Scalability of closed-loop system responses in adaptive control schemes

Simon P. Schatz, Tansel Yucelen, Gerardo De La Torre, Benjamin Gruenwald, Eric Johnson, Florian Holzapfel

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Although adaptive control offers a unique way to control dynamical systems without excessive reliance on their mathematical models, it is well known that nonidentical command profiles result in nonidentical closed-loop responses with such controllers, and therefore, it is of practical interest to study theoretical ways to construct predictable closed-loop responses when adaptive controllers are in the loop. To that end, we introduce a new concept called scalability to adaptive control in this paper. In particular, we analyze how to scale learning rates of adaptive weight update laws of various adaptive control schemes with respect to given command profiles to achieve a predictable closed-loop response. Illustrative numerical examples are provided to demonstrate the proposed concept and a possible utilization for the evaluation of adaptive controllers.

Original languageEnglish (US)
Title of host publication2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509035496
DOIs
StatePublished - Jan 31 2017
Event14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016 - Phuket, Thailand
Duration: Nov 13 2016Nov 15 2016

Other

Other14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016
CountryThailand
CityPhuket
Period11/13/1611/15/16

Fingerprint

adaptive control
feedback control
Closed loop systems
Adaptive Control
Closed-loop
Closed-loop System
Scalability
Controller
controllers
commands
Controllers
Learning Rate
profiles
Dynamical system
Update
dynamical systems
Mathematical Model
learning
Numerical Examples
mathematical models

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Control and Optimization
  • Instrumentation
  • Computer Vision and Pattern Recognition

Cite this

Schatz, S. P., Yucelen, T., De La Torre, G., Gruenwald, B., Johnson, E., & Holzapfel, F. (2017). Scalability of closed-loop system responses in adaptive control schemes. In 2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016 [7838623] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICARCV.2016.7838623
Schatz, Simon P. ; Yucelen, Tansel ; De La Torre, Gerardo ; Gruenwald, Benjamin ; Johnson, Eric ; Holzapfel, Florian. / Scalability of closed-loop system responses in adaptive control schemes. 2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016. Institute of Electrical and Electronics Engineers Inc., 2017.
@inproceedings{bbe07a936de748918acff75f94a522c4,
title = "Scalability of closed-loop system responses in adaptive control schemes",
abstract = "Although adaptive control offers a unique way to control dynamical systems without excessive reliance on their mathematical models, it is well known that nonidentical command profiles result in nonidentical closed-loop responses with such controllers, and therefore, it is of practical interest to study theoretical ways to construct predictable closed-loop responses when adaptive controllers are in the loop. To that end, we introduce a new concept called scalability to adaptive control in this paper. In particular, we analyze how to scale learning rates of adaptive weight update laws of various adaptive control schemes with respect to given command profiles to achieve a predictable closed-loop response. Illustrative numerical examples are provided to demonstrate the proposed concept and a possible utilization for the evaluation of adaptive controllers.",
author = "Schatz, {Simon P.} and Tansel Yucelen and {De La Torre}, Gerardo and Benjamin Gruenwald and Eric Johnson and Florian Holzapfel",
year = "2017",
month = "1",
day = "31",
doi = "10.1109/ICARCV.2016.7838623",
language = "English (US)",
booktitle = "2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
address = "United States",

}

Schatz, SP, Yucelen, T, De La Torre, G, Gruenwald, B, Johnson, E & Holzapfel, F 2017, Scalability of closed-loop system responses in adaptive control schemes. in 2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016., 7838623, Institute of Electrical and Electronics Engineers Inc., 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016, Phuket, Thailand, 11/13/16. https://doi.org/10.1109/ICARCV.2016.7838623

Scalability of closed-loop system responses in adaptive control schemes. / Schatz, Simon P.; Yucelen, Tansel; De La Torre, Gerardo; Gruenwald, Benjamin; Johnson, Eric; Holzapfel, Florian.

2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016. Institute of Electrical and Electronics Engineers Inc., 2017. 7838623.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Scalability of closed-loop system responses in adaptive control schemes

AU - Schatz, Simon P.

AU - Yucelen, Tansel

AU - De La Torre, Gerardo

AU - Gruenwald, Benjamin

AU - Johnson, Eric

AU - Holzapfel, Florian

PY - 2017/1/31

Y1 - 2017/1/31

N2 - Although adaptive control offers a unique way to control dynamical systems without excessive reliance on their mathematical models, it is well known that nonidentical command profiles result in nonidentical closed-loop responses with such controllers, and therefore, it is of practical interest to study theoretical ways to construct predictable closed-loop responses when adaptive controllers are in the loop. To that end, we introduce a new concept called scalability to adaptive control in this paper. In particular, we analyze how to scale learning rates of adaptive weight update laws of various adaptive control schemes with respect to given command profiles to achieve a predictable closed-loop response. Illustrative numerical examples are provided to demonstrate the proposed concept and a possible utilization for the evaluation of adaptive controllers.

AB - Although adaptive control offers a unique way to control dynamical systems without excessive reliance on their mathematical models, it is well known that nonidentical command profiles result in nonidentical closed-loop responses with such controllers, and therefore, it is of practical interest to study theoretical ways to construct predictable closed-loop responses when adaptive controllers are in the loop. To that end, we introduce a new concept called scalability to adaptive control in this paper. In particular, we analyze how to scale learning rates of adaptive weight update laws of various adaptive control schemes with respect to given command profiles to achieve a predictable closed-loop response. Illustrative numerical examples are provided to demonstrate the proposed concept and a possible utilization for the evaluation of adaptive controllers.

UR - http://www.scopus.com/inward/record.url?scp=85015182422&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015182422&partnerID=8YFLogxK

U2 - 10.1109/ICARCV.2016.7838623

DO - 10.1109/ICARCV.2016.7838623

M3 - Conference contribution

AN - SCOPUS:85015182422

BT - 2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Schatz SP, Yucelen T, De La Torre G, Gruenwald B, Johnson E, Holzapfel F. Scalability of closed-loop system responses in adaptive control schemes. In 2016 14th International Conference on Control, Automation, Robotics and Vision, ICARCV 2016. Institute of Electrical and Electronics Engineers Inc. 2017. 7838623 https://doi.org/10.1109/ICARCV.2016.7838623