<p>The trajectory tracking task is an essential aspect of autonomous driving. In addition, the dynamics model is capable of handling high-speed situations and complex road conditions well because it takes into account the force interactions between the vehicle and the ground. Therefore, this paper constructs a model predictive control (MPC) of the vehicle via the model dynamics scheme to address the trajectory tracking problem with the noise considered. Furthermore, an integral enhanced physics-informed neural network (IEPINN) model is constructed to solve the scheme described above. It is worth pointing out that parameters of the proposed scheme are embedded in the IEPINN model, which makes it free of the training time. Then, theoretical analyses and simulations prove the convergence and noise suppression ability of the proposed model. Finally, comparisons with different control schemes validate the effectiveness and feasibility of the proposed scheme.</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

Model predictive control of vehicle with nonlinear dynamics under noisy excitation

  • Shuqiao Wang,
  • Longqi Liu,
  • Long Jin

摘要

The trajectory tracking task is an essential aspect of autonomous driving. In addition, the dynamics model is capable of handling high-speed situations and complex road conditions well because it takes into account the force interactions between the vehicle and the ground. Therefore, this paper constructs a model predictive control (MPC) of the vehicle via the model dynamics scheme to address the trajectory tracking problem with the noise considered. Furthermore, an integral enhanced physics-informed neural network (IEPINN) model is constructed to solve the scheme described above. It is worth pointing out that parameters of the proposed scheme are embedded in the IEPINN model, which makes it free of the training time. Then, theoretical analyses and simulations prove the convergence and noise suppression ability of the proposed model. Finally, comparisons with different control schemes validate the effectiveness and feasibility of the proposed scheme.