Lane Change Trajectory Planning of Intelligent Vehicle Considering Safety and Comfort
Aiming at improving the safety and passenger comfort of intelligent vehicles during lane change,a secondary screening method based on risk field evaluation trajectory was proposed.Firstly,in Frenet coordinate system,the vehicle motion is decoupled into lateral and longitudinal dimensions,and all lateral d-t curve clusters and longitudinal s-t curve clusters are generated based on quintic polynomials.Secondly,based on the dynamic characteristics of the vehicle and the three-circle collision model,the preliminary screening evaluation function of the trajectory is designed,and the qualified trajectory is selected as the candidate trajectory.Finally,referring to the idea of artificial potential field theory,the concept of risk field in the driving process is introduced,and the total loss function is established to evaluate candidate trajectories according to lane change efficiency,lane change risk value and lateral and longitudinal impact degree for secondary screening,and the optimal trajectories are selected and visualized by coordinate conversion.In order to test the feasibility of the algorithm,a two-lane road environment was built,and multiple scenes with different speeds and accelerations of obstacle vehicles were designed to simulate and verify the curve lane change.The results show that the proposed algorithm can meet the safety and comfort requirements of vehicle lane changing.At the same time,in the normal lane change scenario,the passengers are in a comfortable state in 97.5％of the time during the lane change process,and the goal of balancing safety and comfort can also be achieved in the emergency obstacle avoidance scenario.
万方数据
维普
