Model Predictive Trajectory Tracking Control of Automated Guided Vehicle in Complex Environments

Autonomous navigation in a real-world industrial environment is in many ways a challenging task. One of the key challenges is rapid collision-free planning and execution of trajectories to reach any target position and orientation with high accuracy, taking into account the limitations of imperfectness of the vehicle. The model prediction-based motion planners have been successfully used in recent years to generate feasible motions for imperfect vehicles. This paper develops and implements a Model Predictive Control (MPC)-based trajectory controller for path tracking problem in narrow corridors. To evaluate the performance of the proposed method, we designed comparative simulations and experiments. We confirm that the proposed MPC-based controller can track the trajectory precisely and smoothly in specific complex environments. In addition, the proposed methodology can also be a suitable solution to other way-point tracking situations for an industrial mobile robot.