Print Email Facebook Twitter Learning Interactively to Resolve Ambiguity in Reference Frame Selection Title Learning Interactively to Resolve Ambiguity in Reference Frame Selection Author Franzese, G. (TU Delft Learning & Autonomous Control) Celemin, Carlos (TU Delft Learning & Autonomous Control) Kober, J. (TU Delft Learning & Autonomous Control) Date 2020 Abstract In Learning from Demonstrations, ambiguities can lead to bad generalization of the learned policy. This paper proposes a framework called Learning Interactively to Resolve Ambiguity (LIRA), that recognizes ambiguous situations, in which more than one action have similar probabilities, avoids a random action selection, and uses the human feedback for solving them. The aim is to improve the user experience, the learning performance and safety. LIRA is tested in the selection of the right goal of Movement Primitives (MP) out of a candidate list if multiple contradictory generalizations of the demonstration(s) are possible. The framework is validated on different pick and place operations on a Emika-Franka Robot. A user study showed a significant reduction on the task load of the user, compared to a system that does not allow interactive resolution of ambiguities. Subject Active LearningHuman Robot InteractionLearning from DemonstrationsUser-friendly Robot Learning To reference this document use: http://resolver.tudelft.nl/uuid:b26e571f-8018-4170-a524-215116cd01eb ISSN 2640-3498 Source Proceedings of Machine Learning Research, 155, 1298-1311 Event 4th Conference on Robot Learning, CoRL 2020, 2020-11-16 → 2020-11-18, Virtual, Online, United States Part of collection Institutional Repository Document type journal article Rights © 2020 G. Franzese, Carlos Celemin, J. Kober Files PDF franzese21a.pdf 10.29 MB Close viewer /islandora/object/uuid:b26e571f-8018-4170-a524-215116cd01eb/datastream/OBJ/view