A computationally lightweight safe learning algorithm

Dominik Baumann, Krzysztof Kowalczyk, Koen Tiels, Paweł Wachel

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

41 Downloads (Pure)

Abstract

Safety is an essential asset when learning control policies for physical systems, as violating safety constraints during training can lead to expensive hardware damage. In response to this need, the field of safe learning has emerged with algorithms that can provide probabilistic safety guarantees without knowledge of the underlying system dynamics. Those algorithms often rely on Gaussian process inference. Unfortunately, Gaussian process inference scales cubically with the number of data points, limiting applicability to high-dimensional and embedded systems. In this paper, we propose a safe learning algorithm that provides probabilistic safety guarantees but leverages the Nadaraya-Watson estimator instead of Gaussian processes. For the Nadaraya-Watson estimator, we can reach logarithmic scaling with the number of data points. We provide theoretical guarantees for the estimates, embed them into a safe learning algorithm, and show numerical experiments on a simulated seven-degrees-of-freedom robot manipulator.
Original languageEnglish
Title of host publication2023 62nd IEEE Conference on Decision and Control, CDC 2023
PublisherIEEE
Pages1022-1027
Number of pages6
ISBN (Electronic)979-8-3503-0124-3
DOIs
Publication statusPublished - 19 Jan 2024
MoE publication typeA4 Conference publication
EventIEEE Conference on Decision and Control - Marina Bay Sands, Singapore, Singapore
Duration: 13 Dec 202315 Dec 2023
Conference number: 62
https://cdc2023.ieeecss.org/

Publication series

NameProceedings of the IEEE Conference on Decision & Control
ISSN (Electronic)2576-2370

Conference

ConferenceIEEE Conference on Decision and Control
Abbreviated titleCDC
Country/TerritorySingapore
CitySingapore
Period13/12/202315/12/2023
Internet address

Keywords

  • Computer Science - Machine Learning
  • Electrical Engineering and Systems Science - Systems and Control

Fingerprint

Dive into the research topics of 'A computationally lightweight safe learning algorithm'. Together they form a unique fingerprint.

Cite this