基于多模态大模型的机车运行环境感知智能体的研究

打开文本图片集

中图分类号：U291.1；TP183 文献标识码：A 文章编号：2096-4706（2026）05-0106-06

Research on the Intelligent Agent for Locomotive Operation Environment Perception Based on the Multi-modal Large Model

CHEN Zhe， ZHENG Chen， YAN Han， LI Kaixuan， LIU Zheng （CRSC Communication and Information Group Co.，Ltd.，Beijing 10oo71， China）

Abstract：Against the problems of foreign Object Detection and risk recognition inrail transit scenarios，this paper proposes an environment perception intellgent agent system based on the multi-modal large model.The system adopts independent detection pathways for vision and milimeter-waveradar，andon the basis of maintaining the complementary advantages ofsensors，introduces aLarge Language Model intellgent agent module with memory，reasoningandtool-caling capabilities toperformsemanticfusion andrisk analysis onheterogeneous perceptionresults，therebyrealizingamoreaccurate andflexible abnormal eventrecognitionandresponsemechanism.Experimentalresultsshowthatthesystemhas significant advantages insmalltargetrecognition，long-rangedetectionandsemanticjudgmentincomplexscenarios，andcanprovidea system paradigm for high-reliability risk early warning in rail transit scenarios.

Keywords： multi-modal fusion; radar-vision collaboration; environment perception; intelligent agent; track foreigr object detection

0 引言

随着智能交通系统的发展，轨道交通安全面临着日益复杂的运行环境。（剩余8816字）