International Journal of Vehicle Performance (6 papers in press)

Regular Issues

• Research on emotion recognition of drivers peripheral physiological signals based on LSTM  
  by Yaqi Fan, Yutong Lin, Xiaolin Cao, Qin Liang, Jing Chen, Wenchao Xu 
  Abstract: The development of automotive industry and the increase in car ownership have led to increasingly severe issues regarding driving safety. Intervention to adjust the drivers emotions is important to improve driving safety. Effective emotion recognition is a significant prerequisite and foundation for this. Physiological signals, generated by the activity of the human autonomic nervous system, reflect a persons physical and mental state objectively, can be used as feature signals for emotion recognition. In this paper, three peripheral physiological signals [namely electrocardiograph (ECG), blood volume pulse (BVP) and respiration (RSP)] and subjective evaluation data on driver emotions collected from preliminary experiments were used to construct a driver emotion recognition model using long short-term memory (LSTM). The results indicate that compared to Support Vector Machines (SVM), the driver emotion recognition model with LSTM as the core algorithm performs better in terms of both recognition speed and accuracy.
  Keywords: emotion recognition; peripheral physiological signals; driver emotions; long short-term memory; LSTM.
  DOI: 10.1504/IJVP.2025.10072428
   
  
• Identification of intrusion obstacles for underground locomotives based on the fusion of LiDAR and wireless positioning technology  
  by Hongbo Wang, Yang Wang, Yang Shen 
  Abstract: Relying solely on a single camera or radar for obstacle identification is limited under the complex underground mine environment. Given the computational limitations of the hardware and the specific adaptation requirements for underground environments, this paper establishes an experimental platform for underground locomotive intrusion obstacle identification. The coordinate systems of different LiDAR are firstly unified. The height threshold method is improved to perform ground segmentation on the raw data. The Euclidean clustering algorithm is improved with the lidar Based on Scan Line Distribution (BSLD) feature to obtain information such as the outline dimensions and central position of obstacle targets. Ultra wide band (UWB) positioning devices are combined with recorded map information to delineate the region of interest, identify and extract intrusion obstacles within these areas. The experimental results show that the system can reliably identify intrusion obstacles, significantly enhancing the pertinency of the obstacle identification system.
  Keywords: mine automatic driving; underground locomotives; LiDAR; UWB; intrusion obstacle identification.
  DOI: 10.1504/IJVP.2025.10072237
   
  
• Investigation of the optimised control strategy of the loader boom lifting system  
  by Yaodong Yang, Weiwei Yang, Wenming Zhang, Nong Zhang 
  Abstract: This paper studied the optimised control strategy to solve the problems of delayed start and end impact of loader boom lifting. Based on the analysis of the basic principle of the boom lifting system, a simulation model of the hydraulic coupling system was established using AMEsim. An optimised control strategy was proposed by analysing the working characteristics, and a comparative analysis was conducted between the control strategies before and after optimisation. The simulation results showed that the proposed optimised control strategy can reduce the delay time of loader lifting startup by 43.32% and improve the stability of startup by 32.6%, respectively, compared to the original strategy. Meanwhile, it was not affected by changes in oil temperature and working device load. Otherwise, automatic buffering of the boom at the lifting end was achieved, which improved the automation level of the loader.
  Keywords: loader; boom lifting; optimal control; start delay; jerk.
  DOI: 10.1504/IJVP.2025.10072242
   
  
• Development of a nonlinear finite element model for estimating static and dynamic seat cushion characteristics  
  by Darshan Vishnu Dorugade, Subhash Rakheja, Yumeng Yao, Paul-Émile Boileau, Zhou Zheng 
  Abstract: Commercial vehicle seat cushion coupled with occupant are exposed to dynamic forces due to road roughness and vibration transmitted through the vehicle cabin. The mechanical visco-elastic properties of the seat cushion are crucial and are needed to be investigated for enhancing its performance and comfort under static and dynamic conditions. This study investigates the responses of a commercial vehicle seat cushion by measuring its force-deflection characteristics under different pre-loads and excitations. Additionally, a nonlinear finite element seat cushion model was developed using LS DYNA solver to simulate its response under various loading conditions. The simulation results revealed that the force-deflection characteristics of the seat cushion exhibited significant variation depending on the preload, magnitude and rate of excitation. The dynamic simulations provided a reasonably accurate representation of the measured characteristics, showing good agreement with experimental data. However, small discrepancies were observed, particularly with increasing frequency under different loading conditions. This study marked a preliminary attempt at dynamic simulation of seat cushions and verified the validation of the developed model via experimental data. Future work may focus on refining the model to reduce discrepancies and explore additional factors influencing seat cushion performance.
  Keywords: seat cushion; static and dynamic seat cushion characteristics; seat cushion dynamic FEA model.
  DOI: 10.1504/IJVP.2025.10072244
   
  
• Distributed drive 4WS intelligent driving vehicle trajectory tracking and lateral stability control  
  by Wei Gao, Yangyang Zhou, Zhaowen Deng, Minxiang Wei 
  Abstract: To improve the accuracy of trajectory tracking and lateral stability of autonomous vehicles in intelligent driving, a distributed drive four-wheel steering (4WS) trajectory tracking and lateral stability control algorithm with a hierarchical architecture is proposed. Firstly, an upper-level 4WS trajectory tracking controller with feed-forward control is designed based on a linear quadratic regulator (LQR) to determine the front and rear wheel steering angles. A PID-based longitudinal driver model is designed to output the required total driving force for vehicle speed tracking. Secondly, a lower-level linear time-varying model predictive control (LTVMPC) direct yaw moment controller is developed based on a two-degree-of-freedom single-track vehicle model. The optimal additional yaw moment is computed by iteratively solving a constrained optimisation problem, and it is distributed using the axle load ratio method. Finally, a co-simulation platform based on CarSim and MATLAB/Simulink is employed to simulate and analyse the controller's performance. The results show that the 4WS+LQR+LTVMPC strategy effectively reduces the lateral offset, heading angle offset, and sideslip angle while improving the trajectory tracking accuracy and lateral stability of the vehicle.
  Keywords: autonomous vehicles; four-wheel steering; 4WS; trajectory tracking; linear time-varying model predictive control; LTVMPC; co-simulation.
  DOI: 10.1504/IJVP.2025.10072075
   
  
• Optimisation of drag coefficient of a car with integrated canards  
  by Zihou Yuan, Xingren Zheng, Yanming Du, Hongwei Zhang 
  Abstract: The objective of this study is to investigate the impact of canards geometric design on the airflow field at the rear of a sedan, with the aim of reducing the drag coefficient of the vehicle. Four key design variables were analysed: canard angle (θ), length (L1), radius (R), and thickness (H). Using these variables, Latin hypercube sampling generated 50 DOE points, which were simulated in ANSYS Fluent® to calculate the drag coefficient (C_D). A comprehensive performance comparison was performed. Ultimately, a backpropagation neural network (BPNN) coupled with a genetic algorithm (GA) was adopted as the optimal method for canard design optimisation. A random forest model identified θ as the most influential factor on C_D. The optimised design achieved a 21% reduction in drag, minimised the rear vortex region, and significantly accelerated the optimisation process.
  Keywords: computational fluid dynamics; CFD; artificial neural network; ANN; genetic algorithm; GA; automotive aerodynamics; aerodynamic characteristics; parameter optimisation.
  DOI: 10.1504/IJVP.2025.10072245