作者: D.E. Quevedo , J.A. De Dona , G.C. Goodwin
关键词: Invariant (mathematics) 、 Link (knot theory) 、 Exponential stability 、 Constraint (information theory) 、 Model predictive control 、 Control theory 、 Mathematics 、 Stability (learning theory) 、 Context (language use) 、 Horizon
摘要: This paper addresses the issue of dynamics receding horizon linear quadratic control with finite input constraint sets. Under special circumstances it is possible to exploit standard tools from model predictive establish asymptotic stability. However, in general, notion stability too strong when used context alphabet control. We are thus led weaker notions, including that ultimate boundedness closed loop trajectories, which interalia related existence positively invariant utilize a recent characterization link parallel literature on analog-to-digital converters. By borrowing ideas this latter area, we able develop several methods for determining sets, hence ensuring boundedness. The efficacy these illustrated by examples.