Robust Hierarchical Model Predictive Control for Trajectory Tracking with Obstacle Avoidance
Yanchuan Xu, Huarong Zheng, Weimin Wu, Jun Wu
Abstract
A robust hierarchical path planning and trajectory tracking framework is proposed to maintain a collision-free path for autonomous vehicles. For the path-planning, a constrained finite-time optimal problem is solved to generate a feasible and collision-free trajectory considering the vehicle kinematics. For the trajectory-tracking, a motion controller is proposed by solving a constrained model predictive control problem, obtaining the front wheel steering angles. Furthermore, to enhance the robustness of the motion controller against the uneven network-induced time delay and unmodelled lateral vehicle dynamics, an additional error feedback mechanism is introduced in the motion controller. Simulations are conducted when both moving obstacles and static obstacles exist. Simulation results show that the proposed hierarchical control framework can effectively guarantee safe and feasible driving maneuvers.