Graph Neural Planning and Predictive Control for Multi-Robot Communication-Constrained Unlabeled Motion Planning

TL;DR

This paper introduces a hierarchical multi-robot planning framework combining graph attention and predictive control, addressing real-world constraints like dynamics and communication delays, validated in simulation and real quadrotor experiments.

Contribution

It presents a novel hierarchical framework integrating GATP and NMPC for multi-robot motion planning under communication constraints, improving scalability and robustness.

Findings

Enhanced generalization to larger robot teams.

Robustness to communication delays up to 200 ms.

Validated in both simulation and real-world quadrotor experiments.

Abstract

The multi-robot unlabeled motion planning problem of concurrently assigning robots to goals and generating safe trajectories is central in many collaborative tasks. Recent Graph Neural Network methods offer scalable decentralized solutions but rely on simplified dynamics and simulation environments, overlooking key challenges of real-world deployment such as dynamic feasibility and communication constraints. To address these gaps, we propose a hierarchical framework that combines a Graph ATtention Planner (GATP) with a decentralized Nonlinear Model Predictive Controller (NMPC). GATP provides intermediate subgoals through multi-robot cooperation, and the NMPC enforces safety under nonlinear dynamics and actuation constraints. We evaluate our framework in both simulation and real-world quadrotor experiments. Thanks to attention mechanisms and minimal communication requirements, we demonstrate improved generalization to larger teams, robustness to communication delays up to 200 ms and practical feasibility with decentralized on-board inference.