Shared-Roadmap Generation and Evaluator for Multi-Agent Path Planning Using Heterogeneous Graph Neural Network
In this paper, we propose a scalable heterogeneous Graph Neural Network (GNN) framework for the automated generation and evaluation of shared multi-agent roadmaps.
ProofPaper ↗
Key points
- Multi-agent path planning (MAPP) in continuous environments often relies on roadmaps to balance safety and search efficiency.
- Our model covers the representation of waypoints, agent locations, and task locations as distinct nodes in a heterogeneous graph, allowing it to reason over global connectivity and inter-agent interactions.
- By training on occupation density maps aggregated and collected from expert solver trajectories, the GNN learns to identify critical points of interest and prune redundant nodes and edges.
- This process produces a compact, coordination-aware roadmap that is invariant to task permutations and is reusable for multi-agent pick and delivery tasks.
Sources (1)
- [1]Shared-Roadmap Generation and Evaluator for Multi-Agent Path Planning Using Heterogeneous Graph Neural NetworkarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 07:33 PM
In this paper, we propose a scalable heterogeneous Graph Neural Network (GNN) framework for the automated generation and evaluation of shared multi-agent roadmaps.
Multi-agent path planning (MAPP) in continuous environments often relies on roadmaps to balance safety and search efficiency.
Extractive summary: sentences quoted from the sources.