GoSafe: Globally Optimal Safe Robot Learning

Dominik Baumann, Alonso Marco, Matteo Turchetta, Sebastian Trimpe

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

Abstract

When learning policies for robotic systems from data, safety is a major concern, as violation of safety constraints may cause hardware damage. SafeOpt is an efficient Bayesian optimization (BO) algorithm that can learn policies while guaranteeing safety with high probability. However, its search space is limited to an initially given safe region. We extend this method by exploring outside the initial safe area while still guaranteeing safety with high probability. This is achieved by learning a set of initial conditions from which we can recover safely using a learned backup controller in case of a potential failure. We derive conditions for guaranteed convergence to the global optimum and validate GoSafe in hardware experiments.
Original languageEnglish
Title of host publicationIEEE International Conference on Robotics and Automation
PublisherIEEE
DOIs
Publication statusPublished - 1 May 2021
MoE publication typeA4 Conference publication
EventIEEE International Conference on Robotics and Automation - Xi'an, China
Duration: 30 May 20215 Jun 2021

Conference

ConferenceIEEE International Conference on Robotics and Automation
Abbreviated titleICRA
Country/TerritoryChina
CityXi'an
Period30/05/202105/06/2021

Keywords

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

Fingerprint

Dive into the research topics of 'GoSafe: Globally Optimal Safe Robot Learning'. Together they form a unique fingerprint.

Cite this