This article is part of our series highlighting IEEE Fellows in celebration of the Fellow program's 50th-anniversary year.
IEEE Fellows were already working on the technology of driverless cars decades before anyone from Google, General Motors, Nissan, or others gave them much thought.
IEEE Life Fellow Robert Fenton, as a professor of electrical engineering at Ohio State University, in Columbus, worked on autonomous vehicle technology as far back as the 1960s. He began his career searching for ways technology could improve how people drive.
“Instead of trying to make drivers perform like an automated system, I decided to just develop the automated system,” he said in an interview in the school’s OnCampus Online newspaper in April 2003.
According to the article, no researcher before Fenton had pursued automated vehicle technology. In 1962 his team at the university built the first automated vehicle, which is also believed to be the first land vehicle to have a computer. Steering, braking, and speed were controlled via onboard electronics, which filled the trunk, back seat, and most of the front passenger side of the car.
Fenton had his self-driving cars stay on course by following a current-carrying wire laid down in the center of the roadway. A large protuberance packed with electronics to sense the current stuck out from the bumper of his early models, like a trailer hitch in reverse.
Computer control allowed the cars to drive in ways that were beyond the ability of the average human driver. The cars could, for example, automatically steer within 5 centimeters of the lane’s center at speeds approaching 129 kilometers per hour. Microelectronic technology later made it possible to house the control equipment inside the vehicle so that an autonomous car would resemble any other vehicle on the road.
Fenton, who is now retired, and his team are also credited with the research behind an automated highway system, in which a car would be guided by the road rather than by the driver. Sensors and communication devices would link the road and the vehicle to maximize driving performance. Their system also relied on radar and inter-car communications to make the vehicles organize themselves and speed along the roadway without drivers. A lack of government funding forced Fenton to abandon his autonomous vehicle program in the early 1980s.
He was elevated to Fellow in 1986 “for contributions to control systems for automatic control of high-speed highway vehicles.”
IEEE Life Fellow Ümit Özgüner followed in Fenton’s footsteps at Ohio State. A professor of electrical and computer engineering, he and other researchers at the school are working to integrate advances in software, sensing and control, and modeling to address weaknesses in autonomous vehicle design. A key issue in their work is the human component, including understanding how people make decisions and how to predict their actions.
Much of his work considers “specific scenarios where the existence of humans throws a wrench into your nice, automated world and [we must design our] way around that,” said Özgüner in “Building Better Autonomous Vehicles,” published in the school’s Department of Electrical and Computer Engineering’s 2011-12 annual report.
He is also working to scale up the capabilities of fully autonomous vehicles so that they’ll operate with other vehicles in so-called mixed-traffic environments, including ones driven by people. Özgüner co-authored Autonomous Ground Vehicles, published by Artech House Books in July 2011.
For his contributions to the development of intelligent autonomous vehicles, he was elevated to Fellow in 2010.
Newly elevated Fellow Alberto Broggi is working to take humans out of the driver’s seat so as to prevent accidents. Ninety percent of roadway deaths could be avoided, he says, simply by letting cars drive themselves.
A professor of computer engineering at the University of Parma, in Italy, Broggi is also founder and director of the school’s VisLab, a leading laboratory in the study of autonomous vehicles. In an interview with The Institute in January 2012, he said, “Car accidents are purely man-made, and so we should be able to prevent them. We have by developing active safety systems.” These include lane-departure warning systems that sound an alarm to the driver; electronic stability control, which steadies a vehicle as it negotiates a curve; and automated cruise control, which uses radar or cameras to slow a vehicle when it comes too close to the one in front.
Broggi is a pioneer in applying machine vision to unmanned vehicles. He began his research in the early 1990s, when few laboratories were investigating the applicability of artificial vision aboard moving vehicles.
In 2010 Broggi’s group embarked on a test run of two driverless vans, referred to as the VisLab Intercontinental Autonomous Challenge, or VIAC. The electric vans were each outfitted with seven cameras, four laser scanners, a GPS unit, and an inertial sensor suite. Two cameras hanging above the windshield and looking forward provided stereovision, which was used for identifying lane markings and the terrain slope. Three synchronized cameras looking out from behind the windshield stitched their images into a 180-degree panoramic frontal view. The laser scanners—three mono-beam and one four-plane laser beam—detected pedestrians, other vehicles, and obstacles, including potholes. The vans drove for 3 months without drivers, averaging about 160 km per day at a maximum speed of 70 km/hr running four two-hour stretches each. At the time, the trip—from Parma, Italy, to Shanghai, China, a distance of more than 15 000 km—was the longest ever for undertaken by driverless vehicles.
Broggi was elevated to Fellow this year “for contributions to the design of automated vehicles.”
Like Fenton, Broggi believes that an automated highway system will be key to the success of driverless cars.
In an interview published in January in the New York Daily News, Broggi proposed an automated highway with dedicated lanes for autonomous cars. “Much like today’s HOV [high-occupancy vehicle] lanes on busy highway corridors, these proposed ‘self-drive lanes’ would provide a more controlled environment for the first generation of autonomous vehicles.” He explained that cars could detect one another, maintain safe distances, and, theoretically, travel much faster than is possible with drivers on a standard highway. Once the benefits become clearer, Broggi believes, “The market for self-drive cars will grow.”
To learn more about the research on autonomous vehicle technology conducted by the three Fellows, visit the IEEE Xplore Digital Library.