Abstract: In order to improve driving safety of intelligent vehicles, an active obstacle avoidance method by lane changing was worked out. Dynamical and kinetic models were built for four-wheel drive independently vehicle. In the matter of path planning, lane changing path was planned by polynomial function, and combining accelerate threshold and obstacle avoidance, a path adjusting method was provided. In the matter of path tracking, a model prediction controller was designed, and the control law was solved by using the thought of rolling optimization and feedback. By simulation, the controller possesses high tracking accuracy, which means active obstacle avoidance has come true. Tracking accuracy, real-time and stability of controller can be adjusted by adjusting parameters.

Key words: intelligent vehicle, obstacle avoidance by lane changing, polynomial function, rolling optimization, model prediction controller