Human factors in automated driving

ABSTRACT

Within the European Marie Curie project Human Factors of Automated Driving, we are looking into the driver interaction with Automated Vehicles. HFAuto aims to bridge the gap between engineers and psychologists through a multidisciplinary research and training programme. We will combine engineering domains such as simulator hardware, traffic flow theory, control theory, and mathematical driver modelling with psychological domains such as human action and perception, cognitive modelling, vigilance, distraction, psychophysiology, and mode/situation awareness, to optimally address the interdisciplinary domain of human factors.

The PhD at the University Twente investigates human behaviour in HAD using extensive driving simulator studies with a newly developed driver state monitor, supporting the driver in HAD and in transitions between automated and manual driving. In order to study driver behavior in the transition of control from manual to (semi) automated driving and vice versa, we need to study driver behavior in a micro level (that is in the individual driver). This includes topics like: What is the mode awareness of the driver? How can we improve mode awareness etc. The focus of the PhD at the University of Twente is not (only) on isolated automation (one single automated car without cooperative technology) but rather on automation in the context of cooperative driving (shorter time span for implementation) in mixed traffic (i.e. traffic with and without automation). The Examples of research questions to be addressed by the PhD:

1. What is the best approach for the transition from manual control to fully automated control and vice versa?
2. Technical design in cooperation with user: how does the system guarantee a certain level of safety and reliability and what are the failure modes. Or in other words when is the driver allowed to use the system and what happens if a failure occurs.
3. How much time does it take for a driver to retrieve a sufficient level of Situational Awareness in order to take back manual control? And what parameters influence this  time?
4. Can drivers understand the safety modes of the system and how do they respond to changes in this safety (in terms the autonomous reaction of the system like increasing following distance, readiness of the driver to respond, subjective workload, braking, turning off the system etc)? And, as before, how can this be influenced?
5. What is the interpersonal and intrapersonal variability?