Affective Knowledge Representation (AKR) for Cooperative Affective Robots (CARS)

Christine L. Lisetti, Demetrios G. Glinos, Robin R. Murphy, and Russ Turdif

In this article, we describe an Affective Knowledge Representation (AKR) scheme to represent emotion schemata to be used in the design a variety of socially intelligent artificial agents. Our approach in this article and in the applications of our AKR scheme, focuses on the notion of ``social expertise'' of socially intelligent agents in terms of their 1) external behavior and 2) internal motivational goal-based abilities. We claim that social expertise will be critical for the success of human-robot interaction in a variety of applications [6] and that affective phenomena such as emotions and personality are essential in terms of social expertise and autonomous behavior. AKR includes a taxonomy of affect, mood, emotion, and personality, as well as a framework for emotional state dynamics for autonomous behavior. AKR model is being applied to design and implement two collaborating robots which exhibit social expertise as specified above. We have also developed a simulator for our collaborative affective robots in order to generate performance metrics for both real and simulated environments, the Cooperative Affective Robot Simulator (CARS) which we currently describe. CARS is a reconfigurable Java-based simulator for rapid prototyping of control algorithms for cooperating affective robotic agents and it is a tool for evaluating the performance of our autonomous cooperative social robots. In particular, the simulator was configured to represent the robotic agents at the AAAI Mobile Robot Competition’s Hors D'Oeuvres, Anyone? event.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.