Development of Virtual Simulation Experiment Platform for Lane Changing of Autonomous Vehicle

Aiming at the scarcity of virtual simulation experiment teaching platform for autonomous vehicles, a virtual simulation experiment platform for lane changing of autonomous vehicles with friendly user interface and intuitive simulation is developed. Firstly, the human-machine interface design of the platform was completed based on MATLAB GUI, and different models, parameters and working conditions were set independently; lane change decision-making method, trajectory planning method and tracking control method were selected. Secondly, the lane change model was built based on Simulink/CarSim to set parameters and simulate the process. Finally, the effect of lane changing was analyzed according to the simulation results. The results show that the trajectory of lane change based on polynomial is smooth, and the curvature is continuous. The tracking control method based on model predictive control has the advantages of small error, high precision and good stability. The principle of the virtual simulation experiment platform of autonomous vehicle is clear and the simulation analysis is intuitive, which can better improve the students’ autonomous driving control algorithm development ability and innovation ability.