Research In Action

Research In Action

CChIPS autonomous driving research
Human Factors and Emergency Autonomous Vehicle Takeover

Last month I had the pleasure to attend and present at the 2021 World Congress of the Society of Automotive Engineers, a virtual event. While COVID-19 restrictions prevented the Exposition Hall's typical buzz, researchers across the world gathered to share findings on new developments in automotive transportation. 

As a researcher in robotics, I have, for the past five years, been especially interested in the deployment of self-driving technology. This revolution is, according to all observers, a true paradigm shift in transportation technology. Vehicles are now much more than the integration of automotive elements; they are literally computers on wheels for which Human Machine Interfaces must carefully be designed.

It's crucial that owners of highly automated vehicles understand the need to stay alert. 

At the SAE Congress, I had the honor to present on new research, funded by the Center for Child Injury Prevention Studies and the National Science Foundation's Big Data Research Project, that examined the human factors at play as they relate to autonomous driving. The paper, Vehicle Automation Emergency Scenario: Using a Driving Simulator to Assess the Impact of Hand and Foot Placement on Reaction Time, I co-authored with CHOP colleagues Elizabeth Vo-Phamhi, Thomas Seacrist, MBE, and Jalaj Maheshwari, MS and Drexel University's Chris Yang, PhD. 

The Center for Injury Research and Prevention allowed us to leverage a high-fidelity driving simulator in autonomous mode (see above photo) to observe the behaviors of 60 participants in three age groups -- teens (ages 16-19), adults (ages 30-55), and seniors (ages 65+) -- as they turned on various automated features through a number of rides in simulated environments, including highway and rural. 

Participants also wore an eye tracker so that we could assess their level of focus in the various driving scenarios. We also interviewed them about their perceptions of self-driving, both before and after participation.  

Factors to Avoid Crashes

We were especially interested in the position of their hands and feet when a request to intervene occurred with a Time to Collision of 2.2 seconds while navigating a two-way curve road. We found that at the time of the event:

  • Only 12% of participants kept their hands on the steering wheel.
  • Only 64% of participants had their foot close to the pedals.
  • Reaction time was understandably an important factor in the driver's ability to avoid the crash:
    • Participants who reacted within 0.65 seconds of the alert were able to avoid the crash.
    • All participants who reacted after 0.9 seconds of the alert crashed. 

As manufacturers continue to deploy automation, not just on highways, but also in urban environments where pedestrians and bicycles offer formidable challenges to drivers, it’s crucial that owners of highly automated vehicles understand the need to stay alert. They need to be ready to intervene when automated navigation needs to be taken over. Additional training may be required for this type of vehicle to keep everyone safe on the road.