Low‑altitude wireless networks (LAWNs) are emerging as essential infrastructure for heterogeneous unmanned aerial systems that operate concurrently in a shared three‑dimensional airspace. The coexistence of diverse services creates strong coupling among mobility, connectivity, and shared network resources, while each service imposes distinct, time‑varying requirements. These interactions naturally form a dynamic non‑cooperative game whose operating conditions and coordination objectives evolve over time. Conventional optimization or learning‑based controllers that rely on predefined objectives struggle to adapt autonomously to shifting service demands and resource priorities. To address this, we propose a hierarchical hybrid large language model (LLM)‑multi‑agent reinforcement learning (MARL) architecture organized as a dual‑loop structure. The outer adaptation loop employs LLM‑assisted game orchestration to interpret service requirements and operator intent, dynamically reconfiguring objectives and resource priorities. The inner loop runs decentralized, parameter‑conditioned MARL policies under the configured game. A logistics‑monitoring case study demonstrates how the framework enables coordinated coexistence of heterogeneous services, adapting to evolving conditions without retraining the underlying MARL policies. Finally, we outline key challenges and research directions toward scalable, trustworthy, and adaptive agentic LAWNs.
Review