Research on Path Planning Algorithm for Mobile Robots Integrating Improved A* and Dynamic Window Method

The global path planning algorithm is unable to avoid moving obstacles effectively when encountering local sudden states of mobile robots. An improved A * algorithm and dynamic window method fusion algorithm are proposed. Improve the evaluation function of the $A^{*}$ algorithm to enhance the search speed of the fusion algorithm, and introduce the influence and weighting of regional obstacle rate on the evaluation function; When the mobile robot is turning, set a safe distance longer than the robot model to further eliminate unnecessary nodes; Optimizing path folds with Bézier curves, a new path exploration strategy is introduced, and dynamic window method is combined in the local path to achieve emergency obstacle avoidance for random and moving obstacles. The experimental results show that the improved A * algorithm in this paper not only ensures a reasonable global path, reduces exploration nodes, but also reduces path exploration time. By fusing the local dynamic window method, it is possible to successfully avoid random obstacles and locally moving obstacles. The above improved algorithms provide theoretical guidance for practical field applications.