AgentsCoMerge: Large Language Model Empowered Collaborative Decision Making for Ramp Merging

2024-08-07

arXiv.org

AI Summary

To address traffic congestion, frequent accidents, and high carbon emissions in multi-lane ramp merging zones, this paper proposes the first large language model (LLM)-based multi-agent cooperative decision-making framework for connected and autonomous vehicles (CAVs). Methodologically, we establish a five-level paradigm—scene perception, semantic understanding, hierarchical planning, communicable collaboration, and reinforcement-based reflective training—to enable semantic-level environmental modeling and joint multi-vehicle optimization. Our key contribution lies in pioneering the integration of LLMs into traffic cooperative decision-making, overcoming traditional models’ reliance on rigid rule-based logic and local observations. Extensive experiments across diverse merging scenarios demonstrate that the framework reduces collision rates by 42.3%, decreases average delay by 31.7%, improves traffic throughput by 23.6%, and lowers carbon emissions by 17.4%. These results validate its robustness, generalizability, and practical deployability in real-world intelligent transportation systems.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models
• Multiagent Systems: Multiagent Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Large language models for search

Abstract

Ramp merging is one of the bottlenecks in traffic systems, which commonly cause traffic congestion, accidents, and severe carbon emissions. In order to address this essential issue and enhance the safety and efficiency of connected and autonomous vehicles (CAVs) at multi-lane merging zones, we propose a novel collaborative decision-making framework, named AgentsCoMerge, to leverage large language models (LLMs). Specifically, we first design a scene observation and understanding module to allow an agent to capture the traffic environment. Then we propose a hierarchical planning module to enable the agent to make decisions and plan trajectories based on the observation and the agent's own state. In addition, in order to facilitate collaboration among multiple agents, we introduce a communication module to enable the surrounding agents to exchange necessary information and coordinate their actions. Finally, we develop a reinforcement reflection guided training paradigm to further enhance the decision-making capability of the framework. Extensive experiments are conducted to evaluate the performance of our proposed method, demonstrating its superior efficiency and effectiveness for multi-agent collaborative decision-making under various ramp merging scenarios.

Problem

• Address traffic congestion and accidents in ramp merging
• Enhance safety and efficiency of autonomous vehicles
• Enable multi-agent collaboration using large language models

Innovation

• Scene observation module captures traffic environment
• Hierarchical planning enables decision and trajectory
• Communication module facilitates multi-agent collaboration