基于多模态大模型的汽车故障诊断智能体研究

打开文本图片集

中图分类号： 文献标识码： 文章编号：1003-8639（2025）05-0079-04

【Abstract】With therevolutionary influenceof LLMagents，automotive fault diagnosis technology isexpected to undergonew upgradesand changes.Based on thelatest research status of automotive fault diagnosisand artificial intellgence，this paper proposesanautomotive fault diagnosisagent framework basedonmulti-modal LLM.Further， thispaper discusses theconstructionprincipleof automotive fault diagnosisagent.Finall，the key technologiesof automotive fault diagnosisagent are analyzedand elaborated in terms of multi-modalLLM，task planning，memory technology，tol applicationandself-learning，hoping toprovide guidanceand reference for subsequent relevantlanding applications.

【Key words】 automotive fault diagnosis；LLM agent；key technologies

0 引言

根据中国汽车工业协会最新数据，中国汽车产销量于 年继续保持 万辆以上规模并再创新高。（剩余6095字）