MPC Control and Path Planning of Omni-Directional Mobile Robot with Potential Field Method

摘要

The navigation of a omnidirectional mobile robot (OMR) by using a model predictive control (MPC) strategy through the path which is obtained by the potential field method is presented. The OMR is controlled with the three independent stepper motors, which makes it possible to go forward in any expected direction with three Mecanum wheels. The main motivation to use MPC is that it has the an advantage in handling state variable constraints and control variable constraints in practical situations. The past and future control inputs combined with present responses are predicted through the kinematics model of OMR and are optimized at regular interval with respect to a performance index. Simulation results show that OMR moves along the desired and predefined trajectory accurately.