Robust Model Predictive Control for automated trajectory tracking of an Unmanned Ground Vehicle

This work presents a Robust Model Predictive Control (RMPC) approach for trajectory tracking of an Unmanned Ground Vehicle (UGV). In addition to robustness against unknown but bounded disturbances, the controller presented here is also able to deal with constraints on inputs and states due to its formulation as an RMPC. The proposed approach represents an extension of previous Model Predictive Control (MPC) laws based on the concept of constraint restriction (i.e. optimal trajectories via MPC are computed subject to stringent constraints assuming no uncertainty, and a linear local controller ensures that the actual system robustly follows the optimized MPC trajectory). The presented controller carries a low computational complexity overhead, making it attractive for real-time applications. Applying the proposed control approach to the UGV trajectory tracking problem, simulation results demonstrate robust UGV automated trajectory tracking.