Open-TI: Open Traffic Intelligence with Augmented Language Model
Longchao Da, Kuanru Liou, Tiejin Chen, Xuesong Zhou, Xiangyong Luo, Yezhou Yang, Hua Wei
TL;DR
Open-TI addresses the challenge of trustworthy, deployable intelligent transportation analytics by coupling augmented language models with domain-specific traffic tools to perform end-to-end analysis—from map data retrieval to simulation and policy embodiment. The approach introduces a structured prompt design, a pivotal analysis agent, task-specific embodiments (OD optimization and traffic signal control), and a meta-control scheme (ChatZero) that enables language-driven control decisions. Key contributions include the first end-to-end traffic analysis framework that operates from scratch, a modular architecture with an extensible API-style implementation, and extensive ablation studies showing the benefits of prompt design and multi-LLM meta-control. The work has practical impact by lowering barriers for industry practitioners to adopt advanced traffic analytics with explainable, configurable workflows and open community-driven extensions.
Abstract
Transportation has greatly benefited the cities' development in the modern civilization process. Intelligent transportation, leveraging advanced computer algorithms, could further increase people's daily commuting efficiency. However, intelligent transportation, as a cross-discipline, often requires practitioners to comprehend complicated algorithms and obscure neural networks, bringing a challenge for the advanced techniques to be trusted and deployed in practical industries. Recognizing the expressiveness of the pre-trained large language models, especially the potential of being augmented with abilities to understand and execute intricate commands, we introduce Open-TI. Serving as a bridge to mitigate the industry-academic gap, Open-TI is an innovative model targeting the goal of Turing Indistinguishable Traffic Intelligence, it is augmented with the capability to harness external traffic analysis packages based on existing conversations. Marking its distinction, Open-TI is the first method capable of conducting exhaustive traffic analysis from scratch - spanning from map data acquisition to the eventual execution in complex simulations. Besides, Open-TI is able to conduct task-specific embodiment like training and adapting the traffic signal control policies (TSC), explore demand optimizations, etc. Furthermore, we explored the viability of LLMs directly serving as control agents, by understanding the expected intentions from Open-TI, we designed an agent-to-agent communication mode to support Open-TI conveying messages to ChatZero (control agent), and then the control agent would choose from the action space to proceed the execution. We eventually provide the formal implementation structure, and the open-ended design invites further community-driven enhancements.