Robot safety research · standards · engineering
Robot Safety ForumROBOT SAFETY FORUM
한국어English简体中文
← Back to the library

Research paper

GoSafe · Exploration with a learned backup controller

GoSafe expands Bayesian policy optimization beyond an initial safe region by learning conditions from which a backup controller can recover. The abstract reports convergence conditions and hardware experiments.

Reviewed 2026-11-05 · Source and bibliography checked

Bibliographic details and the editorial overview were checked against the original source. Access conditions and editions are identified with the source.
Original title
GoSafe: Globally Optimal Safe Robot Learning
Publisher
IEEE
Authors
Dominik Baumann, Alonso Marco, Matteo Turchetta, Sebastian Trimpe
DOI
10.1109/icra48506.2021.9560738
Conference date
2021-05-30
Source language
en
Edition
2021 IEEE International Conference on Robotics and Automation (ICRA)
Access
Publisher metadata and open author manuscript
Publication status
Published

Source overview

GoSafe expands Bayesian policy optimization beyond an initial safe region by learning conditions from which a backup controller can recover. The abstract reports convergence conditions and hardware experiments.

Use and review

Abstract and bibliographic review only. Guarantees depend on model, uncertainty and constraint assumptions; no ISO13849 safety function or complete installation was validated.

Sources & editions

GoSafe: Globally Optimal Safe Robot Learning ↗

IEEE · 2021 IEEE International Conference on Robotics and Automation (ICRA)

Archivev1abstract/authors verified. IEEE indexed metadata and authorMPIrecord separately identify ICRA2021work, pages4452–4458, indexed2021-10-18; conference starts2021-05-30.