Ming Yao, Xianying Feng, Haigang Deng, Peigang Li, Haiyang Liu, Anning Wang
To further improve the path planning and obstacle avoidance ability of mobile robots, an improved hybrid algorithm is proposed by combining the global path guidance of A-star (A*) algorithm and the real-time collision-free paths selection advantage of the dynamic windows approach (DWA). Firstly, the calculation time of the global path and the number of turning points are reduced by dynamically weighting the heuristic function of the A* algorithm and introducing a bidirectional search structure and a key turning point screening strategy. Secondly, the energy consumption of the drive motors is derived to optimize the score function of the DWA, then the combination weights of the score function are dynamically adjusted through the fuzzy logic control method, which improves the smoothness of the generated paths and the robustness of the algorithm. Finally, the turning points of the global path are utilized as the sub-targets of the improved DWA to reduce the probability of falling into the local optimal solution. Simulations and experiments show that this hybrid algorithm can calculate shorter and smoother paths, and the energy consumption of drive motors is lower. This algorithm is expected to help mobile robots realize the safe and autonomous navigation in applications such as engineering.