A Lane-Changing Trajectory Planning Algorithm Based on Lane-Change Impact Prediction

Current research on lane-changing trajectory planning typically focuses on the benefits of the Subject Vehicle (SV), including factors such as lane-changing efficiency, comfort, and safety. However, this can result in disturbances to the normal driving of the following vehicles. To address this issue, this paper proposes a lane-changing trajectory planning algorithm that takes into account the impact of lane changes. Based on the trajectories of the following vehicles, a gap distance-based evaluation scheme is first proposed to judge whether a vehicle is affected by the lane change of SV. An ordered probit model is then developed to predict the number of vehicles affected, with their efficiency and comfort benefits incorporated into the cost function for SV's lane-changing trajectory planning. Safety evaluation of the planned trajectories is further performed using an improved predictive risk field that can comprehensively quantify the risk posed by the surrounding vehicles. Finally, verification is conducted using the Simulation of Urban Mobility (SUMO) platform, and the results show that the proposed algorithm generates optimal lane-changing trajectories that maximize the overall benefits for multiple relevant vehicles when compared with other existing algorithms.