Learning Rigidity-based Flocking Control with Gaussian Processes

TL;DR

This paper introduces an online Gaussian process-based control method for multi-agent flocking that stabilizes agent motion despite unknown nonlinear dynamics, ensuring high-probability bounded tracking errors.

Contribution

It presents a decentralized, online learning control law using Gaussian processes to stabilize flocking with unknown dynamics in a multi-agent system.

Findings

Exponential stabilization of agent flocking motion.

Probabilistic guarantees on bounded tracking error.

Effective modeling of Reynolds boids behavior.

Abstract

Flocking control of multi-agents system is challenging for agents with partially unknown dynamics. This paper proposes an online learning-based controller to stabilize flocking motion of double-integrator agents with additional unknown nonlinear dynamics by using Gaussian processes (GP). Agents interaction is described by a time-invariant infinitesimally minimally rigid undirected graph. We provide a decentralized control law that exponentially stabilizes the motion of the agents and captures Reynolds boids motion for swarms by using GPs as an online learning-based oracle for the prediction of the unknown dynamics. In particular the presented approach guarantees a probabilistic bounded tracking error with high probability.