Use this url to cite publication: https://hdl.handle.net/20.500.12259/54136
Options
Model-free incremental learning of the semantics of manipulation actions
Type of publication
Straipsnis Web of Science ir Scopus duomenų bazėje / Article in Web of Science and Scopus database (S1)
Author(s)
Aksoy, Eren Erdal | Georg-August-Universität Göttingen, Germany | DE |
LT | ||
Wörgötter, Florentin | Georg August University Göttingen, Germany | DE |
Title
Model-free incremental learning of the semantics of manipulation actions
Is part of
Robotics and autonomous systems. Amsterdam : Elsevier Science, Vol. 71, spec. iss., 2015
Date Issued
Date Issued |
---|
2015 |
Publisher
Amsterdam : Elsevier Science
Extent
p. 118-133
Field of Science
Abstract
Understanding and learning the semantics of complex manipulation actions are intriguing and non-trivial issues for the development of autonomous robots. In this paper, we present a novel method for an online, incremental learning of the semantics of manipulation actions by observation. Recently, we had introduced the Semantic Event Chains (SECs) as a new generic representation for manipulations, which can be directly computed from a stream of images and is based on the changes in the relationships between objects involved in a manipulation. We here show that the SEC concept can be used to bootstrap the learning of the semantics of manipulation actions without using any prior knowledge about actions or objects. We create a new manipulation action benchmark with 8 different manipulation tasks including in total 120 samples to learn an archetypal SEC model for each manipulation action. We then evaluate the learned SEC models with 20 long and complex chained manipulation sequences including in total 103 manipulation samples. Thereby we put the event chains to a decisive test asking how powerful is action classification when using this framework. We find that we reach up to 100% and 87% average precision and recall values in the validation phase and 99% and 92% in the testing phase. This supports the notion that SECs are a useful tool for classifying manipulation actions in a fully automatic way.
Type of document
type::text::journal::journal article::research article
Language
Anglų / English (en)
Coverage Spatial
Nyderlandai / Netherlands (NL)
Description
eISSN: 1872-793X