Using Conceptual Spaces for Object Recognition in Multi-agent Systems

Autor:	João Mário Lopes Brezolin, Sandro Rama Fiorini, Márcia de Borba Campos, Rafael H. Bordini
Rok vydání:	2015
Předmět:	Computer science Process (engineering) Visually impaired business.industry Multi-agent system Cognitive neuroscience of visual object recognition Development (topology) Human–computer interaction Physical space Object model Computer vision Artificial intelligence Agent architecture business
Zdroj:	PRIMA 2015: Principles and Practice of Multi-Agent Systems ISBN: 9783319255231 PRIMA
DOI:	10.1007/978-3-319-25524-8_53
Popis:	The development of a system to support the mobility of blind or visually impaired users requires the development of an agent capable of performing the recognition of perceived objects as well as determining their location within the physical space. Taking into account the principle that conceptual representations can improve the object recognition process, the work presented here proposes the integration of a conceptual-space level into the BDI agent architecture. Such integration is developed on top of the resources of the Jason agent platform and the CSML API for conceptual-space models. In this paper, we present a practical example showing how to integrate conceptual representations into the agent reasoning cycle.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::acf390983463b82045894bd8d23fb8b2 https://doi.org/10.1007/978-3-319-25524-8_53 Zobrazit plný text záznamu