Your search for: 'author:("Pomarlan, Mihai")' has returned 2 results. (0.009s) Save search

Authors: Bateman, John A. | Pomarlan, Mihai | Kazhoyan, Gayane

Article Type: Research Article

Abstract: A fundamental issue concerning the treatment of meaning in context is how to deal with the extremely flexible relationship that appears to hold between descriptions , which are taken as the exchangeable bearers of meaning, and the actual contexts which those descriptions are taken to pick out. Such contexts appear always only to be suggested, or constrained, by descriptions of contexts and so relating levels of description, such as linguistic utterances, to actual contexts of use, such as a situated, fully embodied environment in which language users find themselves, remains an unsolved challenge. The present article sets out a framework …and illustrative implementation of an approach to contextualization that combines ontological engineering principles and situated embodied simulations. For concreteness, we illustrate the approach within an already established architecture for situated robotic agents in order to allow implementation and experimentation in a manner that is not generally accessible when considering linguistic analysis alone. The paper then proposes a hybrid, multi-level architecture that contextualizes linguistic utterances by means of embodied simulations, which then serve further as contextualization constraints on semantic interpretation. Show more

Keywords: Embodiment, simulation, linguistic semantics, ontological analysis, formal ontology

DOI: 10.3233/AO-190218

Citation: Applied Ontology, vol. 14, no. 4, pp. 379-413, 2019

Authors: Höffner, Sebastian | Porzel, Robert | Hedblom, Maria M. | Pomarlan, Mihai | Cangalovic, Vanja Sophie | Pfau, Johannes | Bateman, John A. | Malaka, Rainer

Article Type: Research Article

Abstract: Going from natural language directions to fully specified executable plans for household robots involves a challenging variety of reasoning steps. In this paper, a processing pipeline to tackle these steps for natural language directions is proposed and implemented. It uses the ontological Socio-physical Model of Activities (SOMA) as a common interface between its components. The pipeline includes a natural language parser and a module for natural language grounding. Several reasoning steps formulate simulation plans, in which robot actions are guided by data gathered using human computation. As a last step, the pipeline simulates the given natural language direction inside a …virtual environment. The major advantage of employing an overarching ontological framework is that its asserted facts can be stored alongside the semantics of directions, contextual knowledge, and annotated activity models in one central knowledge base. This allows for a unified and efficient knowledge retrieval across all pipeline components, providing flexibility and reasoning capabilities as symbolic knowledge is combined with annotated sub-symbolic models. Show more

Keywords: Natural language understanding, simulation, grounding, everyday activities, SOMA, robotics

DOI: 10.3233/SW-222973

Citation: Semantic Web, vol. 13, no. 5, pp. 895-909, 2022