This work 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 humanfactors 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 non-adaptive 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.


    Access

    Download


    Export, share and cite



    Title :

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


    Contributors:
    McGuire, S (author) / Furlong, PM (author) / Fong, T (author) / Hecktnan, C (author) / Szafir, D (author) / Julie, SJ (author) / Ahmed, N (author)

    Publication date :

    2019-04-01


    Remarks:

    IEEE Robotics and Automation Letters , 4 (2) pp. 1216-1223. (2019)


    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English



    Classification :

    DDC:    629



    Spy satellites, somebody could be watching you

    Richardson, D.E. | Tema Archive | 1978


    HOW DO WE EDUCATE SOMEBODY MORALLY?

    Bronikowski, M. / European Federation of Sport Psychology / Charles University in Prague | British Library Conference Proceedings | 1999