Title: Radial basis function neural network sliding mode control-based direct yaw moment control for electrically driven vehicles

Authors: Zhaowen Deng; Wenqian Hu; Wei Gao; Yaohua Guo; Qingsong Fan

Addresses: Hubei Key Laboratory of Automotive Power Transmission and Electronic Control, Hubei University of Automotive Technology, Shiyan 442002, Hubei, China; Institute of Automotive Engineers, Hubei University of Automotive Technology, Shiyan, 442002, China ' Hubei Key Laboratory of Automotive Power Transmission and Electronic Control, Hubei University of Automotive Technology, Shiyan 442002, Hubei, China ' Hubei Key Laboratory of Automotive Power Transmission and Electronic Control, Hubei University of Automotive Technology, Shiyan 442002, Hubei, China ' Yutong Bus Co., Ltd., Zhengzhou 450000, China ' Institute of Automotive Engineers, Hubei University of Automotive Technology, Shiyan, 442002, China; State Key Laboratory of Intelligent Manufacturing Equipment and Technology, Huazhong University of Science and Technology, Wuhan 430074, China; School of Key Laboratory of Advanced Manufacturing Technology, Ministry of Education, Guizhou University, Guizhou, 550025, China

Abstract: To improve the lateral stability of distributed electric vehicles, a neural network sliding mode direct yaw moment control (DYC) strategy with hierarchical control was proposed to address the nonlinear dynamic characteristics and poor robustness of the vehicle model under extreme conditions. As well as to control the vehicle to track the expected trajectory, a position-orientation optimal acceleration preview driver model was established based on the optimal acceleration theory of the positional orientation preview. Firstly, a new two-degree-of-freedom (2-DOF) nonlinear control equation of state was derived based on the conventional 2-DOF linear control equation of state. Then, a hierarchical controller was designed in this way, and the radial basis function neural network (RBFNN) sliding mode controller (SMC) was used in the upper layer to calculate the total target additional yaw moment of the vehicle. The lower controller is optimised to minimise tyre adhesion utilisation. It is a torque distribution controller for hub motors that utilises quadratic programming. Finally, simulation tests were performed using CarSim and Simulink. As indicated by the simulation findings, the established controller's better robustness and control accuracy over the SMC and PID controllers improves the vehicle's lateral stability and driving safety.

Keywords: distributed drive electric vehicle; radial basis function neural network; RBFNN; sliding mode control; lateral stability.

DOI: 10.1504/IJVP.2025.145685

International Journal of Vehicle Performance, 2025 Vol.11 No.2, pp.125 - 158

Received: 16 Aug 2024
Accepted: 16 Nov 2024
Published online: 14 Apr 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article