Kinodynamic Planner Dual-Tree RRT (DT-RRT) for Two-Wheeled Mobile Robots Using the Rapidly Exploring Random Tree

Abstract

We propose a new trajectory generation scheme called dual-tree rapidly exploring random tree (DT-RRT), which is designed on the basis of a rapidly exploring random tree (RRT) method. The DT-RRT is suitable for high-speed navigation of a two-wheeled differential mobile robot. The proposed dual tree is composed of a workspace tree and a state tree. The workspace tree finds near sets in the target workspace without considering robot kinematics. Robot trajectories are generated by the extension of the state tree under the consideration of kinematic and dynamic constraints. The proposed scheme allows for different topological structures between the workspace tree and the state tree. Owing to the different structures between two trees, flexible node extensions can be achieved. As a result, the success rate of the node extension can be increased, while the computational cost can be saved. In order to improve the quality of the trajectory, we propose a reconnect-tree scheme that can modify the generated tree structure. The advantage of the reconnect-tree scheme is that the repropagation of the conventional RRT structure is not required. From simulations, the superior performance of DT-RRT was clarified in terms of computing time and success rate. The experimental result supported high quality of the generated trajectory with fast and smooth motion of the robot.Index Terms-Kinodynamic planning, mobile robots, nonholonomic motion planning, rapidly exploring random tree (RRT).

Optimal Path Planning using RRT* based Approaches: A Survey and Future Directions

“…Moon et al [39] presented a kinodynamic variant of RRT* called Dual-Tree RRT (DT-RRT). DT-RRT manages two trees called state tree and workspace tree.…”

“…It also offers reconnect-tree scheme in contrast to rewire operation of RRT* [7]. The reconnect-tree scheme [39] maintains child nodes with reduced computational cost, in order to guarantee asymptotic optimality. The dual tree scheme of DT-RRT [39] …”

“…The reconnect-tree scheme [39] maintains child nodes with reduced computational cost, in order to guarantee asymptotic optimality. The dual tree scheme of DT-RRT [39] …”

“…Recent RRT* based planners have addressed these issues by exploiting search space using different sampling strategies such as direct sampling [34], goal biased sampling [11,56,58], intelligent sampling [28,35,70], two-stage sampling [40], and disc based sampling [39,42]. Few of them are shown in Fig.…”

Annealing-based guidance of a rescue robot for rescue mission with multi-goal navigation

“…Their non-learning-based neural dynamics methodology is extended to unknown workspace by simultaneous navigation and mapping [5]. Most recently, Moon and Chung [6] proposed a dual-tree rapidly exploring random tree approach for motion planning of a two-wheeled differential mobile robot, which fits in high-speed navigation.…”