Using Dynamic Programming for Path Planning of a Spherical Mobile Robot

Spherical mobile robot has good static and dynamic stability, which allows the robot to face different kinds of obstacles and moving surfaces, but the lack of effective control methods has hindered its application and development. In this paper, we propose a direct approach to path planning of a 2DOFs (Degrees of Freedom) spherical robot based on Bellman’s “Dynamic Programming” (DP). While other path planning schemes rely on pre-planned optimal trajectories and/or feedback control techniques, in DP approach there is no need to design a control system because DP yields the optimal control inputs in closed loop or feedback form i.e. after completing DP table, for every state in the admissible region the optimal control inputs are known and the robot can move toward the final position. This enables the robot to function in semior even nonobservable environments. Results from many simulated experiments show that the proposed approach is capable of adopting an optimal path towards a predefined goal point from any given position/orientation in the admissible region. Keywords— spherical mobile robot; path planning; Dynamic Programming.