ADAPTIVE TUBE-MPC FOR NONLINEAR SYSTEMS WITH UNKNOWN NONLINEARITY SATISFYING LIPSCHITZ CONTINUITY

Abstract

This paper proposes a method to design an adaptive-robust model predictive controller for nonlinear systems in which the unknown nonlinearity is assumed to be Lipschitz continuous. MPC is a model-based control strategy, which means the control performance can be severely affected by the uncertainties inside the system. The key idea is that by using the data collected during the operation, we can establish upper bound and lower bound functions of the unknown nonlinearities, which can provide a computable bound for the unknown nonlinearities. With this information, we can formulate the problem into a TubeMPC, which can be solved by current available methods.