Using conceptual spaces for object recognition in multi-agent systems

Abstract

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.