Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
110 tokens/sec
GPT-4o
56 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
6 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

LLMLight: Large Language Models as Traffic Signal Control Agents (2312.16044v5)

Published 26 Dec 2023 in cs.AI

Abstract: Traffic Signal Control (TSC) is a crucial component in urban traffic management, aiming to optimize road network efficiency and reduce congestion. Traditional TSC methods, primarily based on transportation engineering and reinforcement learning (RL), often struggle with generalization abilities across varied traffic scenarios and lack interpretability. This paper presents LLMLight, a novel framework employing LLMs as decision-making agents for TSC. Specifically, the framework begins by instructing the LLM with a knowledgeable prompt detailing real-time traffic conditions. Leveraging the advanced generalization capabilities of LLMs, LLMLight engages a reasoning and decision-making process akin to human intuition for effective traffic control. Moreover, we build LightGPT, a specialized backbone LLM tailored for TSC tasks. By learning nuanced traffic patterns and control strategies, LightGPT enhances the LLMLight framework cost-effectively. Extensive experiments conducted on ten real-world and synthetic datasets, along with evaluations by fifteen human experts, demonstrate the exceptional effectiveness, generalization ability, and interpretability of LLMLight with LightGPT, outperforming nine baseline methods and ten advanced LLMs.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (48)
  1. Reinforcement learning for true adaptive traffic signal control. Journal of Transportation Engineering, 129(3):278–285, 2003.
  2. Grounding large language models in interactive environments with online reinforcement learning. In International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Honolulu, Hawaii, USA, volume 202 of Proceedings of Machine Learning Research, pp.  3676–3713. PMLR, 2023.
  3. Toward A thousand lights: Decentralized deep reinforcement learning for large-scale traffic signal control. In The Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2020, The Thirty-Second Innovative Applications of Artificial Intelligence Conference, IAAI 2020, The Tenth AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2020, New York, NY, USA, February 7-12, 2020, pp.  3414–3421. AAAI Press, 2020.
  4. Driving with llms: Fusing object-level vector modality for explainable autonomous driving. CoRR, abs/2310.01957, 2023.
  5. Prg-light: A novel traffic light control framework with pressurebased-reinforcement learning and graph neural network. In IJCAI 2021 Reinforcement Learning for Intelligent Transportation Systems (RL4ITS) Workshop, August, 2021, virtual, 2021.
  6. Self-organizing traffic lights: A realistic simulation. Advances in applied self-organizing systems, pp.  45–55, 2013.
  7. Drive as you speak: Enabling human-like interaction with large language models in autonomous vehicles. CoRR, abs/2309.10228, 2023.
  8. LLM powered sim-to-real transfer for traffic signal control. CoRR, abs/2308.14284, 2023.
  9. An agent-based learning towards decentralized and coordinated traffic signal control. In 13th International IEEE Conference on Intelligent Transportation Systems, Funchal, Madeira, Portugal, 19-22 September 2010, pp.  665–670. IEEE, 2010.
  10. Torantulino et al. Autogpt. https://github.com/Significant-Gravitas/AutoGPT, 2023. Accessed on 1 December 2023.
  11. Drive like a human: Rethinking autonomous driving with large language models. CoRR, abs/2307.07162, 2023.
  12. The scoot on-line traffic signal optimisation technique. Traffic Engineering & Control, 23(4), 1982.
  13. Survey of hallucination in natural language generation. ACM Comput. Surv., 55(12):248:1–248:38, 2023.
  14. Panda LLM: training data and evaluation for open-sourced chinese instruction-following large language models. CoRR, abs/2305.03025, 2023.
  15. Surrealdriver: Designing generative driver agent simulation framework in urban contexts based on large language model. CoRR, abs/2309.13193, 2023.
  16. Scaling laws for neural language models. CoRR, abs/2001.08361, 2020.
  17. Traffic signal timing manual. Technical report, United States. Federal Highway Administration, 2008.
  18. A preference-aware meta-optimization framework for personalized vehicle energy consumption estimation. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2023, Long Beach, CA, USA, August 6-10, 2023, pp.  4346–4356. ACM, 2023.
  19. Diffusion convolutional recurrent neural network: Data-driven traffic forecasting. In 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30 - May 3, 2018, Conference Track Proceedings. OpenReview.net, 2018.
  20. Practical adversarial attacks on spatiotemporal traffic forecasting models. In Advances in Neural Information Processing Systems, volume 35, pp.  19035–19047. Curran Associates, Inc., 2022.
  21. Robust spatiotemporal traffic forecasting with reinforced dynamic adversarial training. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD ’23, pp.  1417–1428, New York, NY, USA, 2023. Association for Computing Machinery. ISBN 9798400701030.
  22. P Lowrie. Scats-a traffic responsive method of controlling urban traffic. Sales information brochure published by Roads & Traffic Authority, Sydney, Australia, 1990.
  23. Gpt-driver: Learning to drive with GPT. CoRR, abs/2310.01415, 2023.
  24. Human-level control through deep reinforcement learning. Nat., 518(7540):529–533, 2015.
  25. OpenAI. Introducing chatgpt. https://openai.com/blog/chatgpt, 2023a. Accessed on 1 December 2023.
  26. OpenAI. GPT-4 technical report. CoRR, abs/2303.08774, 2023b.
  27. Attendlight: Universal attention-based reinforcement learning model for traffic signal control. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual, 2020.
  28. Toolllm: Facilitating large language models to master 16000+ real-world apis. CoRR, abs/2307.16789, 2023.
  29. Toolformer: Language models can teach themselves to use tools. CoRR, abs/2302.04761, 2023.
  30. Languagempc: Large language models as decision makers for autonomous driving. CoRR, abs/2310.03026, 2023.
  31. Hierarchical reinforcement learning for dynamic autonomous vehicle navigation at intelligent intersections. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2023, Long Beach, CA, USA, August 6-10, 2023, pp.  4852–4861. ACM, 2023.
  32. Llama 2: Open foundation and fine-tuned chat models. CoRR, abs/2307.09288, 2023.
  33. Pravin Varaiya. Max pressure control of a network of signalized intersections. Transportation Research Part C: Emerging Technologies, 36:177–195, 2013.
  34. Voyager: An open-ended embodied agent with large language models. CoRR, abs/2305.16291, 2023.
  35. Presslight: Learning max pressure control to coordinate traffic signals in arterial network. In Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, KDD 2019, Anchorage, AK, USA, August 4-8, 2019, pp.  1290–1298. ACM, 2019a.
  36. Colight: Learning network-level cooperation for traffic signal control. In Proceedings of the 28th ACM International Conference on Information and Knowledge Management, CIKM 2019, Beijing, China, November 3-7, 2019, pp.  1913–1922. ACM, 2019b.
  37. Dilu: A knowledge-driven approach to autonomous driving with large language models. CoRR, abs/2309.16292, 2023.
  38. Efficient pressure: Improving efficiency for signalized intersections. CoRR, abs/2112.02336, 2021.
  39. Transformerlight: A novel sequence modeling based traffic signaling mechanism via gated transformer. In Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2023, Long Beach, CA, USA, August 6-10, 2023, pp.  2639–2647. ACM, 2023.
  40. A novel ramp metering algorithm based on deep reinforcement learning. In 2022 2nd International Conference on Algorithms, High Performance Computing and Artificial Intelligence (AHPCAI), 2022, pp.  128–133. IEEE, 2022.
  41. Drivegpt4: Interpretable end-to-end autonomous driving via large language model. CoRR, abs/2310.01412, 2023.
  42. Cityflow: A multi-agent reinforcement learning environment for large scale city traffic scenario. In The World Wide Web Conference, WWW 2019, San Francisco, CA, USA, May 13-17, 2019, pp.  3620–3624. ACM, 2019.
  43. Reinforcement learning for traffic signal control. http://traffic-signal-control.github.io, 2023a. Accessed on 1 December 2023.
  44. Expression might be enough: representing pressure and demand for reinforcement learning based traffic signal control. In International Conference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA, volume 162 of Proceedings of Machine Learning Research, pp.  26645–26654. PMLR, 2022a.
  45. Trafficgpt: Viewing, processing and interacting with traffic foundation models. CoRR, abs/2309.06719, 2023b.
  46. Multi-agent graph convolutional reinforcement learning for dynamic electric vehicle charging pricing. In Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining, pp.  2471–2481, 2022b.
  47. Irregular traffic time series forecasting based on asynchronous spatio-temporal graph convolutional network. arXiv preprint arXiv:2308.16818, 2023c.
  48. Vision language models in autonomous driving and intelligent transportation systems. CoRR, abs/2310.14414, 2023.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (5)
  1. Siqi Lai (6 papers)
  2. Zhao Xu (47 papers)
  3. Weijia Zhang (52 papers)
  4. Hao Liu (497 papers)
  5. Hui Xiong (244 papers)
Citations (2)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now