在高安全性要求的复杂自动控制系统中,人类操作员的性能崩溃往往会导致严重事故的发生,如何采取有效手段,判断操作员功能状态(Operator Functional States, OFS),避免事故的发生已逐渐为研究者致力解决的难题。
解决该问题的有效途径之一是对操作员在当前任务环境下完成工作任务的性能——OFS进行有效分类。研究表明,操作员的电生理测量数据、操作员的任务性能数据——被控子系统处于目标区域的时间比(Time In Range, TIR)等,可以作为评价OFS的依据。
本文首先采用密封舱内空气管理自动化系统(automation-enhanced Cabin Air Management System, aCAMS)软件,模拟操作员执行太空舱空气调控任务,结合实时采集到的被试者电生理信号和任务性能数据,建立实验平台。应用此平台,将独立分量分析(Independent Component Analysis, ICA)方法用于采集到的脑电(Electroencephalography, EEG)和眼电(Electro-Oculogram, EOG)伪迹的有效分离,在此基础上用支持向量...
【英文摘要】
Safety critical systems under high levels of automation are vulnerable to unpredictable breakdowns in operator performance, which has become a serious concern to society. How to take effective measures to determine the functional status of the operator (Operator Functional States, OFS), to avoid accidents has been gradually researchers to address the problem.
Effective way to solve this problem is the operator to complete,the performance of the current task——the operator functional state (OFS) for effe...