Abstract
The existing off-line observer/controller identification (OCID) method for linear systems is newly extended in this paper for off-line/on-line identification of known/unknown highly nonlinear systems, and a new input-constrained active fault-tolerant tracker is developed, based on the identified linear models. The advantages of the proposed extended on-line OCID method for linear/nonlinear systems are briefly described as follows: (i) Implement novel servo-control-oriented off-line OCID methods in observer and controller canonical forms for highly nonlinear systems; (ii) Is able to overcome the discontinuity induced by the singular value decomposition (SVD) that should be carried out at each sampling instant; (iii) It directly realises the identified parameters in the observer/controller canonical forms; this simplifies the identification process; (iv) Can be practically implemented for the on-line control of an unknown nonlinear system which was constituted by an unknown open-loop plant, an existing but unknown controller and/or an unknown observer; and (v) Can be utilised to develop a new active fault-tolerant controller to compensate the immovable existing controller of the practical operating system. Finally, the servo-control-oriented off-line OCID method for the highly nonlinear PUMA 560 manipulator is shown in the illustrative examples to demonstrate the superiority of the proposed method.
| Original language | English |
|---|---|
| Pages (from-to) | 2632-2662 |
| Number of pages | 31 |
| Journal | International Journal of Systems Science |
| Volume | 50 |
| Issue number | 14 |
| DOIs | |
| Publication status | Published - 2019 Oct 26 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Control and Systems Engineering
- Theoretical Computer Science
- Computer Science Applications
Cite this
}
A novel on-line OCID method and its application to input-constrained active fault-tolerant tracker design for unknown nonlinear systems. / Tsai, Jason Sheng Hong; Yu, Tzu Hsien; Su, Te Jen; Guo, Shu Mei; Shieh, Leang San; Canelon, Jose I.
In: International Journal of Systems Science, Vol. 50, No. 14, 26.10.2019, p. 2632-2662.Research output: Contribution to journal › Article
TY - JOUR
T1 - A novel on-line OCID method and its application to input-constrained active fault-tolerant tracker design for unknown nonlinear systems
AU - Tsai, Jason Sheng Hong
AU - Yu, Tzu Hsien
AU - Su, Te Jen
AU - Guo, Shu Mei
AU - Shieh, Leang San
AU - Canelon, Jose I.
PY - 2019/10/26
Y1 - 2019/10/26
N2 - The existing off-line observer/controller identification (OCID) method for linear systems is newly extended in this paper for off-line/on-line identification of known/unknown highly nonlinear systems, and a new input-constrained active fault-tolerant tracker is developed, based on the identified linear models. The advantages of the proposed extended on-line OCID method for linear/nonlinear systems are briefly described as follows: (i) Implement novel servo-control-oriented off-line OCID methods in observer and controller canonical forms for highly nonlinear systems; (ii) Is able to overcome the discontinuity induced by the singular value decomposition (SVD) that should be carried out at each sampling instant; (iii) It directly realises the identified parameters in the observer/controller canonical forms; this simplifies the identification process; (iv) Can be practically implemented for the on-line control of an unknown nonlinear system which was constituted by an unknown open-loop plant, an existing but unknown controller and/or an unknown observer; and (v) Can be utilised to develop a new active fault-tolerant controller to compensate the immovable existing controller of the practical operating system. Finally, the servo-control-oriented off-line OCID method for the highly nonlinear PUMA 560 manipulator is shown in the illustrative examples to demonstrate the superiority of the proposed method.
AB - The existing off-line observer/controller identification (OCID) method for linear systems is newly extended in this paper for off-line/on-line identification of known/unknown highly nonlinear systems, and a new input-constrained active fault-tolerant tracker is developed, based on the identified linear models. The advantages of the proposed extended on-line OCID method for linear/nonlinear systems are briefly described as follows: (i) Implement novel servo-control-oriented off-line OCID methods in observer and controller canonical forms for highly nonlinear systems; (ii) Is able to overcome the discontinuity induced by the singular value decomposition (SVD) that should be carried out at each sampling instant; (iii) It directly realises the identified parameters in the observer/controller canonical forms; this simplifies the identification process; (iv) Can be practically implemented for the on-line control of an unknown nonlinear system which was constituted by an unknown open-loop plant, an existing but unknown controller and/or an unknown observer; and (v) Can be utilised to develop a new active fault-tolerant controller to compensate the immovable existing controller of the practical operating system. Finally, the servo-control-oriented off-line OCID method for the highly nonlinear PUMA 560 manipulator is shown in the illustrative examples to demonstrate the superiority of the proposed method.
UR - http://www.scopus.com/inward/record.url?scp=85074329158&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85074329158&partnerID=8YFLogxK
U2 - 10.1080/00207721.2019.1672117
DO - 10.1080/00207721.2019.1672117
M3 - Article
AN - SCOPUS:85074329158
VL - 50
SP - 2632
EP - 2662
JO - International Journal of Systems Science
JF - International Journal of Systems Science
SN - 0020-7721
IS - 14
ER -