Approximate Algorithm for Hybrid Model Predictive Control with Time-Varying Reference

Koichi KOBAYASHI  Kunihiko HIRAISHI  Nguyen Van TANG  

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E92-A    No.8    pp.2046-2052
Publication Date: 2009/08/01
Online ISSN: 1745-1337
DOI: 10.1587/transfun.E92.A.2046
Print ISSN: 0916-8508
Type of Manuscript: PAPER
Category: Systems and Control
Keyword: 
offline/online computations,  time-varying reference,  approximate algorithm,  model predictive control,  hybrid systems,  

Full Text: PDF>>
Buy this Article



Summary: 
In this paper, we propose a new approximate algorithm for the model predictive control (MPC) problem with a time-varying reference of hybrid systems. The proposed algorithm consists of an offline computation and an online computation. In the offline computation, candidates of mode sequences are derived. In the online computation, after the mode sequence is uniquely decided among candidates, the finite-time optimal control problem, i.e., the quadratic programming problem, is solved. So by applying the proposed algorithm, the computational amount of the online computation is decreased. First, the MPC problem with a time-varying reference is formulated. Next, the proposed algorithm is explained, and the accuracy of the obtained approximate solution is discussed. Finally, the effectiveness of the proposed method is shown by a numerical example.