摘要
目前,国内部分城市轨道交通运营单位已将知识图谱和语音识别等技术应用于信息查询与票务处理等业务环节,但仍存在着识别手段单一、回答不准确、交互缺乏情感化表达、业务覆盖不全面等问题。为此,文章设计了一种基于语音、视频、环境感知与智能运动等多模态感知技术,并结合大语言模型的AI智能客服机器人。该机器人能够实现票务处理、信息咨询、生物注册、语音对话、远程招援、广告运营、智慧边门、智慧巡检和安全监测等功能,还能够优化现有乘客服务体系,构建由AI智能客服机器人、远程客服与现场客服组成的三级客服服务体系。AI智能客服机器人支持语音和视频融合的票务操作,可实现基于乘客画像的多轮语音对话查询,并在运营期内和运营结束后对站厅进行智能巡检和环境监测。AI智能客服机器人的应用有助于显著提升轨道交通客服智能化水平,并有效降低运营成本。
Currently,several urban rail transit operators in China have incorporated technologies such as knowledge graphs and speech recognition into service areas including information queries and ticketing processes.However,challenges persist,including limited recognition methods,response inaccuracies,lack of emotional expression in interactions,and incomplete business coverage.To address these issues,this paper designs an AI-powered customer service robot that integrates multimodal perception technologies,such as voice,video,environmental sensing,and intelligent motion,with large language models(LLMs).The robot supports various functions,including ticketing,information consultation,biometric registration,voice dialogue,remote assistance,advertising operations,smart access gate management,intelligent inspection,and safety monitoring.It also helps optimize the passenger service system by establishing a three-tier service framework comprising AI robots,remote agents,and on-site staff.The AI robot enables integrated voice and video ticketing operations,supports multi-turn voice dialogues based on passenger profiles,and performs intelligent inspection and environmental monitoring in station halls during and after operational hours.The implementation of this AI customer service robot is expected to significantly enhance the intelligence of rail transit customer services while effectively reducing operational costs.
作者
蔺天震
王翠
徐秋平
崔学广
LIN Tianzhen;WANG Cui;XU Qiuping;CUI Xueguang(Qingdao Boning Futian Intelligent Transportation Technology Development Co.,Ltd.,Qingdao 266111,China;Qingdao Metro Group Co.,Ltd.,Qingdao 266101,China;Qingdao Metro Operation Co.,Ltd.,Qingdao 266101,China)
出处
《智慧轨道交通》
2026年第1期76-80,共5页
INTELLIGENT RAIL TRANSIT
关键词
城市轨道交通
多模态
大模型
智能客服
客服机器人
urban rail transit
multimodal
large language models
intelligent customer service
service robot
