Path Planning Using Improved Hybrid A* Algorithm for Mobile Robots

Abstract

The Hybrid A* algorithm is suitable for Ackermann-type mobile robots with nonholonomic constraints. However, in practical path planning, such issues as long search times, inefficient search efficiency, and excessive planning path changes causing unstable robot motion still exist. To improve the effectiveness of path planning, this paper optimizes the Hybrid A* algorithm by using variable search step size and designing a path evaluation function during node expansion. Path planning experiments are conducted based on simulation and physical experimental platforms. The results show that the improved algorithm reduces the search nodes and planning time, effectively enhancing the algorithm’s search performance and efficiency. In addition, it avoids pose errors caused by reverse driving, ensuring the priority and smoothness of forward driving, and achieves efficient and stable path planning for Ackermann-type mobile robots.