Scholarlink

<p>This paper deals with designing a vision-based finite-time (FT) double-surface integral homogeneous Sliding Mode Controller (SMC) for the lateral control of an Autonomous Vehicle (AV) with unknown vehicle dynamics and nonlinearities of tire forces. Conventional FT controllers are highly sensitive to the initial location of the car relative to the desired path. Instability and loss of tracking accuracy is quite common in these controllers if the nonlinearities of the actuators are not handled carefully during the initialisation. Our proposed homogeneous SMC with the error-and-velocity-dependent degree of homogeneity can handle the limitations of the actuators, namely input saturation and rate limits during the initial transients. The controller eliminates the sensitivity of conventional FT controllers to initial transients while maintaining tracking accuracy and improving the smoothness of the control signal. Stability analyses show that the system’s states converge to the origin in finite time. In addition, phase plane studies show faster convergence of trajectories to the origin compared to conventional algorithms. Extensive simulations are carried out on more than 400 different driving scenarios with different initial conditions and locations on the desired path, and comparisons are made with several conventional FT controllers. The results show that the proposed control algorithm is the only one that is able to successfully navigate the vehicle under all driving scenarios without losing stability or deteriorating tracking accuracy.</p>

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

Double-surface homogeneous sliding mode control for self-driving cars

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年8月5日

查看全文
  • Alireza Hosseinnajad,
  • Navid Mohajer,
  • Mohammad Rokonuzzaman

摘要

This paper deals with designing a vision-based finite-time (FT) double-surface integral homogeneous Sliding Mode Controller (SMC) for the lateral control of an Autonomous Vehicle (AV) with unknown vehicle dynamics and nonlinearities of tire forces. Conventional FT controllers are highly sensitive to the initial location of the car relative to the desired path. Instability and loss of tracking accuracy is quite common in these controllers if the nonlinearities of the actuators are not handled carefully during the initialisation. Our proposed homogeneous SMC with the error-and-velocity-dependent degree of homogeneity can handle the limitations of the actuators, namely input saturation and rate limits during the initial transients. The controller eliminates the sensitivity of conventional FT controllers to initial transients while maintaining tracking accuracy and improving the smoothness of the control signal. Stability analyses show that the system’s states converge to the origin in finite time. In addition, phase plane studies show faster convergence of trajectories to the origin compared to conventional algorithms. Extensive simulations are carried out on more than 400 different driving scenarios with different initial conditions and locations on the desired path, and comparisons are made with several conventional FT controllers. The results show that the proposed control algorithm is the only one that is able to successfully navigate the vehicle under all driving scenarios without losing stability or deteriorating tracking accuracy.