Improved A* Algorithm for Intelligent Navigation Path Planning

Abstract

 For path planning in intelligent navigation, traditional navigation maps currently cannot meet the requirements of autonomous navigation and optimal path search in terms of three-dimensional environmental features and accuracy. Therefore, the study combined multiple sensors of LIDAR and depth camera to construct a 3D simulation environment map model and uses optimized A* Algorithm to improve path planning. The cost proportion factor and improved heuristic function were used to optimize the A* Algorithm. Through experimental comparison before and after optimization, the shortest path and time of the A* Algorithm in the 8×8 grid map before optimization are 12.89 and 0.56s, respectively. It has a shortest path and time of 28.76 and 0.28s in a grid map of 16×16, respectively. The improved A* Algorithm has an optimal path and time of 12.26 and 0.34s on an 8×8 grid map, and a shortest path and time of 26.34 and 0.28s on a 16×16 grid map. These experiments confirm that the improved A* Algorithm improves the search range and efficiency of path planning. This proves its superiority for intelligent navigation path planning and provides technical references for environmental map construction and optimal path planning.