Back to Search Start Over

Self-Supervised Online Learning for Safety-Critical Control using Stereo Vision

Authors :
Cosner, Ryan K.
Jimenez Rodriguez, Ivan D.
Molnar, Tamas G.
Ubellacker, Wyatt
Yue, Yisong
Ames, Aaron D.
Bouman, Katherine L.
Cosner, Ryan K.
Jimenez Rodriguez, Ivan D.
Molnar, Tamas G.
Ubellacker, Wyatt
Yue, Yisong
Ames, Aaron D.
Bouman, Katherine L.
Publication Year :
2022

Abstract

With the increasing prevalence of complex vision-based sensing methods for use in obstacle identification and state estimation, characterizing environment-dependent measurement errors has become a difficult and essential part of modern robotics. This paper presents a self-supervised learning approach to safety-critical control. In particular, the uncertainty associated with stereo vision is estimated, and adapted online to new visual environments, wherein this estimate is leveraged in a safety-critical controller in a robust fashion. To this end, we propose an algorithm that exploits the structure of stereo-vision to learn an uncertainty estimate without the need for ground-truth data. We then robustify existing Control Barrier Function-based controllers to provide safety in the presence of this uncertainty estimate. We demonstrate the efficacy of our method on a quadrupedal robot in a variety of environments. When not using our method safety is violated. With offline training alone we observe the robot is safe, but overly-conservative. With our online method the quadruped remains safe and conservatism is reduced.

Details

Database :
OAIster
Notes :
application/pdf, Self-Supervised Online Learning for Safety-Critical Control using Stereo Vision, English
Publication Type :
Electronic Resource
Accession number :
edsoai.on1313845502
Document Type :
Electronic Resource