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


Current browse context:


Change to browse by:

References & Citations


(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: Safe Model-Based Reinforcement Learning for Systems with Parametric Uncertainties

Abstract: Reinforcement learning has been established over the past decade as an effective tool to find optimal control policies for dynamical systems, with recent focus on approaches that guarantee safety during the learning and/or execution phases. In general, safety guarantees are critical in reinforcement learning when the system is safety-critical and/or task restarts are not practically feasible. In optimal control theory, safety requirements are often expressed in terms of state and/or control constraints. In recent years, reinforcement learning approaches that rely on persistent excitation have been combined with a barrier transformation to learn the optimal control policies under state constraints. To soften the excitation requirements, model-based reinforcement learning methods that rely on exact model knowledge have also been integrated with the barrier transformation framework. The objective of this paper is to develop safe reinforcement learning method for deterministic nonlinear systems, with parametric uncertainties in the model, to learn approximate constrained optimal policies without relying on stringent excitation conditions. To that end, a model-based reinforcement learning technique that utilizes a novel filtered concurrent learning method, along with a barrier transformation, is developed in this paper to realize simultaneous learning of unknown model parameters and approximate optimal state-constrained control policies for safety-critical systems.
Comments: This manuscript has been accepted in Frontiers in Robotics and AI. doi: 10.3389/frobt.2021.733104
Subjects: Systems and Control (eess.SY); Optimization and Control (math.OC)
Cite as: arXiv:2007.12666 [eess.SY]
  (or arXiv:2007.12666v5 [eess.SY] for this version)

Submission history

From: S M Nahid Mahmud [view email]
[v1] Fri, 24 Jul 2020 17:32:18 GMT (1086kb,D)
[v2] Tue, 13 Oct 2020 21:36:56 GMT (740kb,D)
[v3] Tue, 17 Nov 2020 20:20:40 GMT (800kb,D)
[v4] Mon, 1 Feb 2021 06:21:23 GMT (812kb,D)
[v5] Tue, 5 Oct 2021 15:38:29 GMT (1839kb,D)

Link back to: arXiv, form interface, contact.