On receding-horizon approximation in time-varying optimal control

Abstract

The closed-loop stability and infinite-horizon performance of receding-horizon approximations are studied for non-stationary linear-quadratic regulator (LQR) problems. The approach is based on a lifted reformulation of the optimal control problem, under assumed uniform controllability and observability, leading to a strict contraction property of the corresponding Riccati operator. Leveraging this contraction property, a stabilizing linear time-varying state-feedback approximation of the infinite-horizon optimal control policy is constructed to meet a performance-loss specification. Its synthesis involves only finite preview of the time-varying problem data at each time step, over a sufficiently long prediction horizon.