Improved A* and DWA Fusion Algorithm for Mobile Robot Path Planning in Complex Environments

To address the limitations of the A*(A star) algorithm-including low search efficiency and the inability to perform real-time obstacle avoidance in complex environments-this study proposes an Improved A* and DWA (Dynamic Window Approach) fusion algorithm. The algorithm first optimizes the heuristic function to enhance A* search efficiency and applies Bézier curves to smooth the global path, ensuring an optimal trajectory for mobile robots. Subsequently, the improved A* is integrated with the DWA. By refining the robot's initial pose and precomputing dynamic obstacle trajectories based on motion states, the algorithm achieves dynamic obstacle avoidance in complex environments, yielding the optimal path. Simulation results demonstrate that the Improved A* and DWA fusion algorithm rapidly identifies accurate paths in dynamic obstacle scenarios. Compared to the enhanced A* algorithm alone, it reduces average path length by 5.22% and shortens computation time by 18.12%, validating its effectiveness.