On closed-loop stability of model predictive controllers with learning costs

Lukas Beckenbach, Pavel Osinenko, Stefan Streif

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Model predictive controllers are commonly associated with a fixed running and/or terminal cost function. Recently, some possibilities of cost function adaptation inspired by reinforcement learning were investigated. The current study analyzes closed-loop stability of such controllers in a general way. It is shown what constraints on learned running and terminal cost are required for this sake. A particular feature of the suggested control scheme is that, unlike in some common model predictive controllers, an assumed local Lyapunov function does not have to satisfy a decay function not less than the running cost. Relation of the considered control scheme to a baseline model predictive controller and adaptive dynamic programming is discussed. In a case study, it is shown how different cost function adaptation schemes lead to different performance with respect to the infinite-horizon cost.

Original languageEnglish
Title of host publicationEuropean Control Conference 2020, ECC 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages184-189
Number of pages6
ISBN (Electronic)9783907144015
Publication statusPublished - May 2020
Externally publishedYes
Event18th European Control Conference, ECC 2020 - Saint Petersburg, Russian Federation
Duration: 12 May 202015 May 2020

Publication series

NameEuropean Control Conference 2020, ECC 2020

Conference

Conference18th European Control Conference, ECC 2020
Country/TerritoryRussian Federation
CitySaint Petersburg
Period12/05/2015/05/20

Fingerprint

Dive into the research topics of 'On closed-loop stability of model predictive controllers with learning costs'. Together they form a unique fingerprint.

Cite this