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: Adaptive Variants of Optimal Feedback Policies

Abstract: This paper presents a control-theoretic framework which stably combines optimal feedback policies with online learning for control of uncertain nonlinear systems. Given unknown parameters within a bounded range, the resulting adaptive control laws guarantee convergence of the closed-loop system to the state of zero cost. The proposed framework is able to employ the certainty equivalence principle when designing optimal policies and value functions by online adjustment of the learning rate - a mechanism needed to guarantee stable learning and control. The approach is demonstrated on the familiar mountain car problem, where it is shown to yield near-optimal behavior despite the presence of parametric uncertainty.
Comments: Major revision. New theoretical results tested on mountain car domain
Subjects: Systems and Control (eess.SY); Robotics (cs.RO)
Cite as: arXiv:2104.02709 [eess.SY]
  (or arXiv:2104.02709v2 [eess.SY] for this version)

Submission history

From: Brett Lopez [view email]
[v1] Tue, 6 Apr 2021 17:58:38 GMT (485kb,D)
[v2] Thu, 30 Dec 2021 16:28:05 GMT (2085kb,D)
[v3] Tue, 12 Apr 2022 17:03:36 GMT (4293kb,D)

Link back to: arXiv, form interface, contact.