Empirical Inference

Learning Continuous Grasp Affordances by Sensorimotor Exploration

2010

Book Chapter

ei


We develop means of learning and representing object grasp affordances probabilistically. By grasp affordance, we refer to an entity that is able to assess whether a given relative object-gripper configuration will yield a stable grasp. These affordances are represented with grasp densities, continuous probability density functions defined on the space of 3D positions and orientations. Grasp densities are registered with a visual model of the object they characterize. They are exploited by aligning them to a target object using visual pose estimation. Grasp densities are refined through experience: A robot “plays” with an object by executing grasps drawn randomly for the object’s grasp density. The robot then uses the outcomes of these grasps to build a richer density through an importance sampling mechanism. Initial grasp densities, called hypothesis densities, are bootstrapped from grasps collected using a motion capture system, or from grasps generated from the visual model of the object. Refined densities, called empirical densities, represent affordances that have been confirmed through physical experience. The applicability of our method is demonstrated by producing empirical densities for two object with a real robot and its 3-finger hand. Hypothesis densities are created from visual cues and human demonstration.

Author(s): Detry, R. and Baseski, E. and Popovic, M. and Touati, Y. and Krüger, N. and Kroemer, O. and Peters, J. and Piater, J.
Book Title: From Motor Learning to Interaction Learning in Robots
Pages: 451-465
Year: 2010
Month: January
Day: 0

Series: Studies in Computational Intelligence ; 264
Editors: Sigaud, O. and Peters, J.
Publisher: Springer

Department(s): Empirical Inference
Bibtex Type: Book Chapter (inbook)

Address: Berlin, Germany
DOI: 10.1007/978-3-642-05181-4_19
ISBN: 978-3-642-05181-4
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik

Links: PDF
Web

BibTex

@inbook{6621,
  title = {Learning Continuous Grasp Affordances by Sensorimotor Exploration},
  author = {Detry, R. and Baseski, E. and Popovic, M. and Touati, Y. and Kr{\"u}ger, N. and Kroemer, O. and Peters, J. and Piater, J.},
  booktitle = {From Motor Learning to Interaction Learning in Robots},
  pages = {451-465},
  series = {Studies in Computational Intelligence ; 264},
  editors = {Sigaud, O. and Peters, J.},
  publisher = {Springer},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  address = {Berlin, Germany},
  month = jan,
  year = {2010},
  doi = {10.1007/978-3-642-05181-4_19},
  month_numeric = {1}
}