A* Path Planning Algorithm Optimization via Unscented Kalman Filter in Dynamic Environments

The fast advancement of robot technology in recent years has drawn a lot of attention to path planning algorithms in dynamic situations.The classic A* path planning algorithm faces numerous hurdles because of its limited adaptation in dynamic contexts.These challenges include poor real-time performance, a high risk of planning failure, and problems with the heuristic function's adaptability.This study proposes a method that combines the A* algorithm with the Unscented Kalman Filter (UKF).This method uses UKF for state estimation and obtaining dynamic obstacle predictions, providing prior information for path planning, and ultimately enabling the A* algorithm to reduce the expansion of unnecessary nodes.Simulation experiments show that compared with the original A* algorithm, the improved UKF-A* performs better in planning time and failure rate, reducing planning time by 10%-30% and lowering the failure rate.Preliminary results show that this method has great potential.The optimization approach put forth in this study, which combines path planning algorithms with sensor filtering techniques (UKF), expands the field of conventional path planning research by providing a fresh viewpoint on how to handle path planning difficulties in complex dynamic environments.