A Hybrid Algorithm for Dynamic Path Planning of Mobile Robot Using Improved A* and DWA

In recent years, the rapid popularity of indoor mobile robots in a variety of industries has emphasized the importance of effective path planning, which is a key aspect of safe and autonomous navigation. Conventional techniques such as the A* algorithm, despite being widely adopted, reveal deficiencies in areas of generation efficiency and dynamic obstacle handling. This paper proposes a hybrid algorithm that combines the improved A* algorithm with the enhanced Dynamic Window Approach (DWA). A heuristic function is enhanced for better generation, a distance-varying heuristic weight is optimized for the balance between the width-first and depth-first search, and a self-tuning node search scheme reduced the explored nodes efficiently. Additionally, a path optimization method based on the triangle inequality principle is proposed to achieve shorter paths with fewer turning points. The cost function of the DWA algorithm is adjusted by incorporating additional terms to enhance the mobile robot’s path tracking and real-time obstacle avoidance capability. Finally, the proposed approach is validated by using two grid maps with 16 cases. The results reveal considerable time savings of 29.68%, a reduction in path lengths by 3.49%, and a substantial decrease in turning points by 83.26% on a simple map while 8.54, 3.14, 59.59% on a complex map. The proposed hybrid algorithm accomplishes both global and local path planning by generating shorter, and more efficient paths that conform to the kinematic constraints of mobile robots.