Vision-Based Roadway Departure Warning System

Each year in the U.S. alone, there are over 3 million single vehicle roadway departure accidents, resulting in more than 13000 fatalities and an estimated $100 billion in damage. According to the Federal Highway Administration, up to 53% of these accidents could be prevented by systems which warn the driver if the vehicle is drifting out of its lane.

To help avoid these accidents, we have developed a vision-based lateral position estimation system called AURORA (AUtomotive Run-Off-Road Avoidance). Aurora employs a downward looking vision system consisting of a color video camera with a wide angle lens, a digitizer, and a Sun Sparc portable workstation.

By applying a novel template correlation method, it is able to reliably track lane markers on the road at 60 Hz and estimate the vehicle lateral displacement within an average absolute error of 0.8cm. Based on this estimation, the time to lane crossing is calculated for each image field, triggering a warning alarm when it falls below a threshold. Currently there are three warning modalities: visual, audible, and haptic (vibrating the steering wheel).

Home