Position and Orientation Control on Three-step by a State Machine for an Omnidirectional Mobile Robot

Abstract

In the position and orientation control (pose) of a mobile robot, a trajectory is created, and the robot starts from its initial position and ends in a desired position and orientation. Most state-of-the-art solutions employ control laws based on classic Proportional-Integral-Derivative controllers and/or variants, fuzzy controllers, etc., which generate trajectories that can describe sharp curves or undesirable behaviors (for example, going backward). These curves are drastically dependent on whether the desired orientation is in front of the robot's initial pose. Therefore, a three-step control strategy for pose control of a mobile robot is presented. The first task is an orientation control which drives the robot to the desired position, followed by a position control (which can be performed over a single direction), and finally, another orientation control that drives the robot to the desired orientation.