GM’s Pioneering Research Into Causes of Distracted Driving

IEEE member shares what goes into determining safety features

17 April 2015

Photo: General Motors

The latest model of GM's Cadillac offers upgraded safety features, including more responsive brakes to prevent accidents.

GM has been a leader in determining what safety features to install to keep drivers’ attention on the road based on studying what causes accidents in the first place.

While reporting how the latest technologies can keep drivers safe (see “Consumer Electronics May Curb Distracted Driving”), I interviewed IEEE Member Scott Geisler, GM’s vehicle performance owner for driver distraction, who shared some of the company’s studies with me. He’s responsible for internal guidelines, processes, and practices the carmaker uses to keep drivers focused.

“For most people, they generally do things right most of the time,” Geisler said. “Deadly accidents, as catastrophic as they are, are still a rare occurrence. That’s because drivers have trained, adapted, and compensated for things very well, but unfortunate things happen. We are trying to reduce those risks by applying technologies without introducing additional or unacceptable risks.”

Here are some of the studies the company uses.


General Motors Research was the first to develop so-called crash typologies, or classifications, to help it design, develop, and assess potential crash avoidance technologies. In 1991 it came out with what is known as “44-crashes,” which classified 44 of the most common motor vehicle crash situations, which was adopted by other automakers. But the study was limited in several ways, including that it analyzed accidents from only two states.

In 2004, the National Highway Traffic Safety Authority (NTHSA) developed another classification. It replaced GM’s and is now used by the carmaker, said Geisler. Its typology consists of 37 pre-crash scenarios that depict vehicle movements and accident dynamics as well as the critical events occurring immediately prior to crashes involving at least one light vehicle. Called “Description of Light-Vehicle Pre-Crash Scenarios for Safety Applications Based On Vehicle-to-Vehicle Communications, the findings account for 99.4 percent of light-vehicle accidents (passenger cars, vans, minivans, and other vehicles weighing 4.5 metric tons or less) and uses data from the National Motor Vehicle Crash Causation Survey, event data recorder databases, and other sources.

According to Geisler, understanding pre-crash scenarios has led to the creation of advanced-technology safety systems such as brake assist, adaptive cruise control, driver warning systems, head-up displays, and even moving up the center console where many of these features are located.

Today, GM’s active safety groups are working on a system comprised of external forward-mounted cameras, radar, and a specialized version of radar called LIDAR. Known as forward collision warning, the system would identify what’s happening in front of the car to warn drivers they are either following too closely or are at risk of running into the car in front of them.

“We are one of the few manufacturers that do this through cameras as well as radar,” he noted.


Geisler said not all accidents are caused by distractions. Some happen because the driver wasn’t prepared or didn’t comprehend the roadway situation. “Someone could be fully attending to driving but they may have missed something that was relevant that could lead to a crash,” he said. “And that’s what crash avoidance systems help with. All these situations are aggregated into active safety features so, rather than looking at the occupant, we look at solving the problems these actions cause,” he explained.  

Geisler also said that people assume distractions and circumstances leading up to accidents are universal around the world but that’s not the case. “There are driving behaviors that are typical in different regions of the world so the question of distraction is not a single dimension globally,” he explained. “For example, if you’ve driven on roads in China, India, or Italy, lane markings are less of a given than in the United States.”


According to Geisler, GM’s research team also created a measurement in 1967 on eye behavior called PERCLOS that is still used today. PERCLOS stands for proportion/percentage of time in a minute that the eye is 80 percent closed. It is a popular topic of research for assessing driver fatigue.

GM has also studied the role the eyes play in understanding how the driver is processing the environment to maintain control of the vehicle. To that end, its European subsidiary Opel announced in March that it had developed a new type of headlight that tracks the movement of the driver’s eyes. The headlights will light up the areas where the driver is glancing. The system relies on a highly accurate camera that reads the driver’s face more than 50 times per second, scanning prominent points, such as the nose and eyes, to detect movement and thereby the driver’s line of sight. The system then translates the information gathered into data commands for electronically controlled actuators, which align the car’s headlight projectors.

At GM, we’re committed to developing and applying active safety technologies that keep our customers safe,” Geisler said.

IEEE membership offers a wide range of benefits and opportunities for those who share a common interest in technology. If you are not already a member, consider joining IEEE and becoming part of a worldwide network of more than 400,000 students and professionals.

Learn More