Driver Modeling in Transfers of Control from Conditional Automation

We conducted a naturalistic driver study to understand how drivers interact with automated systems in everyday commuting, examining periods where a driver is likely to be highly vigilant along with incidents where driving task may be in low demand.

The Problem 


In order to achieve SAE Level Three vehicle automation, further innovations are needed in driver-to-vehicle and vehicle-to-driver exchange of vehicle control. 

The Question 


How do drivers typically interact with automated vehicle systems and what are typical driver takeover times and patterns?

What We Did 


We conducted a naturalistic driver study to understand how drivers interact with automated systems in everyday commuting, examining periods where a driver is likely to be highly vigilant along with incidents where driving task may be in low demand. A controlled driver study was also conducted at the University of Iowa, with data modeling conducted by the University of Wisconsin. 

The Result 

Through this research we were able to gain a better sense of how far in advance of an potential road issue a vehicle should issue a warning for the driver to take over, or for the vehicle itself to take action, such as slow down or stop.

This is a project in collaboration with University of California, San Diego, University of Iowa – Department of Neurology, University of Iowa Hospitals and Clinics, University of Wisconsin-Madison