[공지사항]을 빙자한 안부와 근황 
Show more

Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

CoordField: Coordination Field for Agentic UAV Task Allocation In Low-altitude Urban Scenarios

Created by
  • Haebom

Author

Tengchao Zhang, Yonglin Tian, Fei Lin, Jun Huang, Patrik P. S uli, Qinghua Ni, Rui Qin, Xiao Wang, Fei-Yue Wang

Outline

In this paper, we propose CoordField, a novel system for efficiently coordinating heterogeneous unmanned aerial vehicles (UAVs) swarms in complex urban environments. CoordField leverages a large-scale language model (LLM) to translate high-level human commands into executable commands (e.g., patrolling, target tracking) for UAV swarms, and distributes and adaptively allocates UAV movements and task selections through a coordination field mechanism. Through 50 comparative experiments in a 2D simulation environment, we demonstrate that CoordField outperforms existing methods in terms of task coverage, response time, and adaptability to dynamic changes.

Takeaways, Limitations

Takeaways:
We present a novel system for efficient coordination of heterogeneous UAV swarms in complex urban environments.
Present an effective method to translate high-level human commands into executable instructions using LLM.
Distributed and adaptive task allocation and UAV control possible through the coordination field mechanism.
The experimental results verify the excellent performance of the proposed system.
Limitations:
Currently, it has only been evaluated in a 2D simulation environment, so performance verification in a real environment is required.
Robustness verification against various types of obstacles and unpredictable situations is required.
Since it depends on the performance of LLM, there is a possibility that the limitations of LLM may affect the system performance.
Lack of experimental results on real UAV swarms.
👍