Motion path planning method based on map reference points
Wenbin Yu,Cheng Fan,Chengjun Zhang,Yadang Chen,Yangsong Li,Yifan Zhang,Na Yin,Xiaolin Cen
DOI: https://doi.org/10.1007/s11042-024-20198-8
IF: 2.577
2024-10-12
Multimedia Tools and Applications
Abstract:In recent years, significant progress has been made in the development of the Rapidly-exploring Random Trees (RRT) algorithm, a probabilistic sampling-based search approach. This algorithm effectively addresses complex challenges in path planning within high-dimensional spaces characterized by intricate constraints. However, the RRT algorithm faces certain limitations, including sluggish computational performance in handling intricate calculations, susceptibility to becoming trapped in confined path regions, and difficulties in generating randomized paths across a given map. To overcome these issues, this study introduces a motion path planning method based on map reference points. Initially, the algorithm selects strategic map reference points, followed by the utilization of an enhanced RRT algorithm for fundamental path selection. Real-world scenario data is collected to empirically evaluate the algorithm's performance. The outcomes of the experiments reveal that, within the same experimental setting, the proposed method outperforms conventional motion path planning algorithms by effectively circumventing obstacles. Furthermore, it substantially reduces the resource wastage caused by global random sampling in the RRT algorithm, leading to expedited path planning and improved pathfinding success rates. Additionally, the proposed approach enhances the RRT algorithm's adaptability to some extent, particularly in negotiating narrow passages and intricate environments.
computer science, information systems, theory & methods,engineering, electrical & electronic, software engineering