Artificial Emotions as Dynamic Modulators of Individual and Group Behavior in Multi-robot System

Abstract

We propose a model for adaptation and implicit coordination in multi-robot systems based on the definition of artificial emotions, which play two main roles: modulators of individual robot behavior, and means of communication among different robots for system-level social coordination. We model emotions as compressed representations of a robot's internal state that are subject to a dynamics influenced by internal and external conditions; they can be efficiently exposed to nearby robots, allowing to achieve local, group-level behavior adaptation and modulation, with minimal computational and bandwidth requirements.