Research on Path Planning Algorithms for Campus Guide Robots

In the intelligent campus scenario, the interpretive service robot is a key device to improve the quality of campus services, and its navigation performance directly affects the user experience. In view of the adaptability defects of conventional mobile robot navigation technology in dealing with complex dynamic campus environments (such as pedestrian flow, vehicle traffic and other moving obstacles), this study proposes a hybrid path planning architecture that integrates global and local planning, focusing on improving the coordination mechanism of the classic A* algorithm and the dynamic window method. The key technical improvements include: using point cloud dimensionality reduction technology to convert three-dimensional lidar data into a two-dimensional grid model, while retaining environmental details while improving computational efficiency; by introducing path safety evaluation indicators and reconstructing heuristic search functions, the smoothness and safety margin of the global path are enhanced; innovatively designing a dual-mode speed optimization strategy to effectively solve the speed oscillation and local deadlock problems that occur in the traditional dynamic window method during trajectory tracking. The robustness and practicality of the algorithm framework in a dynamic environment were verified by comparative tests on the simulation platform and the real campus scene.