Fusion Adaptive Resonance Theory Networks Used as Episodic Memory for an Autonomous Robot

Abstract

Autonomous service robots must be able to learn from their experiences and adapt to situations encountered in dynamic environments. An episodic memory organizes experiences (e.g., location, specific objects, people, internal states) and can be used to foresee what will occur based on previously experienced situations. In this paper, we present an episodic memory system consisting of a cascade of two Adaptive Resonance Theory (ART) networks, one to categorize spatial events and the other to extract temporal episodes from the robot’s experiences. Artificial emotions are used to dynamically modulate learning and recall of ART networks based on how the robot is able to carry its task. Once an episode is recalled, future events can be predicted and used to influence the robot’s intentions. Validation is done using an autonomous robotic platform that has to deliver objects to people within an office area.