A Learnable Safety Measure

Steve Heim, Alexander Rohr, Sebastian Trimpe, Alexander Badri-Spröwitz
Proceedings of the Conference on Robot Learning, PMLR 100:627-639, 2020.

Abstract

Failures are challenging for learning to control physical systems since they risk damage, time-consuming resets, and often provide little gradient information. Adding safety constraints to exploration typically requires a lot of prior knowledge and domain expertise. We present a safety measure which implicitly captures how the system dynamics relate to a set of failure states. Not only can this measure be used as a safety function, but also to directly compute the set of safe state-action pairs. Further, we show a model-free approach to learn this measure by active sampling using Gaussian processes. While safety can only be guaranteed after learning the safety measure, we show that failures can already be greatly reduced by using the estimated measure during learning.

Cite this Paper


BibTeX
@InProceedings{pmlr-v100-heim20a, title = {A Learnable Safety Measure}, author = {Heim, Steve and von Rohr, Alexander and Trimpe, Sebastian and Badri-Spr\"owitz, Alexander}, booktitle = {Proceedings of the Conference on Robot Learning}, pages = {627--639}, year = {2020}, editor = {Kaelbling, Leslie Pack and Kragic, Danica and Sugiura, Komei}, volume = {100}, series = {Proceedings of Machine Learning Research}, month = {30 Oct--01 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v100/heim20a/heim20a.pdf}, url = {https://proceedings.mlr.press/v100/heim20a.html}, abstract = {Failures are challenging for learning to control physical systems since they risk damage, time-consuming resets, and often provide little gradient information. Adding safety constraints to exploration typically requires a lot of prior knowledge and domain expertise. We present a safety measure which implicitly captures how the system dynamics relate to a set of failure states. Not only can this measure be used as a safety function, but also to directly compute the set of safe state-action pairs. Further, we show a model-free approach to learn this measure by active sampling using Gaussian processes. While safety can only be guaranteed after learning the safety measure, we show that failures can already be greatly reduced by using the estimated measure during learning.} }
Endnote
%0 Conference Paper %T A Learnable Safety Measure %A Steve Heim %A Alexander Rohr %A Sebastian Trimpe %A Alexander Badri-Spröwitz %B Proceedings of the Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2020 %E Leslie Pack Kaelbling %E Danica Kragic %E Komei Sugiura %F pmlr-v100-heim20a %I PMLR %P 627--639 %U https://proceedings.mlr.press/v100/heim20a.html %V 100 %X Failures are challenging for learning to control physical systems since they risk damage, time-consuming resets, and often provide little gradient information. Adding safety constraints to exploration typically requires a lot of prior knowledge and domain expertise. We present a safety measure which implicitly captures how the system dynamics relate to a set of failure states. Not only can this measure be used as a safety function, but also to directly compute the set of safe state-action pairs. Further, we show a model-free approach to learn this measure by active sampling using Gaussian processes. While safety can only be guaranteed after learning the safety measure, we show that failures can already be greatly reduced by using the estimated measure during learning.
APA
Heim, S., Rohr, A., Trimpe, S. & Badri-Spröwitz, A.. (2020). A Learnable Safety Measure. Proceedings of the Conference on Robot Learning, in Proceedings of Machine Learning Research 100:627-639 Available from https://proceedings.mlr.press/v100/heim20a.html.

Related Material