Analysis of path planning using RRT* algorithm based on Reeds-Shepp curve for non-holonomic mobile robot

Abstract

In this study, we analyze the optimal path planning method for non-holonomic mobile robot (car-like) while navigating through narrow pathways. Autonomous navigation of a four-wheel non-holonomic mobile robot, limited to forward motion planning, in the narrow pathway environment is challenging. It proves impossible while taking turns. In this kind of environment switch-back (forward and reverse) motion planning is required to succeed the navigation. The path planning is divided in two-phases. The preliminary path is planned by rapidly exploring the random tree algorithm (RRT*) considering the orientation of mobile robot. In the second phase, the planned path is subdivided into smaller parts (local paths) full-filled by continuous curvature. The local path is obtained by Reeds-Shepp curve considering kinematics constraints of mobile robot. The software prototyping tool is Robot Operating System (ROS) from Open Source Robotics Foundation. The path planning model is based on ROS [1] navigation stack.