Ontological representation of vision-based 3D spatio-temporal context for mobile robot applications

Abstract

In this paper, we propose an ontology-base context model that consists of high level context as well as primitive spatial and temporal context. Moreover reasoning tools are used to find out not only simple contextual information such as object location, movement and distance but also hidden contextual information such as some objects disappeared by moving behind bigger objects. Also we use axiomatic rules for resolving uncertainties which might be caused by the mismatches of 3D SIFT key points. Some practical examples will be provided to show the validities of our proposed ontology-based context model. 1