Traffic accidents kill more than 30,000 people each year in the U.S., and David Hurwitz, an assistant professor in civil engineering, is trying to understand why.
“I’m interested in how transportation system users respond to the built environment,” said Hurwitz. “If we can understand drivers’ abilities and limitations, we can design the system more effectively.”
In assessing those abilities and limitations, Hurwitz developed a simulator lab where he and his students can test crash scenarios in a controlled environment. Sophisticated equipment enables him to analyze reaction times, track users’ eye movements and collect a host of other valuable data that may help planners design safer transportation systems.
In the safety of a simulated driving environment, we can expose people to very risky crash scenarios without having them encounter the danger that would be inherent in these phenomena.”
The lab has several unique features, including a specially designed 2009 Ford Fusion that mimics the driving environment. Its equipment includes a rear projection screen and digital computer screens in the side-view mirrors. While virtually in motion, drivers can interact in a realistic three-dimensional setting where visuals refresh at about the same rate humans process information.
“You’re looking at those visuals in real time, just as you might be looking around the room right now,” Hurwitz explained. “It sounds like a bit of a motion odyssey movie ride, but it’s a pretty rigorous research tool.”
Hurwitz and his team also developed a bicycle simulator, integrating it into the lab to allow the two transportation modes to interact. Oregon State has the only lab in the country with such capabilities, and the unique setup will provide important information about how blind spots impact biker safety, how drivers respond to adjacent cyclists and other valuable data.
Although the lab has been running for less than two years, Hurwitz has launched several transportation studies. He and his team looked at the “dilemma zone” — the space before a stoplight intersection where the traffic light turns yellow and the driver isn’t sure whether to stop or go. Drivers can stop too suddenly and risk being rear-ended or speed through the intersection and face a serious crash.
Hurwitz’ team conducted a comprehensive literature review, collecting a spectrum of driver responses such as deceleration rates and the time between perceiving a yellow light and tapping on the brake pedal. Then, in the simulator lab, he replicated intersections described in the literature to see if he could generate similar results. He did.
“We were able to predict what drivers were going to do within a 90 percent threshold,” he said. “The deceleration rate for stopping vehicles and the perception-reaction time of the drivers responding to the yellow light mapped very closely to field-collected data. We think models of that type can be used to change the way we time yellow lights; it could provide information on where we put the advance vehicle detectors and the way we design these systems for safety.”
Hurwitz is also working with construction engineering management professor John Gambatese to study how drivers respond to mobile work zone barriers, which are tractor-trailer trucks driven into place to separate workers from adjacent through traffic. In another study, he is collaborating with civil engineering professor Chris Monsere at Portland State University to assess vehicle-pedestrian accidents that result from flashing yellow arrows — indicating a permissible left turn. Hurwitz would like to understand whether these collisions are caused by incomprehension, inattention or something else.
Given the encouraging results of his yellow-light study, Hurwitz sees a place for driving simulator labs in preventing crashes and helping to improve the built environment.
“As we look toward building transportation alternatives for major projects, I’d like to consider driving simulation as one of the mechanisms for evaluating those designs,” he said. “I think the results that can be acquired in our lab are significant and unique, and there isn’t another way to do that right now. It could be a powerful component to how we build and deliver transportation projects.”
— Abby P. Metzger