Integrating LLMs with ITS: Recent Advances, Potentials, Challenges, and Future Directions

TL;DR

This paper surveys the integration of Large Language Models into Intelligent Transportation Systems, detailing how centralized and decentralized LLMs can enhance traffic prediction, signal control, routing, V2X, and smart-city functions. It systematically reviews GPT, T5, BERT, LLaMA, and FalconLLM, alongside DLLMs like GPT-NeoX, BLOOM, OpenFlamingo, and Petals, and maps their capabilities to ITS applications. Through case studies (TrafficBERT, LLM-Light, STransformer, TransTTE, BERT4ITS) and extensive discussion of challenges—data, computation, ethics, policy, and infrastructure—the paper presents a roadmap for practical, responsible deployment. The work highlights potential gains in efficiency, safety, and sustainability, while stressing data governance, edge computing, and emerging tech (6G, quantum) as critical enablers for real-time, scalable ITS. Overall, it offers a comprehensive blueprint for researchers and practitioners to harness LLMs to evolve ITS toward smarter, more adaptive transportation networks.

Abstract

Intelligent Transportation Systems (ITS) are crucial for the development and operation of smart cities, addressing key challenges in efficiency, productivity, and environmental sustainability. This paper comprehensively reviews the transformative potential of Large Language Models (LLMs) in optimizing ITS. Initially, we provide an extensive overview of ITS, highlighting its components, operational principles, and overall effectiveness. We then delve into the theoretical background of various LLM techniques, such as GPT, T5, CTRL, and BERT, elucidating their relevance to ITS applications. Following this, we examine the wide-ranging applications of LLMs within ITS, including traffic flow prediction, vehicle detection and classification, autonomous driving, traffic sign recognition, and pedestrian detection. Our analysis reveals how these advanced models can significantly enhance traffic management and safety. Finally, we explore the challenges and limitations LLMs face in ITS, such as data availability, computational constraints, and ethical considerations. We also present several future research directions and potential innovations to address these challenges. This paper aims to guide researchers and practitioners through the complexities and opportunities of integrating LLMs in ITS, offering a roadmap to create more efficient, sustainable, and responsive next-generation transportation systems.

Figures (8)

• Figure 1: Projected growth of the LLM market trends by 2030.
• Figure 2: A generic illustration of the LLM transformer.
• Figure 3: Different types of LLM models.
• Figure 4: Basic architecture of the LLaMA-2 model.
• Figure 5: A generic architecture of Falcon LLM.