New study could change what, how and where you drive

A game-changing look at how people drive and at the highways they drive on is coming, as researchers begin to sift through data gathered by a Naturalistic Driving Study (NDS) conducted over the past few years.

The study is part of the Second Strategic Highway Research Program (SHRP 2) and is meant to accelerate “solutions for highway safety, renewal, reliability and capacity” and update research conducted nearly one-half century ago during the country’s “Interstate Decade” (1956-66).

Information gleaned from the NDS may provide answers to important highway safety questions, help engineers design safer roadways, guide the manufacturing of safer vehicles and assist lawmakers in creating traffic safety laws that are more likely to save lives.

“Driver behavior is a critical factor in nearly all traffic crashes,” writes Kenneth L. Campbell, who is chief program officer, safety, with the Second Strategic Highway Research Program.

“Driver impairment – primarily due to alcohol – and driver inattention, distraction, drowsiness, and judgment-related errors are believed to be responsible for significant increases in crash risk,” he explains. “After-the-fact crash investigations, however, cannot determine accurately a driver’s behavior before the crash.”

The naturalistic driving method for gathering research data offers the advantages of detailed and accurate pre-crash information, including driver behavior, and exposure information such as frequency of behaviors in normal driving and the larger context of contributing factors.

“The larger context for exposure enables risk estimates for various driver behaviors and for other contributing factors,” he writes. “The information will support the development of new and improved safety countermeasures to prevent traffic collisions and injuries.”

The study, a product of SAFETEA-LU, began in 2006 at a cost of $170 million for the seven-year project. The research effort was extended by two years – to March 2015 – and granted an additional $60 million for pre-implementation activities.

Campbell called SHRP 2 NDS “the largest study of its kind ever conducted.” Researchers equipped more than 3,100 cars nationwide with cameras, sensors and other tools to capture driving and behavior data.

Efforts were made to have the study volunteers be an accurate representation of typical U.S. drivers. Each participant received an annual incentive of $500, paid in installments. Participation required that the driver provide access to the vehicle, so that the hard drive with accumulated data could be removed and replaced every four to six months. Researchers did not want to include people who enjoyed reckless behavior like drinking and driving.

Besides gathering data on driver behavior, researchers also collected roadway data in the six communities where driver data was being gathered: Bloomington, Ind.; Erie County, N.Y.; Raleigh-Durham, N.C.; Tampa Bay, Fla; Central Pennsylvania and Seattle, Wash.

The hope is that by gathering driver and roadway information simultaneously, the data should show how people interact with their vehicles and how the condition and shape of roadways influence their behavior.

The Virginia Tech Transportation Institute is managing the driver behavior data and the Center for Transportation Research and Education of Iowa State University is developing the roadway information database.

Roadway data include number of lanes, lane type and width, the grade, the superelevation or cross slope, the beginning and end points of a curve, the curve radius, the lighting, the rumble strips, the median type, the width of the paved shoulder, the speed limit signs and their locations, the location of intersections, the number of approaches, and the traffic-control devices.

Approximately 12,000 miles of urban and rural roadway were measured in both directions for the six research sites.

Four petabytes of raw data have been gathered so far, equaling four million, one-gigabyte flash drives. Full analysis of the data could take years because the ability to capture raw data has increased dramatically, but the ability to analyze that data is still a slow process.

Current design of pavements and bridges largely follow the results of the landmark American Association of State Highway Officials (AASHO) Road Test conducted from 1958 to 1960 on a specially built, seven-mile test site in Ottawa, Ill. At a cost of $27 million, it was the largest road experiment of its time.

Published 12-20-13