Everybody Needs Somebody Sometimes: Validation of Adaptive Recovery in Robotic Space Operations

This letter assesses an adaptive approach to fault recovery in autonomous robotic space operations, which uses indicators of opportunity, such as physiological state measurements and observations of past human assistant performance, to inform future selections. We validated our reinforcement learnin...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE robotics and automation letters Ročník 4; číslo 2; s. 1216 - 1223
Hlavní autoři:	McGuire, Steve, Furlong, P. Michael, Fong, Terry, Heckman, Christoffer, Szafir, Daniel, Julier, Simon J., Ahmed, Nisar
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	2230 Support IEEE 01.04.2019 Institute of Electrical and Electronics Engineers The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Adaptation models Algorithms Autonomy Computer simulation Cybernetics, Artificial Intelligence and Robotics Human performance Human-centered robotics learning and adaptive systems Machine learning Monitoring NASA Recovery Resource management Robotics Robots Space missions space robotics and automation Task analysis Learning And Adaptive Systems Task Analysis Human-Centered Robotics Space Robotics And Automation Nasa Adaptation Models Robots Resource Management Space Missions Monitoring
ISSN:	2377-3766, 2377-3766
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	This letter assesses an adaptive approach to fault recovery in autonomous robotic space operations, which uses indicators of opportunity, such as physiological state measurements and observations of past human assistant performance, to inform future selections. We validated our reinforcement learning approach using data we collected from humans executing simulated mission scenarios. We present a method of structuring human-factors experiments that permits collection of relevant indicator of opportunity and assigned assistance task performance data, as well as evaluation of our adaptive approach, without requiring large numbers of test subjects. Application of our reinforcement learning algorithm to our experimental data shows that our adaptive assistant selection approach can achieve lower cumulative regret compared to existing nonadaptive baseline approaches when using real human data. Our work has applications beyond space robotics to any application where autonomy failures may occur that require external intervention.
Bibliografie:	2230 Support 2230 NIHMS1522439 ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
ISSN:	2377-3766 2377-3766
DOI:	10.1109/LRA.2019.2894381