Learning probabilistic decision making by a service robot with generalization of user demonstrations and interactive refinement

Sven R. Schmidt-Rohr* (Corresponding Author), Fabian Romahn, Pascal Meissner, Rainer Jäkel, Rüdiger Dillmann

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingChapter

1 Citation (Scopus)

Abstract

When learning abstract probabilistic decision making models for multi-modal service robots from human demonstrations, alternative courses of events may be missed by human teachers during demonstrations. We present an active model space exploration approach with generalization of observed action effect knowledge leading to interactive requests of new demonstrations to verify generalizations.At first, the robot observes several user demonstrations of interacting humans, including dialog, object poses and human body movement. Discretization and analysis then lead to a symbolic-causal model of a demonstrated task in the form of a preliminary Partially observable Markov decision process. Based on the transition model generated from demonstrations, new hypotheses of unobserved action effects, generalized transitions, can be derived along with a generalization confidence estimate. To validate generalized transitions which have a strong impact on a decision policy, a request generator proposes further demonstrations to human teachers, used in turn to implicitly verify hypotheses.The system has been evaluated on a multi-modal service robot with realistic tasks, including furniture manipulation and execution-time interacting humans.
Original languageEnglish
Title of host publicationFrontiers of Intelligent Autonomous Systems
Subtitle of host publicationStudies in Computational Intelligence
EditorsS Lee, KJ Yoon, J. Lee
PublisherSpringer-Verlag Berlin Heidelberg
Pages309-322
Number of pages14
Volume466
ISBN (Electronic)978-3-642-35485-4
ISBN (Print)9783642354847
DOIs
Publication statusPublished - 2013
Event12th International Conference of Intelligent Autonomous Systems - Jeju
Duration: 26 Jun 201229 Jun 2012

Publication series

NameAdvances in Intelligent Systems and Computing
PublisherSPRINGER-VERLAG BERLIN
Volume194
ISSN (Print)2194-5357

Conference

Conference12th International Conference of Intelligent Autonomous Systems
CityJeju
Period26/06/1229/06/12

Cite this

Schmidt-Rohr, S. R., Romahn, F., Meissner, P., Jäkel, R., & Dillmann, R. (2013). Learning probabilistic decision making by a service robot with generalization of user demonstrations and interactive refinement. In S. Lee, KJ. Yoon, & J. Lee (Eds.), Frontiers of Intelligent Autonomous Systems: Studies in Computational Intelligence (Vol. 466, pp. 309-322). (Advances in Intelligent Systems and Computing; Vol. 194). Springer-Verlag Berlin Heidelberg. https://doi.org/10.1007/978-3-642-35485-4_26