Scholarlink

In this paper, a backstepping adaptive neural network (BANNC) active suspension control strategy is proposed to improve vehicle ride comfort. The concept of differential isomorphism is introduced, the state space equations containing body displacement and suspension dynamic travel are constructed, and the obstacle Lyapunov function is utilized to limit the range of suspension dynamic travel. Aiming at the problem that the tire stress is complex and difficult to calculate, the neural network function is used to estimate the nonlinear stress of the tire, and combined with the adaptive control algorithm, so that the control strategy of the system can still maintain good control effect when the tire stress changes with the system state. Finally, the effectiveness of the algorithm in this paper is proved by joint simulation. The BANNC algorithm in this paper shows an improvement of approximately 14% in spring deflection and acceleration compared to the passive suspension, and an improvement of around 22% in the dynamic/static load ratio.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Adaptive Control for Vehicle Active Suspension with Limited Working Space Based on Tire Stress Estimation

  • 会议论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年

查看全文
  • Zhenjun Lin,
  • Shuang Liu,
  • Mingxuan Li,
  • Dingxuan Zhao,
  • Nayu Su,
  • Cong Zhang

摘要

In this paper, a backstepping adaptive neural network (BANNC) active suspension control strategy is proposed to improve vehicle ride comfort. The concept of differential isomorphism is introduced, the state space equations containing body displacement and suspension dynamic travel are constructed, and the obstacle Lyapunov function is utilized to limit the range of suspension dynamic travel. Aiming at the problem that the tire stress is complex and difficult to calculate, the neural network function is used to estimate the nonlinear stress of the tire, and combined with the adaptive control algorithm, so that the control strategy of the system can still maintain good control effect when the tire stress changes with the system state. Finally, the effectiveness of the algorithm in this paper is proved by joint simulation. The BANNC algorithm in this paper shows an improvement of approximately 14% in spring deflection and acceleration compared to the passive suspension, and an improvement of around 22% in the dynamic/static load ratio.