Dual-Attention Heterogeneous GNN for Multi-robot Collaborative Area Search via Deep Reinforcement Learning

TL;DR

This paper introduces a dual-attention heterogeneous graph neural network trained with deep reinforcement learning to improve multi-robot collaborative area search by effectively modeling complex relationships and balancing exploration with target coverage.

Contribution

It proposes a novel DA-HGNN that models heterogeneous entities and their relationships, enabling better task balancing in multi-robot search tasks, validated through extensive 3D simulation experiments.

Findings

Outperforms existing methods in scalability and generalization.

Effectively models complex spatio-temporal relationships.

Decouples exploration and coverage tasks using dual-attention mechanisms.

Abstract

In multi-robot collaborative area search, a key challenge is to dynamically balance the two objectives of exploring unknown areas and covering specific targets to be rescued. Existing methods are often constrained by homogeneous graph representations, thus failing to model and balance these distinct tasks. To address this problem, we propose a Dual-Attention Heterogeneous Graph Neural Network (DA-HGNN) trained using deep reinforcement learning. Our method constructs a heterogeneous graph that incorporates three entity types: robot nodes, frontier nodes, and interesting nodes, as well as their historical states. The dual-attention mechanism comprises the relational-aware attention and type-aware attention operations. The relational-aware attention captures the complex spatio-temporal relationships among robots and candidate goals. Building on this relational-aware heterogeneous graph, the type-aware attention separately computes the relevance between robots and each goal type (frontiers vs. points of interest), thereby decoupling the exploration and coverage from the unified tasks. Extensive experiments conducted in interactive 3D scenarios within the iGibson simulator, leveraging the Gibson and MatterPort3D datasets, validate the superior scalability and generalization capability of the proposed approach.