What Happens When Autonomous Vehicle Technology Fails?

Autonomous Vehicle (AV) technology is quickly expanding its market and becoming more widespread on US roads. Current vehicles still require the presence of a human driver and are thus regarded as "semi-autonomous." The role of the driver is an important one, as it requires both the constant monitoring of the vehicle's operation and outside environment, as well as the capability to immediately takeover control of the vehicle should a failure of the autonomous technology happen. But can a driver react in a safe and efficient manner after an autonomous technology failure scenario?

The Mineta Transportation Institute recently published a new report, Analysis of Disengagements in Semi-Autonomous Vehicles: Drivers’ Takeover Performance and Operational Implications that explores the answer to such a question.

The study placed human drivers in a high-fidelity integrated car simulator capable of handling both human and autonomous driving. A population of 40 individuals was tested to understand how quickly and efficiently drivers could takeover vehicle control if the autonomous technology (AT) failed. “In the safety-critical situation of an AT disengagement, it is important to ensure that the human driver has enough time to react and respond effectively to the vehicle’s request for human control,” says Dr. Francesca Favaro, the project’s Principal Investigator in a short video interview about her collaboration with the Mineta Transportation Institute.  

“We tested control takeover metrics including response times and vehicle drift from centerline under several different scenarios,” noted Favaro. “Despite a small sample size, our results have important operational implications.”

Key results included:

• Of the two speed settings selected for the study (high speed of 65 mph and low speed of 55 mph), the low-speed category yielded better performance for all test subjects.
• Three age groups were tested (18-34 years, 35-54 years, and 55+ years) and the older drivers performed best in terms of both maximum drift and comparison between conventional manual driving and driving after AT failure.
• More than three-quarters of drivers resorted to acceleration and steering rather than braking and decreasing the vehicle’s speed after they took over the vehicle control, and one-third of those actually thought they had braked instead of accelerating when asked to recall their reaction to the disengagement.
• All participants received both an auditory and visual warning for the disengagement, but half of them reported not seeing the visual warning on the central console.

From a regulatory standpoint, the preliminary results point to the importance of setting thresholds for maximum operational speed of vehicles driven in autonomous mode when the human driver serves as back-up, perhaps warranting a lower speed limit than conventional vehicles. This research shows that the establishment of an operational threshold could reduce the maximum drift and lead to better control during takeover. Unintentional drift also attests to the need for discussions on possible dedicated lane usage for autonomous vehicles and separation from conventional traffic, as well as for the possibility of increasing lane width in dedicated lanes for semi-autonomous vehicles. With regard to the age variable, neither the response times analysis nor the drift analysis provide support for any claim to limit the age of drivers of semi-autonomous vehicles.

ABOUT THE MINETA TRANSPORTATION INSTITUTE

At the Mineta Transportation Institute (MTI) at San Jose State University (SJSU) our mission is to increase mobility for all by improving the safety, efficiency, accessibility, and convenience of our nation's’ transportation system. Through research, education, workforce development and technology transfer, we help create a connected world. MTI was founded in 1991 and is funded through the US Departments of Transportation and Homeland Security, the California Department of Transportation, and public and private grants. MTI is affiliated with SJSU’s Lucas College and Graduate School of Business.

ABOUT THE AUTHORS

Dr. Francesca Favaro is an Assistant Professor in the Department of Aviation and Technology in the College of Engineering at San Jose State University. Dr. Favaro has authored several journal publications and conference proceedings on a variety of topics, and has been interviewed by multiple media outlets including the Wall Street Journal, Wired Magazine, and Verge Tech forum as a leading expert in the field of automation safety. Dr. Favaro is an FAA Aviation certified Advanced Ground Instructor, a certified Remote Pilot for drone commercial operations, and a solo-endorsed pilot. Sky Eurich, Syeda Rizvi, Shivangi Agarwal, Sumaid Mahmood, and Nazanin Nader are all students at San Jose State University.

###

Contact:

Irma Garcia, MTI Communications & Workforce Development Coordinator

408.924.7560

irma.garcia@sjsu.edu