论文标题
一种积极的方法来监视和增强驾驶员的功能 - 亚当·科格特（Adam Cogtec）解决方案
An active approach towards monitoring and enhancing drivers' capabilities -- the ADAM cogtec solution
论文作者
论文摘要
在给定时刻，驾驶员的认知能力是评估驾驶员安全的最难以捉摸的变量。与其他身体状况（例如，短相或手动残疾认知能力）相反。安全法规试图通过消除诸如饮酒或吸毒之类的风险因素，禁止发短信等次要任务等风险因素来降低与驾驶员的认知能力有关的风险，并敦促驾驶员在感到疲倦时休息。但是，一个人不能调节影响驾驶员认知的所有因素，此外，在大多数情况下，驾驶员的瞬时认知能力甚至是秘密的。 在这里，我们介绍了一种主动方法，旨在监视受所有这些因素原因影响的特定认知过程，并直接影响驾驶员在驾驶任务中的性能。我们依靠卡尔·弗里斯顿（Karl Friston）构建的科学方法（Friston，2010年）。我们开发了一个封闭的环路方法，其中记录了驾驶员对视觉探测的反应。对机器学习 - 矩阵进行了对警惕状况的眼部反应的训练，并能够检测到由于疲劳和药物滥用而能够降低能力。我们的结果表明，无论损害原因如何，我们都设法将受损和未损害的认知过程分类（77％的准确性，5％的错误警报）。
Driver's cognitive ability at a given moment is the most elusive variable in assessing driver's safety. In contrast to other physical conditions, such as short-sight, or manual disability cognitive ability is transient. Safety regulations attempt to reduce risk related to driver's cognitive ability by removing risk factors such as alcohol or drug consumption, forbidding secondary tasks such as texting, and urging drivers to take breaks when feeling tired. However, one cannot regulate all factors that affect driver's cognition, furthermore, the driver's momentary cognitive ability in most cases is covert even to driver. Here, we introduce an active approach aiming at monitoring a specific cognitive process that is affected by all these forementioned causes and directly affects the driver's performance in the driving task. We lean on the scientific approach that was framed by Karl Friston (Friston, 2010). We developed a closed loop-method in which driver's ocular responses to visual probing were recorded. Machine-learning-algorithms were trained on ocular responses of vigilant condition and were able to detect decrease in capability due fatigue and substance abuse. Our results show that we manage to correctly classify subjects with impaired and unimpaired cognitive process regardless of the cause of impairment (77% accuracy, 5% false alarms).
