Optimal trajectory planning based on bidirectional spline-RRT* for wheeled mobile robot

Abstract

This paper proposes a trajectory planning of a mobile robot using bidirectional-Rapidly-exploring Random Tree star [RRT star] algorithm with spline technique. In the proposed method, the basic bidirectional-RRT star algorithm is combined with the spline technique to generate smooth trajectories, which are important for curved path navigation of a wheeled mobile robot. The bidirectional-RRT star tree structure is extended by using a spline method based on a cubic Ferguson's curve. Trajectories that are been generated using the proposed bidirectional spline-RRT star algorithm satisfies direction constraints approach on both source and target positions. This makes the proposed algorithm remarkably unlike from other trajectory planning algorithms. As a result, the paths produced by the mobile robots are sub-optimal, dynamically and geometrically feasible, and satisfy direction constraints approaches. Simulation results that are performed affirm these bidirectional spline-RRT star algorithm properties and show the validity of the proposed algorithm, implying that it can be efficiently applied to trajectory planning of wheeled mobile robot operating in real-time environments.