Challenges Faced by Large Language Models in Solving Multi-Agent Flocking (2404.04752v2)
Abstract: Flocking is a behavior where multiple agents in a system attempt to stay close to each other while avoiding collision and maintaining a desired formation. This is observed in the natural world and has applications in robotics, including natural disaster search and rescue, wild animal tracking, and perimeter surveillance and patrol. Recently, LLMs have displayed an impressive ability to solve various collaboration tasks as individual decision-makers. Solving multi-agent flocking with LLMs would demonstrate their usefulness in situations requiring spatial and decentralized decision-making. Yet, when LLM-powered agents are tasked with implementing multi-agent flocking, they fall short of the desired behavior. After extensive testing, we find that agents with LLMs as individual decision-makers typically opt to converge on the average of their initial positions or diverge from each other. After breaking the problem down, we discover that LLMs cannot understand maintaining a shape or keeping a distance in a meaningful way. Solving multi-agent flocking with LLMs would enhance their ability to understand collaborative spatial reasoning and lay a foundation for addressing more complex multi-agent tasks. This paper discusses the challenges LLMs face in multi-agent flocking and suggests areas for future improvement and research.
- Evaluating multi-agent coordination abilities in large language models, 2023.
- Decentralized multi-agent active search and tracking when targets outnumber agents, 2024.
- Multi-agent consensus seeking via large language models. arXiv preprint arXiv:2310.20151, 2023.
- Why solving multi-agent path finding with large language model has not succeeded yet. arXiv preprint arXiv:2401.03630, 2024.
- Scalable multi-robot collaboration with large language models: Centralized or decentralized systems?, 2024.
- Improving factuality and reasoning in language models through multiagent debate, 2023.
- Stochastic nonlinear ensemble modeling and control for robot team environmental monitoring. arXiv preprint arXiv:2212.11447, 2022.
- Large language models and medical knowledge grounding for diagnosis prediction. medRxiv, pages 2023–11, 2023.
- Metagpt: Meta programming for a multi-agent collaborative framework, 2023.
- Camel: Communicative agents for ”mind” exploration of large language model society, 2023.
- Flocking of decentralized multi-agent systems with application to nonholonomic multi-robots. IFAC Proceedings Volumes, 41(2):9344–9349, 2008. 17th IFAC World Congress.
- Discrete-time flocking control in multi-robot systems with random link failures. IEEE Transactions on Vehicular Technology, pages 1–15, 2024.
- Chameleon: Plug-and-play compositional reasoning with large language models. arXiv preprint arXiv:2304.09842, 2023.
- Divide and conquer for large language models reasoning. arXiv preprint arXiv:2401.05190, 2024.
- Reza Olfati-Saber. Flocking for multi-agent dynamic systems: Algorithms and theory. IEEE Transactions on automatic control, 51(3):401–420, 2006.
- Clipswarm: Generating drone shows from text prompts with vision-language models, 2024.
- Adversar: Adversarial search and rescue via multi-agent reinforcement learning, 2022.
- Flocking in complex environments – attention trade-offs in collective information processing. arXiv preprint arXiv:1907.11691, 2019.
- Craig W. Reynolds. Flocks, herds and schools: A distributed behavioral model. In Proceedings of the 14th Annual Conference on Computer Graphics and Interactive Techniques, SIGGRAPH ’87, page 28, New York, NY, USA, 1987. Association for Computing Machinery.
- Flocking and swarming in a multi-agent dynamical system, 2023.
- David Schlangen. Dialogue games for benchmarking language understanding: Motivation, taxonomy, strategy, 2023.
- A survey of decision-theoretic approaches for robotic environmental monitoring. arXiv preprint arXiv:2308.02698, 2023.
- Language conditioned traffic generation, 2023.
- Stable flocking of mobile agents part i: dynamic topology. In 42nd IEEE International Conference on Decision and Control (IEEE Cat. No. 03CH37475), volume 2, pages 2016–2021. IEEE, 2003.
- Planbench: An extensible benchmark for evaluating large language models on planning and reasoning about change, 2023.
- Flocking for multiple subgroups of multi-agents with different social distancing. IEEE Access, 8:164705–164716, 2020.
- Chain-of-thought prompting elicits reasoning in large language models. arXiv preprint arXiv:2201.11903, 2022.
- Mathchat: Converse to tackle challenging math problems with LLM agents. In ICLR 2024 Workshop on Large Language Model (LLM) Agents, 2024.
- Investigating the effectiveness of task-agnostic prefix prompt for instruction following. arXiv preprint arXiv:2302.14691, 2023.
- Multi-robot flocking control using multi-agent twin delayed deep deterministic policy gradient. In 2022 19th International Conference on Electrical Engineering, Computing Science and Automatic Control (CCE), pages 1–5, 2022.
- Automatic model selection with large language models for reasoning. arXiv preprint arXiv:2305.14333, 2023.
- Progressive-hint prompting improves reasoning in large language models. arXiv preprint arXiv:2304.09797, 2023.
- Multi-robot flocking control based on deep reinforcement learning. IEEE Access, 8:150397–150406, 2020.