Context Aware Cognitive Architecture for Adaptive Socially Assistive Robots.

Abstract

Socially Assistive Robots need to be aware of their context to be useful. Acceptance and usefulness come from adapting the robot’s behavior to the user and the situation. This paper aims to achieve this enhanced social awareness. It presents a cognitive architecture designed to include the whole context and to be shared by all agents in a smart environment. The architecture is provided with an ontology to facilitate the extension of the robot’s functionalities and the description of its behavior. It can adapt this behavior by toggling the execution of use cases and by modifying the execution of each use case according to the context. To test the proposed cognitive architecture, a set of use cases has been defined following a participatory design involving residents and caregivers from a retirement home. These use cases were incorporated into the architecture and analyzed in a one-month pilot study conducted in the retirement home. Results discussed in this paper show the potential of the solution to increase robot efficiency through adaptability.