A* Algorithm Path Planning based on Geometric Optimization

Efficient path planning in unmanned warehouse systems is challenged by dynamic, densely-obstructed environments, often causing traditional methods to become trapped in local optima. This paper proposes a geometric optimization method merging the A* search algorithm with an APF. Voronoi diagrams and quadtree decomposition are integrated to optimize obstacle representation and enhance heuristic accuracy, while the APF dynamically adjusts forces to ensure smoothness and adaptability. Experimental results show a 37% average reduction in expanded nodes and a 60% increase in planning efficiency compared to traditional A* and Dijkstra algorithms, indicating significant practical potential for unmanned warehouse distribution.