Data-Driven Model Predictive Longitudinal Control for Autonomous Vehicles With Unknown Chassis

In this paper, a data-driven model predictive control (MPC) strategy is proposed for longitudinal control of an autonomous vehicle with an unknown chassis. The data-driven MPC strategy is developed to handle multiple constraints and the unknown chassis based on input-output data with measurement noise. A semi-empirical composite control (SECC) algorithm is developed to deal with variable slope roads and motor nonlinearities by a feedforward controller and an empirical model from throttle opening and steady speed. Both recursive feasibility and input-to-state stability are analyzed for the autonomous vehicle based on the data-driven model predictive controller. Experiment results under variable speed and constant speed scenes are presented to demonstrate effectiveness and superiority of the data-driven MPC and SECC strategy.