Neuromorphic Robot Dream

TL;DR

This paper introduces a neurobiologically inspired framework for autonomous robots that alternates between real-time rule-based control during the day and neural network updates during the night, enhancing emotional and behavioral capabilities.

Contribution

It proposes a novel day/night cycle approach for robotic systems, integrating real-time control with neural network updates for more realistic emotional behaviors.

Findings

Effective real-time control during the day phase.

Neural network updates improve emotional responses.

Framework supports autonomous decision-making.

Abstract

In this paper we present the next step in our approach to neurobiologically plausible implementation of emotional reactions and behaviors for real-time autonomous robotic systems. The working metaphor we use is the "day" and the "night" phases of mammalian life. During the "day phase" a robotic system stores the inbound information and is controlled by a light-weight rule-based system in real time. In contrast to that, during the "night phase" information that has been stored is transferred to a supercomputing system to update the realistic neural network: emotional and behavioral strategies.