Progressive optimization path planning method for intelligent mobile agent in unstructured environment

Due to the presence of cluttered obstacles and irregular road edges in most operation scenarios, this poses significant challenges for the autonomous mobile operation of intelligent agents. To address the challenges of such unstructured scenarios, this study proposes a progressive optimization path planning method for intelligent mobile agents. Firstly, to eliminate the blind search behavior of the traditional RRT algorithm, a dynamic target bias function is introduced to guide the RRT to search as close to the target point as possible, obtaining the initial planned path. Secondly, to improve the comprehensive performance of the path planning, an iterative artificial potential field optimization algorithm is developed to iteratively update the initial planned path to obtain a progressively optimized path. Finally, different scale scenarios are designed to verify the effectiveness of the proposed method. The result shows that the method proposed in this study can significantly improve compared to traditional methods.