VTS-LLM: Domain-Adaptive LLM Agent for Enhancing Awareness in Vessel Traffic Services through Natural Language

2025-05-02

AI Summary

Existing Vessel Traffic Service (VTS) systems suffer from weak spatiotemporal reasoning and unintuitive human–AI interaction, particularly when interpreting multi-source heterogeneous data and supporting collaborative decision-making. Method: This paper introduces the first domain-adaptive large language model (LLM) agent for VTS, framing high-risk vessel identification as a knowledge-enhanced Text-to-SQL task. It establishes the first VTS-specific Text-to-SQL benchmark and proposes four key innovations: (1) NER-driven relational reasoning, (2) a semantic algebraic intermediate representation, (3) a query rethinking mechanism, and (4) an agent-based domain-knowledge injection framework. It also conducts the first systematic analysis of linguistic style—imperative, operational, and formal—on Text-to-SQL performance. Results: Experiments demonstrate consistent superiority over both general-purpose and SQL-specialized baselines across all three query types. The approach significantly improves spatiotemporal risk identification accuracy and enhances decision interpretability in real-world VTS scenarios.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Multiagent Systems: Agent Communication

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Vessel Traffic Services (VTS) are essential for maritime safety and regulatory compliance through real-time traffic management. However, with increasing traffic complexity and the prevalence of heterogeneous, multimodal data, existing VTS systems face limitations in spatiotemporal reasoning and intuitive human interaction. In this work, we propose VTS-LLM Agent, the first domain-adaptive large LLM agent tailored for interactive decision support in VTS operations. We formalize risk-prone vessel identification as a knowledge-augmented Text-to-SQL task, combining structured vessel databases with external maritime knowledge. To support this, we construct a curated benchmark dataset consisting of a custom schema, domain-specific corpus, and a query-SQL test set in multiple linguistic styles. Our framework incorporates NER-based relational reasoning, agent-based domain knowledge injection, semantic algebra intermediate representation, and query rethink mechanisms to enhance domain grounding and context-aware understanding. Experimental results show that VTS-LLM outperforms both general-purpose and SQL-focused baselines under command-style, operational-style, and formal natural language queries, respectively. Moreover, our analysis provides the first empirical evidence that linguistic style variation introduces systematic performance challenges in Text-to-SQL modeling. This work lays the foundation for natural language interfaces in vessel traffic services and opens new opportunities for proactive, LLM-driven maritime real-time traffic management.

Problem

• Enhancing vessel traffic safety via adaptive LLM agents
• Improving spatiotemporal reasoning in maritime traffic systems
• Enabling intuitive human interaction with multimodal VTS data

Innovation

• Domain-adaptive LLM agent for VTS decision support
• Knowledge-augmented Text-to-SQL for vessel identification
• NER-based reasoning and semantic algebra for context awareness