Motion Planning and Control of an Autonomous Mobile Robot

Abstract

This paper introduces an efficient motion planning method for autonomous mobile robots, which is based on the rapidly exploring random tree (RRT) algorithm. RRT belongs to the class of sampling-based algorithm and is widely used to solve the motion path planning problem of mobile robots. In this context, a motion planning method for a differential driven mobile robot is discussed which includes an improved RRT variant with a closed loop postprocess for the generation of control sequence for the robot. The motion planning subsystem is integrated with the associated controller. The RRT based motion planning subsystem runs an onward simulation using a system model and the controller to determine the state trajectory. It uses the controller in two different ways: 1) To Eliminate the error in the position between reference and actual path 2) To eliminate the error in the orientation. It utilises a PI controller and a pure pursuit controller. Thus, the controller not only facilitates the execution of the motion plan but also ensures reduction in error in the predicted trajectory ensuring acceptable performance.