Skip to main content

NC State News

Automobile Control Research Opens Door To New Safety Features

Researchers from North Carolina State University have created a computer program that allows a car to stay in its lane without human control, opening the door to the development of new automobile safety features and military applications that could save lives.

“We develop computer vision programs, which allow a computer to understand what a video camera is looking at – whether it is a stop sign or a pedestrian. For example, this particular program is designed to allow a computer to keep a car within a lane on a highway, because we plan to use the program to drive a car,” says Dr. Wesley Snyder, a professor of electrical and computer engineering at NC State and co-author of a paper describing the research. “Although there are some vision systems out there already that can do lane finding, our program maintains an awareness of multiple lanes and traffic in those lanes.”

Researchers have written a program that uses algorithms to sort visual data and make decisions related to finding the lanes of a road, detecting how those lanes change as a car is moving, and controlling the car to stay in the correct lane.

Researchers have written a program that uses algorithms to sort visual data and make decisions related to finding the lanes of a road, detecting how those lanes change as a car is moving, and controlling the car to stay in the correct lane.

Specifically, Snyder and his co-authors have written a program that uses algorithms to sort visual data and make decisions related to finding the lanes of a road, detecting how those lanes change as a car is moving, and controlling the car to stay in the correct lane.

“This research has many potential uses,” Snyder says, “such as the development of military applications related to surveillance, reconnaissance and transportation of materials.

“This computer vision technology will also enable the development of new automobile safety features, including systems that can allow cars to stay in their lane, avoid traffic and gracefully react to emergency situations – such as those where a driver has fallen asleep at the wheel, had a heart attack or gone into diabetic shock. This can help protect not only the car that has the safety feature, but other drivers on the road as well. That’s a next generation of this research.”

A paper describing the research will be presented in Anchorage, AK, May 4-6 at the IEEE International Conference on Robotics and Automation, which is chaired by Snyder. The paper, “Concurrent visual multiple lane detection for autonomous vehicles,” is co-authored by Snyder and NC State Ph.D. students Rachana Gupta and Shepherd Pitts. The research was sponsored by Lotus Engineering.

-shipman-

Note to Editors: The presentation abstract follows.

“Concurrent visual multiple lane detection for autonomous vehicles”

Authors: Rachana Ashok Gupta, Wesley Snyder, W. Shepherd Pitts, North Carolina State University

Presented: May 4-6, 2010, at the IEEE International Conference on Robotics and Automation, Anchorage, AK.

Abstract: This paper proposes a monocular vision solution to simultaneous detection of multiple lanes in navigable regions/urban roads using accumulator voting. Unlike other approaches in literature, this paper first examines the extent of lane parameters required for continuous control of any vehicle manually or autonomously. The accumulator-based algorithm is designed using this fundamental control knowledge to vote for the required lane parameters (position of lanes and steering angle required) in the image plane. The novel accumulator voting scheme is called “Parametric Transform for Multi-lane Detection.” This paper not only adapts predictive control in the image plane, but also detects multiple lanes in the scene concurrently in the form of multiple peaks in the accumulator. This method is robust to shadows and invariant to color, texture, and width of the road. Finally, the method is designed for dashed/continuous lines.

Leave a Response

Your email address will not be published. Required fields are marked *