Active lane change trajectory planning method for autonomous vehicles in emergency obstacle avoidance situations

In this paper, a new active lane change trajectory planning method for autonomous vehicles is proposed to improve the smoothness of the lane change trajectory and the success rate of obstacle avoidance. Firstly, by integrating sensors such as LiDAR and in vehicle cameras, environmental information can be comprehensively perceived. Secondly, analyze the motion status information of the vehicle and the obstacle vehicle, determine the time distance of collision danger, and plan the obstacle avoidance path based on this. Finally, by fitting the starting and ending points with a fifth degree polynomial curve, combined with jump optimization and Euler Lagrange equation, a smooth and comfortable lane change trajectory for autonomous vehicles is generated. The experimental results show that the curvature value of the method proposed in this paper is stable between 0.20 and 0.25, and the highest obstacle avoidance success rate reaches 98.5%.