Abstract

Path planning is an important aspect in robotics, allowing a robot to reach its destination safely and efficiently. The Informed Rapidly-exploring Random Tree Star (Informed RRT*) algorithm has become one of the popular path planning methods due to its efficiency and ability to handle complex workspaces. Since its introduction in 2014, various variants of Informed RRT have been developed to improve its performance. However, there has been no comprehensive study comparing the performance of these variants. This research aims to conduct a comparative study of the performance of several Informed RRT variants in solving path planning problems. The algorithms compared are APF-IRRT, BI-RRT, Informed RRT* + DWA, Informed RRT*-Connect, Informed RRT*-PSO, MI-RRT-Connect, and Informed RRT* with wrapping** procedure. Algorithm performance is evaluated based on computation time and path length metrics. Experiments were conducted on various path planning scenarios with different complexities. The results show that certain Informed RRT variants perform better than others in different metrics. APF-IRRT and Informed RRT-Connect* generally have faster computation time, Informed RRT-PSO* and Informed RRT with wrapping procedure produce smoother paths. This research provides a better understanding of the strengths and weaknesses of Informed RRT variants, enabling the selection of appropriate algorithms for specific path planning applications.