Year: 2026
Type: article
Abstract: Path planning methods based on the Rapidly-exploring Random Tree (RRT) algorithm generate paths composed of straight-line segments. This can make it difficult to account for a robot’s turning radius, and the physical constraints of the robot may result in infeasible paths. To address this, converting these straight paths into smooth curves is essential for effective robot path planning. General techniques for this conversion include Bezier-splines and B-Splines. In this study, we propose a path planning method that simplifies RRT-generated paths by removing unnecessary segments and actively avoids collisions with obstacles. In order to ensure that the generated curved path avoids obstacles, we applied the Particle Swarm Optimization (PSO) algorithm to adjust the positions of the B-Spline control points. Simulation results demonstrate that our proposed method successfully generates smooth, collision-free curved paths, significantly enhancing the navigability for robots with turning radius constraints.
Source: Discover Computing
Authors: Sungmin Jo, Jeongju Choi
Institution: Dong-A University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,520
Funder: Dong-A University
OpenAlex ID: W7154730071