Multi-Agent Recurrent Rendezvous Using Drive-Based Motivation

TL;DR

This paper extends the Motivation Dynamics framework to multi-agent systems, enabling persistent rendezvous behaviors and revealing complex recurrent dynamics through bifurcation analysis and numerical simulations.

Contribution

It introduces a multi-agent extension of the Motivation Dynamics framework and analyzes its bifurcation properties for rendezvous behaviors.

Findings

The extended framework encodes persistent rendezvous in multi-agent systems.

The system exhibits complex recurrent behavior similar to strange attractors.

Bifurcation analysis characterizes the decision-making dynamics.

Abstract

Recent papers have introduced the Motivation Dynamics framework, which uses bifurcations to encode decision-making behavior in an autonomous mobile agent. In this paper, we consider the multi-agent extension of the Motivation Dynamics framework and show how the framework can be extended to encode persistent multi-agent rendezvous behaviors. We analytically characterize the bifurcation properties of the resulting system, and numerically show that it exhibits complex recurrent behavior suggestive of a strange attractor.