; real time myoelectric control; real time control; myoelectric control; nonlinear regression; electromyography; EMG; human-computer interface; rehabilitation engineering.

  • Analysis of topological features in spatial distribution pattern using GIS and remote sensing   Order a copy of this article
    by Bingxin Li, Anbang Yu, Tong Liu, Sulin Guo 
    Abstract: This research offers a creative perspective from topological features for coordinated development between rural residents and the environment, as well as the intensive use of rural lands in Jilin province, China, so as to relief Jilin province from the bottleneck stage of the urbanisation process. Using the data of Landsat8 remote sensing image and four-dimensional graph in 2018, the study analysed the distribution pattern of rural settlements of Jilin province according to its topographical features from the characteristics of density and scale perspective, by taking Antu from the east, Dehui from the middle and the Tongyu from the west as the typical county unit as research objects, with the help of human-computer interaction interpretation and GIS-based spatial analysis method from the influencing factors perspective for the first time. It is found that the rural settlement was in a radiating spatial distribution pattern along Ha-Chang urbanisation development belt on the whole, and its density shows central > East > West. Though the West had poor density, it had a large scale of rural settlement, which showed great potential in improving the urbanisation development of Jilin province. The spatial distribution pattern of rural settlements is obviously affected by a radiation effect from the road system and an aggregation effect of orchard fields. The eastern and western rural settlement distribution patterns were mainly affected by the road and water system from the environmental aspect, as well as the intensive orchard fields land use from the production aspect. The central region was mainly affected by the slope grade and road system from the environmental aspect, as well as the upland use from the production aspect.
    Keywords: rural settlement; GIS/RS analysis method; spatial distribution; topological features.

  • The factor impact to the development of prefabricated construction industry chain based on structural equation models   Order a copy of this article
    by Yun Cui Fan, Dan Dan Chen 
    Abstract: The development of prefabricated construction is an important direction for the transformation and upgrading of the construction industry, and the prefabricated construction industry chain is the foundation for the stable development of the prefabricated construction industry. Based on the theory of industry chain, this paper studies the factor impact to the prefabricated construction industry chain, and analyses the factors influencing the development of the prefabricated construction industry chain through the structural equation models. The results show the five influencing factors of enterprise participation willingness, enterprise capability, inter-enterprise coupling degree, structure level and external environment, which have significant positive impacts on the development of the prefabricated construction industry chain. Therefore, it is necessary to promote the healthy and stable development of the prefabricated construction industry chain from the aspects of enhancing enterprises' participation willingness, improving enterprises' capabilities, strengthening the coupling ability between enterprises, optimising the structure of the industry chain, and creating a good external environment.
    Keywords: prefabricated construction; industry chain; influencing factor; structural equation models.

  • Tutorial on robotic olfaction via gas-sensitive UAV: technologies and methodologies   Order a copy of this article
    by Lei Cheng, Zhihao Cai, Qingxue Zeng, Jialiang Zhong, Yuao Li 
    Abstract: When used in a wide range of environments for sniffing tasks, gas-sensitive drones will perform better than common ground-based mobile robots owing to their flexibility. In recent years, with the improvement of various related theories, gas-sensitive unmanned aerial vehicles (UAV) have become a hot research direction. Generally speaking, gas-sensitive UAV platform consists of electronic nose and commercial UAV, which can be divided into different application directions according to the configuration of electronic nose module. The research on its application mainly focuses on gas localisation and gas distribution mapping. In this paper, the design and application of gas-sensitive UAV are reviewed, the electronic nose applicable to gas-sensitive UAV is summarised, and the design methods, existing problems and its solutions of gas-sensitive UAV are introduced. After that, the research on gas localisation and gas distribution mapping is summarised and analysed, and the possible research directions in the future are proposed.
    Keywords: robotic olfaction; gas-sensitive UAV; gas-localisation methods; gas mapping methods.

  • Adaptive deep neural network tracking controller augmented using only one-neuron hidden layers for nonlinear systems subject to high constraints and unknown uncertainties   Order a copy of this article
    by Hamou Ait Abbas 
    Abstract: A new control strategy, adaptive tracking controller augmented using deep learning hybrid method, is proposed in the current paper to achieve excellent tracking performances of nonlinear systems in the presence of structured and unstructured uncertainties. First, the designed controller employs feedback linearisation control (FLC) to linearise the partially known NLS, then the dynamic compensator (Dcom) is involved to stabilise the linearised system. However, conventional controllers suffer from limitations due to the presence of these high uncertainties. The main objective of the current paper is to demonstrate the feasibility of applying deep learning (DL) algorithm as an approximator for neglected dynamics and uncertain parameters that often exist in NLSs. Notice that the training process of the designed deep neural network (DNN) algorithm consists of an unsupervised pretraining (UPT) process and a supervised fine-tuning (SFT) process. Furthermore, the performed input information of the Dcom is used as input data and the term of uncertainties (inversion error) as a target of the learning algorithm (LA). Thus, the key idea of the developed adaptive FLC augmented using DNN is to both replace the conventional controller Dcom and compensate adaptively the effect of modelling errors and parametric uncertainties for highly uncertain NLSs. Notice that the weight adaptation rule of the DNN is derived from the Lyapunov stability analysis, which ensures the boundedness of the error signals of the studied system. Simulations of the proposed adaptive controller based DNN are conducted then compared with the Dcom without DNN, PI controller, and adaptive controller based SHLNN to demonstrate its practical potential. Furthermore, both the feasibility and efficiency of involving DL in the control area have been successfully confirmed through a robustness test.
    Keywords: nonlinear systems; unmodelled dynamics; parametric uncertainty; feedback linearisation control; dynamic compensator; neural network; deep learning; deep neural network; auto-encoder; stacked auto-encoder; unsupervised pretraining process; supervised fine-tuning process; hidden layer; layer-wise non-supervised learning algorithm; backpropagation; adaptive control; stability analysis; Lyapunov direct method; robustness test.

  • Model-based fault tolerant control and fault isolation through a bipartite graph approach   Order a copy of this article
    by Seyed Mohamad Kargar 
    Abstract: This paper presents an integrated fault detection and diagnosis (FDD) and fault-tolerant control (FTC) approach. The fault is detected and transmitted to the fault-tolerant control block, where a predefined bank of controllers is designed based on the quasi model predictive control approach. A bipartite graph of system equations, variables, and faults is introduced to detect the faults. Upon a fault detection and based on the fault detection and diagnosis block's information, the controller is switched to the appropriate mode. This proposed fault detection approach is subject to structural analysis. No hardware is necessary for this proposed approach, which is considered as its advantage. This approach is applicable without having a deep knowledge of the physical system. The contribution of this paper is twofold. First, the proposed integrated FDD/FTC can simultaneously detect and diagnose the faults. Moreover, the proposed FTC block guarantees the stability of the closed-loop system. The simulation results indicate that this approach accurately detects and isolates the faults. The fault-tolerant controller preserves the stability and performance of the system upon fault occurrence.
    Keywords: fault-tolerant control; fault detection; model predictive control; bipartite graph; Dulmage-Mendelsohn decomposition; residual.

  • Modelling influence of big data analytic capability on enterprise performance: mediating effect of strategic flexibility   Order a copy of this article
    by Hua Zhang, Ligang Liu, Hongji Yang 
    Abstract: It is undeniable that the technological storm brought by big data is transforming our life, work and thinking with the rapid development of information technology and internet applications. Nowadays, the development of big data has gradually moved from the initial stage of hot speculation to the practical stage. Many researchers and entrepreneurs have begun to pay attention to the important impact of big data analytic capability on enterprise performance. In this paper, a large-scale questionnaire survey was conducted among over 420 employees of 370 enterprises in the major economic circles of China, and 223 valid samples were obtained. Through the mediating regression analysis and moderating regression analysis using SPSS 23.0 and AMOS 24.0, the paper found that big data analytic capability has a significant positive influence on corporate performance, and that strategic flexibility plays a mediating effect on it. Environmental uncertainty has a positive moderating effect on the path joining big data analytic capability and strategic flexibility and the path joining strategic flexibility and enterprise performance.
    Keywords: big data analytic capability; strategic flexibility; corporate performance; environmental uncertainty.

  • Disturbance rejection decoupling control for a wastewater treatment process   Order a copy of this article
    by Wei Wei, Xinyu Cai, Min Zuo 
    Abstract: Concentrations of the dissolved oxygen (DO) and nitrate nitrogen (NN) directly affect the effluent quality. Keeping those concentrations at a desired level is critical in a wastewater treatment process. However, numerous undesirable factors as well as couplings between DO and NN pose a challenge in keeping concentrations of both DO and NN at a desired level. To achieve satisfactory performance, rather than model those issues accurately, they can be viewed as disturbances, and a disturbance rejection control (DRC) is designed. By estimating those disturbances in real time, couplings are addressed effectively, and robustness of the closed-loop system is also promoted. Numerical results based on the Benchmark Simulation Model No.1 (BSM1) shows that the DRC is able to reject the disturbances satisfactorily and meet the control requirements. It may be a practical and feasible wastewater treatment control approach.
    Keywords: disturbance rejection control; dissolved oxygen; nitrate nitrogen; BSM1; wastewater treatment processes.

  • Regular form based sliding mode control design on a two-wheeled inverted pendulum   Order a copy of this article
    by Yankun Yang, Xinggang Yan, Konstantinos Sirlantzis, Gareth Howells 
    Abstract: In this paper, a Lagrangian-based dynamics is employed for a two-wheeled inverted pendulum with the consideration of unknown matched and unmatched uncertainties that are bounded by known nonlinear functions. The model is linearised and further transformed into a regular form to facilitate the analysis and design. A sliding surface is designed, and a set of conditions is developed such that the resulting sliding motion is uniformly ultimately bounded from a practical perspective. Further, a sliding mode control scheme is proposed such that the system is driven to the sliding surface in f inite time and maintained on it thereafter. Finally, both simulation and experiment are presented using practical model parameters data to demonstrate the effectiveness and robustness of the regulation control.
    Keywords: sliding mode control; regular form; wheeled inverted pendulum; nonlinear system; mobile robots.

  • Revisited Adaptive Sliding Mode Control of Underactuated Mechanical Systems with Real-Time Experiments   Order a copy of this article
    by Afef Hfaiedh, Afef Abdelkrim 
    Abstract: In this paper, the stabilisation problem of a second-order underactuated mechanical system, namely the inertia wheel inverted pendulum, is addressed. Stabilising control design for this system is quite challenging since the control approaches designed for fully actuated systems could not be straightforwardly applied to such a system. We propose a new approach based on a first robust sliding mode controller (SMC) and an adaptation law that does not require the knowledge of uncertainties and perturbations bound and guarantees sufficiently small gain. To resolve the problem of underactuation, the proposed approach is based on a global change of coordinates, transforming the system into a cascade interconnection of linear and nonlinear subsystems. Through this transformation, two new desired trajectories are proposed for the design of the controller. Extensive real-time experiments are conducted to validate the effectiveness of this approach, where the first-order sliding mode controller and the adaptive sliding mode scheme are compared. The experimental results show that the proposed adaptive approach outperforms the original sliding mode controller in terms of external disturbance rejection.
    Keywords: adaptive sliding mode control; underactuated mechanical systems; strict-feedback form; inertia wheel inverted pendulum.

  • Ice clamping system in manufacturing systems as a cyber-physical system towards Industry 4.0   Order a copy of this article
    by Alexandra Mironova, Paolo Mercorelli, Andreas Zedler 
    Abstract: Clamping devices, as a part of a manufacturing system, have the role to fix workpieces during machining operations in order to hold the piece in place to maintain the position locally as well as to ensure a secure grip to withstand the process forces applied to the piece. The quality of the parts produced depends not only on the forces applied by the tool action, but results at the same time from the fixture action itself. In order to prevent the machined part from slipping or being totally released in the presence of the dynamic forces which are acting on the tool-workpiece-clamping system, the clamping forces must be high enough. This paper gives an overview on the most important literature on the topic ice clamping systems for new generation manufacturing in the context of Industry 4.0. Beside this overview prospective and future challenges are given.
    Keywords: ice clamping; manufacturing; data integration; computer supported collaborative work.

  • Mixed H2/H control for two quadrotors transporting a cable-suspended payload   Order a copy of this article
    by Minhuan Guo, Dongbing Gu 
    Abstract: In this paper, we design a mixed $H_2/H_{infty}$ controller for two quadrotors transporting a cable-suspended payload. The task for this system is to control the payload to track a pre-defined trajectory in a confined space. There are two challenging issues to be addressed when performing such a task: one is the robustness of the controlled system as the system is under-actuated and very easy to be affected by exogenous disturbances, and the other is the constrained operating environment in which the system is asked to pass a door without causing any collision. Additionally a pre-defined trajectory should be generated to allow the system to pass the door with a smoothed transient behavior. We develop a mixed $H_2/H_{infty}$ controller for this task to strike a balance between smoothing the tracking trajectory and resist the disturbance. The developed controller also possesses the ability to cope with the constraints imposed on inputs and outputs, and it is implemented by using a Linear Matrix Inequality (LMI) approach. The trajectory generation is based on an off-line optimization approach by considering the output flatness of the system. Finally, numerical simulation results are provided to evaluate the system performance and the comparison results against individual $H_2$ or $H_{infty}$ controller are also presented.
    Keywords: quadrotor transportation; mixed $H_2$/$H_{infty}$ control; constrained optimal control; robust control; linear matrix inequality.

  • Feedforward gain tuning with genetic algorithm   Order a copy of this article
    by Maher Ben Hariz, Faouzi Bouani 
    Abstract: This paper deals with the design of a fixed low order controller that ensures some closed loop temporal specifications, such as settling time and overshoot. Determining the controller parameters involves the resolution of a non-convex optimisation problem. Since the problem is not convex and in order to reach the global optimum, a global optimisation method has been applied. In order to annul the steady state error, especially in presence of system parameters variation, two methods are proposed to compute, online, the feedforward gain. The first method is based on the genetic algorithm and the second one is founded on the classic gradient method. By adopting the proposed procedure, users have only to specify the desired closed loop performances. Simulation results are proposed to shed light on the efficiency of the controller.
    Keywords: low order controller; temporal specifications; genetic algorithm; online gain tuning.

  • MPPT Based Backstepping Control For a grid connected-Fuel Cell System Using High Step-up Converter   Order a copy of this article
    by Fadwa El Otmani, Abdelmajid Abouloifa, Meriem Aourir, Karim Noussi, Hanane Katir, Fatimazahraa Assad, Ibtissam Lachkar 
    Abstract: This paper aims at studying fuel cell energy production as a promising solution to replace fossil fuels. The management of the fuel cell stack (FCS) while connected to the grid is a real challenge due to FCSs high nonlinearity and the complexity of the power conditioning stage (PCS). Therefore, a nonlinear controller is designed to handle the energy between the FCS and the electrical grid. In fact, the FCS energy is supplied to the electrical grid through two stages (a high-gain DC-DC and DC-AC converters) and an LCL filter. The proposed controller is elaborated based on the backstepping approach, then a maximum power point tracking (MPPT) block and a fuel flow rate regulator are implemented to extract and exploit the maximum power. The suggested controller has been tested in Matlab software, where the MPPT and the power factor correction are ensured.
    Keywords: proton exchange membrane fuel cell; three-level boost; interleaved boost; electrical grid; backstepping approach; maximum power point tracker; power factor correction; power conditioning stage.

  • Emery particles identification under contour extraction with maximum entropy approaches   Order a copy of this article
    by Peiyi Zhu, Ying Ding, Ya Gu 
    Abstract: The emery line is kind of a cutting line electroplated with emery particles, which is used to cut hard materials such as silicon wafer, etc. The emery line cutting capacity depends on the number, size and density of emery particles on the line. Traditional detecting by workers is replace by the technology of machine vision, which can improve efficiency of detecting and avoid the waste of resources. However, the emery particle identification is difficult for the adhesive particles on the cutting line using existing methods. The method of contour extraction with maximum entropy approaches is proposed. Image binarisation is adopted with maximum entropy threshold method. The initial evolution of the level set curve is present to process the outline of image, which can simplify the process of evolution from initial curve to emery particles contour. Adaptive weight coefficient was used for adaptively evolving according to the picture information, which can improve the adaptability of model identification. Experimental results show that the proposed algorithm has higher accuracy in terms of emery particle identification compared with the threshold segmentation algorithm and the traditional level set algorithm.
    Keywords: emery particles; contour extraction; maximum entropy threshold method; edge detection; level set.

  • Identification of external force for slender flexible structures via sparse modelling with an under-relaxation scheme   Order a copy of this article
    by Weijie Song, Qian Qiao, Yoshiki Nishi 
    Abstract: This study proposes a method for identifying unknown external forces that act on slender flexible bodies. In this method, the external force is represented as an expansion in the Chebyshev polynomials. The expansion coefficients are determined to minimise the number of nonzero coefficients and the difference between the measured and estimated quantities. An under-relaxation scheme is applied to enhance the robustness of computations for the identification. Static and dynamic beam problems are solved using the method to examine how well it can reconstruct the external forces. Numerical computations and analyses of the results demonstrate that the method generally produces a satisfactory external force reconstruction, and its accuracy is sensitive to the choice of measuring device.
    Keywords: external force identification; Chebyshev polynomials; sparse modelling; under-relaxation; beam.

  • Data-driven bearing fault detection using hybrid autoencoder-LSTM deep learning approach   Order a copy of this article
    by Pooja Kamat, Rekha Sugandhi, Satish Kumar 
    Abstract: Artificial Intelligence (AI) and its sub-domains of machine learning and deep learning have kindled the interests of both industry practitioners and academicians. Its contribution to the manufacturing industry in making intelligent predictions about a machines health and its working has seen a huge surge in research in recent years. Nowadays, AI in manufacturing is popularly applied for the efficient fault detection of machinery using data analytics. Traditional fault predictive classification and further diagnosis have pitfalls such as low prediction accuracy, poor feature extraction and susceptibility to noise. To overcome these disadvantages, this paper proposes the deep-learning based hybrid Autoencoders (AE) Long-Short Term Memory (LSTM) framework for fault detection. The dimensionality reduction with automatic latent feature extraction by autoencoders and temporal feature consideration by LSTM help to achieve high fault diagnosis accuracy. The empirical results show that fault detection of roll bearings based on the proposed hybrid AE-LSTM deep learning technique achieved superior results in comparison to the traditional K-means clustering technique.
    Keywords: bearings; fault detection; deep learning; autoencoder; LSTM; K-means.

  • Double integral sliding mode control of a boost PEMFC system   Order a copy of this article
    by Ouarda Heddad, Lahcene Ziet, Tewfik Bekkouche 
    Abstract: In this article, we will discuss and treat the control strategy which is based on Double Integral Sliding Mode based on the Pulse Width Modulator (PWM-DISM). In this study, the simple integral Sliding Mode (PWM-SISM) control/command was firstly used to show the Steady State error (SSE) of the boost. Secondly, the PWM-DISM control was used for the first time on a proton exchange membrane fuel cell (PEMFC) system to compensate for SSE. The PWM-DISM control was used to track the reference output of the fuel cell system composed of a 7 kW fuel cell supplying a 24 to 96 ? resistive variable load and a 48 V DC bus voltage via a boost converter. The control signal is performed to effectively control the cyclic ratio adjustment of the pulse width modulator (PWM) applied to the boost converter switch formed by an IGBT or MOS transistor. The PWM-DISM control is compared with the conventional PWM-SISM using the SimPower Systems models implemented in Matlab/Simulink. The results of the compared simulation between the considered controls confirm a better efficiency and performance for the PWM-DISM controls with regard to: steady state error for a maximum load between 550 mV and 5 mV, steady state error for a minimum load between 170 mV and 2 mV, response time for a maximum load 16.8 ms and response time for a minimum load of 20.6 ms; therefore the improvement and stabilization of the PEMFC system affected by a large variation of these variables: hydrogen supply pressure and load variation in free or connected mode.
    Keywords: pulse width modulation; polymer exchange membrane fuel cell; power converter; double integral sliding mode; steady state error; pressure hydrogen supply.

  • Predictive functional control based on particle swarm optimisation algorithm for MIMO process with time delay   Order a copy of this article
    by Hamid Ghadiri, Hamed Khodadadi, Ehsan Razavi 
    Abstract: The present study aims to propose a method for designing a predictive functional controller (PFC) based on the particle swarm optimisation (PSO) algorithm for the multi-input multi-output (MIMO) process with time delay. Owing to the cross-coupling, time delay, and other MIMO industrial process challenges, the classical controller could not overcome the system challenges. PFC is a model-based predictive controller that uses the state-space equations of the system. According to the results, the optimisation problem is resolved using the PSO for determining the optimal controller parameters. The incorporated PFC and PSO's performance as the proposed controller is evaluated using simulations on the chamber pressure coke furnace as a MIMO process with time-delay. The simulation results demonstrate that the PFC variables tuned by the PSO approach have better performance than the conventional PFC, especially in the presence of disturbances and uncertainty.
    Keywords: MIMO; time delay; predictive functional control; particle swarm optimisation; chamber pressure.

  • Design of fault-tolerant observer for brushless DC motor under rotor eccentricity fault   Order a copy of this article
    by Shihao Liu, Kexin Xing, Wei Dong, Yegui Lin 
    Abstract: The traditional sliding-mode observer (SMO) cannot accurately observe the back-electromotive force (back-EMF) values when the brushless DC motor (BLDCM) has a rotor eccentricity fault. In order to solve the problem, the influence of a static eccentricity fault on inductance and back-EMF is analysed, and a fault-tolerant observer is designed in this paper. The fault-tolerant observer uses the Lyapunov stability theory to identify the phase-to-phase inductance parameters online, and is based on the traditional SMO. According to the phase-to-phase inductance identification values, the SMO coefficient matrix is adjusted in real time. Simulation results show that the designed fault-tolerant observer can accurately observe the phase-to-phase back-EMF values of BLDCM under the rotor eccentricity fault conditions.
    Keywords: brushless DC motor; rotor eccentricity fault; fault-tolerant observer; sliding-mode observer; parameter identification.

  • Accurate positioning for the centre of a circular array target based on vanishing points   Order a copy of this article
    by Lijun Sun, Shaokui Ma, Tianfei Chen, Ming Yan, Xiaodong Liu 
    Abstract: Circular array targets are widely used in vision measurement systems. The detection of circular targets is usually regarded as ellipse detection after projective transformation. However, the centre of the ellipse is not the actual projection of the centre of the target circle. For this reason, this paper proposes an algorithm for locating the imaging centre of the target circle based on the vanishing points. After completing the extraction and filtering of the ellipse edge, the projective transformation matrix is used to obtain the two vanishing points referring to the coordinate axis direction of the target plane. Then, using the principle of perspective invariance and the two vanishing points, two pairs of tangent points on the ellipse edge are determined. Finally, the cross point between the lines connecting the two pairs of tangent points is the actual projection point of the target circle. Both simulation experiments and actual experiments demonstrate that the proposed algorithm has the characteristics of simple operation, faster running speed, and higher precision.
    Keywords: circular array target; projective transformation; vanishing points; homography matrix.

  • Machine learning approaches for modelling a single shaft gas turbine   Order a copy of this article
    by Hamid Asgari, Emmanuel Ory 
    Abstract: In this study, machine learning-based models of a single shaft gas turbine (GT) are developed. For this purpose, recurrent neural networks (RNN) are employed to train the datasets of the GT variables in both Python programming environment (by using Pyrenn Toolbox) and MATLAB software. Thirteen significant variables of the gas turbine are considered for the modelling processes. The resulting models are validated against the test datasets. The results demonstrate that the RNN models are capable of performance prediction of the system with a high reliability and accuracy. However, in this study, the overall results demonstrate that the RNN model set up in MATLAB has a better performance with a higher accuracy compared with the model developed in Python.
    Keywords: gas turbine; machine learning; modelling; simulation; artificial intelligence; recurrent neural network; black-box model.

  • Research on fault isolation method for composite fault in power module of modular multilevel converter   Order a copy of this article
    by Yue Mo, Ming Liu, Feng Zheng, Zetao Li 
    Abstract: The fault diagnosis technology of the modular multilevel converter has important engineering significance. Isolation of composite faults is a difficult problem in fault diagnosis. In this paper, the isolation method of composite faults in the modular multilevel converter power module is studied. The structural fault is modelled. The isolation problem of parametric fault superimposed with structural faults is discussed. A composite fault isolation method for the power module of the modular multilevel converter is established. Aiming at the problem of gradually closing of the fault isolation window caused by structural fault, an off-line fault isolation method for power module composite faults based on dynamic trajectory is proposed. Simulation experiments show the effectiveness of the proposed method.
    Keywords: modular multilevel converter; power module; fault isolation; multiple faults; structural faults; observer.

  • Study on low flocculation and sterilisation technology of oilfield produced water based on GA-RBF optimisation algorithm   Order a copy of this article
    by Fengbin Yang, Yujiang Wang 
    Abstract: There are a lot of sulphate-reducing bacteria in produced water. A variety of cationic bactericides can be added to solve the problems of water corrosion and blocking. The anionic polyacrylamide in the produced water of chemical flooding reacts with bactericide to produce flocculation sedimentation, which leads to the increase of sediment. Considering the bactericidal performance and compatibility, a kind of non-ionic low flocculation bactericide was developed, and its process parameters were optimised, which was proved by experiments. A genetic algorithm radial basis function (GA-RBF) neural network optimisation processor is proposed. The linear and nonlinear analyses of the orthogonal test parameters are carried out. Based on the genetic algorithm, the weights and thresholds of the neural network are optimised to complete the prediction of test samples and training samples.
    Keywords: produced water; bactericide; experimental evaluation; RBF neural network; genetic algorithm.

  • Genetic algorithm enhanced cloaking design of silicon multi-layers in dielectric background   Order a copy of this article
    by Zhenzhong Yu, Yizhi Wang, Xingliu Hu, Zhong Yang, Xiaomin Tian, Yan Zhang, Shanshan Gu 
    Abstract: The optimal cloaking design for a cylindrical structure in a dielectric medium is proposed. For an electrically small conducting cylinder, a free-space shell can be used to suppress the main scattering. With the introduction of magnetic material in the shell, one can further greatly reduce the scattering. For a larger object in the dielectric background, we design a multi-layered cloak made up of alternating layers of free space and silicon. Genetic algorithm helps to find the optimal thickness of the covering layers to minimise the total scattering. For a conducting cylinder with its radius compared to wavelength, we find that six optimised layers can eliminate most of the scattering. The change in the thickness of the free-space layer does not obviously affect the cloaking efficiency. The non-resonance characteristics make the multi-layered cloak possible to achieve a moderate broadband cloaking.
    Keywords: genetic algorithm; Mie theory; invisibility cloak; optimisation method.

  • A novel friction model for predicting nonlinear friction dynamics   Order a copy of this article
    by Anthony Nnaji, William Holderbaum, Victor Becerra 
    Abstract: In this paper a new dynamic friction model capable of predicting observed friction dynamics is presented. The model incorporates a pre-sliding friction function with non-local memory hysteresis features. Simulations showed the models ability to predict known friction features. An experimental test-bed was developed and used for friction characterisation through a carefully designed set of experiments. The new model has one parameter more than the LuGre model, which is easy to estimate through system identification. System identification was performed to determine the parameters of the proposed friction model as well as the LuGre and GMS models for the purposes of model performance comparison. Experimental and simulation results of the new dynamic friction model with the identified model parameters exhibited a strong correspondence with the results of the characterisation experiments, capturing known friction features. The new friction model, being simple in both structure and implementation, demonstrated better capability for modelling friction phenomena than the LuGre and GMS friction models.
    Keywords: friction regimes; friction dynamics; friction models; characterization; Stribeck effect; system identification.

  • Optimal synthesis of planar eight-link walking leg mechanism using genetic algorithm   Order a copy of this article
    by B.V. Raghavendra, Anandkumar R. Annigeri 
    Abstract: A walking leg mechanism robot is a type of locomotion that operates by means of legs and/or wheels on rough terrain or flat surface to a motor attached to the mechanism. These types of robot are effective in rough terrains such as large agricultural sector for farming, mobility in open fields, and dangerous and rescue areas for humans. The dimensions of the links, number of links, pivot point and rpm of the motor play critical roles in deciding the foot profile. The performance of the walking leg mechanism is evaluated based on stride height, stride length, velocity, acceleration and stability of the robot. In this paper, the optimisation of link lengths for the foot profile of a planar single degree of freedom eight link walking leg mechanism using an evolutionary approach (genetic algorithm) is attempted. The lengths of the eight links are optimised for stride length and stride height of the foot trajectory for nine different configurations of the walking leg mechanism.
    Keywords: eight-link coplanar mechanism; genetic algorithm; optimisation; synthesis of mechanism.

  • Dynamic modelling of a hovering type autonomous underwater vehicle with ballast tank   Order a copy of this article
    by Abdollah Karimi, Reza Hasanzadeh Ghasemi 
    Abstract: In this article, dynamic modelling of a hovering type autonomous underwater vehicle (HAUV), equipped with thruster and ballast tanks, is presented. Ballast tanks are among the equipment available for controlling AUVs. They provide high manoeuvring and stability capabilities for the robot. In the past, the use of ballast tanks in HAUV was not prevalent, and it was merely used in AUVs of torpedo shape type. The number and arrangement of the tanks for the robot are selected by considering that all the feasible degrees of freedom should also be achievable here by changing the buoyancy roll, pitch, and heave. Having the equations of motion rewritten based on the variable of the mass of the tanks, a nonlinear model of the desired robot was obtained. Eventually, several motion-time graphs were obtained by applying various kinds of input that indicate the significant capability of the tanks in exerting a force on the robot.
    Keywords: hovering type autonomous underwater vehicle; ballast tank; dynamic modelling.

  • Model-free fractional order sliding mode control of permanent magnet synchronous motor   Order a copy of this article
    by Ashraf Hagras, Abdelnasser Nafeh 
    Abstract: This paper proposes Model-Free Fractional Order Sliding Mode Control (MFFOSMC) for Permanent Magnet Synchronous Motor (PMSM). This approach proposes a new simple switching control law characterised by simplicity of design and flexibility of control to add more flexibility and performance in the controller design. The equivalent control of SMC is based on Model-Free Control (MFC), which doesnt need process model and is free of heavy parameter tuning. The model-free control is based on Radial Basis Function (RBF) Neural Network (NN), which doesnt require known bounds of uncertainties. The stability of the proposed approach was analysed and guaranteed using Lyapunov stability theory. This new method achieved the required goals of adaptive control not only against large parameter variations, external disturbances and measurement noise but also against higher saliency of the rotor. Simulation results using MATLAB/SIMULINK validated the improved performance of the proposed approach by comparing it with the intelligent PI controller and conventional PI-based vector control.
    Keywords: fractional order SMC; intelligent PI controller; model-free control; RBF neural network; reaching law.

  • Soft sensor model for monitoring and online control based on a dynamic model and local instrumental variable technique   Order a copy of this article
    by Roja Parvizi Moghadam, Jafar Sadeghi, Farhad Shahraki 
    Abstract: The aim of this paper is two data-based soft sensors design for accurate prediction of isopropyl benzene concentration in an industrial unit. One of the important difficulties in this unit is the quality measurement by lab analyzer which is time-consuming that can improve by design of a soft sensor. The first proposed sensor is based on the state-dependent-parameter model and a local instrumental variable (LIV) method relying on the static data. The main novelty of this work is focused on the second soft sensor that is introduced to compensate the time lag ignorance in the first proposed soft sensor. A dynamic model is considered between predicted values of LIV-based soft sensor and simulated concentration. Their performances are evaluated by industrial and simulated data. The results of these non-parametric models show a very low error percentage and good agreement with prediction quality from the rigorous model and compare to other models.
    Keywords: online monitoring; quality control; data-based soft sensor; local instrumental variable; dynamic model.

  • Nonlinear optimal stabilising control of a two-wheel robot   Order a copy of this article
    by Surapong Kokkrathoke, Andy Rawsthorne, Hongwei Zhang, Xu Xu 
    Abstract: The stabilisation of a two-wheel robot is a classical benchmarking problem for determining the effectiveness of a control technique. In this paper, a nonlinear optimal control technique is applied to a two-wheel robot which demonstrates excellent control performance comparing against the linear quadratic regulator technique. Simulation results demonstrated that this nonlinear optimal controller can achieve accurate tracking of wheel angular displacement and effective stabilisation of the robot from a very large range of initial pitch angles. Practical factors such as maximum motor voltages are considered and analysed using an extended state-space model to embed such control saturations. Significantly, the two-wheel robot can be balanced from a body pitch angle of up to 88.0
    Keywords: nonlinear system; optimal stabilising control; controllability; two-wheel robot.

  • Control-oriented observer for cylinder pressure estimation of SI engine using frequency response function   Order a copy of this article
    by Syed Abbas Ali, Samir Saraswati 
    Abstract: This paper presents a closed-loop control-oriented observer (COB) for in-cylinder pressure estimation, which uses a one-zone combustion model in the feed-forward loop and a frequency response function (FRF) model in the feedback loop. The model uses the phase difference in firing frequency of crankshaft speed as a measure to update COB parameters. The update rate of control parameters in the COB is one engine cycle. The COB is found observable at the top dead centre (TDC). The applicability of the proposed model is illustrated by simulation and experimental results. Further, a comparative study using closed-loop COB, FRF, and Sliding Mode Observer (SMO) methods is also presented.
    Keywords: frequency response functions; control-oriented observer; cylinder pressure.

  • A neural network enhanced detection method for GPS with slowly growing error   Order a copy of this article
    by Weina Chen, Zhong Yang, Yizhi Wang, Shanshan Gu, Yujuan Tang 
    Abstract: Aimed at the problem of early detection of the slowly growing error in GPS positioning, an improved detection method based on neural network has been proposed which is aided by the inertial navigation system. The state and measurement model of the integrated navigation system has been established. The back propagation neural network model is constructed for fault detection according to the mapping relationship between the residual test statistic and the fault. And the robust estimation theory is adopted to decrease and eliminate the impact of abnormal data on the fusion results. The simulation results and the field test show that the method proposed by the paper has a better detection performance compared with the autonomous integrity monitoring by extrapolation method and the rate detector algorithm. The new method can ensure the accuracy and integrity of the INS/GPS integrated navigation system in required navigation performance.
    Keywords: global positioning system; slowly growing error; integrity monitoring; neural network; integrated navigation.

  • Towards a global nonlinear control strategy for a DFIG-based wind turbine in a high wind energy penetrated system   Order a copy of this article
    by Karim Noussi, Abdelmajid Abouloifa, Hanane Katir, Ibtissam Lachkar, Fadwa El Otmani 
    Abstract: Variable speed wind turbines offer several advantages compared with those with fixed speed, especially as far as it concerns their ability to capture energy efficiently. To benefit from these features, the power transmitted to the electrical grid must reach its optimum. In this context, this paper proposes a nonlinear controller, so as to control the overall Variable Speed Wind Energy Conversion System (VS-WECS). This latter consists of a wind turbine rotor, a Doubly Fed Induction Generator (DFIG), an AC/DC/AC converter, and the electrical grid. The proposed controller is based on the sliding mode technique aiming at controlling the stator powers, the grid side reactive power and regulating the DC-link voltage, in order to manage the flow of energy between the wind turbine system and the power grid as well as to improve the dynamic system and transient stability. The performance of the 2 MW grid-connected DFIG wind turbine system is checked with Matlab/SimPowerSystems environment and examined under fixed and variable wind speed profiles to confirm the reliability and validity of the control strategy.
    Keywords: power control; DFIG; wind energy conversion system; sliding mode control; Lyapunov stability.

  • Control engineering solutions during epidemics: a review   Order a copy of this article
    by Jafar Tavoosi, Mohammad Amin Shirkhani, Amirreza Azizi 
    Abstract: The widespread and sudden outbreak of the Coronavirus in 2019, known as covid-19, caused almost all countries in the world to face many problems. Over time, with the rapid spread of the virus, COVID-19 became a serious concern to people all over the world. The outbreak of the virus prompted many cities and countries to enact lockdown laws to try to control the epidemic. These global lockdowns have done a lot of damage to the world economy. For this reason, different countries of the world have tried to use various technologies, such as the Internet of Things (IoT), blockchain, artificial intelligence, drones, etc., to reduce the prevalence of covid-19 and minimize the damage to the economy by telecommuting methods. In this paper, we briefly review the impact of using these technologies during the outbreak of epidemics such as covid-19.
    Keywords: internet of things; artificial intelligence; blockchain; tele-medicine; drones; control engineering; covid-19.

  • Novel adaptive balanced control of humanoid robot type NAO robot   Order a copy of this article
    by Mahmoud Farhat, Brahim Brahmi, Maarouf Saad, Mohammad Rahman 
    Abstract: This paper presents a novel adaptive tracking controller based on the ModifiedFunction Approximation Technique (MFAT). The proposed approach includes a sliding mode with MFAT to approximate the unknown dynamic model of the NAO robot on which it mainly relies. The impact policy and unknown dynamics of the system are considered when implementing the control approach to improve the NAOs robot locomotion. Unlike conventional FAT, the basis functions in the dynamic parameters approximation updated law are avoided in the proposed MFAT. The Lyapunov function is presented a stability analysis according to which all error signals are uniformly ultimately bounded in a closed loop. Simulations were performed to validate the design MFAT approach for the NAO robot by carrying out a comparative study with a conventional model-based controller.
    Keywords: adaptive control; humanoid robot; function approximation technique; NAO robot; biped locomotion.

  • Convergence analysis of cuckoo search algorithm based on Martingale theory   Order a copy of this article
    by Lijun Sun, Xiaodong Liu, Tianfei Chen, Ming Yan, Shaokui Ma 
    Abstract: Cuckoo Search (CS) algorithm is an emerging kind of biologically inspired algorithm that has been successfully applied in several areas. However, its mathematical theory has not yet been fully established, and the theoretical analysis of convergence is relatively inadequate. In this article, the martingale theory is introduced for the first time to prove the convergence of CS, which replaces the ergodic analysis of the Markov chain. First of all, according to the basic principles of CS, we establish the mathematical model of the Markov chain and group state sequence, and its Markov properties are analysed, and then the optimal group state set is obtained. After that, the supermartingale of the cuckoo evolution sequence with an optimal fitness value is derived. In the end, based on the convergence theorem of supermartingale, it is demonstrated that the CS algorithm ensures global convergence.
    Keywords: cuckoo search; Markov chain; convergence; martingale theory.

  • Design and implementation of deviation rectification control model for heavy-duty automated guided vehicle   Order a copy of this article
    by Shanshan Gu, Yekun Bi, Zhong Yang, Yizhi Wang, Weina Chen 
    Abstract: With the development of Automated Guided Vehicle (AGV) technology, aiming at the characteristics of heavy load AGV with large size and high load capacity, this paper studies the improved PID controller based on speed control to rectify the vehicle deviation. Firstly, the position and attitude of AGV are measured. Then the PID controller based on vehicle motion model is designed, and the incremental PID control based on integral separation is used to realize the vehicle deviation rectification. The best PID parameters are obtained by simulation experiment. And the real vehicle experiment demonstrates that the deviation rectification control model designed in this paper can meet the actual requirements.
    Keywords: heavy-duty AGV; improved PID; rectification control; magnetic sensor.

  • Modelling of a GSA-based framework of an agent-based system within parameter boundaries: applications to robotics and automatic voltage regulator systems   Order a copy of this article
    by Arabinda Ghosh, Anjan Kumar Ray, Omkar Singh 
    Abstract: The enormous potentials of an agent-based nature-inspired algorithm named the gravitational search algorithm (GSA) remains unfulfilled, which is the aim of this work. Existing GSA has provided ways to select the super-agents or prune the poor performing agents. Here, a holistic framework is proposed by introducing a survival boost which re-establishes self-significance of a poor-performing agent and liberates all agents from an infinity trap for acceleration. It restricts the agents within the search space whose violation produces infeasible solutions. The analysis is extended to solve a number of benchmark test functions, an inverse kinematic problem of a 7 DOF robot manipulator, and a voltage regulation problem of an automatic voltage regulator system. The results and comparative studies show better performances in compare with some of the popular methods from the literature (e.g. existing GSA, chaotic GSA, different variants of particle swarm optimisation, artificial bee colony, differential evolution, Jaya algorithm).
    Keywords: gravitational search algorithm; chaotic GSA; multi-agent system; robot manipulator; inverse kinematics; automatic voltage regulator.

  • Observer design for linear systems with external disturbance and intermittent measurements   Order a copy of this article
    by Xu Cai, Xuyang Lou 
    Abstract: This paper deals with the state estimation of linear systems with external disturbance for which measurements of the output are available intermittently. An observer with triggered jumps is designed within a hybrid systems framework. By employing the stability results in hybrid systems, the global uniform asymptotic stability of a closed set related to the estimation errors, which contains all coincident points of the system state and its estimated value, is achieved. Based on the Finsler lemma, conditions for the global uniform asymptotic stability of the estimation errors are established in the form of linear matrix inequalities. Finally, numerical simulations including a damped mass-spring system and a DC motor system are provided to illustrate the results.
    Keywords: observer; hybrid systems; external disturbance; intermittent measurement.

  • Observer/Kalman filter identification with support vector machines   Order a copy of this article
    by Sinchai Chinvorarat, Chien-Hsun Kuo 
    Abstract: This paper presents the novel identification filter by using the SVM technique with the OKID to identify nonlinear dynamic systems. The proposed filter increases the identifiability and accuracy of the state space realisation model. By integrating the SVM with the OKID in a single identification block, the rich training input and output data from the dynamic system are fed into the identification block and start computing the hyperplane classifier with a radial basis kernel function as a nonlinear mapping function. The algorithm can determine nonlinear signals (noise) from the hyperplane parameters. The modified input-output data determined by filter out nonlinear signals from the dynamic system output are used for the OKID filter. Since the proposed SVM/OKID filter identifies the discrete model from the nearly noiseless signals, it demonstrates high accuracy in identifying the nonlinear dynamic system over the regular OKID.
    Keywords: SVM/OKID filter; kernel function; Markov parameters.

  • Decentralised nonlinear predictive control for twin rotor multi-input multi-output system   Order a copy of this article
    by Achour Benchabane, Noureddine Bali 
    Abstract: Abstract: The paper proposes a manner for building a decentralized nonlinear predictive control of the MIMO system (TRMS). The control is based on minimizing the quadratic objective functions consisting of an error between the desired reference and the predicted output. This control is purpose to make the TRMS pursue a specific trajectory or stabilize its angles (pitch and yaw ) in definite position. The decentralized control structure has been proposed in this work. This involves breaking down the complex structure of the TRMS into interconnected subsystems vertical and horizontal, each of which is a menu of sensors and actuators that are specific to it, and controlled by a local control unit, the coupling effect are considered as perturbation to each other. In order to validate the control structure developed for the TRMS, performance and robustness tests were performed in simulation. The results confirmed the efficacy of the control method.
    Keywords: decentralized control; nonlinear predictive control; MIMO system (TRMS); predicted output.

  • Adaptive variable structure controller design of uncertain nonlinear systems based on state estimation: a new scheme   Order a copy of this article
    by Ruiping Xu, Zhen Liu 
    Abstract: In this work, state-estimation problem of uncertain nonlinear systems is studied within a new variable structure control framework. An observer is first presented to estimate the unavailable state variables, via which a novel adaptive controller is developed for ensuring the effective estimation of the internal state. Then, based upon the designed observer, a novel linear sliding surface and a new adaptive variable structure controller are shown for the underlying systems. Further, exponential stability analysis of the system during the sliding mode is carried out by virtue of LMI technique. At last, two demonstrated examples are shown to validate the efficiency and advantage of the given design.
    Keywords: state estimation; exponential stability; adaptive variable structure control.

  • Stock closing price prediction based on combined model of PCA-IMKNN   Order a copy of this article
    by Xin Yue, Yaojian Zhou, Chenxun Yuan 
    Abstract: This paper proposes an Improved Multidimensional K Nearest Neighbors (IMKNN) method based on Principal Component Analysis (PCA), which uses multi-dimensional time series to predict the closing price of stocks. PCA converts multiple indicators related to the closing price into several principal components, whereas little informationrnlost, thereby reducing the dimensionality of the input multi-dimensional series and improving the efficiency of analysis. The IMKNN method is proposed based on MKNN. Compared with MKNN, its distance value has more possibilities, which can overcome rnthe difficulty of sorting caused by too limited possibilities of distance values in the existing MKNN method. Four representative stocks and four general evaluation criteria were used to test the newly proposed PCA-IMKNN method. The results show that, compared with the MKNN method, IMKNN method, and PCA-MKNN method, the proposed PCA-IMKNN model has higher prediction accuracy.rn
    Keywords: principal component analysis; K-nearest neighbors; PCA–IMKNN; forecast; closing price.

  • Data-driven optimal control for stochastic systems with non-Gaussian disturbances   Order a copy of this article
    by Lanlan Lai, Liping Yin, Yue Hong, Tao Li 
    Abstract: In this paper, a data based algorithm is applied to optimize the performance index and search for global solution for non-Gaussian systems. The control objective is to track a desired probability density function (PDF). The control law is obtained through optimizing performance index. The well-known kernel density estimation (KDE) technique is employed to estimate the output PDFs because the output PDFs are immeasurable for many industrial processes, and the performance index function is established based on the stochastic distribution control theory. The established performance index function is optimized by using an intelligent optimization algorithm with more simple formulation and less computation load than existed results. Furthermore a new global optimal control strategy can be obtained through a data based control algorithm. Two numerical examples are given to demonstrate the effectiveness of the control algorithm.
    Keywords: kernel density estimation; performance index function; stochastic distribution control; probability density functions.

  • On parameter separation of the Wiener system identification   Order a copy of this article
    by Shaoxue Jing 
    Abstract: The parameter separation in the Wiener system parameter identification is considered in this paper. Firstly, the Wiener system to be identified is parameterized into a pseudo linear model based on the key term separation technique; secondly, the parameter estimation algorithm is used to get the parameter estimation value containing the product term; To get the unique model, the parameter separation is needed. Four parameter separation methods, i.e., fixing single nonlinear parameter method, fixing single linear parameter method, fixing linear block energy method, and fixing nonlinear model energy method, are proposed. Several numerical simulation examples show that the performance of the fixing linear block energy (FLE) method is stable, and the prediction error of which is small. After a little modification, the FLE can be applied to the parameter separation of other nonlinear models.
    Keywords: Wiener system; parameter separation; key term separation; estimated model.

  • XGBoost-based prediction modelling and analysis for health literacy assessment   Order a copy of this article
    by Yan Hong, Xiaoda Zhang, Jinxiang Chen 
    Abstract: Big data analysis and XGBoost modelling for health literacy prediction for health literacy prediction are investigated in this paper, which gives a new idea for health literacy assessment. 750 residents in Tongliao, Inner Mongolia, China were tested to answer 68 questions in three questionnaires about health literacy. A big dataset with 742 samples was constructed first. Every sample had 68 characteristics. Based on the dataset, BPNN and XGBoost prediction are established, respectively. R2 score obtained by XGBoost model is 0.97553, which is higher than one solved by BPNN. The relative error rates and absolute errors obtained by using XGBoost are less than 5% and 10 points, respectively. Therefore, XGBoost model is more effective than BPNN for predicting peoples' health literacy. The influence of every feature on the residents' health literacy score is calculated by XGBoost, which can help to analyse the influence of every question on residents health literacy assessment.
    Keywords: health literacy; prediction model; XGBoost; characteristics analysis; big dataset.

  • Global stabilisation of control systems using a new sliding mode control method and its application to a general type of synchronisation for chaotic systems   Order a copy of this article
    by Sundarapandian Vaidyanathan 
    Abstract: This research work details the design of a new sliding mode control (SMC) technique for the global stabilisation of control systems. As a simple consequence, this work derives a new sliding mode control method for a general type of synchronisation called generalised projective synchronisation (GPS) between chaotic systems. Next, we present the model of a new multistable two-scroll chaotic system. The new two-scroll chaotic system has three unstable balance points. We detail how to stabilise the new chaotic system using the new sliding mode control. We also illustrate the GPS results for the new two-scroll chaotic system using the proposed SMC design.
    Keywords: stabilisation; sliding mode control; nonlinear control; synchronisation; chaotic systems; chaos; multistability; Lyapunov exponents.

  • Output feedback Takagi-Sugeno fuzzy model predictive control through linear matrix inequalities approaches   Order a copy of this article
    by Marcus Vinicius Silverio Costa, Thalita Brenna Da Silva Moreira, Fabricio Gonzalez Nogueira 
    Abstract: The present paper proposes an output feedback control scheme combining the Takagi-Sugeno (T-S) fuzzy method with a model predictive control (RMPC) technique, using parallel distributed compensation (PDC) and linear matrix inequalities (LMIs). The study presents an algorithm of relaxed RMPC, considering a nonlinear varying parameters rule-based T-S fuzzy model. Moreover, a new stability criterion is proposed considering an online observer-based output feedback T-S fuzzy model predictive control (FMPC). The aforementioned criterion is implemented through LMIs constraints ensuring system's robust stability. This procedure assembles the aforementioned techniques and apply them in a benchmark problem. The obtained results evidence the better performance of the proposed method in comparison with the benchmark controller, considering analysis of time responses, objective function and performances indices.
    Keywords: fuzzy control; model predictive control; FMPC; T-S fuzzy model; LMIs; output feedback control; robust stability criterion.

  • Minimise pruning cost of a node-weighted directed acyclic graph on applications of management   Order a copy of this article
    by Zhi-Ming Chen, Cheng-Hsiung Lee, Yu-Feng Lin 
    Abstract: A novel optimisation problem is proposed in this research. In the proposed model, a corporation is represented as a directed acyclic graph (DAG) with weight. A directed edge in the DAG represents the relationship between a division and its subdivision. The weight denotes the pruning cost of each node. The objective is to partition the graph into two parts so that one of the parts would be pruned to minimise total pruning cost. The proposed model can be formulated as an integer linear programming problem which is hard to find the optimal solution. In this research, we show that it can be solved in polynomial time by using a general linear programming solver. Furthermore, we propose an improvement method which the optimal solution can be solved much more quickly than only using a general LP solver.
    Keywords: optimisation; integer programming; pruning cost; node-weighted directed acyclic graph; 2-partition; partial order set.

  • Comparison between Real-Time Auto-Tuning PID and Conventional PID Controller for a Dairy Industrial Evaporation Process   Order a copy of this article
    by Qingbo Meng, Hongwei Zhang, Martin Howarth 
    Abstract: Evaporation processes are commonly applied in the food industries to concentrate the liquor for future treatments. It presents challenges to the modelling and control approaches due to the system complexity, especially for the multi-effect evaporators. In this study, an industrial milk evaporation process is introduced and a mathematical model of a three-effect falling film evaporator is developed using MATLAB/SIMULINK with added disturbances. A real-time closed-loop auto-tuning PID and a conventional PID are both presented and applied to the model as the control strategies for the evaporation process. The simulation results are compared to illustrate the improvements from the auto-tuning PID controller.
    Keywords: multi-effect falling-film evaporator; auto-tuning PID controller; industrial milk evaporation process.

  • Modified predictive control of continuum manipulators with learning-based model   Order a copy of this article
    by Aida Parvaresh, S. A. A. Moosavian 
    Abstract: Continuum manipulators are considered as systems with high intrinsic complexities, nonlinearities, and uncertainties, which encounter several problems in accurate modelling and control purposes. However, model predictive control (MPC) strategy, which is a popular control scheme in robotics, is highly dependent on the precision and accuracy of the associated model used by the controller. Accordingly, any mismatch in the parameters/structure of the model would result in the deteriorated performance of MPC. To alleviate the accuracy-related problems, data-driven adaptive MPC (DD-AMPC) is proposed in this paper. This approach is based on deriving an explicit linearized model of the system through a data-driven identification procedure, which is appropriate for model-based control purposes. In this approach, the continuum manipulator can be modelled and controlled without the requirement for comprehensive knowledge of the physical system. Moreover, to reduce the computational cost, the traditional receding horizon concept is modified by parametrizing the control trajectory. The proposed control scheme is implemented and its performance and functionality in point-to-point tracking, path following, and presence of disturbances are assessed. Additionally, its superiority over conventional control schemes is deduced by comparing the results of proposed approach with the results of conventional controllers. The obtained results confirm the effectiveness of proposed scheme in different scenarios.
    Keywords: adaptive model predictive control; learning-based modelling; continuum manipulators; system identification; data-driven identification.

  • Ridge regression and lasso regression based least squares algoritm for a time-delayed rational model via redundant rule   Order a copy of this article
    by Zili Zhang, Jing Chen, Yawen Mao 
    Abstract: This paper proposes a ridge regression based least squares algorithm (LS-RR) and a three stage lasso regression based least squares algorithm (TS-LS-LR) for a rational model with unknown time-delay. By using redundant rule method, the time-delayed rational model is turned into a new model. In order to identify the parameters of this new model, the LS-RR algorithm is proposed. The parameter vector of this model contains two parts, redundant parameters and true parameters. To pick out the redundant parameters, the lasso regression based least squares algorithm (LS-LR) is proposed. Furthermore, the TS-LS-LR is introduced to improve the estimation accuracy. The numerical simulation shows the effectiveness of the proposed algorithms.
    Keywords: polynomial non-linear system; time-delayed model; rational model; lasso regression; ridge regression; redundant rule.

  • Front-end Matching Optimized Algorithm of Cartographer with Multi-resolution Layered Search Strategy
    by Yiqun Di, XIANGHUA MA, DONG GUO 
    Abstract: Cartographer is used in LiDAR SLAM (Simultaneous localization and mapping), and it has a stable mapping results and utilizes loop closure detection to eliminate accumulated errors. However, considerable calculation is needed to guarantee the accuracy of mapping results and the decrease of accumulated errors when CSM (Correlative Scan Matcher) is used to carry out the global search of the front-end matching. In order to resolve the above problem, a structure, multi-resolution layered search strategy (MLSS) is proposed in the frond-end of cartographer. MLSS structure makes use of layered strategy and the full-number locally optimal principle to decrease accumulated errors from sources and time costs of front-end matching. Verification and comparison studies demonstrate the effectiveness of the proposed structure.
    Keywords: SLAM, multi-resolution layered search strategy, front-end matching, full-number locally optimal principle

  • MRFCNN: the optimisation method of convolutional neural network for underwater target recognition   Order a copy of this article
    by Hongbin Wang, Pengming Wang, Shengchun Deng, Zhenghao Gu 
    Abstract: In the field of underwater target recognition, with the increase of various sensors, information and underwater noise, underwater target recognition is becoming more and more complicated. Therefore, traditional methods can no longer meet the current needs, and neural network has obvious advantages in dealing with the classification problems with complicated environmental information and vague background knowledge. In this paper, we discuss the optimization problem from two aspects: feature extraction and target classification. Then we propose a correlation optimization method based on convolutional neural network, and carry out related underwater simulation experiments. The experimental results show that the optimization improvement has a certain improvement compared with the previous accuracy, which fully proves the effectiveness of the proposed optimization method.
    Keywords: underwater target recognition; convolutional neural network; feature extraction; target classification.

  • System identification of rover dynamics: a comparison of three model structures   Order a copy of this article
    by Christina Ivler, Norma Gowans, Cole Marfise 
    Abstract: Dynamic testing of a small instrumented ground vehicle was conducted in order to identify an accurate state-space model for simulation of autonomous vehicles. This paper describes the application of frequency domain system identification to model the yaw/steering response of a small-scale rover. Several model structures of varying levels of complexity were adapted and applied to a small-scale rover: Dynamic Bicycle model, Roll-Yaw model, and a Lumped (based on Taylor series expansion) model. In comparing these three model structures, it was found that the Dynamic Bicycle model provided a simple model structure with good performance but cannot model roll dynamics. The Roll-Yaw model gave the most accurate model and better prediction for a range of vehicle speeds but is significantly more complex. Finally, the Lumped model gave a highly accurate model at the identified speed condition; however, it cannot be accurately extrapolated to other speeds.
    Keywords: system identification; rover; steering dynamics; bicycle model; frequency domain methods; frequency sweeps; roll-yaw model; ground vehicle; coherence weighted; composite windowing; identification of inertia.

  • Control of non-minimum phase system using inverse response compensator with different approximations   Order a copy of this article
    by Manish Yadav, Hirenkumar. G. Patel 
    Abstract: The objective of this paper is to control non-minimum phase systems with dead time in the presence of uncertainty and disturbances. For the motivation of such problems, the series cascade control scheme is used, specifically for slow process dynamics and actuator nonlinearities. The Internal Model Control (IMC) is used for the inner loop controller while the outer loop controller is designed via fractional-filter-based IMC strategy with an inverse response and dead time compensator. The uniqueness of this work lies in that the Taylor series, Pade and All pole approximation are used for outer loop controller design for approximation of time delay. The robustness is carried out via sensitivity analysis. Two non-minimum phase dead time systems are used to demonstrates the usefulness of the proposed control strategy.
    Keywords: series cascade control; dead time compensator; inverse response compensator; fractional-filter.

  • Optimisation of group batch scheduling in flexible flow shop based on multi-player cooperative Ggame   Order a copy of this article
    by Zhonghua Han, Xusheng Bian, Ziyao Ding, Sun Dechang 
    Abstract: In a flexible flow shop, if there are group batch processes, the processing task needs to form multiple batches for production in the batch operation. In order to make up batch processing, there will be job waiting time thereby extending the total working hours of the production process, and the conflict between various production indexes is further intensified. Therefore, this paper first establishes the Group Batch Scheduling of Flexible Flow Shop (GBSFFS) mathematical model, and proposes a local scheduling method based on multi-player cooperation and complete information static game, establishes a method of game elements such as game information, game player, game strategy, game payoff for group batch processing, and uses the method for behaviour prediction of various game players, that is, the valuation indexes, which reduces the evaluation indexes conflict in the GBSFFS problem. On the premise of reducing the waiting time, other evaluation indexes have been increased.
    Keywords: flexible flow shop; production jam; multi-player static cooperative game; batch scheduling.

  • Outlier detection algorithm based on deviation characteristic   Order a copy of this article
    by Yong Wang, Hongbin Wang, Pengcheng Sun, Xinliang Yin 
    Abstract: Outlier mining focuses on researching rare events through detection and analysis to dig out the valuable knowledge from them. In the static data set environment, the traditional LOF algorithm calculates the local outlier factor through the whole data set and requires a lot of computing time. To solve this problem, the algorithm divides the data space into grids, and calculates the local outlier factor based on the centroids of the grids. Since the grid number is less than data point number, the time complexity is obviously reduced under acceptable error. When the new data points are added, it can rapidly detect outliers. The contrast experiment results show that the new algorithm can reduce the computation time and improve the efficiency, while achieving comparable accuracy.
    Keywords: outlier detection; local outlier factor; deviation characteristic; fast LOF detection algorithm.

  • Distributed control of linear partial integro-differential equations based on the input-output linearization approach   Order a copy of this article
    by Ahmed Maidi, Jean-Pierre Corriou 
    Abstract: In this paper, the input-output linearisation control approach is extended to distributed parameter systems whose dynamical behaviour is described by a partial integro-differential equation. The design of the infinite dimensional state feedback controller is achieved using the late lumping approach, i.e. using the partial integro-differential equation model without any prior reduction or approximation. Thus, based on the notion of the characteristic index as a generalisation of the relative degree, a distributed state feedback controller is designed by evaluating the successive time derivatives of the controlled output. The designed controller yields in closed loop a first order lumped parameter system where the time constant is a design parameter that fixes the desired dynamic behaviour. The stability of the closed loop system is investigated, based on semi-group theory, by employing the perturbation theorem of the bounded linear operators, and the sufficient condition for exponential stability in $L^2$-norm is derived. This condition yields the upper bound for the design parameter, i.e. the time constant. Both output tracking and stabilisation capabilities of the developed state feedback are demonstrated through numerical simulation by considering three application examples: Volterra, Fredholm and Fredholm-Volterra PIDEs. The effectiveness of the developed controller is shown by simulation.
    Keywords: distributed parameter system; partial integro-differential equation; input-output linearisation; semi-group theory; perturbation theorem; exponential stability.

  • Maximum power harvesting from a PV system using an improved two-stage MPPT scheme based on incremental conductance algorithm and integral controller   Order a copy of this article
    by Mostufa Atia, Noureddine Bouarroudj, Aimad Ahriche, Abdelhamid Djari, Yehya Houam 
    Abstract: This article deals with a novel strategy called two-stage incremental conductance (INC) algorithm for maximum power point tracking (MPPT) for photovoltaic (PV) systems. This method is proposed to tackle the shortcomings of classical incremental conductance (INC) method such as oscillations, imperfect MPPT, slow tracking of MPP, etc. These demerits stem from the use of a one-stage INC algorithm with a fixed duty cycle step size, which cannot provide the optimal duty cycle especially under changing weather conditions (temperature and irradiance) and resistive load. The first stage of the proposed approach is used for providing the reference voltage using the INC algorithm; and the second one is an integrator controller tuned by Routh's criterion used to ensure the stability of the voltage closed loop control. Simulation and numerical results in different cases confirm the superiority of the proposed two-stage INC algorithm over the classical one based on one stage, with an efficiency of more than 93.75%.
    Keywords: PV module; boost converter; MPPT; INC algorithm; one stage; two stages; Routh's criterion.

  • Hyperparameter optimisation of ensemble classifiers and its application for landslide hazards classification   Order a copy of this article
    by Jiuyuan Huo, Hamzah Murad Mohammed Al-Neshmi 
    Abstract: Along with assessing the landslide hazards taking into consideration the faced difficulties and the consumed time when determining the algorithm configurations and parameters manually, the primary aspiration of this study is to optimise the parameters of two ensemble-based machine learning algorithms using Particle Swarm Optimisation (PSO), Genetic Algorithm (GA), and Bayesian optimisation so that the optimised algorithms can identify and classify landslides more efficiently and accurately. Random Forest Classifier and XGBoost models were used and the ADASYN was implemented to overcome the shortage of imbalanced data. In the experiments, it was clearly shown that the hypered ensemble-based models along with the PSO and GA successfully surpassed the single models on classifying the landslide triggers, sizes, and types. The experimental results demonstrated that the hyperparameters optimisation can greatly improve the accuracy of the ensemble classifiers, thus it can provide accurate classification results and decision support for the disaster prevention and mitigation management departments.
    Keywords: landslide; optimisation; random forest; XGBoost; ADASYN.

  • Robust adaptive SMC for uncertain singular delayed systems via observer   Order a copy of this article
    by Xiaoliang Tang, Zhen Liu 
    Abstract: This article is focused on state-estimation-based adaptive control design for uncertain singular systems subject to state delay and uncertain nonlinearity by employing sliding mode technique. Firstly, the unmeasured state variables are generated by a particular observer without any inputs, and a new switching surface function of linear type is presented. In view of linear matrix inequality technique, the motion of the closed-loop system on the sliding surface is analysed, and a new admissibility criteria is deduced. Then a switching controller with adaptive rules is synthesised to ensure the established sliding surface can be attained in finite moment. Finally, an illustrative example is proposed to demonstrate the feasibility of the theoretical method.
    Keywords: singular delayed systems; state-estimation; adaptive sliding mode control; admissibility.

  • Multi Stopping Criterion Multi Feature based Multiobjective Cohort Intelligence Algorithm for Thermoacoustic Engine Optimization
    by Mukundraj Patil, Satish Kumar 
    Abstract: Aim of this research is to investigate the performance characteristics of thermoacoustic engine using multi stopping criterion multi feature based Multiobjective Cohort Intelligence (MOCI) algorithm. MOCI and the state-of-the-art algorithms are applied to study performance characteristics of a thermoacoustic engine (TAE). Exploratory and statistical analyses revealed better performance of the MOCI algorithm on qualitative and quantitative performance metrics. Post optimality analysis showed a better region of interest for an analyst and the desirable working ranges for each variable of TAE design. Pressure-frequency relationship showed high correlation and it is useful for future study and detailed investigation of thermoacoustic phenomenon. MOCI established competitive results which are useful in benchmarking TAE performances in future researches. The effective design of TAE using MOCI algorithm aids in the sustainable development of society in terms of affordable and clean energy, clean climate and responsible consumption and production.
    Keywords: multi objective; cohort intelligence; thermal device optimization; stopping criterion; thermo acoustic; thermoacoustic prime mover; best algorithm; performance benchmarking;

  • Hover autopilot design for an uncrewed helicopter using static output feedback controller   Order a copy of this article
    by Femi Thomas, Mija S J 
    Abstract: the design of a linear matrix inequality based static output feedback controller for an uncrewed helicopter in hover flying mode is presented. The six degrees of freedom linear time-invariant state-space model of the vehicle is developed analytically from first principles considering the force and moments acting on it without the usual simplifying approximations. Since access to full state information is not a situation in practice, output feedback based controller is an appealing solution for the autopilot design. Here, two separate static output feedback controllers are developed for the fast inner-loop and the slow outer-loop dynamics of the vehicle. As the number of variables to be fed back is reduced, the proposed scheme is simpler compared with conventional state feedback controller. Numerical simulation studies validate that the proposed controller exhibits fast transient performance and robustness when subjected to wind disturbances acting in the three fuselage axes during the hover flight.
    Keywords: uncrewed helicopter; hover flight; static output feedback controller; static state feedback controller; linear matrix inequality; Lyapunov equation.

  • Hybrid fuzzy level set approach for multiple sclerosis lesions assessment in magnetic resonance brain images
    by Chaima DACHRAOUI, Aymen Mouelhi, Cyrine Drissi, Salam Labidi 
    Abstract: Multiple Sclerosis is a neurological autoimmune disease characterized by progressive degeneration due to the myelin attack on the central nervous system. The diagnosis is based essentially on clinical features and additional examinations mainly magnetic resonance imaging findings. The diagnosis of multiple sclerosis requires all defined criteria that aim to study spatial and temporal dissemination. Thus, in this work, the automatic segmentation of multiple sclerosis plaques is opted in order to computerize the process and the follow-up. This approach is a hybrid method allowing to combine Fuzzy C-Means method with geodesic models until obtaining an automatic task. This is a retrospective study in which data were collected from the National Institute of Neurology in Tunisia. The proposed method is for medical neuroradiology research. The eventual results are improved after some pre-treatments, therefore, the interest in pre-processing. High accuracy was achieved for the models discussed in this paper (93% - 84%). Accordingly, the suitability and practical usefulness of the "simple" pre-treatments to achieve multiple sclerosis classification are demonstrated.
    Keywords: Multiple sclerosis, Brain, MRI, Automatic segmentation, Hybrid approach, geodesic contour, fuzzy c-means, lesions segmentation, T2 FLAIR

  • An optimal control problem associated with Lorentz group SO(3; 1)
    by Archana Tiwari, Kishor Chandra Pati 
    Abstract: Lorentz group is the group of transformation of spatial and time coordinates associated with special theory of relativity. It is both a group and admits a topological description as a smooth manifold. Hence, Lorentz group can act as a configuration manifold of control systems. This opens up the scope to study the controllability and optimal control problems of control systems on Lorentz group. Here, a left invariant, driftless control system is de fined on the group. An optimal control problem is formulated with an objective to minimize the cost function and satisfy the given dynamical constraints. Stability of the system around equilibrium points are studied. Two unconventional numerical integrators, Kahan's and Lie-Trotter integrator and conventional Runge-Kutta integrator is implemented to study the system dynamics and their corresponding trajectories are shown.
    Keywords: Lorentz group; control system; optimal control; stability.

  • PFOID-SMC approach to Mitigate the effect of Disturbance and Parametric Uncertainty on the Quadcopter
    by Sanjay Kumar, Lillie Dewan 
    Abstract: This paper investigates the possible types of disturbance viz aerodynamic factors, sensor noise, random noise, wind effect, and parametric uncertainty acting on an Unmanned Aerial Vehicle quadcopter and their adverse effect on the performance. To ensure the desired performance and increase quadcopters' stability, accuracy, and task reliability, disturbances and uncertainties detection and diagnosis are very important. Proportional-Fractional order-Integral-Derivative controller surface-based Sliding Mode Controller is proposed to mitigate the effect of disturbances and uncertainties. The system's stability is proved using the Lyapunov criterion, and performance is validated by simulation. Results of PFOID surface-based SMC are compared with Proportional-Integral-Derivative surface-based Sliding Mode Controller
    Keywords: Non-linear systems, Quadcopter Dynamic, Aerodynamic effects, Parametric Uncertainty, Sensor Noise, Wind Noise, Proportional-Integral-Derivative (PID), Proportional Fractional-order integral derivative (PFOID), sliding surface, Sliding mode control (SMC).

  • Investigation and Realization of PID and LQR Control Methods in Parrot Mambo Minidrone
    by Mohamed Okasha, Jordan Kralev, Maidul Islam 
    Abstract: A quadcopter is multivariate and unstable, highly nonlinear dynamic system, which requires a proper controller to ensure the stability and performance of the system. This study aims to investigate different types of control methods for Parrot Mambo minidrone. In this study, different control methods used on quadcopters such as Proportional Integral Derivative (PID) and Linear Quadratic Regulator (LQR) are investigated and implemented. First, the Parrot built-in PID controller is tested in simulation and experimentally validated using MATLAB and Simulink, followed by the design of the LQR controller. For both controllers, the operating point is selected such that the minidrone can hover along the vertical dimension. The design and tuning of the LQR is carried out by giving weight on the inertial coordinates and on the motor signals, which determine the performance of the minidrone with minimization of quadratic cost function. The LQR controller shows that the system tends to have less overshoot in vertical trajectory. In many testing scenarios, the LQR controller shows better overall performance compared to PID controller in both simulation and experimental testing.
    Keywords: PID, LQR, UAV, Parrot Mambo Minidrone, Control Methods

Special Issue on: SCC-2019 Recent Advances in Systems and Electrical Engineering

  • Multi-weld defects detection based on Gabor filter and Hough transform   Order a copy of this article
    by Chiraz Ajmi 
    Abstract: Weld defect detection is an important application in the field of Non-Destructive Testing (NDT). These defects are mainly due to manufacturing errors or welding processes. In this context, image processing especially segmentation is proposed to detect and localize efficiently different types of defects. It is a challenging task since radiographic images have deficient contrast, poor quality and uneven illumination caused by the inspection techniques. In this article, we propose a robust and automatic method based on the combination of Gabor and Hough transform to detect two major defect types from mono- or multi-weld defect images. It consists of two main stages which are the preprocessing and the segmentation step. The first one is based on the Gaussian filter and contrast stretching. The segmentation step is performed using a sequence of tools starting with Gabor filter and Otsu binarisation method to isolate the weld region from the background and non-weld coming to the Canny detector to extract edges of the spherical defect shape, and finally a modified Hough transform technique for detection, location of linear defects and limiting the welding area. The experimental results show that our proposed method gives good performance for industrial radiographic images.
    Keywords: weld defect; radiography; NDT; Hough transform; Canny detector; Gabor filter.

  • Adaptation of deep learning auditory event recognition and detection in audio surveillance systems   Order a copy of this article
    by Sara Alsubhi, Safiah Endargiri, Ahad Alkabsani, Kaouther Laabidi 
    Abstract: The work of this paper focuses on the idea of adapting computerised machines with the sophisticated abilities to relatively comprehend and act upon an auditory input of natural linguistic nature. We emphasise the addition of acoustic-based audio inputs to the current CCTV systems for the goal of compensating any incomplete data and to reach the maximum usage of the current surveillance systems. In this model, we apply the isolated word technique on a dataset of 8000 audio inputs dedicated to different individuals through the application of two distinct neural networks. The algorithm provides event-based detection capabilities by allowing the detection of unauthorised access through the automatic recognition of each spoken input, together with the identity of the speaker. The proposed algorithm obtained accuracy rates of 84.1% and 80.1% for the recognition by the speakers identity and by the spoken input, respectively. In addition, it showed its superiority over the SVM-based model.
    Keywords: classification; automatic speech recognition; natural language processing; artificial neural network; deep learning; convolutional neural network; audio detection; speaker recognition; speech recognition; Mel spectrogram; CCTV.

  • Design of robust non-integer controllers for fractional order MIMO systems   Order a copy of this article
    by Emna Ouhibi, Maher Ben Hariz, Faouzi Bouani 
    Abstract: The main objective of this paper is to design a robust non-integer order controller for Multi Input Multi Output (MIMO) systems. A linear parametric uncertainties fractional order model is used to describe the system's dynamic behaviour. Assuming that the system can be decoupled and in order to avoid interactions between the systems inputs and outputs, the simple decoupling method has been adopted. In the presence of model parameter uncertainties, the controller parameters are obtained by minimising a min-max non-convex optimisation problem. The optimised cost function is expressed using the closed loop system and the desired characteristic polynomial coefficients. So, the proposed controller ensures some closed loop temporal specifications in spite of the presence of model parameter uncertainties. Simulation examples are carried out to show the performance and the efficiency of the proposed controller.
    Keywords: robust non-integer controller; fractional order MIMO system; simple decoupler; temporal specifications.

  • The harmful influence of employing a large number of pilot resources on the performance of massive MIMO systems   Order a copy of this article
    by Abdelfettah Belhabib, Mohamed Boulouird, Moha M’Rabet Hassani 
    Abstract: One of the severe constraints that limit the implementation of Massive Multi-Input Multi-Output (M-MIMO) systems is called Pilot Contamination (PC). To overcome this problem, a Soft Pilot Reuse (SPR) strategy is proposed, which aims to properly allocate the Pilot Sequences (PS) to the users based on their properties, e.g. Quality-of-Service (QoS). Despite its flaws, which can damage the spectral efficiency, the SPR strategy has been heavily employed as a decontaminating strategy. To reveal the weakness of this approach, this paper adopts a SPR based-strategy that separates the users within cells based on their large scale fading coefficients; therefore, because the edge-users (EUs) are mostly affected by the PC compared to the centre-users (CUs), a set of orthogonal pilots is allocated to the EUs of the overall cells, while the CUs of each cell are obliged to reuse the same set of orthogonal pilots as the CUs of adjacent cells.
    Keywords: massive MIMO; pilot contamination; soft pilot reuse; pilot overhead; conventional strategy; 5G wireless communications.

  • A sensorless mixed DFIM control strategy based on fuzzy-PI speed controller and current sliding mode controller for electric vehicles   Order a copy of this article
    by Mouna Zerzeri, Adel Khedher 
    Abstract: This paper discusses Doubly-Fed Induction Motors (DFIM) integration possibilities in Electric Vehicle (EV) propulsion systems. The motivation behind this work is to develop an appropriate control law for DFIM to be suitable for EV applications. From this point of view, the main contributions of this study are: (i) a mixed control approach of an electric vehicle propulsion system based on a DFIM which is composed of a Fuzzy-PI controller for the mechanical mode and a sliding mode controller for the electrical one; (ii) an active power distribution law of both stator and rotor circuits connected to space vector pulse width modulation voltage source inverter; (iii) a new method to estimate speed by associating an MRAS estimator with an extended Luenberger observer in order to ensure the sensorless command and to compensate the load torque impacts. The effectiveness of the developed sensorless mixed control algorithm is proved by the obtained simulation results for different DFIM operating ranges.
    Keywords: electric vehicle propulsion system; DFIM; fuzzy-PI controller; sliding mode controller; power distribution; MRAS; Luenberger observer.

  • Stability analysis and fault-tolerant control of neutral systems   Order a copy of this article
    by Rabeb Benjemaa, Aicha Hsoumi, Saloua Bel Hadj Ali Naoui 
    Abstract: This work studies the problem of stability analysis and fault-tolerant control of neutral systems. Improved stability criteria are obtained by applying the Lyapunov method and the Linear Matrix Inequality (LMI) technique. An adaptive-observer is designed to detect and estimate faults, and a new control law is proposed to achieve fault compensation. A numerical example is given to illustrate the theoretical results.
    Keywords: neutral time-delay system; stability; fault detection; fault estimation; fault-tolerant control; adaptive observer; linear matrix inequality.

  • Open-switch fault detection scheme in wind energy conversion system based on rotor currents analysis   Order a copy of this article
    by Bilel Touaiti, Hechmi Ben Azza, Mohamed Jemli 
    Abstract: To deal with the performance and service continuity of Doubly Fed Induction Generator (DFIG) based wind energy conversion systems during open-switch Insulated Gate Bipolar Transistor (IGBT) fault, a fault-tolerant Voltage Source Inverter (VSI) is proposed. The fault-tolerant VSI is designed with a redundancy leg to replace the faulty leg after open-switch fault occurrence. In this investigation, we propose a new topology of DC-bus tied DFIG based on an uncontrolled rectifier in the stator side. A Field Oriented Control (FOC) scheme is used to control the DFIG. The open-switch IGBT fault is studied, and a fault detection scheme based on rotor currents analysing within healthy and faulty conditions is proposed. Case study results show that the proposed fault-tolerant VSI can ensure the service continuity of DFIG-DC system. Experimental results are presented in this paper to validate the proposed fault detection scheme.
    Keywords: doubly fed induction generator; fault-tolerant voltage source inverter; redundancy leg; DC bus.

  • Legged robot design and Van der Pol oscillator-based control approach   Order a copy of this article
    by Riadh Zaier, Omer Eldirdiry 
    Abstract: It is widely accepted within the field of animal locomotion that the gait is generated by a central pattern generator (CPG). This type of locomotion pattern generator is usually modelled as a nonlinear oscillator. Many CPGs had been developed in the last few decades and applied as a gait generator of multi-legged robots. Yet, some issues related to the robustness of the locomotion controller still need to be addressed, such as the tuning of the oscillators parameters and its strategy of coupling with the sensory feedback. Therefore, this paper starts by proposing a mechanical design of a bio-inspired legged robot with a passive toe joint. The leg must have a similar gait to that of humans, and hence, the structure is made of light material with dimensions proportional to that of human leg. The paper then elaborates on how to control the leg upon the key characteristics of the CPG generating the rolling motion of the gait. In contrast with the literature, which set the locomotion controller as an oscillator, we consider here the overall control system as a Van der Pol oscillator, and the controller is then deduced accordingly. The legged robot is modelled as an inverted pendulum with a few control parameters that can be tuned to modulate the rolling motion and make it adaptive along with the stride. The locomotion controller is structured so that the overall closed loop system exhibits a stable limit cycle. Finally, the validation of the simulation platform and the implementation results of the designed robot are reported.
    Keywords: biomechanical legs; Van der Pol oscillator; central pattern generator; limit cycle; nonlinear controller.

  • Parameter identification for a photovoltaic module: comparison of PSO, GA and CS metaheuristic optimisation algorithms   Order a copy of this article
    by Ines Ben Ali, Mohamed Wissem Naouar, Eric Monmasson 
    Abstract: Modelling photovoltaic (PV) cells is important for understanding its operation and evaluating its performances under various conditions. For that, it is mandatory to estimate with good accuracy the parameters of a PV cell circuit. In the last decade, metaheuristic optimisation algorithms became widely used for parameter identification of PV modules. In this context, this paper presents a comparative study of three metaheuristic optimisation algorithms: the genetic algorithm GA, the particle swarm optimisation PSO and the cuckoo search CS. These optimisation algorithms were used for PV cell parameter estimation. The used estimation process is based on: (i) a simplified analytical model of a PV cell and (ii) only three I-V curve points that are always available from technical data. The comparative study showed that GA is not appropriate to be used in parameter extraction of the PV model. The CS algorithm exhibited its superior efficiency over other algorithms in terms of estimation accuracy and ease of implementation. However, PSO and particularly the hybrid PSO combined with pattern search algorithm (PSO-PS) appeared to be the most promising in terms of computational efficiency by offering faster speed convergence to the global optimum solution with few tuned parameters.
    Keywords: photovoltaic generator; parameters identification; cuckoo search; particle swarm optimisation; genetic algorithm.

  • Design and characterisation of a new GaN/AlGaN HEMT transistor   Order a copy of this article
    by Naji Guedri, Naoufel Ismail, Rached Gharbi 
    Abstract: AlGaN/GaN HEMT is considered as a backbone of both optical and microwave high power electronic applications. Gallium nitride (GaN) is a wide band-gap semiconductor material with excellent properties for high power, high frequency, and high temperature electronics. In this paper, we designed and discussed the structure of a high electron mobility transistor HEMT based on GaN. To this end, we designed the original two-dimensional architecture composed of COMSOL Multi-physics software. The model evolved in this work is characterised by its very small geometric dimensions 0.52
    Keywords: new structure AlGaN/GaN HEMT; temperature effects and simulation; COMSOL Multi-physics; I-V characteristics.

  • Polynomial chaos for the design of robust controllers with time specifications: a comparison of two approaches   Order a copy of this article
    by Amani Added, Maher Ben Hariz, Faouzi Bouani 
    Abstract: This paper deals with the design of a robust controller for stochastic dynamic systems. A new method is proposed to compute the controller parameters to achieve the desired time requirements. It is based on the polynomial chaos expansion, which is used to provide a high order deterministic state space model. Then, the controller parameters are obtained by minimising a cost function, which depends on the closed loop and the desired characteristic polynomials. The proposed method is compared with a classical method founded on an uncertainty parametric model. The control law problem is formulated as a min-max optimisation. The developed controllers ensure desired closed loop performances, such as settling time and overshoot. Simulation results are presented and compared to highlight the efficiency and the performances of the proposed controllers.
    Keywords: robust controller; low order controller; stochastic uncertainties; polynomial chaos; min-max optimisation; GGP; closed-loop performances.

Special Issue on: ICECOCS'20 Intelligent Control for Future and Complex Systems

  • Digital Implementation of Model Predictive Control of an Inverter for Electric Vehicles Applications
    by khawla Gaouzi, Hassan El Fadil, Zakariae El Idrissi, Abdellah Lassioui 
    Abstract: The control of DC-AC converters with output LC filter has a special importance in applications intended for electric vehicle. However, the controller design becomes more complicated. This paper proposes a Model Predictive Control (MPC) of dc-ac power converter for electric vehicles applications. The control objective is to regulate the output voltage of a three-phase inverter with output LC filter to its desired constant values. Once the control law is elaborated, several simulations are performed using Matlab/Simulink tools and the results show that the output voltages perfectly track its references. Experimental results are also given, which show the effectiveness of the model predictive controller.
    Keywords: Model predictive control; three phase voltage source inverter; electric vehicle; Matlab/Simulink

Special Issue on: Modelling and Simulation Techniques for Industry Applications

  • Hybrid projective synchronisation of fractional-order network system   Order a copy of this article
    by Lingzhong Zhang, Xiangli Li 
    Abstract: The hybrid projective synchronisation (HPS) of memristor-based fractional order delayed neural networks system (MFDNNs) is considered. By using the definitions and properties of differential inclusions and set-valued map, fractional order correlation inequalities are established. The hybrid projective synchronisation of MFDNNs is investigated by designing a feedback controller, and several criteria are derived under the frame of Razumikhin-type stability theorem and Lyapunov function method. A numerical example is given to verify the conclusion.
    Keywords: hybrid projective; networks; delay; synchronisation; memristor.

  • Synchronisation of inertial reaction-diffusion complex networks with mixed time delays via spatiotemporal sampling control   Order a copy of this article
    by TianE Chen, ZaiHe Cheng 
    Abstract: This paper addresses the global synchronisation problem of a class of inertial complex networks with reaction-diffusion terms and mixed time-varying delays. First, the spatiotemporal sampling control scheme using incomplete measurements is proposed to ensure the synchronisation of the delayed inertial reaction-diffusion complex networks, which deduce the update rate of the synchronisation controller. Secondly, by taking new appropriate LyapunovKrasovskii function, using improved Wirtinger Jensen inequalities, new synchronisation criteria are derived that depend on the relationship among the sampling time interval, the feedback control gain and the sampling space interval. Finally, two numerical simulation results substantiate the effectiveness of the theoretical results.
    Keywords: reaction-diffusion; inertial complex networks; incomplete spatial measurements; spatiotemporal sampling control.

  • Parameter and time delays estimation based on compressed sensing for MISO CARMA system modelling   Order a copy of this article
    by Taiyang Tao, Qingxi Xu 
    Abstract: This paper considers estimating parameters and time-delay for multiple input single output (MISO) controlled auto-regressive moving average (CARMA) models with unknown time delays. Inspired by compressed sensing method and hierarchical idea, an orthogonal matching pursuit algorithm based iteration is presented for the MISO CARMA system. The presented algorithm can estimate time-delay and parameters of the MISO CARMA system with limited sampled data. An example is given to show the effectiveness of the algorithm with a few sampled data.
    Keywords: parameter estimation; time-delay; MISO CARMA system; hierarchical idea; orthogonal matching pursuit.

  • Fixed-order H-inf control of an electric power steering system   Order a copy of this article
    by Ai Ping Pang, Hui Zhu, Hong Bo Zhou, Dao Yin Long, Tao Qin, Jing Yang 
    Abstract: The electric power steering (EPS) system has a very small phase margin and it is particularly easy to start oscillation. Therefore, a stable controller is needed to improve the stability margin. In addition, the gain of the torsional moment diagram of the EPS system in different situations causes parameter perturbation, which requires the control design to have a certain robustness of performance and consider all the stability margin, robust performance of the system and the bandwidth requirement of the system. In this paper, the fixed-order H-inf control design is adopted to form an H-inf optimisation matrix with multi-dimensional performance output and a third-order fixed-order H-inf controller is designed for. The controller has good robustness performance, high stability margin, and has a relatively low order compared with the conventional H-inf controller design, which is convenient for practical applications.
    Keywords: fixed-order controller; H-inf control; electric power steering system; robust stability.

  • The algebraic solution parameter estimation methods for control systems based on frequency responses   Order a copy of this article
    by Ling Xu 
    Abstract: System identification is a basic technique which can construct system models for designing system controller. In order to reduce the complexity of the iterative or recursive algorithms and develop easy ways to identify system models, this paper is concerned with the problem of parameter estimation for control systems from observed input and output data of system frequency characteristics. Because the system parameters are contained in the real frequency and imaginary frequency characteristics, we design several algebraic solution methods that use only seldom measurements to determine system parameters in terms of the first-order inertial system and second-order pole systems. In this study, the two-point method, three-point method and four-point method are derived by using the discrete observations of the real frequency characteristic or imaginary frequency characteristic or combining the real frequency and imaginary frequency characteristics. Finally, the effective of the proposed algebraic solution methods for estimating parameters of control systems is verified by a few numerical examples.
    Keywords: parameter estimation; system identification; algebraic solution; transfer function.

  • Design of FPC board assembly control system based on robot integration technology   Order a copy of this article
    by Yan Liu, Xianrong Jiang, Y.A. Di, Chao Liu 
    Abstract: Assembling flexible printed circuit board FPC to printed circuit board PCB is a multi-process and refined work. Using two EPSON C4-A601S robots to work cooperatively to insert both ends of FPC board into electronic circuit board PDF can not only realise the connection of the two circuits, but also achieve the purpose of high-precision assembly. The design takes RC700-A robot controller as the core, and a machine vision system is introduced to accurately judge the positional relationship between the PCB board and FPC board, so as to realise the accurate positioning of FPC board. The force sense control part is introduced to detect and monitor the insertion force of FPC to realise the control of inserting into place. The PLC control part is introduced to complete the loading and unloading of the material platen. Ethernet or remote I/O communication between various parts is established to realise data or signal state sharing. The robot controller cooperates with vision and force sense controllers to control the working path and posture of the two robots and realise FPC board assembly. In the design, EPSON RC+7.0 is used as the software platform to write the robot motion control program. Using Vision Guide as software platform, image processing, visual positioning and other operations are carried out. Force Guide is used as a software platform to set force control parameters and monitor the force in real time. The application of this system in electronic circuit board assembly can not only improve efficiency and reduce production cost, but also realise high precision, intelligence and informatisation in the assembly process.
    Keywords: robot; FPC board assembly; visual positioning; force monitoring; control system.

  • Seismic data reconstruction method based on morphological component analysis   Order a copy of this article
    by Jianhong Yao, Jicheng Liu, Yongxin Chou, Ya Gu 
    Abstract: The real seismic data is usually undersampled in the space domain because of the physical or economic limitations. So the incomplete seismic data needs to be reconstructed before the subsequent processing. A MCA-based method is discussed in this paper, which uses the curvelet dictionary and LDCT dictionary to reconstruct the smooth components and the singular components respectively. The BCR algorithm is adopted to complete the sparse optimisation. The validity of the proposed method was tested by numerical experiments on synthetic and real data. As the method based on curvelet combining with POCS is widely used in practice, we compare its reconstructed results with the MCA-based method. The numerical results validate that the proposed method has higher reconstruction performance.
    Keywords: seismic data reconstruction; MCA; curvelet; LDCT.

  • Multi-cell multi-Bernoulli tracking method based on fractal measurement model   Order a copy of this article
    by Jihong Zhu 
    Abstract: For detecting and tracking directly from image observations without the need for any separate target detection, a novel multi-cell tracking method based on multi-Bernoulli filter using local fractal feature estimation is proposed. The Hurst coefficient estimated by the rescaled range analysis method in this paper is considered as the local fractal feature, which is one of most important parameters in the application of fractal theory. The local fractal feature can offer two advantages for multi-Bernoulli filter. One is that the observation model is easy to establish, the other is that the computation of likelihood function is simple. Experiment results show that our proposed method could achieve an accurate and joint estimate of the number of cells and their individual states, especially in the case of the number of cell population varying and the cellular morphology changing. Furthermore, it shows equivalent tracking accuracy against other tracking methods.
    Keywords: multi-cell tracking; multi-Bernoulli filter; local fractal feature; Hurst coefficient; rescaled range analysis.

  • Study on temperature rise and thermal deformation of rotor caused by eddy current loss of magnetic-liquid double suspension bearing   Order a copy of this article
    by Liwen Chen, Dianrong Gao, Jianhua Zhao, Jisheng Zhao 
    Abstract: Magnetic-Liquid Double Suspension Bearing (MLDSB) is composed of electromagnetic supporting system and hydrostatic supporting system. Owing to greater supporting capacity and static stiffness, it is a great choice under the occasion of middle speeds, overloading and frequent starting. As the MLDSB works, the rotor will rotate at high speed and cut the magnetic induction line to produce eddy current loss (ECL), which will increase the temperature of rotor and lead to thermal deformation. Gaps between magnetic poles and magnetic sleeve are small, so thermal deformation of rotor has a clear impact on oil film thickness, bearing capacity and operation stability. Therefore, the simulation model of MLDSB was established, and the simulation of current loss, temperature rise and thermal deformation of rotor under maximum load condition was carried out. The result showed that eddy current loss will be aggravated by the increase of coil current, number of turns of coils and rotor speed, which will cause temperature rise and thermal deformation of rotor. The research in this paper can provide the theoretical reference for ECL of MLDSB.
    Keywords: magnetic-liquid double suspension bearing; eddy current loss; temperature rise; thermal deformation.

  • Identification of MISO Wiener systems using the LMI algorithm   Order a copy of this article
    by Lincheng Zhou, Xiangli Li 
    Abstract: This paper focuses on a new identification method for multiple-input single-output (MISO) Wiener nonlinear systems, in which the static nonlinear block is assumed to be a polynomial. The basic idea is to establish a MISO Wiener nonlinear identification model with polynomial nonlinearities by means of the key term separation principle. Then, a new identification method based on Levenberg-Marquardt iterative (LMI) search techniques, which can make full use of all the measured input and output data, but also automatically change the search step-size according to the change values of the quadratic criterion function, is derived to obtain an accurate and fast parameter estimation of the model. Finally, the simulation results demonstrate the efficacy of this method.
    Keywords: LMI search; parameter estimation; iterative algorithm; multiple-input single-output system; Wiener nonlinear system.

  • A Novel Double-mGBDT-based Q-Learning   Order a copy of this article
    by Qiming Fu, Shuai Ma, Dawei Tian, JianPing Chen, Zhen Gao, Shan Zhong 
    Abstract: This paper proposes a novel double-mGBDT-based Q-learning algorithm. Compared with traditional deep reinforcement learning, the proposed algorithm uses the mGBDT to replace the DNN, where the mGBDT is introduced as the function approximator. In the learning process, based on the state, we use the Bellman equation to construct the target value, which is used to train the mGBDT in an online manner. Like DQN, we also adopt two mGBDT frameworks, which are used to address the problem of easy divergence. To verify performance, we apply the proposed algorithm DQN and mGBDT to the traditional benchmark problems in CartPole and MountainCar. The results show that the proposed algorithm can converge to the optimal policy, and compared with DQN, the proposed algorithms stability is much better after convergence.
    Keywords: deep learning; reinforcement learning; mGBDT.

  • Back-propagation algorithm to estimate the parameters of auto-regressive exogenous model   Order a copy of this article
    by Tianyang Xu, Jing Chen, Yingjiao Rong 
    Abstract: This paper proposes a back-propagation (BP) algorithm to estimate the parameters of Auto-Regressive eXogenous (ARX) models. By using the SAG method, the proposed algorithm identifies the weights/parameters of the neural network constructed for the ARX model. Furthermore, in order to decrease the oscillation phenomenon in the SAG algorithm, two modified SAG algorithms are developed. A simulation experiment is presented to verify the effectiveness of the proposed methods.
    Keywords: system identification; SAG; sliding data window; weighted; BP neural network.

  • Research on pole placement approach of non-standard S-state-space model with case study   Order a copy of this article
    by Yizhi Wang, Zhong Yang, Shanshan Gu, Weina Chen, Zhenzhong Yu 
    Abstract: Pole placement is a very significant method to improve system performance regarding state feedback control system based on state space approach. However, the criteria to apply pole placement are rather restrictive, especially a MIMO object, requiring a model fully controllable and in standard controllable form. This paper aims to increase the restriction of pole placement regarding those with non-standard form but proved controllable, by introducing a matrix K into state feedback path for matrix transmission. Additionally, the proposed method is applied to an industrial case study based on its state-space model. Simulation results show that, after successful pole placement, system stability and rise time have been improved significantly.
    Keywords: pole placement method; non-standard form model; state feedback approach; matrix transformation.

Special Issue on: Recent Trends of Adaptive Control and its Applications for Unmanned Systems

  • Controlling of lower order dead system by implementing using adaptive RST algorithm   Order a copy of this article
    by Yanzhu Guo 
    Abstract: Since the advent of time, mankind has made numerous technological advancements. Every system build is time varying in nature as well as nonlinear, hence they intend to vary with time. A dead or slow system tends to vary at a slow rate, hence controlling them is more challenging compared with other nonlinear systems. In order to control such systems, we implemented an RST-based control algorithm that can track the output response of the unknown system and control it with minimal error. (?) is a quantity measured in time and because it is a time-varying parameter for lower order systems, this is the parameter that will change with time and in turn change an entire system response. Since the system is time varying, it is bound to change its coefficients at certain points in time, which changes the system, hence the previously applied control parameters might not be suitable to control the system. Least Squares (LS) algorithm is used to track the changes occurring in the system with time and thus changing the control parameter to complement the changes occurring in the system. The proposed algorithm of RST control design of a slow time-varying system is compiled and simulated in MATLAB environment. The proposed method is compared with the conventional PID controller and it tracks the signal faster with the same amount of work force. It provides better performance and it is more robust and effective than the traditional PID controller.
    Keywords: RST control algorithm; slow time-varying system; dead systems; least squares algorithm; adaptive control; self-tuning regulator.

  • Research on vibration compensation control of electromagnetic bearings rotor   Order a copy of this article
    by Wen Ji, Dongyuan LV, Jianjun Yuan, Qichao LV, Lijun RONG, Lili Zhu 
    Abstract: Aiming at the micro vibration caused by mass imbalance of high-speed magnetically suspended rotor, a novel electromagnetic force vibration compensation control method based on adaptive notch filter has been studied. Considering the influence of mass unbalance disturbance, the closed-loop control system model is established, and the internal relationship between mass imbalance and electromagnetic force disturbance is analysed. The simulation results show that the method can effectively suppress the influence of electromagnetic force induced by mass imbalance and reduce the micro vibration of the whole system.
    Keywords: magnetically suspended rotor; mass unbalance; notch filter; vibration control; force compensation.

  • Comparative analysis of simulator tools for unmanned aerial vehicle communication networks   Order a copy of this article
    by Haque Nawaz 
    Abstract: In recent years the real world problems are going to be solved by researchers using simulation method for experiments in the field of networks and communication. However, the simulator tools are generally used to assess the theories, ideas and hypothesis by research society. The Unmanned Aerial Vehicle Communication Networks (UAVCN) is an emerging area, having many issues which could be explored. The development of a testbed to explore such issues is an expansive proposition. Hence simulation tools are required to study UAVCN. Therefore, it is more important task for researchers to select an appropriate simulator tool for specific research experiment. This paper presents the simulator tools such as NS2, NS3, JSim, MatLab, OMNeT++, OPNET, QualNet, GloMoSim and AVENS. These various simulator tools' features and their pros and cons are discussed. Comparative analysis of features of these tools is highlighted in tabular form, and selected simulator tools are compared by implementing a routing protocol. In addition, simulator tools' performance is evaluated in terms of memory use and computation time with respect to increasing the density of network nodes. Consequently, it has perceived that NS3 is more efficient than NS2, OMNeT++ and GloMoSim. In addition, it has been found that through the support of these tools, researchers can explore and design the testbed scenarios and evaluate the performance of UAVCN.
    Keywords: UAV; UAVCN; NS2; NS3; JSim; MatLab; OMNeT++; OPNET; QualNet; GloMoSim; AVENS; simulator tools; unmanned aerial vehicles.

  • Non-singular terminal sliding mode control of converter-fed DC motor system with mismatched disturbance compensation   Order a copy of this article
    by Arshad Rauf 
    Abstract: In this paper, the problem of angular velocity trajectory tracking for converter-fed DC motor system with both matched and mismatched disturbances is addressed using a non-singular terminal sliding mode control. The combination of DC-DC power converters with DC motors for generating so-called smooth start of drives has numerous practical applications. Achieving high performance in such systems is however limited by the influence of unknown multiple disturbances. Furthermore, the conventional mathematical model used to express the dynamics of converter-fed motor systems effectively complicates the ability to compensate the multiple disturbances directly with control signal. In this paper, a control structure with nonsingular terminal sliding mode controller and a finite-time disturbance observer is proposed to address these practical issues. First, a special state transformation is applied, aggregating the acting disturbances/uncertainties in a sole perturbing term of the system expressed in new coordinates. Then, the observer estimates in real-time the information about the lumped disturbances based on already available input/output signals and the obtained estimated signals (and their high order time-derivatives) are used to construct a sliding surface. Finally, the sliding mode controller is applied to achieve high performance of the resultant plant dynamics and to robustify the governing scheme against modeling discrepancies. The efficiency of the proposed control method is validated through multi-criteria numerical simulations.
    Keywords: converter-fed DC motor; finite-time disturbance observer; mismatched disturbances; sliding surface.

  • Single dimension based fuzzy sliding mode control design for the stabilisation of underactuated unmanned underwater vehicle   Order a copy of this article
    by Ghulam E. Mustafa Abro, Saiful Azrin Bin Mohd Zulkifli, Muhammad Sadiq Ali Khan, Vijanth Sagayan Asirvadham 
    Abstract: This paper provides a comparative analysis of traditional fuzzy-based sliding mode controller (F-SMC) and Single Dimension based Fuzzy Sliding mode controller (SDF-SMC) for the stabilisation of an underactuated unmanned underwater vehicle (UUV). It has been observed that the sliding mode control design stabilises the underactuated mechatronic system very smartly, i.e. quadrotor craft, but it causes a high number of oscillations known as Zeno phenomenon. This effect can easily be eliminated using fuzzy version of sliding mode controller (F-SMC) but it also takes long processing time owing to fuzzification, inference and then defuzzification steps. The major reason for consuming such a huge process time is the two-dimensional rule base table. Thus, this research work will turn that two-dimensional table of rules into a single dimension and propose single input based fuzzy sliding mode controller (SDF-SMC) generating crisp input. In addition to this, the paper presents the comparative simulation work between conventional Fuzzy-based SMC (F-SMC) and Single Dimension based Fuzzy SMC (SDF-SMC), performed on MATLAB/Simulink software to shorten the processing time. In order to validate the effectiveness of proposed control scheme, the SDF-SMC algorithm is also implemented using proper hardware and experimental setup.
    Keywords: sliding mode control; fuzzy-based SMC; single dimension based fuzzy SMC; unmanned underwater vehicle; processing time.

  • Manoeuvring control of an underactuated single rotor aircraft   Order a copy of this article
    by Boni Liu 
    Abstract: The purpose of this paper is to control the manoeuvres of a model of a single rotor unmanned aircraft (SRUA). SRUA delivers a more flexible and attractive solution with a characteristic of vertical takeoff and landing (VTOL) and hovering ability. This study also designs the control algorithm of this aircraft. The designed control algorithm is applied to the model with six degrees of freedom (DOF). Simulations results show that the designed control scheme has better stability and behavior. A new configuration of rotorcraft as the research object, to realise autonomous flight, according to the existing relevant practical experience, combining the theory of flight mechanics, automatic control, computer technology, and other related disciplines to design the rotorcraft flight control system. Analysis of the flight principle and structure of the rotorcraft is evaluated and carry out the analysis of force and torque, establishing the mathematical model. In combination with the state equations, analyse the coupling analysis and stability of aircraft and describe its control difficulties. A flight control strategy for aircraft is presented and the control loop is designed, including the high control loop, speed control loop, and attitude control loop. After the simulation of several subjects, the proposed control strategy can realise the autonomous flight of the unmanned aircraft.
    Keywords: unmanned aerial vehicle; control strategy; flight control system; autonomous flight; simulation.

  • Robust model reference adaptive control for five-Link robotic exoskeleton   Order a copy of this article
    by Saim Ahmed 
    Abstract: In this paper, H based model reference adaptive control (MRAC) for controlling the five-link lower limb robotic exoskeleton is developed during the single support phase with uncertainties and unknown external disturbances. Unlike classical MRAC, the presented scheme using H control and MRAC with adaptive error gain is proposed under uncertain parameter vector and external disturbance. For robustness, H performance is applied to attenuate the impact of perturbation. Moreover, robust asymptotic stability of the overall system is investigated by Barbalat Lemma selecting a suitable Lyapunov functional candidate. Finally, the efficacy of the developed scheme is validated by relevant example.
    Keywords: adaptive control; H∞ performance; asymptotic stability; exoskeleton.

  • An AI-driven automotive smart black box for accident and theft prevention   Order a copy of this article
    by M.Kashif Shaikh, Sellappan Palaniappan, Touraj Khodadadi 
    Abstract: This paper proposes an automotive Smart Black Box (SBB) for accident and theft prevention using artificial intelligence (AI). The SBB is a versatile device and can work with any type of vehicle including electric and conventional cars. It has five smart features including constant facial recognition through AI to identify the driver's face and drowsiness detection. Drowsiness detection will help avoid catastrophic disasters by alarming the drivers if they are about to fall asleep. The SBB also has a 24/7 voice-recording feature that can be used to identify the reasons that caused the accident. Another feature is real-time vehicle tracking using Global System for Mobile Communication (GSM) technology. The SBB would immediately notify the owner if there is any abnormal vehicle movement and help prevent theft. The main contribution of this study is to design an SBB that contains the five aforementioned features and synergising these features without any of them malfunctioning. The black box will be a Raspberry Pi working on OpenCV for monitoring the car and the details will be sent to the user via a network connection to his android application for a more efficient approach to notify the user. Experimental results prove the efficacy of the designed SBB in recording the relevant information and helping prevent both accidents and thefts of vehicles.
    Keywords: face recognition; AI; drowsiness detection; GPS tracking; GSM module; smart black box.

  • Formation control of multiple UAVs using PID control approach   Order a copy of this article
    by Muhammad Shafiq, Abid Muhammad Khan 
    Abstract: This paper studies the formation control of multiple UAVs using PID algorithm. In formation control flight mode, the system uses a control law for keeping the UAVs in formation within the core of multi-UAV formation itself. At present, most of the formation researches are using the particle model, or the autopilot model. This paper contains details of establishing UAV models, flight controllers and the track formation to keep system design consistency in closed loop formation. The full formation of the system is to maintain control law, which is in order to promote a consistent message. We study construction of the team, its model, formation of geometric description, collaborative model variables that are used in unmanned aerial vehicles. We investigated detailed design collaboration of flight and track control algorithm, not only for formation of a stable internal formation, also letting the whole formation can suppress the predetermined route.
    Keywords: formation control; multiple UAVs; PID control.

  • Path planning of hovercraft using an adaptive ant colony with an artificial potential field algorithm   Order a copy of this article
    by Zain Anwar Ali, Han Zhangang 
    Abstract: This study designs a novel strategy by combining the adaptive ant colony optimization (ACO) method with the artificial potential field (APF) algorithm. The classic ACO algorithm has multiple limitations, such as falling into local optimum, slow convergence rate, etc. This hybrid strategy aims to counter the aforementioned problems. This study discusses the previous and current works in the concerned research area to better understand the solutions available and then try to improve them further. Then, this paper presents the mathematical model of the hovercraft. Afterwards, this study designs the novel hybrid method by using the adaptive ACO in conjunction with the APF method. We use two different scenarios in simulation to test the validity of the designed strategy. First, we test the hybrid method in an environment with predetermined obstacles. Secondly, we use a dynamic mission area with shifting obstacles to further prove the efficiency of the designed method. The simulation results prove that the designed strategy is more effective and robust than traditional ACO. It converges quicker and finds the most optimal path.
    Keywords: adaptive ant colony; artificial potential field; path planning; hovercraft.

Special Issue on: Machine Learning in Bio-Signal/Image Analysis

  • Classification of magnetic resonance images of brain using concatenated deep neural network
    by Abhishek Das, Mihir Narayan Mohanty 
    Abstract: Medical image classification is an ongoing research topic in the field of medical science. Deep learning is playing a vital role in image analysis due to its ability of auto feature extraction. Various works have been developed for brain image classification with models with high complexity but less performance. In our work, we have explored the deep learning techniques in the ensemble and stacking approach with less complexity and improved performance. Convolutional Neural Network, Recurrent Neural Network, and Long Short Term Memory are used as base classifiers for feature extraction and first stage classification. The predictions of the base classifiers are fed to the Multilayer Perceptron model for second stage training and classification. The performance of the proposed model is verified with a brain magnetic resonance image dataset online available at Kaggle. F1-score, recall, precision, sensitivity, specificity, and accuracy are calculated for evaluation of the proposed method and the effectiveness of choosing this method is discussed in the result section. 97% classification accuracy is achieved in the proposed method on the brain MRI dataset which is representing a competitive result concerning the state-of-the-art methods to the best of our knowledge.
    Keywords: Brain image classification; convolutional neural network; recurrent neural network, long short term memory; multilayer Perceptron; ensemble learning

  • Relevant Gene Selection Using ANOVA-Ant Colony Optimization Approach for Malaria Vector Data Classification
    by Micheal Olaolu Arowolo, Joseph Bamidele Awotunde, Peace Ayegba, Shakirat Oluwatosin Haroon-Sulyman 
    Abstract: Recent progress in gene expression data research makes it possible to quantify and identify several thousand genes' expressions simultaneously. For malaria infection and transmission, gene expression data classification using dimensionality reduction is a standard approach in gene expression data analysis and proposed for this study. A major problem occurs in the reduction of high dimensional data, it plays a significant role in improving the precision of classification, allowing biologists and clinicians to correctly predict infections in humans by choosing a limited subclass of appropriate genes and deleting redundant, and noisy genes. The combination of a novel Analysis of Variance (ANOVA) with Ant Colony Optimization (ACO) approach as a hybrid feature selection to select relevant genes is suggested in this study to minimize the redundancy between genes, and SVM is used for classification. The proposed method's efficacy was shown by the experimental outcomes based on the high-dimensional of gene expression data.
    Keywords: Malaria Vector; Gene Expression; Analysis of Variance; Ant Colony Optimization; Support Vector Machine; Machine learning

  • A novel framework for segmentation of uterus fibroids in ultrasound images using machine learning models
    by Dilna K T, Anitha J, Jude Hemanth 
    Abstract: A tumor of noncancerous structure that appears in uterus during child-bear years are uterine fibroids. Thus, it is necessary to design a fibroid detection system for the fibroid ablation. Various methods developed for the detection of fibroids are easily affected by the image artefacts as they do not take into consideration the spatial information and have lower efficiency problems for fibroid segmentation. This paper put forward a method for segmentation for fibroid detection. The proposed segmentation model overcomes the drawbacks of existing methodologies of fibroid detection in all stages. Here, the speckle noise existing in the noisy input image can be removed by using IGDT-DWT method and EMD-GCLAHE method. After contrast enhancement, the segmentation of the contrast-enhanced image is done using a novel clustering algorithm namely PC-KMA. The proposed segmentation algorithm effectively detects the fibroids, which are experimentally proved by comparing it with existing classifiers.
    Keywords: Uterus fibroid, ultrasound scanned images, DWT, Kmean algorithm

  • A Hybrid Model for the Identification and Classification of Thyroid Nodules in Medical Ultrasound Images
    by Rajshree Srivastava, Pardeep Kumar 
    Abstract: Ultrasonography(USG) is one of the leading diagnostic methods for accurately distinguishing the early-stage of thyroid nodules. ANN-SVM hybrid model is proposed for the identification and classification of thyroid nodules in medical ultrasound images. After feature extraction using Grey level co-occurrence matrix method, two experiments are performed. In the experiment-1, five different machine learning (ML) classifiers like Random forest (RF), Support vector machine (SVM), Decision tree (DT), Artificial neural network(ANN) and K- nearest neighbor (KNN) are used for classification. While in experiment-2, the two best classifiers based on the performance are hybrid together. The proposed hybrid model has achieved 84.12% accuracy, 85.14% sensitivity and 82.95% specificity on the public dataset having 295 USG images and 90% accuracy, 91.66% sensitivity and87.5% specificity on the local dataset having 654 thyroid USG images. It has shown an improvement of 2% to 5% in the performance evaluation in comparison with the other state-of-the-art methods.
    Keywords: Ultrasonography, Hybrid, Artificial Neural Network, Machine Learning, Thyroid Nodule, Classification

  • Mathematical Modeling for prediction of spread of COVID-19 & AI/ML based technique to Detect SARS-CoV-2 via Smartphone Sensors
    by DIGVIJAY PANDEY, Sumeet Goyal, Harjinder Singh, Joginder Singh, Rahul Kakkar, P Naga Srinivasu 
    Abstract: An infectious and communicable disease named COVID-19 is a novel coronavirus which originated from China. The virus is known as severe acute respiratory syndrome coronavirus 2 abbreviated as SARS-CoV-2 and generally known as COVID-19. Its epicentre is in the Wuhan city of China. In this paper, mathematical model (SEIR) for the prediction of infectious diseases have been described and modelled which can be used to predict the cases and a framework is proposed to detect SARS-CoV-2 from home is also proposed based on artificial intelligence, machine learning and smartphone embedded sensors. There are various sensors embedded in the smartphones such as proximity sensor, light sensor, accelerometer, gyroscope and fingerprint sensors which have very fast processors and memory space making it easy to read the symptoms or activity and scan the CT images, compare and can be used to detect COVID-19. The proposed framework is based on AI, cloud and ML
    Keywords: SARS-CoV-2 ,Modeling,AI/ML,Sensor

    by Moolchand Sharma, Shubbham Gupta, Himanshu Aggarwal, Tarun Aggarwal, DEEPAK GUPTA, Ashish Khanna 
    Abstract: Alzheimer's disease is a type of brain cancer, similar to coronary artery disease. Alzheimer's disease (AD) is a progressive neurological disorder that impairs memory, thinking abilities, and behavior. Thus, early detection of the condition is critical, as there is no cure. We conducted a comparative analysis of various evolutionary algorithms for extracting meaningful information from the Alzheimer's dataset, which is then utilized to predict whether or not a patient has the illness. We attained an accuracy of 78-85 percent using machine learning methods. When we utilized various evolutionary algorithms to perform feature selection, we observed an increase in accuracy of 5-10%, with Grey Wolf and Quantum Grey Wolf optimization (qGWO) achieving the highest accuracy of 92.8 percent and 94.5 percent, respectively, using a random forest classifier. The model was evaluated using three metrics: the increase in accuracy, the time required to execute, and the number of features eliminated. Additionally, the testing revealed that certain characteristics are replicated across multiple models and might be regarded as critical in the process of identifying Alzheimer's disease.
    Keywords: Alzheimer's Disease (AD); Machine Learning; Neurodegenerative Disease; Bio-Inspired Algorithms; qGWO

  • An improved ensemble learning approach for the prediction of cardiovascular disease using majority voting prediction
    by Sibo Prasad Patro, Neelamadhab Padhy, Rahul Deo Sah 
    Abstract: Coronary heart disease (CHD) is one of the most common heart disease types in the world. Heart disease is a critical health issue today. It is one of the most frequent causes of mortality due to a lack of proper medical diagnosis, technology, and a healthy lifestyle. Machine learning is a typical application to predict an outcome from existing data. The machine learns the patterns from an existing dataset, and it applies different rules to the dataset to predict the outcome. Classification becomes a powerful machine learning technique for prediction. Few classification algorithms give satisfactory results, and some others produce limited accuracy. In this work, we propose a new ensemble classification model by combining multiple classifiers for improving the accuracy of weak algorithms. An ensemble classifier was applied by using a majority vote-based technique for cardiovascular disease prediction and classification. The performance of this model is implemented on the Cleveland dataset from the UCI repository has applied a three-dimensionality approach on the dataset, and the average accuracy of each method is calculated as PCA(0.8636), K-PCA(0.8630), and LDA(0.90). As the PCA and K-PCA provide the same accuracy, whereas LDA gives higher average accuracy. So LDA is used as the best dimensionality reduction technique. The results show that Model-6 based on AdaBoost with base estimator SVC gives an accuracy of 0.9230 and Model-7 based on AdaBoost with base estimator RFC gives testing accuracy 1.0 respectively.
    Keywords: Ensemble methods, Voting classifier, Coronary heart disease, Bagging classifier, Stacking classifier, Adaboost classifier

  • The statistical analysis in the era of big data   Order a copy of this article
    by Wang Zelin, Liu Xinke, Zhang Weiye, Zhi YingYing, Cheng Shi 
    Abstract: In the big data environment, the traditional machine learning algorithm for data processing is somewhat inadequate. Therefore, machine learning algorithms adapted to big data environment have become a research hotspot. At the time of the marriage of big data and machine learning, it is necessary to predict the related challenges and opportunities. This paper mainly analyses and summarises the current research status of machine learning algorithms for processing big data, and discusses the new opportunities and challenges that the machine learning paradigm will face in the era of big data. It explores the new technology breakthrough that machine learning will produce in the era of big data.
    Keywords: big data; machine learning; deep learning; integrated learning; transfer learning.