Liu, Di (BYD Auto Industry Company Limited), Dong, Wei (BYD Auto Industry Company Limited), Zhong, Yihe (BYD Auto Industry Company Limited)

Compensation of Electric Power Steering System Based on Neural Network

Scheduled for presentation during the Invited session "Modeling, control and optimization of vehicle and renewable energy system" (FrB4), Friday, November 1, 2024, 11:50−12:10, Room T4

7th IFAC Conference on Engine and Powertrain Control, Simulation and Modeling, Oct 30 - Nov 1, 2024, Dalian, China

This information is tentative and subject to change. Compiled on January 2, 2025

Abstract

Error introduced by Electric Power Steering (EPS) dead zones and delays in Advanced Driver Assistance Systems (ADAS) significantly impacts driving safety and performance. In ADAS applications, vehicles demand heightened precision and reliability. This study addresses EPS system errors by employing a BP neural network for compensation. Initially, EPS system data pertinent to road tests was collected and subjected to correlation analysis to identify variables strongly influencing EPS system output torque. Subsequently, a neural network model of the inverse EPS system, integrating error characteristic information, was trained using BP neural networks. The model was then implemented in Simulink for validation. Comparative analysis was conducted by replaying data to contrast the model developed in this study with those derived from conventional methods. Furthermore, the model was integrated as a compensation module into the ADAS control system and tested in practical road conditions using vehicle-mounted chips. Experimental findings substantiate the effectiveness and superiority of the proposed error compensation method for engineering applications.