Practical Probabilistic Trajectory Planning Scheme based on the Rapidly-Exploring Random Trees for Two-Wheeled Mobile Robots

Abstract

The RRT (Rapidly Exploring Random Tree) based planners using probabilistic sampling approaches have been receiving significant attention because of their ability to deal with high-dimensional planning problems efficiently. However, it is still a challenge to generate trajectories for a mobile robot, given the kinematic and dynamic constraints. In this paper, we present an RRT node extension scheme using an asymptotically stable controller for a two-wheeled mobile robot. The proposed algorithm can generate dynamically feasible trajectories. The simulation results show that the proposed scheme can deal with the narrow regions efficiently. The computational time of the simulation results shows that the proposed scheme is twice as fast as the conventional approach.