Crosswords Sudoku and Comics
Science

Scientists Build Single Model That Predicts Human Crash Avoidance Behavior

Researchers at Delft University combined perception, decision-making and steering response into one framework, and Waymo is already using it.

WASHINGTON (July 11, 2019) — The locations of external sensors and communication devices that guided the autonomous shuttle involved in the Nov. 8, 2017, collision with a commercial truck are detailed in this graphic. The Navya shuttle was operating in Las Vegas as part of a pilot program when a com
WASHINGTON (July 11, 2019) — The locations of ext…      Autonomous Vehicle Sensor    NTSBgov / Wikimedia Commons (Public domain)
By Free News Press Editorial Team
Published June 27, 2026 at 1:31 PM PDT

When a car in front of you brakes without warning, most drivers respond within fractions of a second. Whether they brake, swerve, or do both has been difficult to model accurately until now.

Scientists at Delft University of Technology, working with the autonomous vehicle company Waymo, have built a model that predicts how human drivers respond to dangerous traffic situations with humanlike accuracy. The results were published in Nature Communications on June 10. For the first time, different types of collision avoidance behavior are combined into a single model.

Previous tools only captured parts of the problem. According to Phys.org, existing models typically described only reaction time or steering behavior in isolation. The new model brings perception, decision-making and physical execution together into one coherent framework.

Arkady Zgonnikov, an assistant professor at Delft University of Technology, explained what sets the model apart. "Existing models typically describe only part of this process, such as reaction time or steering behavior," Zgonnikov said. "Our new model brings all these components together."

The researchers tested the model against real human behavior in three scenarios: a lead vehicle braking suddenly, an oncoming car entering the lane unexpectedly, and a vehicle failing to yield at an intersection. The model received exactly the same information as the human drivers in each case. "The model showed realistic braking reaction times and made similar choices between braking and steering," Zgonnikov said. The model also incorporates human limitations, so its responses look like something a person would actually do.

Waymo is already applying the model. Mauricio Peña, chief safety officer at Waymo, said the model can help the sector "move toward a shared, scientifically grounded approach to assessing collision avoidance."

Beyond testing autonomous vehicles, the model may have regulatory uses. Zgonnikov said it "can help address whether autonomous vehicles are safer than human drivers, a key question in regulation," and that it makes it possible to set clear, measurable requirements for manufacturers.

The research was led by Julian F. Schumann and colleagues at Delft. The paper, titled Active inference as a model of collision avoidance behavior in human drivers, appears in Nature Communications.

A Gavia Scientific Autonomous Underwater Vehicle dries on the deck of the Coast Guard Cutter Healy after recovery from a simulated oil in ice exercise in the Arctic Aug. 20, 2014. The AUV is outfitted with sonar and radar sensors which allow responders to locate oil trapped beneath or within the ice
A Gavia Scientific Autonomous Underwater Vehicle …      Autonomous Vehicle Sensor    Petty Officer 1st Class Shawn Eggert / Wikimedia Commons (Public domain)