International Journal of Systems, Control and Communications (19 papers in press)

Regular Issues

• A smart home system for elderly living alone based on an improved DS algorithm and multi-sensor data fusion  
  by Qianqian Hu, Heguang Wang, Bing Wang 
  Abstract: A smart home system design method is proposed to address the challenges faced by the elderly living alone using smart home systems. By introducing a fuzzy control algorithm, the stability and fault tolerance of the system are enhanced, and the accuracy of device coordination is improved. The experimental results show that the proposed method has an average absolute error of 1.483 and a root mean square error of 1.976, which are lower than the traditional algorithms by 1.012 and 1.486, respectively. The data processing efficiency is improved by 23%, and the system response time is shorter than that of other algorithms, and the overall operation efficiency is enhanced. In addition, the system operates stably with a packet loss rate of 0.174% and a high data transmission speed, which proves that the method has a large potential in practical applications.
  Keywords: multi-sensor; data fusion technology; DS; fuzzy control algorithm; smart home system.
  DOI: 10.1504/IJSCC.2025.10071099
   
  
• Chattering free fractional-order sliding mode control for a buck converter based on an extended state observer: design, analysis and experiments  
  by Tao Zheng, Juan Li, Yi Xiao, Kaiwen Cao, Shengquan Li 
  Abstract: This paper presents a fractional-order non-singular terminal sliding mode control (SMC), based on an extended state observer (ESO) to mitigate load resistance disturbances in the buck converter. First, a novel modelling method is developed to convert the matched and mismatched disturbances caused by load resistance variations into a uniform matched total disturbance. An extended state observer is then designed to estimate and compensate the system state variables and total disturbance. Second, to overcome the issues of slow convergence and chattering phenomenon in traditional sliding mode control, a fractional-order non-singular terminal sliding mode control method is designed to achieve stable tracking of the reference voltage. Third, the stability of the closed-loop system is proved by Lyapunov theorem. Finally, both simulation and experimental results show that the proposed fractional-order SMC exhibits excellent voltage tracking performance and effective rejection of load disturbances.
  Keywords: buck converter; fractional-order calculus; disturbance estimation; non-singular terminal sliding mode; extended state observer; ESO.
  DOI: 10.1504/IJSCC.2025.10071127
   
  
• A power image autonomous recognition method based on improved regional full convolution network  
  by Shuhua Liang, Yansong Sun, Dalei Wu, Xian Yang, Jiaying Li, Lei Gao 
  Abstract: In view of the poor recognition effect caused by the interference of electromagnetic wave, external environment and other factors in the process of power image acquisition, an autonomous recognition method of power image based on improved regional full convolution network is proposed. Firstly, the power image is collected and the interference factors are analysed. Based on this, the image pre-processing is completed; Secondly, the offset and weight are introduced to increase the receptive field of the standard grid to improve the regional full convolution network; Then, the improved regional full convolution network is applied to construct the power image autonomous recognition model to realise the power image recognition function; Ultimately, empirical trials are conducted to substantiate the progressiveness of the suggested approach. The outcomes reveal that the detection precision of the proposed method for power imagery surpasses 94.32%, and it exhibits superior accuracy and recall rates.
  Keywords: full convolutional neural network; power image; autonomous recognition; smart grid.
  DOI: 10.1504/IJSCC.2025.10071258
   
  
• Fault diagnosis method for substation relay protection equipment based on CNN-SVM model  
  by Bing Tang, Zhenguo Ma, Tianlei Xia, Yuming Huang 
  Abstract: With the increasing complexity of power systems, diagnosing equipment faults has become increasingly challenging. Traditional methods often struggle to handle noise and nonlinear issues in power data effectively. To address these limitations, a fault diagnosis model for substation relay protection equipment was developed using a support vector machine (SVM), enhanced with a convolutional neural network (CNN) and a channel attention mechanism for further performance optimisation. Experimental results demonstrated that with a dataset size of 2,000, the proposed model achieved an accuracy of 97.2% and a false positive rate of 2.8%. Additionally, the model effectively diagnosed various fault types, attaining an average accuracy of 85% with a diagnosis time of approximately 1.3 seconds. These findings highlight the models superior fault diagnosis capabilities, including reduced false alarm rates and stable performance with large-scale data, providing robust technical support for the reliable and efficient operation of power systems.
  Keywords: relay protection equipment; RPE; fault diagnosis; convolutional neural network; CNN; support vector machine; SVM.
  DOI: 10.1504/IJSCC.2025.10071737
   
  
• High resolution network combined with PnP algorithm for pose estimation of aerobics robot  
  by Yan Liu, Yan Zhao, Bingyan Yu 
  Abstract: The traditional pose estimation method for aerobics robots has problems of low accuracy and computational efficiency. This study utilises high-resolution networks to extract local features and uses a transformer for multi-scale feature fusion to design a pose estimation method for aerobics robots. This method achieves an adaptive fusion of multi-scale features by constructing a transformer-enhanced HRNets feature extraction module, effectively capturing local and global features of robot joints. At the same time, a deformable attention mechanism is introduced to reduce the complexity of feature processing, and the EPnP algorithm with sparse control point constraints is used to establish the mapping relationship between 3D point clouds and 2D images. The precise solution of pose parameters is achieved through Levenberg-Marquardt optimisation. The proposed fitness robot pose estimation based on a high-resolution network and PnP algorithm can effectively improve the precision and efficiency of robot pose estimation, and reduce computational costs. This study is meaningful for the practical application of robot vision and attitude control.
  Keywords: high-resolution network; PnP algorithm; transformer; aerobics robot; feature fusion.
  DOI: 10.1504/IJSCC.2025.10071785
   
  
• Development of an integrated intelligent burnishing tool holder  
  by Ruize Tan, Zhipeng Yuan, Xuanyi Lin, Jialiang Zhu, Zhongyu Piao 
  Abstract: Cutting force measurement and vibration measurement are fundamental requirements in the burnishing process. Hence, various methods of measuring the cutting force and vibration have been proposed by many researchers. In this study, a novel integrated intelligent burnishing tool holder was designed to measure three-dimensional cutting forces and vibrations during the burnishing process. Using the positive piezoelectric effect of quartz crystal, a cutting force sensing unit was designed on a standard CNC tool holder to measure cutting forces, while three-axis vibrations were simultaneously monitored using a commercial A27F100 three-axis piezoelectric accelerometer during the burnishing process. Then the cross-sectional dimensions of the handle are determined by calculating the dimensions of the quartz crystal group. Finally, the three-way cutting force measurement part of the tool holder was calibrated. The calibration results show that the tool holder has high sensitivity and low nonlinear error, repeatability error and cross-axis interference.
  Keywords: cutting force measurement; vibration signal measurement; piezoelectric quartz crystal force measuring unit; integrated intelligent burnishing tool holder.
  DOI: 10.1504/IJSCC.2025.10071811
   
  
• Intelligent tea picking model integrating YOLOV5 and Fast R-CNN algorithm  
  by Yafei Li, Xuanzhang Zhu 
  Abstract: Due to the limitations of traditional manual tea picking, an intelligent tea picking model is proposed to enhance efficiency. The model pre-processes images using Gaussian filtering and two colour spaces for tender leaf and background segmentation, optimised by Otsus algorithm. An improved watershed algorithm segments the tea leaves, while the Zhang refinement algorithm and Shi-Tomasi corner detection determine picking points. Combining YOLOv5 and Fast R-CNN, with ResNet-50 and CBAM for feature extraction, ensures accurate tea recognition. A binocular vision system provides 3D coordinates, and a robotic arm performs precise picking. Results show the YOLOv5s model achieved over 0.8 in accuracy, recall, and average precision, with 97.2% segmentation accuracy, and CBAM enhanced model performance. This model offers a robust solution for intelligent, automated tea picking, supporting the mechanisation of tea production.
  Keywords: tea; YOLOV5; Fast R-CNN; RGB; image; binocular vision.
  DOI: 10.1504/IJSCC.2025.10071914
   
  
• Sliding mode control of wheel hub motors for electronic differential in electric vehicles  
  by Zhengjie Li 
  Abstract: A vehicle dynamics model based on sliding mode variable structure is proposed. The final load generated by the control process is smaller than that of the rear wheel independent drive mode, indicating that the four-wheel independent drive mode has more precise control over the motor. Compared with PI control, the wheel speed tracking accuracy of four-wheel independent drive electric vehicles is significantly improved, and in comparative experiments, it is shown that the longitudinal force error of the electric vehicle is controlled within 1%. At the same time, comparing the lateral force of the two, the difference value approaches 0, and there is a slight difference of about 1% in vertical load. The handling stability of the whole vehicle is also improved, and compared with the traditional sliding mode variable structure, the four-wheel independent drive model shows better control performance and higher accuracy in wheel speed control.
  Keywords: electric vehicle; variable structure control; hub motor; PI control; coordinated control algorithm.
  DOI: 10.1504/IJSCC.2025.10071975
   
  
• Product surface defect detection algorithm based on transfer learning and DL  
  by Xitao Sun, Shuo Xue 
  Abstract: This study addresses the critical challenge of enhancing surface defect detection accuracy in industrial manufacturing through an optimised deep learning framework. We propose a hybrid model integrating convolutional block attention module (CBAM) and embedded inverse residual block (EIRB) into the U-Net architecture, combined with meta-transfer learning for structural optimisation. The enhanced network demonstrates superior performance: achieving 93.6% classification accuracy, 81% recall rate, and 8.98-second average detection time. Cross-part testing on six industrial components shows 92.5% accuracy for gears and 90.3% for valves. Notably, under varying lighting conditions, it maintains the highest F1-score of 0.89 compared to conventional models. This approach balances computational efficiency with detection robustness, making it suitable for real-time industrial applications requiring high precision and adaptability. The results confirm that our method provides both technical superiority and practical feasibility for advanced defect detection systems in manufacturing environments.
  Keywords: industrial products; surface defect detection; transfer learning; meta transfer learning; MTL; U-Net.
  DOI: 10.1504/IJSCC.2025.10071976
   
  
• Mobile network data collection based on A2S-Det and DSANS algorithms  
  by Zeyang Xu 
  Abstract: Aiming at the problems of high energy consumption and delay in sensor data collection in mobile networks, the proposed optimisation algorithm in the research achieves the optimisation of energy consumption and delay through adaptive anchor point selection and detection, combined with the non uniform step-size distributed sub-gradient algorithm. When the number of anchor points is 9, the energy consumption is as low as 1.63 J, and when the coverage parameter is 4, the transmission delay is as low as 125 s. The transmission delays of static and dynamic data collection in winter are 149 s and 123 s respectively, and the energy consumption is 0.48 J and 0.31 J. Under 200 nodes, the energy consumption of the algorithm is 1.51 J, which is superior to 1.99 J of the greedy algorithm. This algorithm effectively reduces latency and energy consumption and improves the efficiency of data acquisition in mobile networks.
  Keywords: wireless sensor network; WSN; data collection; A2S-Det; DSANS; transmission delay; energy consumption.
  DOI: 10.1504/IJSCC.2025.10072073
   
  
• Decentralised adaptive fuzzy sliding mode control for robotic arms using a voltage control approach in workspace  
  by Li Wang 
  Abstract: The paper presents a novel decentralised adaptive fuzzy sliding mode control (AFSMC) strategy with voltage-based control for robotic arms operating in the workspace. Traditional torque-based methods require precise dynamic modelling and are often too computationally intensive for real-time or embedded applications. The proposed approach directly manipulates motor voltage inputs, simplifying control law derivation while ensuring robustness against uncertainties, unknown dynamics, and external disturbances. By integrating fuzzy logic approximators within the sliding mode framework, the method effectively compensates for structural and non-structural uncertainties, eliminating the need for accurate dynamic models. A hyperbolic tangent function is employed to reduce chattering and achieve smoother control signals. Furthermore, the workspace-based design addresses end-effector trajectory limitations inherent in joint-space controllers. Simulation results for a three-degree-of-freedom manipulator demonstrate high tracking precision, excellent disturbance rejection, and lower computational demand, making the proposed voltage-based AFSMC highly suitable for real-time industrial and collaborative robotic applications.
  Keywords: robotic arm control; voltage control tactic; adaptive fuzzy sliding mode control; AFSMC.
  DOI: 10.1504/IJSCC.2025.10072074
   
  
• Sign language recognition using improved 3D convolutional neural networks  
  by Hrithik Paul, Soubhik Acharya, Priti Paul, Bitan Misra, Nilanjan Dey 
  Abstract: Sign language recognition (SLR) plays an important role in enabling communication for those who are hard to hear or deaf. SLR involves recognising and translating signs into natural language, and this task can be enhanced by employing deep learning methods. The proposed approach uses 3D convolutional neural networks (3D CNNs) to extract features. Through this method, improvements in the accuracy and real-time performance of SLR can be achieved. In this experimental study, a 3D CNN along with an LSTM architecture is implemented for feature extraction in SLR systems, and their advantages and limitations over 3D CNN and 2D CNN and 3D CNN combined models are highlighted. Compared with the traditional 3D CNN architecture, the 3D CNN-LSTM model can effectively interpret the spatiotemporal features of sign language expression, which is crucial for accurately recognising signs. Additionally, various strategies for optimising the architecture of 3D CNN-LSTM to achieve better performance are discussed in this article. Finally, some remaining challenges and future research directions in this area are highlighted. The analysis of the outcomes indicates that the 3D CNN-LSTM architecture has excellent potential for enhancing the accessibility of SLR systems and facilitating communication for individuals who communicate via sign language.
  Keywords: 3D convolutional neural network (3D CNN); deep learning; sign language recognition; SLR; long short-term memory; LSTM; 2D convolutional neural network; 2D CNN.
  DOI: 10.1504/IJSCC.2025.10072174
   
  
• A study on classification of small sample based on stochastic configuration networks  
  by Saixian Yuan, Xuemei Yao, Yan Tang, Hongmei Zou 
  Abstract: Due to the limited availability of samples and the high cost of annotation, small sample classification presents significant challenges. Traditional models often struggle with poor generalisation and inadequate inter-class separability. To tackle these issues, this paper introduces an ensemble model called RS-SCN, which combines stochastic configuration networks (SCN) with the random subspace (RS) method. In this approach, the feature space is partitioned into random subspaces, each used to train an independent SCN model. The outputs of these models are then integrated through majority voting. This strategy reduces dependence on large-scale datasets while enhancing generalisation performance. Experimental results demonstrate that RS-SCN surpasses traditional methods in both generalisation and robustness. As a result, this approach enhances the applicability of SCN to both small-sample classification and function approximation tasks, offering an effective solution to the generalisation challenges posed by limited data.
  Keywords: stochastic configuration networks; SCN; classification of small sample; random subspace method; subset of features; majority voting.
  DOI: 10.1504/IJSCC.2025.10072175
   
  
• Non-fragile H_∞ control for 2D Markov jump systems with partially unknown probabilities and its application in metal rolling process  
  by Peng Cui, Zhenghao Ni, Feng Li 
  Abstract: This article studies the asynchronous H_∞ non-fragile control problem of two-dimensional Markov jump systems with imperfect probability information based on the Roesser systems. First, to make the system better simulate actual engineering applications, its transition probability and observation probability are considered to be completely known, partially unknown and completely unknown. Second, the hidden Markov model addresses the asynchronous phenomenon between the controller and the system, given that the system mode might not be precisely achieved in the actual environment. Simultaneously, non-fragile control is used to overcome the influence of controller gain fluctuation during the action process. Furthermore, a suitable Lyapunov function is constructed, and the associated theorem is used to deduce adequate requirements for the closed-loop system's stability. Lastly, the industrial steel rolling process and the Darboux equation example are used to confirm the feasibility of the suggested asynchronous non-fragile controller.
  Keywords: Roesser systems; hidden Markov model; HMM; non-fragile control; partially information; 2D systems.
  DOI: 10.1504/IJSCC.2025.10071155
   
  
• Multimodal comparative learning chip defect detection algorithm based on GLIP guidance  
  by Ziyi He, Bingqi Wang, Li Ma, Jingjing Fang 
  Abstract: In the production of semiconductor chips, the existing process technology and the working environment have an impact on the quality of the chip, so defect detection on the chip surface is crucial. However, in real-world environments, it is challenging to collect a sufficiently large and highly representative sample of defects. In this paper, we propose a multimodal comparative learning approach with GLIP for location guidance and Multi-scale fusion modules for different multiscale fusions to localise defect locations of different shapes and sizes. In the testing phase, samples from different chip types in the training set were used to demonstrate the good generalisation ability and accuracy of our model. Data was tested on the MVTEC dataset to demonstrate the superiority of our method, where the image-level and pixel-level accuracies on our privately owned chip dataset can reach 91.3 and 92.6, and the pixel-level accuracy on the MVTEC is 92.3.
  Keywords: defect detection; visual language model; zero-sample inference; comparative learning; transfer learning.
  DOI: 10.1504/IJSCC.2025.10071156
   
  
• A Q-learning-based approach in clock synchronisation in wireless sensor networks  
  by M. Muthumalathi, P.B. Pankajavalli, A. Priya Dharshini 
  Abstract: In this paper, clock synchronisation in wireless sensor networks through Q-learning-based approach is investigated. This paper proposes a distributed approach that leverages distributed Q-learning-based synchronisation (DQS) to optimise packet transmission decisions for enhancing clock synchronisation. The proposed DQS method dynamically adjusts transmission behaviours based on historical experiences, aiming to minimise packet loss and energy consumption while synchronising clocks among sensor nodes. In particular, DQS approach presents a more efficient communication path with only fewer updates and transmission which result in less energy consumption for enhancing clock variance. The experimental result reveals that the proposed DQS approach has 3% to 7% reduction in energy consumption when compared to traditional distributed synchronisation (TDS) and gradient time synchronisation (GTS). Further, the packet loss and logical clock variance are reduced to 31% and 33% comparing with TDS and GTS with 500 rounds of synchronisation.
  Keywords: wireless sensor networks; WSNs; clock synchronisation; Q-learning; reinforcement learning; packet transmission and energy consumption.
  DOI: 10.1504/IJSCC.2025.10071158
   
  
• Mobile robot control based on high-order sliding mode differentiator  
  by Cuiping Pu, Yi Wu, Dongchen Dai, Ruixin Li 
  Abstract: This paper proposes a dual loop sliding mode control method based on high-order sliding mode differentiator to address the problem of chattering in trajectory tracking of sliding mode controlled mobile robots. The sliding mode control based on high-order sliding mode differentiator avoids high-frequency switching of control signals in traditional sliding mode control, thereby suppressing jitter. High order sliding mode differentiators can achieve fast convergence within a finite time, and mobile robots can reach the desired trajectory in a shorter period of time. High order sliding mode differentiators can accurately estimate the system state, effectively compensate for unknown system dynamics and external disturbances, and improve the robustness and stability of the system. Simulation analysis shows that this control method effectively reduces chattering, has stronger anti-interference ability, and better trajectory tracking effect.
  Keywords: high-order sliding mode differentiator; mobile robots; trajectory tracking; dual loop control.
  DOI: 10.1504/IJSCC.2025.10071249
   
  
• Event-triggered adaptive PID control for nonlinear dynamic process  
  by Cong Xu, Wuyu Zhou 
  Abstract: Proportional-integral-derivative (PID) control has been extensively employed for nonlinear dynamic process due to its simple control mechanism, high reliability, and easy implementation. However, it is difficult to determine the control parameters of the conventional PID controllers, which makes it difficult to adapt to the changes in nonlinear dynamic process. To address the challenges of low precision and excessive updates in nonlinear dynamic process control, an innovative event-triggered adaptive PID (EAPID) control method is proposed in this paper. Firstly, an adaptive PID controller based on the long short-term memory neural network is designed to enhance control precision. The network parameters are updated online using the back-propagation through time (BPTT) algorithm and the momentum term is introduced to update the controller parameters to improve the control accuracy. Secondly, an event-triggered mechanism is exploited to ensure the stability of the system, so that the controller is updated only when the triggering mechanism is violated, reducing computational resource consumption. Finally, the effectiveness of the proposed control method is validated by two numerical examples. The comparison results with other methodologies demonstrate the effectiveness and superiority of the proposed EAPID control method.
  Keywords: PID control; long short-term memory network; LSTM; event-triggered control; nonlinear dynamic process.
  DOI: 10.1504/IJSCC.2025.10070804
   
  
• Calibration between gimbal camera and UWB anchor based on non-convex optimisation  
  by Jun Ma, Yaojun Zhou, Fei Lu, Xiaoming Xu, Yang Li, Keke Zhang 
  Abstract: In recent years, with the development of mobile robot technology, ultra-wide band (UWB) sensors consisting of anchors and tags, along with cameras, have been increasingly and widely used in the perception modules of robots. In response to the field of view (FOV) problem caused by the commonly used fixed cameras, this paper proposes a camera system with a one-dimensional gimbal, aiming to solve the data fusion problem between the gimbal camera and the UWB system. In this study, we address the limited (FOV) challenge posed by gimbal cameras in mobile robot perception systems. Here, we present an offline calibration method for the camera and UWB sensors, which utilises an Aruco (Ar) marker-based calibration plate and an optimisation method to obtain the external parameters of the sensors. The contributions of this work include the advancement of sensor calibration for mobile robotics and the enhancement of perception capabilities through the fusion of gimbal camera and UWB data.
  Keywords: ultra-wideband sensor; gimbal camera; external parameter calibration; optimisation; azimuth of arrival base station.
  DOI: 10.1504/IJSCC.2025.10071159