This paper presents a framework that enables a robot to discover various object categories through interaction. The categories are described using action-effect relations, i.e. sensorimotor contingencies rather than more static shape or appearance representation. The framework provides a functionality to classify objects and the resulting categories, associating a class with a specific module. We demonstrate the performance of the framework by studying a pushing behavior in robots, encoding the sensorimotor contingencies and their predictability with Gaussian Processes. We show how entropy-based action selection can improve object classification and how functional categories emerge from the similarities of effects observed among the objects. We also show how a multidimensional action space can be realized by parameterizing pushing using both position and velocity. ; QC 20160422
A sensorimotor learning framework for object categorization
2016-01-01
ISI:000388682400003
Aufsatz (Zeitschrift)
Elektronische Ressource
Englisch
DDC: | 629 |
Learning semantic object parts for object categorization
British Library Online Contents | 2008
|A cortical framework for invariant object categorization and recognition
BASE | 2011
|Gaussian Processes for Object Categorization
British Library Online Contents | 2010
|