Distributed Stochastic Search for Multi-Agent Model Predictive Control

TL;DR

This paper introduces a distributed stochastic search framework for multi-agent model predictive control that enhances scalability and effectiveness in navigating complex, nonconvex environments, outperforming existing methods in simulations.

Contribution

It proposes a novel distributed sampling-based optimization method using ADMM for scalable multi-agent MPC in complex environments.

Findings

Successfully navigates nonconvex environments in simulations.

Achieves 100% task completion with zero collisions in a 64-agent formation.

Outperforms distributed IPOPT in complex multi-agent tasks.

Abstract

Many real-world multi-agent systems exhibit nonlinear dynamics and complex inter-agent interactions. As these systems increase in scale, the main challenges arise from achieving scalability and handling nonconvexity. To address these challenges, this paper presents a distributed sampling-based optimization framework for multi-agent model predictive control (MPC). We first introduce stochastic search, a generalized sampling-based optimization method, as an effective approach to solving nonconvex MPC problems because of its exploration capabilities. Nevertheless, optimizing the multi-agent systems in a centralized fashion is not scalable as the computational complexity grows intractably as the number of agents increases. To achieve scalability, we formulate a distributed MPC problem and employ the alternating direction method of multipliers (ADMM) to leverage the distributed approach. In multi-robot navigation simulations, the proposed method shows a remarkable capability to navigate through nonconvex environments, outperforming a distributed optimization baseline using the interior point optimizer (IPOPT). In a 64-agent multi-car formation task with a challenging configuration, our method achieves 100% task completion with zero collisions, whereas distributed IPOPT fails to find a feasible solution.