The road to the autonomous vehicle will develop gradually through the introduction of new
assistive and automatic systems in the coming years. The closest target is to liberate drivers
of the need of permanent monitoring and to allow, for limited periods and driving contexts, to
carry out other activities while driving (SAE Level 3).
In connection with all kinds of partially automated driving there is the question of decision
making: Who should, and when, take over control of single functions or of the entire system –
the human or the vehicle. Should the system take over it would need to know about the current
state of the driver at any time, in order to carry out a controlled handover to the driver or
to initiate emergency measures. It is essential to assess whether a person is inattentive, occupied
with other things, sleepy or even sleeping, or if there is some other reason why
he/she is not able to take over the control of the function. Currently, there are already invehicle
systems that warn of drowsiness. They are based on different technical measurement
approaches, e.g. the observation of the steering movements, tracking or eye blinking.
However, we are of the opinion that the accuracy of these systems is yet insufficient to determine
the driver's attention or state of alertness. Therefore, our effort will be to warn as
precisely and as early as appropriate, before the state of the driver leads to incorrect responses
viz. before he/she falls asleep.
The first objective of the project is to merge data of measured autonomous arousal, camera
monitoring and driver behaviour. The aim is to warn about impaired fitness to drive with
the primary goal of drowsiness detection and with the potential also to send information to
the traffic infrastructure by means of appropriate communication.
The second objective of the project is to gain the necessary information through fully embedded,
up-to-date sensors and by intelligent integrated processing. On the market monitoring
sensor technology is only accepted if it works unobtrusively and without additional effort
for the users.
The third objective of the project is to identify changes in the attention, when people are
aware of using automated driving, and to identify possible safety-relevant effects.
The project WACHsens collects the necessary data in a gender-specific driving simulator
study. An innovative step is the integration of multiple data sources to one whole system that
is expected to be robust against errors and malfunctions of individual sensors.
The results of the project will be a system of sensors, algorithms and a classification-model,
which is able to characterise earlier and more reliably the state of the driver. This will be based
on comprehensive data concerning age and gender, suitable for integration into the existing