Skip to Main content Skip to Navigation
Journal articles

An Expectations Framework for Domestic Robot Assistants

Abstract : Robots that are supposed to work in everyday environments are confronted with a wide variety of situations. Not all such situations can be taken into account by a programmer when implementing the system. This is why robots often show strange behavior when they encounter situations that their programmer has not expected. We propose a knowledge-based approach to explicitly represent expectations in the robot program. Comparing those expectations to the current situation allows the robot itself to detect unusual situations and react appropriately. Our general framework can incorporate expectations from different knowledge sources and offers a flexible combination of different expectations. We demonstrate the feasibility of the approach in the context of a household robot in simulation and in a real environment. Finally we discuss the adequacy of our proposed solution and open questions for further research.
Complete list of metadata

Cited literature [19 references]  Display  Hide  Download
Contributor : Alexandra Kirsch <>
Submitted on : Monday, January 15, 2018 - 1:26:09 PM
Last modification on : Wednesday, January 29, 2020 - 4:12:04 PM
Long-term archiving on: : Sunday, May 6, 2018 - 11:12:53 PM


Files produced by the author(s)


  • HAL Id : hal-01684304, version 1


Michael Karg, Alexandra Kirsch. An Expectations Framework for Domestic Robot Assistants. Advances in Cognitive Systems, Cognitive Systems Foundation, 2013, pp.77-92. ⟨hal-01684304⟩



Record views


Files downloads