Nonlinear Model Predictive Path-Following for Mecanum-Wheeled Omnidirectional Mobile Robot

摘要

This paper presents a control framework to drive a mecanum-wheeled omnidirectional mobile robot (MWOMR) to reach and follow a given path. To this end, a nonlinear model predictive control (NMPC) for the path following problem is designed such that the robot moves along the path while all the physical constraints are satisfied. For the purpose of implementing the state feedback NMPC, the robot is equipped with three types of sensor to get the robot position. All those information are managed on embedded systems with ROS (Robot Operating System). The control performance is evaluated for several paths and both cases for fixed and varying heading directions are tested. The simulation and experiment results are given to show the effectiveness of the proposed robot control framework.