Title :
Stability Analysis of Quadratic MPC With a Discrete Input Alphabet
Author :
Aguilera, Ricardo P. ; Quevedo, D.E.
Author_Institution :
Sch. of Electr. Eng. & Comput. Sci., Univ. of Newcastle, Newcastle, NSW, Australia
Abstract :
We study stability of Model Predictive Control (MPC) with a quadratic cost function for LTI systems with a discrete input alphabet. Since this kind of systems may present a steady-state error, the focus is on practical stability, i.e., ultimate boundedness of solutions. To derive sufficient conditions for practical stability and characterize the ultimately invariant set, we analyze the one-step horizon solution and adapt tools used for convex MPC formulations.
Keywords :
discrete systems; linear systems; predictive control; set theory; stability; LTI systems; convex MPC formulations; discrete input alphabet; linear time-invariant systems; model predictive control; one-step horizon solution; practical stability; quadratic MPC; quadratic cost function; stability analysis; steady-state error; sufficient conditions; ultimate boundedness; Asymptotic stability; Cost function; Hidden Markov models; Mathematical model; Quantization (signal); Signal processing algorithms; Stability analysis; Finite sets; practical stability; predictive control; quantized systems;
Journal_Title :
Automatic Control, IEEE Transactions on
DOI :
10.1109/TAC.2013.2264551
