|
" Practical Issues in Nonlinear Model Predictive Control: "
Toshiyuki Ohtsuka Affiliation: Department of Systems Innovation, Graduate School of Engineering Science, Osaka University,
| Document Type
|
:
|
BL
|
| Record Number
|
:
|
783518
|
| Doc. No
|
:
|
b603531
|
| Main Entry
|
:
|
Toshiyuki Ohtsuka Affiliation: Department of Systems Innovation, Graduate School of Engineering Science, Osaka University,
|
| Title & Author
|
:
|
Practical Issues in Nonlinear Model Predictive Control: : Real-Time Optimization and Systematic Tuning\ Toshiyuki Ohtsuka Affiliation: Department of Systems Innovation, Graduate School of Engineering Science, Osaka University,
|
| ISBN
|
:
|
9783642010934
|
|
|
:
|
: 9783642010941
|
| Abstract
|
:
|
In this paper, we discuss two important practical issues in nonlinear model predictive control (NMPC): real-time optimization and systematic tuning. First, we present a couple of efficient algorithms based on the assumption that the sampling period is sufficiently short. Real-time algorithms are obtained in a unified manner as initial-value problems of ordinary differential equations (ODEs) for unknown quantities. A brief survey is given on applications of such ODE-type real-time algorithms in mechanical systems. Furthermore, as a first step toward systematic tuning of a performance index, we propose combining feedback linearization with NMPC. The proposed performance index can be tuned with only <i>one</i> parameter to adjust the output response and the magnitude of the control input. The effectiveness of the proposed method is demonstrated in numerical examples.
|
| Added Entry
|
:
|
Kohei Ozaki Affiliation: Department of Mechanical Engineering, Graduate School of Engineering, Osaka University,
|
|
|
:
|
Toshiyuki Ohtsuka Affiliation: Department of Systems Innovation, Graduate School of Engineering Science, Osaka University,
|
| |