Maximally informative interaction learning for scene exploration

Herke Van Hoof, Oliver Kroemer, Heni Ben Amor, Jan Peters

Research output: Chapter in Book/Report/Conference proceedingConference contribution

24 Scopus citations

Abstract

Creating robots that can act autonomously in dynamic, unstructured environments is a major challenge. In such environments, learning to recognize and manipulate novel objects is an important capability. A truly autonomous robot acquires knowledge through interaction with its environment without using heuristics or prior information encoding human domain insights. Static images often provide insufficient information for inferring the relevant properties of the objects in a scene. Hence, a robot needs to explore these objects by interacting with them. However, there may be many exploratory actions possible, and a large portion of these actions may be non-informative. To learn quickly and efficiently, a robot must select actions that are expected to have the most informative outcomes. In the proposed bottom-up approach, the robot achieves this goal by quantifying the expected informativeness of its own actions. We use this approach to segment a scene into its constituent objects as a first step in learning the properties and affordances of objects. Evaluations showed that the proposed information-theoretic approach allows a robot to efficiently infer the composite structure of its environment.

Original languageEnglish (US)
Title of host publication2012 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2012
Pages5152-5158
Number of pages7
DOIs
StatePublished - 2012
Externally publishedYes
Event25th IEEE/RSJ International Conference on Robotics and Intelligent Systems, IROS 2012 - Vilamoura, Algarve, Portugal
Duration: Oct 7 2012Oct 12 2012

Publication series

NameIEEE International Conference on Intelligent Robots and Systems
ISSN (Print)2153-0858
ISSN (Electronic)2153-0866

Other

Other25th IEEE/RSJ International Conference on Robotics and Intelligent Systems, IROS 2012
Country/TerritoryPortugal
CityVilamoura, Algarve
Period10/7/1210/12/12

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Software
  • Computer Vision and Pattern Recognition
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Maximally informative interaction learning for scene exploration'. Together they form a unique fingerprint.

Cite this