Computer Science · Computer Vision and Pattern Recognition
Planning for Mobile Robots Based on Fusion Improved Algorithm and DWA Algorithm
See on the mapAbstract
To address the needs of mobile robot path planning, which requires simultaneously considering path optimality, dynamic obstacle avoidance, and smoothness of the motion path, this paper proposes a path planning algorithm that integrates an improved A* algorithm and the Dynamic Window Approach (DWA) algorithm. Firstly, the algorithm simplifies the calculation steps by using a convex corner point-based search method to address the issue of the traditional A* algorithm having many calculation nodes. It also optimizes the calculation method of the cost function in the A* algorithm, significantly reducing the number of nodes traversed and improving the efficiency of path planning. Finally, based on global path planning, an improved DWA algorithm based on key nodes is introduced to achieve the optimality of global path planning and the ability of local obstacle avoidance. By the use of MATLAB to construct test scenarios such as static environments and dynamic cross environments, the results show that the improved A* algorithm based on convex corner point optimization search increases the calculation speed by 60% compared to the traditional A* algorithm in global path planning, and significantly enhances the robot's obstacle avoidance ability in dynamic environments.