Output-Lifted Learning Model Predictive Control

Abstract We propose a computationally efficient Learning Model Predictive Control (LMPC) scheme for constrained optimal control of class nonlinear systems where the state and input can be reconstructed using lifted outputs. For considered systems, we show how to use historical trajectory data collected during iterative tasks construct convex value function approximation along with safe set in space virtual These constructions are iteratively updated used synthesize predictive policies. that proposed strategy guarantees recursive constraint satisfaction, asymptotic stability, non-decreasing closed-loop performance at each policy update. Finally, simulation results demonstrate effectiveness on kinematic unicycle.