Tuning, Control and Path Planning of a Spherical Robot using Stochastic Signals

Abstract

Mobile robots have been deployed in civilian applications. From this family of robots, there is one with interesting properties: the spherical robot. Controlling this type of robot is challenging; even if non-holonomic, it could navigate uneven terrains where other wheeled robots would get stuck. Given that their components are embedded inside the body of the robot, they can be deployed in hostile environments. Motivated by the latter, the design and implementation of a PID controller for a spherical robot with two degrees of control is presented, namely, lateral and longitudinal motion. Our controller uses stochastical signals to tune and control lateral and longitudinal motion in a robust manner. Our approach has been tested in path planning tasks, moving from one point to another, avoiding obstacles on the way. We use semi-circle trajectories to resolve the path planning and present results in simulations to demonstrate the effectiveness of our approach.