We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

eess.SY

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Electrical Engineering and Systems Science > Systems and Control

Title: Towards Data-driven LQR with KoopmanizingFlows

Abstract: We propose a novel framework for learning linear time-invariant (LTI) models for a class of continuous-time non-autonomous nonlinear dynamics based on a representation of Koopman operators. In general, the operator is infinite-dimensional but, crucially, linear. To utilize it for efficient LTI control, we learn a finite representation of the Koopman operator that is linear in controls while concurrently learning meaningful lifting coordinates. For the latter, we rely on KoopmanizingFlows - a diffeomorphism-based representation of Koopman operators. With such a learned model, we can replace the nonlinear infinite-horizon optimal control problem with quadratic costs to that of a linear quadratic regulator (LQR), facilitating efficacious optimal control for nonlinear systems. The prediction and control efficacy of the proposed method is verified on simulation examples.
Comments: This work has been submitted to 6th IFAC Conference on Intelligent Control and Automation Sciences for possible publication. arXiv admin note: text overlap with arXiv:2112.04085
Subjects: Systems and Control (eess.SY); Machine Learning (cs.LG); Dynamical Systems (math.DS); Optimization and Control (math.OC)
Cite as: arXiv:2201.11640 [eess.SY]
  (or arXiv:2201.11640v1 [eess.SY] for this version)

Submission history

From: Petar Bevanda [view email]
[v1] Thu, 27 Jan 2022 17:02:03 GMT (369kb,D)

Link back to: arXiv, form interface, contact.