Behavioral entropy as a measure of driving performance

Driving Assessment 2001: The First International Driving Symposium on Human Factors in Driver Assessment, Training and Vehicle DesignFederal Highway Administration; Federal Motor Carrier Safety Administration; Ford Motor Company; National Highway Traffic Safety Administration; University of Iowa; KQ Corporation; Nissan Technical Center North America; Transportation Research Board; TRL(2001)

引用 53|浏览5
暂无评分
摘要
Delayed event detection and degraded vehicle control are observed when drivers fuel their need to perform extra-driving activities. Vehicle control and event detection are shown to degrade most if the in-vehicle task requires spatial cognitive resources and/or if the activity requires visual perception and/or manual control manipulation. In-vehicle tasks with auditory input and/or voice output that primarily demand low levels of verbal cognitive resources appear to affect event detection only to a small degree and seem to have no effect on vehicle control. A theory-based approach to measure, analyze, and interpret these performance assessments is presented. Results from SAE paper #1999-01-0892 are used as a vehicle to demonstrate that steering entropy (a measure of vehicle control) in conjunction with reaction times to unpredictable peripheral events (a surrogate measure for event detection) offer clear insight into the safety consequences of various in-vehicle tasks. These results are here discussed in the context of a simple linear predictive model that is based on Wickens' theory of multiple resources. The model is shown to offer useful predictions about the interpretations of the effects that various in-vehicle tasks have on driving performance in general and driver distraction in particular.
更多
查看译文
关键词
reaction time,behavior,mathematical models,cognition,theory,performance,visual perception,attention,steering
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要