In addition, their initial reactions were analyzed in a typical complex scenario following a takeover request. This scenario was preceded by a level 3 drive during which the tiredness of the drivers was rated at regular intervals by trained assessors on the basis of valid tiredness indicators.
60 subjects aged from 18 to 87 took part in the study. 48 percent of this random sample were younger drivers aged from 18 to 35, 25 percent were in the middle age range of 36 to 55, and 27 percent were older drivers (55+). 38 people in the random sample were male (63 percent), and 22 of them were female (37 percent).
After a drive in which they had reached a moderate level of tiredness and were not distracted, 90 percent of the drivers with the automation system looked at the road again for the first time after 1 second, had their hands on the steering wheel and their feet on the pedals after 3-4 seconds and had switched off the automation system after 6-7 seconds. However, if the first glance at the mirror and the glance at the speedometer are taken as indicators of awareness of the situation, you see that 12-15 seconds were required. These reactions, which are required in order to understand the current traffic situation, were thus delayed by up to 6 seconds compared to the reactions of drivers with control of the vehicle in the same situation. Except for the first glance at the road, the values of tired drivers with an automation system are comparable to those of alert drivers on a short, automated drive in which they are strongly distracted visually, cognitively and in terms of motor activity by an engaging secondary task.
A level 3 drive without secondary activities for the driver should not exceed 15 minutes. Longer drives without breaks have to be considered unsafe, since drivers are unable to monitor the situation during an automated drive for an extended period without getting tired, and the condition of the driver cannot be assessed with certainty by technical systems. If drivers are given a sufficiently long advance warning period, even when tired or after a long drive, they are able to handle driving situations after a takeover request about as well as those driving without an automation system. 12 to 15 seconds is a sufficient warning time for the majority of drivers after a drive of more than 30 minutes, or for a tired driver after a drive of more than 15 minutes.
It is thus essential that the driver’s condition is monitored during the automated drive and taken into account when a decision is taken about how long it takes to resume manual control safely and comfortably. In addition, the risks of drivers with automation systems being distracted by secondary tasks should be weighed against the risk of falling asleep. In contrast to drivers of conventional vehicles, it may make sense for drivers with an automation system to be distracted by a secondary task that is controlled by the vehicle to prevent tiredness setting in.