Robust Tube-MPC Based Steering and Braking Control for Path Tracking at High-Speed Driving

Abstract

This paper proposes a steering and braking control strategy for stable path following at high speed in a robust model predictive control (MPC) framework. In severe driving situations, model uncertainties can cause control errors and lead to control failures. This paper introduces a method to minimize uncertainty and a controller that is robust against uncertainty. To minimize uncertainty, the proposed multi-point linearization reduces linearization error. Moreover, the used model minimizes the prediction errors due to the delays and lags of the actuators. The proposed controller, based on tube MPC, ensures the satisfaction of road friction limits for uncertain systems. Conditions to satisfy the feasibility of the MPC are proposed and proven. The proposed controller generates steering and braking inputs that ensure that none of the tires exceeds its friction limit. Simulation and experimental results in various test scenarios show the effectiveness of the proposed approach.