International Journal of Vehicle Design (48 papers in press)

Regular Issues

• Refined modelling of thin-walled beam, plate and joint for automobile frame  
  by Jiantao Bai, Wenjie Zuo 
  Abstract: At the conceptual design stage, the simplified frame is extensively applied in the body-in-white (BIW) structure to rapidly calculate its performances. However, it is difficult to acquire an accurate simplified frame of the BIW structure for the calculation of the bending stiffness, torsional stiffness and frequencies. This paper proposes a simplified modelling method by using the thin-walled beams (TWBs) with complex sections, semi-rigid elements and cross beam structures to create the simplified frame. Compared with the traditional modelling method, the TWBs contain more types of the complex section, the semi-rigid elements can describe various deformations, and the plate structures are further considered. Firstly, the properties of the complex sections are summarised. Especially, the torsional moments of inertia of the multi-cell sections are derived. Secondly, the semi-rigid beam element is reduced to a super element, which is composed of one beam element and three translational and three rotational springs. Among them, the spring stiffness of the semi-rigid element is obtained by solving the detailed finite element model of the TWBs, which can be accurately and rapidly solved by using this method. Thirdly, the cross beam structure is introduced to replace the plate structure by the equivalence of the mass and central deflection for the first time. This method can further improve the accuracy of the simplified frame. Lastly, a numerical example demonstrates that the simplified frame can accelerate the conceptual design of the BIW structure.
  Keywords: conceptual design; frame structure; complex section; semi-rigid element; cross beam structure.
  
  
• Dual-redundancy multi-mode control of high safety reliability steering wheel system  
  by Junnan Mi, Tong Wang, Xiaomin Lian 
  Abstract: The steering wheel system of a steer-by-wire system (SBW) offers steering torque to driver. In order to improve its safety and reliability, a dual-redundancy steering wheel system is proposed. This research proposes a dual-redundancy multi-mode control method for this system, which is divided into different control modes, namely the angle alignment mode, the pivot steer with resistance mode, and the uniform approach to middle position mode, so that the system can adapt to different vehicle running states and drivers manipulation habits. At the same time, a torque balance control method for two channel motors is proposed to balance their torques, and an order-reduce and reforming control method is proposed so that the system can still work even if one motor fails. Finally, this study built a vehicle test platform to verify the control method and prove its effectiveness.
  Keywords: steer-by-wire; safety; steering wheel; redundancy; control; steering torque ;SBW.
  
  
• Estimation method of vehicle sideslip angle considering coupled longitudinal and lateral dynamics  
  by Qian Zhang, Hongliang Zhou, Zhiyuang Liu, Linhui Zhao 
  Abstract: The coupled longitudinal and lateral dynamics control plays an important role in improving the vehicle motion performance. It is very complex to describe the lateral tyre force with a coupled longitudinal and lateral dynamics, thus making it difficult to establish a lateral dynamics model and design a sideslip angle observer. In this paper, based on the piecewise affine (PWA) description of the lateral force, a novel lateral dynamics modelling method is proposed considering coupled longitudinal dynamics. Firstly, a nominal PWA model of the lateral force is established. Secondly, by regarding the slip ratio and the dynamic load as varying parameters, we obtain a novel PWA lateral dynamics model which can describe the influence of the slip ratio on the lateral dynamics. Finally, the switched T-S observer is presented. Simulation results show the effectiveness of the proposed PWA model and the designed observer.
  Keywords: coupled longitudinal and lateral dynamics; piecewise affine model; vehicle sideslip angle; lateral force; slip ratio; switched T-S observer.
  
  
• Automotive airbag stiffness evaluation: a simulation-based conceptual design  
  by Javad Marzbanrad, Vahid Rastegar, Esmail Dehghani 
  Abstract: In this research, a model and a particle simulation method were used to analyse the stiffness of a fully deployed airbag that is related to the severity of the occupant injury in a frontal crash. First, a mathematical model based on a three-dimensional airbag geometry was introduced. Then, quasi-static and dynamic test setups were defined, including an external impactor to simulate an actual crash. A simulation was also designed by the corpuscular method with the same initial conditions to validate the airbag model and to assess the effect of parameters at the out-of-position impact. Finally, airbag stiffness characteristics were calculated for different airbag capacities and inflator performances as a function of internal pressure and volume. The evaluated time-dependent stiffness characteristics were linearized by the proposed method. The estimated coefficients were correlated to airbag geometry and inflator performance and can predict occupant injury in a high-speed crash.
  Keywords: automotive airbag; in position; out of position; stiffness; linearization; corpuscular method; conceptual design; occupant safety; crash injury; inflator performance.
  
  
• Design and evaluation of a driver intent based mobile control interface for ground vehicles  
  by Chengshi Wang, Kim Alexander, Philip Pidgeon, John Wagner 
  Abstract: The advent of drive-by-wire technologies has enabled exceptional mobility options for a greater number of drivers (e.g., age, experience, physical limitations). To extend the capability of current advanced drive-by-wire vehicles, an alternative vehicle driving mechanism must be considered to accomplish different driving manoeuvres while adapting to specific needs and constraints of certain drivers. In this study, a cellphone-inspired portable human-machine-interface (HMI) that incorporated the directional control of the vehicle as well as the brake and throttle functionality into a single holistic device is presented. A nonlinear adaptive control technique and an optimal control approach based on driver intent are also proposed to accompany the mechatronic system for combined longitudinal and lateral vehicle guidance. Assisting the disabled drivers by excluding extensive arm and leg movements ergonomically, the device has been tested in a driving simulator platform. Human test subjects evaluated the mechatronic system with various control configurations through obstacle avoidance and city road driving test, and a conventional set of steering wheel and pedals were also used for comparison. Subjective and objective results from the tests demonstrate that the mobile driving interface with the proposed control scheme can enhance the drivers performance by up to 55.8% when compared with the traditional driving system during aggressive manoeuvres. The systems superior performance during certain vehicle manoeuvres and approval received from the participants implies its potential as an alternative driving adaptation for disabled drivers.
  Keywords: longitudinal and lateral dynamics; vehicle dynamics; nonlinear control; adaptive control; optimal control; state flow control; mobile control interface; portable HMI; emulated cellphone driving device; driver intent; human subject testing; ground vehicles; drive-by-wire.
  
  
• Analysis of consumer preferences regarding automated driving technology and driving environment: the case of South Korea  
  by Seungryong Jeon, Kyuho Maeng, Youngsang Cho 
  Abstract: This study analysed consumers preferences regarding the technological attributes of automated vehicle (AV) and the driving environment. We conducted a conjoint survey among 1008 respondents in Korea. Based on the collected data, a utility function was estimated using a mixed logit model, considering the heterogeneity of the respondents preferences. In the conjoint survey design, we considered six core technological and economic attributes of automated driving: road conditions, weather conditions, ambient lighting conditions, fuel efficiency, automated driving level, and price. The empirical results show high heterogeneity in consumers attitudes towards AV technologies. By comparing the relative importance of these six attributes, we found that the price of the automated driving option is the most important attribute, followed by road conditions. Furthermore, the predicted market share of AVs, considering their technological evolution, shows that affordable price is an important factor for their diffusion.
  Keywords: automated vehicles; consumer preference; conjoint analysis; mixed logit; willingness to pay.
  
  
• Optimal design and experimental research on a new hybrid electromagnetic actuator with energy reduction for vehicle suspension systems  
  by Renkai Ding, Ruochen Wang, Xiangpeng Meng, Long Chen 
  Abstract: The linear electromagnetic actuator (LEMA) fails to balance the energy regeneration/electric energy reduction and dynamic performance of suspension systems. A new hybrid electromagnetic actuator (HEMA) integrating a linear motor and a hydraulic damper is put forward and designed to address the problem in this study. On the basis of the preliminary determination of the design and constraints of the HEMA, a modified skyhook control is employed as the control method, and the performance parameters are optimised for different road conditions. The urban road condition is used as an example, the performance requirement that should be achieved in this driving condition is considered as the optimisation goal of the structural parameters, and the theoretically achievable performance of HEMA is obtained. Then, a prototype is developed for characteristic and dynamic control tests. Results show that the HEMA can ensure the dynamic performances while recovering vibration energy compared with the LEMA. In the active control process, the former can significantly reduce the electric energy consumption and achieve the same dynamic performance control. Moreover, the HEMA can significantly improve riding comfort while keeping the dynamic tyre load within a reasonable range. The results demonstrate that the HEMA can effectively coordinate the vibration energy regeneration/electric energy reduction and dynamic performance of suspension systems.
  Keywords: HEMA; vibration energy regeneration/electric energy reduction; dynamic performance; optimal design; experimental research.
  
  
• Power distribution methods of ultra-capacitor/battery hybrid power source for vehicular applications  
  by Benxiang Lin, Guizhou Ren 
  Abstract: Compared with traditional single power sources such as batteries, hybrid power source (HPS) for vehicular applications can reduce energy consumption and improve energy efficiency. The development of high-efficiency HPS for electric vehicle (EV) faces a huge challenge to meet the elevated energy demands under complex and variable working conditions. A HPS consisting of ultra-capacitor (UC) with high power density and battery with high energy density is recognized as having strong potential to meet this challenge. Reasonable power distribution control and optimization methods can significantly improve the working efficiency and application performance of UC/battery HPS. However, various control methods have different performance characteristics. It is not possible to claim that any one of them is the most effective method, usually making it difficult to quickly and accurately determine the most suitable control method for a specific application of UC/battery HPS. This paper aims to provide a comprehensive and transparent description of the known power distribution control methods that recently applied in various UC/battery HPSs. The article starts with an overview of typical energy storage technologies for HPS. After that, the power distribution control methods of UC/battery HPS are analysed and discussed systematically, such as explicit rule-based, fuzzy rule-based, model prediction, dynamic programming (DP), convex optimization, particle swarm optimization (PSO), and genetic algorithm (GA). Following this, a new UC/battery HPS is proposed, its operation principle is described, rule-based and fuzzy-based power distribution methods are initially formulated. The results of a simulation example verify the validity of the designed scheme.
  Keywords: ultra-capacitor/battery; hybrid power source; electric vehicle; power distribution method.
  
  
• Fully hydraulic synchronous steering control of a vehicle based on gain fuzzy sliding mode variable structure  
  by Guang Xia, Jianshan Chen, Xiwen Tang, Linfeng Zhao, Shaojie Wang 
  Abstract: Aiming at the problem that the expected steering curve determined by the existing methods has poor traceability and cannot achieve synchronous steering, a method for determining the expected steering curve and its feasible area is proposed. Aiming at the uncertainty of system disturbance and the nonlinearity of oil leakage, based on the combined reaching law, sliding mode control is adopted, and the saturation function is introduced instead of the sign function, which imposes certain constraints on the control system. Aiming at the problem that the combined approximation law lacks gain adaptation, a fuzzy rule table based on wheel angle and angular velocity is formulated, and the gain fuzzy sliding mode control is realised. Simulation and experimental results show that the expected steering curve based on steering efficiency has good tracking performance; the gain fuzzy sliding mode variable structure control has excellent dynamic response characteristics and control precision. It can effectively eliminate steering deviation and realise full hydraulic synchronous steering of steering system.
  Keywords: full hydraulic steering system; synchronous steering; fuzzy sliding mode control; expected turning curve; steering deviation.
  
  
• Multi-objective optimisation of composite coil spring for vehicle suspensions  
  by Xiaokai Chen, Chao Li 
  Abstract: This work presented a general form of spring rate and strength prediction model and an optimisation method for developing a suspension FRP coil spring. The prediction model can deal with an arbitrary sequence of layers instead of widely used
  Keywords: fire-reinforced plastic; coil spring; stiffness prediction model; strength prediction model; optimisation design.
  
  
• Coupled electrothermal model and thermal Fault diagnosis method for lithium-ion battery  
  by Qiuting Wang, Wei Qi 
  Abstract: The dynamic behaviour of the lithium-ion battery system is analysed. The nonlinear parameters, residual errors and thermal faults of the battery model are studied. In our study, a coupled electrothermal model is established based on the cell electrical dynamic characteristics. The parameter identification algorithm is presented based on Lyapunov observer. The core temperature and surface temperature of the battery cell are calculated based on extended Kalman filter. We propose a new thermal fault diagnosis method and the residual generation scheme. The UDDS (Urban Dynamometer Driving Schedule) dynamic working condition is used to verify our electrothermal model and fault diagnosis method. The experimental results indicate that the thermal characteristics can be described and the thermal fault can be diagnosed more accurately.
  Keywords: lithium-ion battery; coupled electrothermal model; fault diagnosis; Lyapunov observer; extended Kalman filter; UDDS.
  
  
• The optimisation design of CFRP bumper beam based on ply compatibility  
  by Jing Chen, Sen Xu, Zhen Liu, Aotian Tang 
  Abstract: This study aimed to develop an optimization approach for carbon fibre-reinforced polymer bumper beams. First, we established a specific finite element model and conducted a high-speed crash simulation of a steel bumper beam to verify the accuracy of the model. Then, the shape and parameter of the section of the bumper beam were determined using an orthogonal experimental design, and the concept of ply compatibility constraint was proposed. Subsequently, the ply sequence and thickness were optimised using the multi-island genetic algorithm and the multi-objective particle swarm optimisation algorithm based on kriging surrogate model with adding-point strategy. Results showed that the section force of the energy-absorbing box of the bumper beam was reduced after optimisation. The indexes of intrusion and energy absorption were also improved to some extent compared with the original ones. A real vehicle experiment proved that the optimised bumper system met all the requirements on strength and crashworthiness, with 47.61% weight reduction.
  Keywords: bumper beam; carbon fibre reinforced polymer; ply compatibility; multi-island genetic algorithm; KMOPSO; kriging surrogate model.
  
  
• Investigation of drivetrain dynamics on low- ground using the brush model  
  by Jianing Yang, Georg Jacobs, Achim Kramer 
  Abstract: This article aims to investigate the influence of the low-friction ground on off-highway drivetrain dynamics. A comprehensive vehicle simulation model is established for this purpose. This vehicle model consists of three parts: the torsional drivetrain model, the vehicle body model of longitudinal dynamics and the tyre model with slip behaviour. To correctly reflect tyre slip behaviour, the brush model, which features a physical description of the tyreground contact patch, is applied. A group of simulations is carried out on the complete vehicle model, mainly from two perspectives: drivetrain eigen-modes and drivetrain transient response. For the drivetrain eigen-mode analysis, different tyre slip states and the transition process between these two states are taken into consideration. For the drivetrain transient response, the vehicle is assumed to drive over a patch of low-
  Keywords: low-µ ground; tyre slip model; brush model; torsional drivetrain model; drivetrain eigen-modes; drivetrain transient response.
  
  
• Multi-PMSM adjacent cross-coupling iterative learning synchronisation control  
  by Yadi Zhou, Mingzhu Xu, Shaohua Li 
  Abstract: For the current widely used adjacent cross-coupling control strategy, when the number of motors is large, the control structure is complex and the calculation is difficult which makes it hard to access the demand of simple and efficient high-precision synchronisation control. The adjacent cross coupling control structure is studied, and an iterative learning controller is designed. A new multi motor speed synchronisation control strategy is proposed, which reduces the complexity of the control structure. The simulation and experimental data prove that the control algorithm has strong robustness. Compared with the traditional adjacent cross-coupling control method, the synchronisation error of the system can be reduced by 78.17% and the tracking error can be reduced by 62.40%, which effectively improves the control accuracy of multi-motor synchronisation operation.
  Keywords: permanent magnet synchronous motor; vector control; synchronous control; adjacent cross coupling control;.
  
  
• Derivation of vehicle dimensions at the early concept stage based on occupant posture  
  by Igor William Santos Leal Cruz, Thomas Gänsicke, Julian Francisco Sandiano, Christian Raulf, Thomas Vietor 
  Abstract: This paper presents a new method for deriving dimensional concepts for vehicles at the early concept stage. The derivation is centred on the posture of the occupants, which is calculated with statistical models and used to determine spatial requirements. Dimensions that do not follow directly from the posture calculations are determined with dimensional chains. The number of independent parameters is reduced by specifying dimensions that are assumed constant within a given vehicle segment. The application of the method is exemplified with different vehicle concepts. Vehicle length, width, height, and wheelbase calculated in the examples are comparable to those of existing vehicles. The use of non-proprietary, closed-form equations and the low number of independent parameters simplify implementation and application of the method.
  Keywords: vehicle dimensions; dimensional concept; concept stage; vehicle layout; driver posture; passenger posture.
  
  
• On the aerodynamic effects of the optimised DrivAer fastback model car under adverse situations  
  by I-Ying Chiang, Tung Wan 
  Abstract: In this study, the benchmark DrivAer model is employed for optimisation work of vortex generator locations with digital side mirrors through numerical simulation. Aiming for the best drag reduction via various installation positions by the kriging surrogate method, results show that for the best-achieved location, it is observed that the positive effect of drag reduction owing to an increase in the intensity of vortices; its wake also becomes weaker. In addition, in order to cope with fast-changing severe weather, the impacts of important physical phenomena such as gusty crosswind, heavy rain, and cornering are simulated and compared as well through detailed analyses of different physical quantities, force coefficients, and Strouhal number. Our simulations show that the gusty crosswind and cornering have the most negative influences on car aerodynamic performance, thus proving the effectiveness of the current approach in automobile design and safety. In addition, the combination of gusty and rain conditions is also elaborated in our work, and results show a similar trend as previous simulations.
  Keywords: DrivAer model; drag reduction; vortex generator; kriging model; gusty crosswind; heavy rain; cornering.
  
  
• Alternative designs for caster shims in commercial vehicles  
  by Sagar Jambukar, C. Sujatha 
  Abstract: Currently, in vehicles with front solid axles and leaf spring suspension arrangement, caster angles for the front steered wheels are incorporated using caster shims. The caster shim, a wedge-shaped structure, is inserted between the leaf springs and the axle, providing the necessary tilt of the kingpin axis known as the caster angle. Herein, the authors investigate the use of an alternative arrangement for introducing caster angle by incorporating the required tilt in the axle design, thereby eliminating the use of additional components, namely caster shims. Comparative numerical analysis for symmetric static loading and peak asymmetric loading for the existing and two other alternative designs has been discussed in detail in this work. The structural integrity of the proposed designs was found to be equally good, thus resulting in acceptability for practical use. Further, the proposed designs provide added flexibility in the choice of kingpin inclination (KPI) and caster angle, which will result in the improved kinematic and dynamic response of these vehicles. This has been achieved without compromising the functionality of the axle, with the advantage of eliminating a component and, therefore, notable cost savings to the automotive manufacturer.
  Keywords: caster angle; caster shims; alternative design; structural analysis.
  
  
• The effect of camber control on power consumption during handling manoeuvres  
  by Herman Hamersma, Janu Botha, Andries Peenze, Schalk Els 
  Abstract: Vehicle power consumption is receiving widespread attention in the industry. One of the approaches is to include camber control to reduce power loss during cornering. This approach changes the camber angle during cornering, reducing the steering angle needed. The published literature is limited to semi-empirical tyre models and simplified vehicle models. This investigation uses a physics-based tyre model and a full vehicle model to verify if the power savings reported in the literature are indeed achievable. The simulation results indicate that the reported power savings are indeed possible, but that the power saved during normal driving is limited. It is concluded that camber control is not suited to reducing power consumption during normal driving. A deeper investigation revealed that the use of radial tyres is one explanation for this finding. Camber control is thus not recommended as a power-saving strategy for practical applications using traditional radial tyres.
  Keywords: camber angle control; cornering losses; vehicle power saving; handling.
  
  
• Parametrisation of a rolling resistance model for extending the brush tyre model  
  by Lisa Ydrefors, Martin Åsenius, Hugo Jansson, Sogol Kharrazi, Mattias Hjort, Jan Åslund 
  Abstract: A rolling resistance model has been created and parametrised with the purpose of modelling tyre rolling resistance within complete vehicle dynamics simulations. The rolling resistance model is based on a combination of the Masing and Zener model to simulate the Payne effect and the viscoelastic properties of rubber. The parametrised model is able to recreate the relationship between the rolling resistance and the tyre deformation well and it has a low computational power requirement. Today the model is limited to simulation of free-rolling tyres on a flat surface, but it can be extended to also include the effects of changes in operating conditions such as wheel angles or road surface.
  Keywords: rolling resistance; parametrisation; vehicle dynamics simulation; wheel load; tyre deformation; tyre modelling; tyre temperature; Zener model; Masing model.
  
  
• Precise pressure adjustment of electro-hydraulic brake system based on pressure-position cascade control and compensation  
  by Yiping Liu, Xiaofei Pei, Xuexun Guo, Shihao Zhen, Zhiwei Sun 
  Abstract: A control architecture consisting of feedforward compensation, friction compensation, pressure outer loop, and position inner loop is proposed for pressure control of an electro-hydraulic brake system (EHB). The pressure and position loops are cascaded to take full advantage of the accuracy of the pressure feedback and the rapidity of the position feedback. To ensure the tracking accuracy of the master cylinder pressure and actuator position, both the inner and outer loops are designed as adaptive sliding mode controllers according to the nonlinear model. Based on the cascade architecture, an online friction compensation module is constructed using motor speed and position to improve the tracking accuracy of the inner-loop controller. The feedforward compensation module is used to improve the pressure response speed and to reduce the adjustment burden of the pressure outer loop. The test results show that the control strategy can achieve accurate pressure-position tracking and has excellent fault tolerance performance.
  Keywords: electro-hydraulic brake system; adaptive sliding mode; cascade control; friction compensation; feedforward compensation; fault tolerance.
  
  
• How does an AR-HUD system affect driving behaviour? evidence from an eye-tracking experiment study  
  by Yunuo Cheng, Xia Zhong, Liwei Tian 
  Abstract: In recent years, the continuous development of IVIS (in-vehicle information systems) has immensely enriched the driving experience, while also occupying the driver's cognitive resources to varying degrees, causing driving distraction. Therefore, studying the influence of the AR-HUD (Augmented Reality Head-up Display, AR-HUD) system on driving behaviour is of great significance for the in-vehicle information system to manage the complexity of information and enhance driving safety. This experiment applied five typical risky driving scenarios under two lighting conditions (daytime and night), as stimulus materials. The effects of eye movement behaviour and risk reaction time indicators during driving were comprehensively analysed. The experiment results showed that the AR-HUD system can significantly improve the subjects' attention to risky AOIs (Area of Interest, AOI) in night driving situations, as well as reduce the difficulty of processing information in risky driving scenarios, thus reducing cognitive load; In terms of reaction time, the AR-HUD system can significantly reduce the driver perception time for risk driving scenarios and thus respond more quickly to high-risk situations. The experimental conclusions validate the role of AR-HUD technology in improving driving safety and driving behaviour and provide a new direction for further development of in-vehicle information systems.
  Keywords: AR-HUD; driving behaviour; risk driving; eye movement; area of interest.
  DOI: 10.1504/IJVD.2023.10056084
   
  
• Motion control of autonomous vehicles after a tyre blow-out, based on differential-flatness-MPC  
  by Yuhai Wang, Yanhui Xing, Yanfeng Cong 
  Abstract: This paper presents a vehicle safety evaluation index after a tyre blow-out event. By applying safety indicators, the movement characteristics of the vehicle are analysed according to different driving operation strategies, and then give the driving operation strategy for the vehicle after a tyre blow-out. In addition, through the analysis of the performance of the flat tyre wheel, the vehicle dynamics model with a flat tyre is presented. According to the operation strategy of the flat tyre, the MPC method based on differential flatness is used to investigate the motion control problem of vehicles with flat tyres when driving on a motorway. The simulation results show that the method can control the movement behaviour of the vehicle with a flat tyre well.
  Keywords: MPC; differential-flatness; tyre blow-out; motion control; autonomous vehicle; Simulink.
  
  

Special Issue on: Multi-Objective Design and Structural Optimisation of Vehicle Components with Nature-Inspired Optimisation Algorithms

• Integrated optimisation of two-speed powertrain parameters and shifting strategy for energy in electric vehicle  
  by Daoguang Zhu, Congbo Li, Lingling Li, Ying Tang 
  Abstract: In order to improve the economic performance and extend the range of electric vehicles (EV), an integrated optimisation method for the design and optimisation for powertrain parameters and shifting strategy are proposed. Firstly, the powertrain parameters are matched to ensure the dynamic performance of electric vehicles and the shifting strategy with comprehensive performance is designed based on the analysis of dynamic and economy performances. Secondly, a multi-objective integration model of powertrain parameter and shifting strategy optimisation is proposed to take the minimum energy consumption as the optimisation objective without sacrificing dynamic performance, which is solved by a multi-objective particle swarm optimisation algorithm. Finally, to verify the energy-saving performance of the proposed multi-objective integration problem, case studies have been conducted and a whole vehicle simulation model is proposed based on Matlab/Simulink platform. The simulation results show that the proposed method can effectively reduce the energy consumption and extend the range of electric vehicles under different driving cycle.
  Keywords: electric vehicle; powertrain parameter; shifting strategy; multi-objective integrated optimisation; particle swarm optimisation algorithm.
  
  

Special Issue on: New Energy Vehicles' NVH and Lightweight and Control Technologies

• Comparison of deep learning methods for predicting charging energy of power batteries  
  by Xuefeng Zhu, Guoliang Xie 
  Abstract: Accurate prediction of the electric vehicle charging energy is essential for power grid companies to rationally allocate power resources, customise appropriate tariffs and select the location of charging piles. Currently, machine learning methods have been widely applied in this field. Aiming to predict Electric Vehicles (EV) charging energy more precisely, this paper compares several machine learning methods and concludes that Long Short-Term Memory (LSTM) neural network has better behaviour. As the initial training data was incomplete, we supplemented the training data with MissFrorest neural network. We compared Back Propagation (BP), Xtreme Gradient Boosting (XGBoost), and LSTM networks for the prediction of charging energy, and found that LSTM has the best prediction effect, XGBoost has the second best, and BP has the worst effect. LSTM addresses the issue of gradient dispersion due to introducing time series, and thus has a better prediction effect. The experimental results show that Mean Absolute Error (MAE) and Root-mean-square error (RMSE) indices for four of five experimental vehicles using the LSTM algorithm are smaller than those using BP and XGBoost methods. Compared with the BP, XGBoost algorithms, the average reduction of MAE is 42.79%, 23.48%, and RMSE is reduced by 43.42%, 19.65%.
  Keywords: predicting charging energy; deep learning; power batteries; electric vehicles.
  
  

Special Issue on: Cyber Security in Internet of Vehicles

• Stabilisation of traffic flow by considering multiple information based on vehicle-to-vehicle communication  
  by Qian Li, Haiyang Wang, Dongfan Xie 
  Abstract: The rapid development of advanced technologies means that vehicles can share information with each other based on vehicle-to-vehicle (V2V) communication. It is expected that the V2V information can improve efficiency and stability of traffic flow, which has attracted much attention in traffic flow theory. To this end, this study develops an alternative car-following model with the consideration of V2V information. By using the linear stability theory, stability analysis is performed, and the string stability condition is obtained. The results indicate that V2V information can improve the stability of traffic flow, and traffic fluctuations can thus be suppressed. Case studies are carried out based on numerical simulations, and the results coincide with the theoretical ones.
  Keywords: vehicle-to-vehicle; communication; car-following model; stability analysis; traffic congestion.
  
  

Special Issue on: The Modelling and Control of Automated Guided Vehicle Systems

• Dynamic characteristics analysis of spatial suspension mechanisms based on instantaneous screws  
  by Guofeng Zhou, Yafei Wang, Zhisong Zhou, Jingkai Wu 
  Abstract: Multi-link, MacPherson and double-wishbone suspensions are commonly used in passenger cars. The dynamic characteristics of a vehicle are strongly affected by these independent suspensions. This paper presents a unified quarter-vehicle model which incorporates both the spatial suspension kinematics and the tyre dynamics for the dynamic analysis of the three spatial suspensions. In the quarter-vehicle model, it consists of a spatial suspension mechanism and a ground-wheel contact model. For a vehicle running straight ahead on the uneven road, the dynamic characteristics analysis using the quarter-vehicle model is divided into three steps. Firstly, for the quarter-vehicle model the corresponding instantaneous screws are determined at any instant. Then, the theory of screws is used to describe the kinestatics of the quarter-vehicle model. Finally, on basis of the kinestatic relations, the dynamic equations derived using the Lagrangian function are applied to the dynamic analysis. As to the feasibility of the theoretical method, different road disturbances are considered for the numerical examples. The simulation results respectively of the theoretical method and the widely used Adams/View software are compared to verify the numerical performances of the dynamic analysis.
  Keywords: spatial suspension mechanisms; quarter-vehicle model; instantaneous screws; kinematics; statics; dynamics; theory of screws.
  DOI: 10.1504/IJVD.2021.10056213
   
  
• Parameter-space-based robust control of heterogeneous platoon with stochastic packet dropout  
  by Jiawei Wang, Fangwu Ma, Sheng Zhu, Yu Yang, Sukru Yaren Gelbal, Bilin Aksun-Guvenc, Levent Guvenc 
  Abstract: This paper presents a parameter-space-based multi-objective, robust CACC platooning controller for a heterogeneous vehicular platoon with stochastic packet dropout during inter-vehicular communication. The parameter space approach is adopted to optimise the gains of the robust controller, to handle the multiplicative uncertainty, and to project the robust performance requirements. The feasible region where internal stability and preceding vehicle following accuracy are satisfied is then visualised in the parameter space. Subsequently, the robust controller, which combines the feedforward loop and feedback loop, is developed by selecting the gain from the feasible solution area. The simulation results of a six-vehicle heterogeneous platoon are presented and evaluated to verify the efficiency of this control algorithm. The results show that string stability and comfort are well guaranteed by this robust controller, even for the heterogeneous platoon with stochastic packet dropout and the following error is limited to an acceptable range.
  Keywords: cooperative adaptive cruise control; stochastic packet dropout; heterogeneous platoon; parameter space approach.
  
  
• Active roll control for rollover prevention of semi-trailers with robust invariant set  
  by Echuan Yang, Chuanren Xie, Jian Ou, Senlin Zhang, Liang Qin 
  Abstract: To improve the roll stability of semi-trailers, a robust model predictive controller (RMPC) is designed. To analyse the vehicle dynamic behaviour, a nonlinear seven-degrees of freedom (7-DOF) vehicle model is defined. Based on the robust invariant set theory, and taking the uncertainty of the drivers driving behaviour into account, the maximal robust control invariant (RCI) set is calculated and its robustness is analysed. The N-step controllable sets of the vehicle are also solved. An anti-roll controller considering multiple constraints is designed based on the robust model predictive control theory. Simulation results show that the controller can keep the lateral load transfer rate within 0.7 and make the state variables converge. In addition, the controller can reduce the lateral acceleration by 50% in the step steering input test.
  Keywords: anti-roll control; invariant set; robust control; model predictive control; semi-trailer.
  
  
• Design of remote control smart car with two-way information communication  
  by Hongyan Qin, Ben Zhang 
  Abstract: This paper takes Freescale's remote wireless control smart car as the research object, completes the design of the main control board hardware circuit and the upper controller remote control software, and realises two-way information communication. The smart car collects the vehicle position signal by the CMOS image sensor, and transmits the image to the controller in real time by the APC220 wireless module. The XS128 microcontroller receives the control instructions sent by the controller, and drives the DC motor to control the smart car. The single-chip microcomputer collects the pulse signal of the wheel speed through the photoelectric encoder, which is captured by the MCU for PID closed-loop control. Tests show that the smart car and the controller can achieve error-free data transmission in a long-distance, multi-interference environment, and have good stability and obtain real-time control information.
  Keywords: wireless remote control; smart car; two-way information communication; anti-jamming design.
  
  
• Real-time estimation of the vehicle moment of inertia based on IMU measurements  
  by Xiongshi Wang, Jiahua Qi, Steffen Mueller 
  Abstract: Identifying the vehicles moment of inertia for vehicle control in any driving condition is very important, e.g. for a good continuous performance of autonomous driving. In this study, an identification method is proposed for the pitch and roll moments of inertia. The performance of the proposed estimation method was evaluated through experiments using a 4-post test rig and IMU measurement system under different road conditions. Several scenarios have been investigated to verify the accuracy and robustness of the estimator in real-time. Our work shows very promising results with acceptable convergence time for a variety of manoeuvres.
  Keywords: vehicle dynamics; parameter estimation; state estimation; moment of inertia.
  
  

Special Issue on: Vehicle Design Processes

• Design of BLDC motor drive system using alternative controllers for performance evaluation in electric vehicle applications  
  by Mohamraj Nandakumar 
  Abstract: Electric vehicles have emerged as a promising and important alternative means of transportation, replacing IC engine driven automobiles, which use fossil fuels. One important feature to be incorporated in an electric automobile is regenerative braking for extending the operating range. Accordingly, BLDC motor-based power electronic drive systems capable of energy regeneration have become attractive. This paper discusses the automobile dynamics, covering road friction, aerodynamic forces, transmission systems and calculation of tractive force. The time dependant profile of the reflected torque and speed variables at the motor shaft are computed and used as reference data for an intelligent controller. Here, a multi-loop control scheme has been developed, whose gain parameters are tuned based on two alternate algorithms viz., PSO (Particle Swarm Optimisation) technique and MNFIS (Multiple Neuron Fuzzy Inference System) for comparison. Simulation results for two power circuit regeneration topologies with the above control algorithms are presented. The experimental part is a representation of profile based operation, which deals with the application of the PMBLDC motor-based drive system for powering a passenger car and the evaluation of the system performance. Regeneration of power is validated in a laboratory setup and recorded.
  Keywords: BLDC motor; electric vehicle; MNFIS; PSO.
  
  

Special Issue on: Advanced Safety Design and Control for Electric Vehicles

• Steering and braking combined path tracking control considering the reserve potential of tyre force  
  by Shaosong Li, Feng Wang, Haitian Xu, ZongHao Li, Gaojian Cui 
  Abstract: The coordination control of the steering and braking system is the key technology for vehicle stability control under high-speed and low adhesion limit conditions. The concept of equal reserve capacity of tyre longitudinal and lateral forces is proposed in this paper, which provides a novel idea for tyre force distribution. On this basis, a steering and braking combined path tracking control method based on model predictive control is proposed to guarantee vehicle path tracking performance. Research results show that the proposed method enables the vehicle to steadily and accurately track the planned trajectory under high-speed and low-adhesion condition. Compared with the schemes in other methods, the reserve margin of the tyre force is greatly improved in the proposed method, thus enabling the tyre to exert greater control potential.
  Keywords: trajectory tracking; steering and braking combination; model predictive control; reserve potential of tyre force.
  
  
• A new torque ripple suppression strategy based on the CSA for PMHM of electric vehicles under New European Driving Cycles  
  by Yao Zhang, Xiaodong Sun 
  Abstract: This paper presents a new torque ripple suppression strategy for a permanent magnet hub motor (PMHM) of electric vehicles' drive. With the complex characteristics such as nonlinear time delay and multi-dimension presented by the PMHM, the traditional PID controller has been unable to meet the requirements of the control system. Thus, the cat swarm algorithm (CSA) is introduced to improve the accuracy of PID parameters thanks to its good global search ability. Moreover, it is found that the proposed CSA-PID in the outer loop can obtain better performance such as smaller torque ripple and faster dynamic response both in steady and dynamic state compared with the traditional PID controller. Finally, the strategy proposed in this paper was applied to the vehicle model through HIL test platform. The possibility of applying the strategy proposed to EVs was verified under the New European Driving Cycle.
  Keywords: permanent magnet hub motor; cat swarm algorithm; torque ripple suppression; New European Driving Cycle.
  
  
• Research on crashworthiness and lightweight of frame body based on load path and material selection  
  by Tingting Wang, Ruoyan Dong, Yuechen Duan, Dongchen Qin 
  Abstract: In order to effectively optimise the frame body structure and match the performance of lightweight materials with the function of body structure, a material-structure optimisation framework of multi-material frame body is proposed to improve the lightweight and collision safety at the same time. Firstly, in order to improve the crashworthiness of the frame, the equivalent static load method is used to analyse the load path of the frame body to obtain the optimal structure. Secondly, the crashworthiness evaluation method based on evolutionary structural optimisation method is used to evaluate each member of frame body, which provides the basis for material selection. Finally, the material index is introduced to establish the material library. According to the deformation evaluation results, the material selection method based on bubbling method is used to select materials orderly to match the function of members, and the multi-material frame with the objective selection scheme is obtained. In this study, the proposed method is demonstrated by the lightweight of racing car body. The results show that the body mass is reduced by 25.60 kg after the lightweight design, and the crash safety is improved. Therefore, the proposed optimisation framework of multi-material frame body.
  Keywords: frame body; lightweight; load path; material selection.
  
  

Special Issue on: Emerging Technologies in New Energy Vehicles for Better Safety and Economic Performance

• Multi-objective optimisation design for brushless electrically excited synchronous machines in electric vehicles  
  by Naixi Xu, Xiaodong Sun, Ming Yao 
  Abstract: This paper proposes a new multi-objective optimization method to obtain the best performance of brushless electrically excited synchronous machines (BEESMs). First, the initial design of the BEESM is carried out and the parametric model is established. After the optimization problem of BEESM is transformed into a mathematical model, all design parameters are divided into three subspaces by sensitivity analysis method and multivariate analysis of variance method. Each sub-space is optimized based on approximate models and multi-objective optimization methods. Select the optimal solution from the Pareto solution set according to the filter conditions. The three subspaces are optimized in order, and the iterative process will continue until the convergence condition is satisfied. The experimental results based on the finite element model (FEM) show that this method can reduce the computational cost of BEESM optimization and improve optimization efficiency. The motor optimized by this method has better performance.
  Keywords: finite element model; brushless electrically excited synchronous machine; multivariate analysis of variance; Pearson correlation coefficient; multi-objective optimisation.
  
  
• Comparative analysis of multi-level inverter driven PMSM drive in automotive application  
  by Rakesh Shriwastava, Jagdish Chaudhari, Mohan Thakre 
  Abstract: This paper presents the comparative hardware analysis of multi-level inverter PMSM drive on the basis of two level and Twelve Switch Neutral point clamped Inverter (TSNPCI) driven Permanent Magnet Synchronous Motor (PMSM) drive using the AVR microcontroller. Multilevel inverters offer several advantages compared with the conventional three-phase bridge inverter in terms of lower dv/dt stresses, lower electromagnetic compatibility, smaller rating and better output features. The PMSM can be a better candidate in automotive application owing to many advantages, including high power-to-weight ratio, high efficiency, and rugged construction. The comparative harmonic features of the inverter-driven PMSM drive with SVM were analysed. The proposed AVR microcontroller based TSNPCI driven PMSM drive gives better control performance of speed and torque response, good dynamic response, less distorted output, less harmonic distortion and lower costs. Owing to good speed response and less torque ripple, it is found suitable in automotive application.
  Keywords: twelve switch neutral point clamped inverter; multilevel inverter; space vector modulation; total harmonic distortion; AVR microcontroller.
  
  
• Optimisation of braking energy recovery for rear-drive electric vehicles based on fuzzy control  
  by Li Ye, Zhihao Wu, Wenjing Lv 
  Abstract: To further improve the use rate of braking energy recovery and ensure the stability and safety of braking, this paper comprehensively considers the braking dynamics and regenerative braking structure to optimise the control strategy of braking energy recovery. Fuzzy controller, regenerative braking module, and regenerative braking model of electric vehicle are established by Matlab/Simulink and AMEsim, respectively. The fuzzy controller is designed, and the classic series and parallel control strategy is compared with the braking control strategy based on fuzzy control. Research shows that compared with the traditional series and parallel control strategies, the fuzzy control strategy has a better effect on braking energy recovery. The braking conversion rate is increased by 11.9% and 21.2%, and the braking recovery rate is increased by 2.78% and 4.96%, respectively, with a better recovery capacity of braking energy, and can increase the vehicle driving range and improve the vehicle energy use.
  Keywords: electric vehicle; braking energy recovery; fuzzy control; joint simulation.
  DOI: 10.1504/IJVD.2022.10053242
   
  
• Power-on downshift analysis of a seamless two-speed transmission in electric vehicles  
  by Yang Tian, Boyu Tong, Shilei Zhou, Xiaoyu Ji, Xiaorun Tian, Yahui Zhang, Xun Shen, Guilin Wen 
  Abstract: Electric vehicles (EVs) have many advantages against internal combustion engine (ICE) vehicles and have caught the interest of academics and industry together. The two-speed transmission can improve the overall power and expand the mileage of EVs. In this paper, the power-on downshifting tactics of a seamless two-speed transmission are studied. The seamless two-speed transmission is controlled by a one-way clutch and two brakes for shifting gears. To achieve smooth downshifting process, three control strategies are proposed to study the process of the power-on downshifts. Firstly, the mathematical model of the two-speed transmission for electric vehicles is constructed. Then, three downshift strategies are analysed based on a simplified mathematical model. In this study, the difference of vehicle jerk and friction work in different downshifting strategies is explored by keeping the gearshift time unchanged. After analysing the simulation results, the advantages and disadvantages of different gear downshift strategies are clearly summarised.
  Keywords: electric vehicles; two-speed seamless transmission; downshift control strategies; gearshift quality.
  
  
• A comprehensive study for range-extended electric vehicles from the perspectives of the technology, policy, market, and design concept  
  by Fuquan Zhao, Xinglong Liu, Han Hao, Zongwei Liu 
  Abstract: This paper investigates the technical characteristics, policy positioning, and market trend of range-extended electric vehicles (REEVs) and explores the design concept for REEVs. The results show that REEV is the transition solution in the near to medium term to address the consumers range anxiety about battery electric vehicles (BEVs) and meet the dual credit policy for automotive manufacturers. The market penetration of REEVs will increase due to the subsidies withdrawn and battery raw material prices rising until the battery cost drops to a predictable value. The technical configuration of that engine generating electricity to drive the traction motor directly is the technology development direction for REEVs. The technical solutions of REEVs (component parameters) should be selected according to the vehicle segment and market positioning. To maximize the technical advantages of REEV, it should develop dedicated hybrid engines (DHEs) and optimize the energy management strategy for the specific REEV.
  Keywords: range-extended electric vehicle; technical characteristic; policy positioning; market trend; design concept.
  DOI: 10.1504/IJVD.2023.10055543
   
  
• Stability control for distributed drive electric vehicles confronting differences of phase angle between coaxial motors  
  by Yueying Zhu, Weiyan Wei, Chao Xing, Chengcong Zhen, Huyi Song 
  Abstract: To enhance the straight driving stability of a distributed drive electric vehicle (DDEV) actuated by switched reluctance motors (SRMs), a stability control strategy for the vehicle confronting wheel torque discordance caused by differences of phase angle (DPA) coming from motors on two sides is proposed in this paper. The potential influence of the DPA between coaxial motors on vehicle stability is studied based on the established SRM and DDEV dynamics models. Then, the stability control strategy eliminating DPA is proposed by introducing feedback control for vehicle yaw rate, after DPA calculator and control selector are designed. Finally, the stability control system including DPA control system, vehicle dynamics model, driver model, and SRM drive system is founded and carried out under conditions of two-wheel and four-wheel driving modes. The co-simulation results prove the stability control strategy can effectively eliminate DPA such that the straight driving stability is greatly enhanced in both driving modes.
  Keywords: electric vehicles; torque coordination control; differences of phase angle.
  
  
• Research on multi-objective intelligent shifting schedule of electric vehicle AMT considering ride comfort and economy  
  by Donghui Lv, Bo Zhu, Lin Yuan, Zhidong Liu, Xue Bai 
  Abstract: Considering the current shifting strategy of multi-speed AMT separates the steady-state shifting from the transient shifting process in the pure electric vehicle, it is difficult to a comprehensive improvement of shifting quality, dynamic performance, and driving economy. In this paper, taking advantage of the artificial intelligence technology, a Fuzzy Neural Network (FNN) based T-S model is established via obtaining the training data from skilled drivers' experience and expert knowledge. A two-speed AMT pure electric vehicle model is used to investigate the fuzzy shifting strategy performance. According to the co-simulation results of AMESim and SIMULINK, the average jerk of 10.006 is recorded, compared to the value of 16.472 based on an ordinary shifting schedule. The results show that FNN-based schedule fully reflects drivers shifting intensions in pursuing shifting smoothness, at the same time, improving vehicle dynamic performance with negligible economic performance loss.
  Keywords: two-speed AMT; gear shift smoothness; fuzzy neural network; intelligent shifting schedule.
  
  
• Influence of critical turning speed of pure electric mining dump truck on rollover and sideslip  
  by Weiwei Yang, Wenming Zhang, Nong Zhang 
  Abstract: Because vehicles often violate the speed limit on curved roads, rollover and sideslip accidents occur on curved roads frequently. Designing the speed limit value of the curve road is an important measure to improve the driving safety of vehicles on curved roads. Considering how sharply the mine road turns, the safe driving speed of vehicles on curved roads is proposed in this paper based on the critical state of vehicle rollover and sideslip. Through the dynamic analysis of the vehicle curve driving, the relationship between the maximum speed limit of the curve and the parameters such as the lateral slope, the lateral force coefficient, and the turning radius under different curve radii is discussed to avoid lateral sideslip when the vehicle is driving on the curve. The curve road's speed limitation under each parameter's influence is obtained by changing the relevant parameters.
  Keywords: pure electric mining dump truck; curved road section; steering speed; driving stability; uphill and downhill.
  
  
• Four-mass-PWA model-based robust H-infinity coordinated control strategy for MTP of DM-PHEV  
  by Cong Liang, Xing Xu, Feng Wang, Shaohua Wang 
  Abstract: The multiple actuators of plug-in hybrid electric vehicles (PHEVs) participating in mode transition process (MTP), including engine start, clutch slipping and engagement. However, the engine start and clutch slip process involved in the MTP may lead to complex dynamics. In this paper, a thirteen degree of freedoms dynamic model of dual motor plug-in hybrid electric vehicle (DM-PHEV) is built and according to the results of sensitivity analysis, the model can be simplified as a four-mass model which is more friendly to the design of controller. Furthermore, the non-linearities caused by the engagement of clutch are handled with a piecewise affine (PWA) modelling method, and the vibration attenuation is realized by designing a PWA model-based robust H1 coordinated control strategy. Simulation and hardware-in-the-loop (HiL) test demonstrate that compared with the controller based on conventional two-mass model, the vehicle jerk controlled by proposed controller based on four-mass model can reduce the vehicle by 89.38%.
  Keywords: mode transition; sensitivity analysis; four-mass model; piecewise affine; coordinated controller.
  
  
• Internal temperature prediction of solar car based on IAGA-BP neural network  
  by Tingting Zhang 
  Abstract: The instability of the interior temperature of the solar car during winter heating leads to a lot of unnecessary energy consumption. This paper proposes to predict internal temperature based on the improved adaptive genetic algorithm-back propagation neural network (IAGA-BP). Firstly, the traditional adaptive genetic algorithms crossover and mutation probability are improved to get the improved adaptive genetic algorithm. The improved genetic algorithm is used to optimise the BP neural network to overcome the disadvantages of the BP network algorithm, such as easy falling into the local extremum, low learning efficiency, and slow convergence speed. Then, an internal temperature prediction model based on IAGA-BP neural network is established. Finally, the results of IAGA-BP are compared with results based on the particle swarm optimisation-back propagation neural network model (PSO-BP). The experimental results show that the mean absolute and square errors of IAGA-BP temperature prediction are 0.2810 and 0.1070. However, the mean absolute and square errors of PSO-BP temperature prediction are 1.6087 and 0.5191. The IAGA-BP network has better prediction accuracy than the PSO-BP network. Therefore, IAGA-BP neural network temperature prediction model can reasonably predict the car temperature to achieve the purpose of energy-saving.
  Keywords: new energy vehicle; economic and environmental protection; adaptive genetic algorithm; BP neural network; temperature prediction.
  
  
• Multiobjective optimisation of an energy-efficient magnet track-eddy current composite brake for high-speed trains  
  by Benzhen Guo, Desheng Li, Bin Wang, Zequn Li, Tong Zhao 
  Abstract: An energy-efficient magnetic track-eddy current composite brake (MT-ECB) for high-speed trains is proposed in this study. It has two modes: the magnetic track working mode and eddy current working mode. First, the structure and operating principle of the composite brake are introduced. Subsequently, a multiobjective optimisation process of the major structural parameters was conducted with the consumed PM volume, electrical excitation, and magnetic attraction force as optimisation objectives. A kriging surrogate model for the MT-ECB was established. Optimal Pareto front solutions were acquired by adopting the non-dominated sorting genetic algorithm II (NSGA-II), and the optimisation results were verified via a 3D finite element simulation. After the optimisation of the MT-ECB, the consumed PM volume decreased by 18%, and the power consumption was reduced by 21%. The electric energy consumed per 1000 N of the eddy current braking force decreased from 4 kW to approximately 2 kW, resulting in a remarkable energy-efficient effect.
  Keywords: magnetic track; eddy current; brake; high-speed train; multiobjective optimization.
  
  
• Collaborative optimisation of lane change decision and trajectory planning based on double-layer deep reinforcement learning  
  by Yang Shen, Jiankun Peng, Cheng Kang, Siyu Zhang, Fengyan Yi, Jinming Wan, Xingmao Wang 
  Abstract: In this paper, the lane change process of a vehicle is divided into two stages: lane change decision and lane change movement. The paper proposes the depth of a two-layer reinforcement learning architecture, the use of upper structure DQN exchange to control decision, and send the lane changing information to the lower DDPG of vehicle trajectory control. When the lane changing process is finished, the collaborative optimisation of DQN is completed through the feedback of vehicle position information before and after lane change. The collaborative optimisation of DQN is then complete. The results show that the proposed two-layer deep reinforcement learning architecture can increase the average speed of the agent vehicle by 2-5% and reduce the average lateral speed and lateral acceleration by 12.5% and 12.2%, respectively, in the lane-changing process. Compared with no collaborative optimisation, the optimal lane change timing of the effectively co-optimised two-layer architecture is 34.64%.
  Keywords: deep reinforcement learning; lane change decision; lane change action; collaborative optimization.
  
  
• Planning and control of autonomous driving in lane-change manoeuvre based on MPC: a framework and design principles  
  by Zhiwei Zhao, Kang Yuan, Jiatong Du, Yulei Wang, Yanjun Huang, Hong Chen 
  Abstract: To realise a safe trajectory planning and control performance based on Model Predictive Control (MPC) and investigate the effects of MPC horizons on the system performance, this paper proposes a longitudinal-lateral decoupled trajectory planning method with a safety guarantee principle, and develops an MPC tracking controller using the vehicle dynamics model. Besides, the effects of the planning horizon and the control horizon on lane-changing performance are investigated. Finally, a comprehensive evaluation indicator considering safety, comfort, and efficiency is proposed to provide recommended prediction horizons for different styles of driving behaviour. Several cases are studied and the results validate the effectiveness of the proposed method.
  Keywords: autonomous vehicle; motion planning; model predictive control.
  
  
• Energy consumption optimisation of a battery thermal management system for electric vehicles considering different cooling modes  
  by Yunfeng Hu, Nuo Deng, Xun Gong, Yao Sun, Hong Chen 
  Abstract: For electric vehicles, the energy consumption of the battery thermal management system is one of the key factors affecting driving mileage. To reduce the consumption in the summer, an energy optimization method for battery thermal management systems, in which the liquid-cooling mode and air cooling mode are directly considered actuators, is presented in this paper. First, the influence of the vehicle speed on the air-cooling efficiency is considered in the modelling process of the battery thermal management system. Second, the weighted sum of the energy consumption and temperature tracking error of thermal management system is considered as the optimisation objective function. Third, the high dimensional dynamic programming algorithm is used to solve the nonlinear optimisation problem to obtain the optimal control trajectory. Finally, the effectiveness is verified under different driving conditions.
  Keywords: dynamic programming; electric vehicles; battery thermal management system; energy optimisation; cooling mode; control.
  DOI: 10.1504/IJVD.2022.10054962