Real-time Autonomous Vehicle Maneuver Generation Algorithm Based on Hierarchical Planning
WenHao Zhai · 2024
Global path planning, whether based on sampling-based algorithms or heuristic search algorithms, can search for a global path from the start point to the destination in a relatively short time. However, they often struggle to avoid local obstacles and have limited adaptability to local dynamic environments, making it difficult to meet the requirements of obstacle avoidance in unknown environments. On the other hand, the local path planning DWA algorithm can perform online real-time planning and has excellent obstacle avoidance capabilities. However, in large-scale and complex environments, it may fail to obtain the optimal path or even reach the target point if there is a lack of intermediate waypoint guidance. Therefore, to leverage the strengths of both approaches, an instruction generation algorithm based on a hierarchical plan approach is introduced. This algorithm utilizes the global path points as relay points for the DWA algorithm, enabling real-time obstacle avoidance tasks. Finally, the superiority of the fusion algorithm is verified through comparative simulation experiments.