Longitudinal Tracking Control of Autonomous Vehicle Based on Improved Model Predictive Control

Longitudinal tracking control is an important part of the autonomous vehicles motion control. Aiming at the problem of position tracking error caused by speed delay, an improved model predictive control (MPC) longitudinal high-precision tracking control method is proposed. Firstly, the state equation of vehicle longitudinal control system is established by using vehicle position, velocity and acceleration. Secondly, the objective function considering tracking position error, speed error and driving comfort is established. The upper level controller uses quadratic programming to solve the optimal desired acceleration, and the lower level controller switches the vehicle drive and brake modes according to the desired acceleration. Finally, MATLAB/Simulink and vehicle simulation software Carsim are used for simulation. The results show that the improved MPC longitudinal tracking controller can achieve high-precision position and speed tracking.