基于眼动和脑电特征的高速铁路行车调度员疲劳状态判别研究  

Research on Fatigue State Discrimination of High Speed Railway Dispatchers Based on Eye-movement and EEG Features

作　　者：张光远[1,2] 王灿 陈诚[3] ZHANG Guangyuan;WANG Can;CHEN Cheng(School of Transportation and Logistics,Southwest Jiaotong University,Chengdu 610031,Sichuan,China;National andRegional Joint Engineering Laboratory of Comprehensive Intelligent Transportation,Southwest Jiaotong University,Chengdu 610031,Sichuan,China;Transportation&Economics Research Institute,China Academy of RailwaySciences Corporation Limited,Beijing 100081,China)

机构地区：[1]西南交通大学交通运输与物流学院,四川成都610031 [2]西南交通大学综合交通运输智能化国家地方联合工程实验室,四川成都610031 [3]中国铁道科学研究院集团有限公司运输及经济研究所,北京100081

出　　处：《铁道运输与经济》2024年第11期196-204,212,共10页Railway Transport and Economy

基　　金：四川省自然科学基金项目(24NSFSC0485);中国国家铁路集团有限公司科技研究开发计划课题(Q113623S04003);西南交通大学本科教改项目(20221103,2103065)。

摘　　要：针对当前高速铁路行车调度员疲劳状态判别方法尚未结合多源信息的优势,提出了一种基于多源信息的高速铁路行车调度员疲劳状态判别方法,在利用支持向量机对眼动和脑电特征进行软分类作为决策层输入的基础上,结合上一时间点疲劳状态判别结果,利用改进的D-S证据理论进行决策层融合得到最终的判别结果。通过采集20名行车调度员的模拟实验数据对模型进行检测,结果表明:改进后的研究方法在疲劳状态判别方面准确率达到93.75%。可以看出,信息融合判别的准确率高于单纯依靠眼动和脑电特征的方法,增加上一时间点的疲劳判别结果有助于提高模型的鲁棒性和可靠性,并且改进的模型对高速铁路行车调度员疲劳状态判别具有有效性。Given the fact that the current methods for the discrimination of fatigue state of high speed railway dispatchers doesn’t consider the advantages of combining multi-source information,an identification method based on multi-source information was proposed.Based on the soft classification of eye movement and EEG features using support vector machine as the input of decision level,combined with the judgment result of fatigue state at the last point of time,the final judgment result was obtained by using the improved D-S Evidence Theory.By collecting the simulation experiment data of 20 dispatchers,the model was tested.The results show that the accuracy of the improved research method is up to 93.75%.It follows that the accuracy of information fusion discrimination is higher than that of the method solely relying on eye�movement and EEG features.The addition of fatigue discrimination results at the last point of time is helpful to improve the robustness and reliability of the model,and the improved model is effective in discriminating the fatigue state of high speed railway dispatchers.

关 键 词：眼动和脑电特征 高速铁路行车调度员 支持向量机 D-S证据理论 疲劳状态判别 

分 类 号：U293.12[交通运输工程—交通运输规划与管理]