Robot movement path planning integrating A* algorithm and dynamic window

To address the coupled problem of insufficient global path geometric smoothness and slow local dynamic obstacle avoidance response in mobile robots operating in unstructured and complex environments, a hierarchical path planning model integrating the improved A* algorithm and the fuzzy adaptive dynamic window method is proposed. At the global level, the obstacle potential field factor is introduced to optimize the heuristic search, and a cubic B-spline curve is used for trajectory smoothing to eliminate the geometric discontinuity of the discrete grid path; at the local level, an evaluation function weight adaptive adjustment mechanism based on fuzzy logic is constructed to enable the dynamic window method to reconfigure the heading and obstacle avoidance priority online according to the real-time environmental risk level, avoiding the oscillation or deadlock caused by fixed weights. Simulation results show that the proposed method significantly improves the navigation performance in mixed dynamic scenarios: the cumulative turning cost of the global path is reduced by more than 70%, and the path curvature is achieved continuous smoothness; under dense dynamic interference, the obstacle avoidance success rate exceeds 95%, the average navigation time is shortened by approximately one-third, and the path smoothness cost is reduced by more than 60%. This hierarchical framework significantly alleviates the problem of "suboptimal planning and unstable obstacle avoidance" in traditional algorithms under complex environments through the engineering integration of improvement modules.