Trajectory Tracking Control of Intelligent Vehicle Based on Model Predictive and Genetic Algorithm

Aiming to improve the performance of the model predictive control (MPC) controller of intelligent vehicle trajectory tracking, a method for setting the weight coefficient of MPC controller based on genetic algorithm-based method was proposed. Based on the lateral dynamics model of monorail vehicle, the MPC controller of intelligent vehicle trajectory tracking was designed. The weighting coefficient of lateral deviation was taken as an example to analyze its influence on the performance of MPC controller under the condition of tracking a straight line. Taking the minimum of the response time and the deviation of intelligent vehicle tracking the target trajectory as the goal, the problem of determining the weighting coefficients of MPC controller was transformed into a multi-objective optimization problem, which was solved by the genetic algorithm to obtain the optimized weighting coefficient. MATLAB/Simulink and Carsim software were used to simulate the effect of vehicle tracking the target trajectory before and after weight coefficient optimization under different vehicle speeds, and the response time and tracking deviation were analyzed. The results show that the weight coefficient has a great influence on the response time and tracking deviation of vehicle tracking target trajectory, and the optimization of weighting coefficient can improve the performance of MPC controller, and improve the response speed and accuracy of vehicle tracking target trajectory.