Lane‐change trajectory planning method for connected autonomous vehicles in freeway ramp merging areas

Freeway entrance ramp merging areas are often prone to traffic congestion and accidents, primarily due to the complex interactions between vehicles. The development of connected autonomous vehicles (CAVs) offers significant potential to mitigate these issues. This paper proposes a motion trajectory planning strategy that simultaneously enhances trajectory quality, computational efficiency, and driving comfort. The proposed method comprises three core modules: a hierarchical finite state machine‐based lane‐change decision model that accelerates decision‐making by quantifying risks and opportunities in the S–T diagram; a lateral trajectory planning module that generates lateral paths using an improved grid sampling approach; and a longitudinal velocity planning module that employs a pruned dynamic programming algorithm to speed up the search process. The resulting lateral and longitudinal trajectories are then coupled to form an initial solution, which is further optimized by an iterative linear quadratic regulator under multi‐objective constraints. Working in concert, these modules enable CAVs to perform real‐time, smooth, and safe lane‐change maneuvers. The experimental results indicate that the optimized lane‐change method reduces time cost by approximately 20%, compared to the five‐polynomial trajectory planning method. Moreover, the velocity and acceleration profiles are smoother than those produced by conventional dynamic programming algorithms, ensuring both trajectory smoothness and control precision. These improvements significantly enhance the safety and efficiency of the lane‐change process.