Even though drivers may perform driving-unrelated activities during automated driving, they must be ready to resume the driving task dependent on the level of automation. With the ultimate aim of guaranteeing safe takeovers, automated driving systems at certain levels of automation need to assess drivers’ readiness to take over preliminary to a transition based on driver monitoring systems. In order to optimize current methods of drivers’ state detection, the present work tries to investigate the idea of a human-based driver observation. It is strived to find first answers to the question, whether and which benefits such an approach would offer, i.e. whether criteria identified by human observation can serve for drivers’ state detection systems. In this context, it was attempted to identify and define possible levels of involvement into the driving task. Furthermore, a study was conducted to find first criteria based on human observation for the levels’ detection that in the future might be examined using existing technical methods.
[1]
Kristie Lee Young,et al.
Sources of Distraction Inside the Vehicle and Their Effects on Driving Performance
,
2009
.
[2]
Anna Anund,et al.
Detecting Driver Sleepiness Using Optimized Nonlinear Combinations of Sleepiness Indicators
,
2011,
IEEE Transactions on Intelligent Transportation Systems.
[3]
Maximilian Schwalm,et al.
Uncanny and Unsafe Valley of Assistance and Automation: First Sketch and Application to Vehicle Automation
,
2017
.
[4]
Keiichi Uchimura,et al.
Driver Inattention Monitoring System for Intelligent Vehicles: A Review
,
2009,
IEEE Transactions on Intelligent Transportation Systems.