T-S-Fuzzy-Model-Based Output Feedback Steering Control in Autonomous Driving

In this paper, a model predictive controller based on T-S fuzzy model is investigated for the path-tracking of autonomous vehicles. Due to complicated road conditions, it is the case that the vehicle velocity changes during cruising, and moreover, some physical states of the vehicle are unavailable for control. By adopting fuzzy modeling techniques, a T-S fuzzy model of the vehicle lateral dynamics is proposed by considering the varying longitudinal velocities. The robust positively invariant (RPI) set for the state estimation error and observer gain are determined through offline computation, then a fuzzy model predictive controller is derived by solving a Min-max optimization problem. Finally, some validation results with Car-sim/Simulink joint simulation are given to demonstrate the effectiveness of the proposed methods.