On-board Mission Replanning for Adaptive Cooperative Multi-Robot Systems

2025-06-06

AI Summary

This paper addresses the challenge of collaborative task re-planning for multi-robot systems in dynamic remote environments, formalizing it as the Collaborative Multi-Robot Planning (CMRP) problem—a novel variant of the Multiple Traveling Salesman Problem (mTSP) supporting multiple starting locations, explicit inter-robot collaboration, and variable-duration tasks. Method: We propose a graph attention-based encoder-decoder architecture integrating Graph Attention Networks (GATs), sequence-level attention, and lightweight neural combinatorial optimization, specifically optimized for edge deployment (e.g., on Raspberry Pi). Contribution/Results: CMRP is the first formalization enabling decentralized, asynchronous initialization, and real-time onboard re-planning. Experiments in UAV coordination scenarios achieve a 90% re-planning success rate, solution quality exceeding 90% of LKH3’s optimal solutions, and 85–370× inference speedup over conventional solvers—demonstrating strong trade-offs among collaboration fidelity, timeliness, and embedded-system feasibility.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Multiagent Systems: Multiagent Planning
• Intelligent Robots: Multi-Robot Systems

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Cooperative autonomous robotic systems have significant potential for executing complex multi-task missions across space, air, ground, and maritime domains. But they commonly operate in remote, dynamic and hazardous environments, requiring rapid in-mission adaptation without reliance on fragile or slow communication links to centralised compute. Fast, on-board replanning algorithms are therefore needed to enhance resilience. Reinforcement Learning shows strong promise for efficiently solving mission planning tasks when formulated as Travelling Salesperson Problems (TSPs), but existing methods: 1) are unsuitable for replanning, where agents do not start at a single location; 2) do not allow cooperation between agents; 3) are unable to model tasks with variable durations; or 4) lack practical considerations for on-board deployment. Here we define the Cooperative Mission Replanning Problem as a novel variant of multiple TSP with adaptations to overcome these issues, and develop a new encoder/decoder-based model using Graph Attention Networks and Attention Models to solve it effectively and efficiently. Using a simple example of cooperative drones, we show our replanner consistently (90% of the time) maintains performance within 10% of the state-of-the-art LKH3 heuristic solver, whilst running 85-370 times faster on a Raspberry Pi. This work paves the way for increased resilience in autonomous multi-agent systems.

Problem

• Develops on-board replanning for multi-robot systems in dynamic environments
• Addresses cooperative TSP challenges with variable task durations
• Enables fast decentralized replanning without central communication

Innovation

• Graph Attention Networks for cooperative replanning
• Attention Models for efficient mission adaptation
• On-board Raspberry Pi deployment for fast execution