GA-DWA and A *-APF Path Planning Algorithms for Wheel-legged Robots

Path planning is crucial for autonomous mobility and task execution, but there is currently a lack of research focused on wheel-legged robots. Two fusion algorithms are proposed for path planning of wheel-legged hybrid robots in this paper. Based on the motion characteristics of wheel-legged robots, step and slope obstacle grids are designed, and their penalty rules are formulated to establish the map model. As the global path planning algorithm, the classic genetic algorithm is optimized through improvements to the fitness function and genetic operations, while the A * algorithm is optimized by adopting a new heuristic function and incorporating an obstacle cost function. Subsequently, the optimized genetic algorithm and A * algorithm are respectively integrated with local path planning algorithms DWA and APF, resulting in the development of the GA-DWA and A *-APF fusion algorithms. Simulations are conducted to validate the proposed algorithms. The results show that both the GA-DWA algorithm and the A *-APF algorithm can effectively enable the wheel-legged robot to avoid dynamic obstacles and reach the target point. Furthermore, the GA-DWA algorithm exhibit good global optimality, while the A *-APF algorithm performs better in time efficiency.