• CN: 11-2187/TH
  • ISSN: 0577-6686

Journal of Mechanical Engineering ›› 2026, Vol. 62 ›› Issue (14): 325-337.doi: 10.3901/JME.260482

Previous Articles     Next Articles

Stochastic Tube MPC Based Automated Vehicle Path Tacking Control Considering Terminal Constraint

YONG Haonan, LU Shaobo, XIE Wenke, CUI Tianbei, YANG Fan   

  1. College of Mechanical and Vehicle Engineering, Chongqing University, Chongqing 400044
  • Received:2025-08-01 Revised:2026-01-02 Published:2026-08-29

Abstract: Reliable path tracking function is the key to the implementation of autonomous driving technology, but model uncertainty will inevitably cause path tracking performance to degrade. tube-based model predictive control(TMPC) is an effective solution, but it faces the limitations of being overly conservative and having difficulty coping with large-value disturbances due to the bounded disturbance assumption. To this end, a preview stochastic TMPC(PSTMPC) method considering terminal constraints is proposed to reduce its conservatism and expand its scope of application. This method considers the probability distribution characteristics of uncertain disturbances based on the bounded assumption and introduces probabilistic reachable sets to tighten the nominal system constraints. At the same time, infinite time domain and terminal inequality constraints are combined under extreme conditions to improve the controller's ability to handle large disturbances. In addition, basedt on the preview idea, a strategy that regards the model curvature term as a known control sequence is proposed to make full use of the curvature information and enhance the tracking performance. Numerical calculation results show that the probability reachable set can effectively broaden the feasible domain of the nominal system. Simulation results verify that the PSTMPC method significantly improves path tracking performance; hardware-in-the-loop experimental results further prove the effectiveness of the method.

Key words: autonomous vehicle, path tracking control, model uncertainty, model predictive control, chance constraints

CLC Number: