It looked like something out of a science fiction movie when the first U.S. test of "driver-disengaged transit" occurred last August just north of San Diego. There, along a 7.6 mile stretch of highway closed to midday traffic, drivers simply took their hands off the wheel, removed their feet from the pedals, and cruised along -- guided by a combination of technologies that kept them within designated lanes, at a specified speed, and at a safe distance from other vehicles and obstacles. As their cars moved along the "intelligent" highway, people read the newspaper or did paperwork, but not one of them crashed, broke down, or were involved in one of the 6.5 million highway accidents that occur in the United States each year.
The recent demonstration, conducted by the National Automated Highway System Consortium (NAHSC), was the latest step toward a goal mandated in the 1991 Intermodal Surface Transportation Efficiency Act (ISTEA). The goal is to get automated highways in use by the general public by 2020, and NAHSC -- a working group of state and federal government agencies, private industry and academia -- has taken the lead in making that a reality.
What's An AHS
Automated highway systems, when launched, are expected to pack as many as three times more cars onto existing roads, sparing the expense of building new roadways while simultaneously decreasing congestion. Dick Bishop, Federal Highway Administration manager for the automated highway system program, said while it now costs between $1 million and $100 million to build a mile of new highway, it would cost less than $10,000 to equip that same mile with automated vehicle technology.
The view from the passenger's perspective inside a vehicle under automated control
Additionally, computer-steered vehicles are expected to eliminate up to 90 percent of the car accidents that occur on U.S. roads. "The solution technology holds is we can make roads safer," said Kyle Nelson, chief information officer at the California Department of Transportation (Caltrans). "Statistics show 90 percent of all accidents are caused by driver error, so anything you can do to give the driver a little more time, alert them they're about to have a collision or have a computer take the controls and assist, is going to save a lot of lives."
NAHSC plans to build special automated vehicle lanes on existing highways. Drivers will then have a choice of whether they want to use the system or not. If they choose to use it, they will enter a special lane that merges them onto the automated system. At that point, the car's technology interacts with the technology on the highway and takes over the driving. The automated vehicles will travel in a closely packed group, allowing them to "draft" quickly along the road, saving fuel at the same time. A driver who wants to get off the highway would enter a "transition" lane, slowly disengage the system and merge off into normal traffic.
NAHSC -- which includes organizations such as the U.S. Department of Transportation, the University of California, Bechtel, Caltrans, General Motors, and Lockheed Martin -- conducted the recent demonstration of the technology that could eventually comprise the $200 million automated system.
NAHSC examined many systems; among them a hybrid system from the American Honda Motor Co. Honda's offering used cameras and radar in rural areas and then switched to under-bumper sensors guided by iron magnets embedded in urban roadways. Another system, from Ohio State University, teamed radar with magnetic strips in the road.
But according to Nelson, the demonstration was more about combining technologies that already exist rather than testing new technologies. "The equipment being used is pretty much off-the-shelf stuff," he said. "What's new are the combinations of different technologies. Some of the vehicles out there were guided by magnets in the pavement, some of them were guided by magnetic tape, others were guided by video systems that literally track the white lines down the side of the road to keep the vehicle positioned. The idea is that we'll start determining which technologies work the best, and which are going to become the standards for the highways of the future."
Some of the components that make automated driving possible are stored in the trunks of the test vehicles.
Jim Rillings, program manager at NAHSC, said the demonstration helped the organization's members learn about what will work and what won't. "Almost everything we tried worked well, which surprised us," he said. "But we did learn that -- at least in the current state of development -- the vision-based technologies are not as reliable as other types of sensing technologies, like magnetic markers."
At the same time, NAHSC was searching for inexpensive technologies that will make automated vehicles affordable to the average consumer. They hope to reduce the costs of the equipment down to the $1,000 range, or about the same as some other high-end options offered on new cars.
Choosing A System
NAHSC won't decide which automated system it will go with until 2000. And, as Nelson points out, it may not be just one system that's chosen. "It may be decided that for highly-congested urban areas, the magnet technology is the way to go; while on long rural stretches of two-way highways, the visual system is the best."
"I think it will end up being a basket of options, with a range of technologies," said Rillings. "Those options will be available to regional transportation agencies to select which best suit their own surface transportation needs. But we do have to maintain national interoperability -- so trucks, for example, that run on an automated system in New York, need to be able to use the same equipment on a system in California." That means NAHSC will have to set certain standards, particularly when it comes to communication between the vehicles and the infrastructure.
Once a system is decided on, NAHSC plans to put together a prototype highway somewhere in the United States by 2002. After that, officials in Washington must decide whether to promote the system for installation by state highway departments.
"Right now, the most important things are to continue the technical development, continue making the systems more capable of operation in a wider variety of conditions, and continue working on ways to make it less expensive," said Rillings.
Will It Fly?
Despite the many advantages automated highway systems afford, the question remains whether Americans -- traditionally viewed as lovers of the automobile and the freedom that goes along with it -- will be willing to use an automated system.
"Yes, because it
will be entirely their choice," said Rillings. "I think that people that want to be chauffeured automatically will do so some of the time and won't other times."
Nelson agreed. "They'll like it once they realize the advantages. And it's not like all of a sudden tomorrow the public is going to find that they're in a fully automated vehicle. We see this as a very evolutionary process. Some of this equipment -- like GPS or smart cruise-control systems that use radar to apply the brakes when you're too close to the vehicle in front of you -- are available now. Every few years there will be new electronic and automated upgrades available to the public in the vehicles they get. That will make for a smoother transition down the line."
As far as California is concerned, Nelson said if even half the population uses the system, it will make an enormous difference. "There are some real problems with congestion on California highways," he said. "We got involved in this because we knew there had to be solutions that are less expensive than adding new lanes to freeways."
So far, NAHSC feels it's on track for success with the project. "We do feel we'll be successful, and a real reason for the success is the joint participation by both the public and the private sector," said Rillings. "We have private-sector companies involved, but we also have both state and federal government agencies involved. That really makes a difference to have everyone working together toward a common goal."
October Table of Contents