AC-MASAC: An Attentive Curriculum Learning Framework for Heterogeneous UAV Swarm Coordination

2026-02-12

AI Summary

This work proposes a multi-agent reinforcement learning framework integrating role-aware heterogeneous attention mechanisms with structured curriculum learning to address key challenges in cooperative path planning for heterogeneous unmanned aerial vehicle (UAV) swarms, including asymmetric agent dependencies, sparse rewards, and catastrophic forgetting. The proposed approach models asymmetric inter-agent dependencies through a tailored attention mechanism and mitigates sparse reward and catastrophic forgetting issues via hierarchical knowledge transfer and a phased proportional experience replay strategy. Experimental results on a custom simulation platform demonstrate that the method significantly outperforms existing approaches in terms of task success rate, formation retention ratio, and weighted task completion time, while substantially improving training stability and collaborative efficiency.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Multiagent Systems: Multiagent Planning
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Cooperative path planning for heterogeneous UAV swarms poses significant challenges for Multi-Agent Reinforcement Learning (MARL), particularly in handling asymmetric inter-agent dependencies and addressing the risks of sparse rewards and catastrophic forgetting during training. To address these issues, this paper proposes an attentive curriculum learning framework (AC-MASAC). The framework introduces a role-aware heterogeneous attention mechanism to explicitly model asymmetric dependencies. Moreover, a structured curriculum strategy is designed, integrating hierarchical knowledge transfer and stage-proportional experience replay to address the issues of sparse rewards and catastrophic forgetting. The proposed framework is validated on a custom multi-agent simulation platform, and the results show that our method has significant advantages over other advanced methods in terms of Success Rate, Formation Keeping Rate, and Success-weighted Mission Time. The code is available at \textcolor{red}{https://github.com/Wanhao-Liu/AC-MASAC}.

Problem

• heterogeneous UAV swarm
• asymmetric inter-agent dependencies
• sparse rewards
• catastrophic forgetting
• cooperative path planning

Innovation

• attentive curriculum learning
• heterogeneous UAV swarm
• asymmetric dependencies
• catastrophic forgetting
• multi-agent reinforcement learning