|
|
|
Actual Hands-off Steering:
And Other Wonders of the Modern World
by Bob Bryant
 |
| In the platoon scenario, eight cars in a tight formation - 6.5 meters apart - at nearly 105 km/h traveled the demo course as coordinated unit with the vehicles "communicating" with each other 50 times per second. (Photo courtesy of California PATH) |
It's magic! Or so it seems -- cars driving themselves.
Well, we know, it's not magic. It's very technical and explainable -- the logical culmination of years of study, development, and testing. But that doesn't detract from the wonder of it all.
It's like the tricks of the master magicians and illusionists; you know that there is a logical explanation for all the apparently supernatural feats, but it is still exciting.
You know the magician did not really saw the woman in half, and the Statue of Liberty did not really disappear. But we're still amazed. Even if it's not magic, the cars did drive themselves -- at least without the help of human drivers.
I know that is true hecause I saw them. I even rode in three of the automated vehicles at Demo '97, the demonstration of automated high-way system (AHS) technologies in San Diego on Aug. 7 to 10, 1997. I rode in a car, a minivan, and a bus, and it was exhilarating to barrel down that 12.2-km segment of the Interstate 15 high-occupancy-vehicle lanes at 105 km/h with the drivers' feet tucked under their seats and their hands in their laps -- truly "hands-off, feet-off" driving. Even though it is a very overused cliche, I couldn't help thinking, "Look Ma, no hands -- or feet!"
Demo '97, -- put on by the National Automated Highway System Consortium (NAHSC), an industry-government-academia collaboration -- was a congressionally mandated demonstration to prove that it is technically "feasible" to use these AHS technologies to significantly alleviate several of the most enduring transportation problems in the United States -- and in the rest of the world as well.
AHS addresses three major concerns, explained Bill Stevens, the NAHSC Program technical director. One is safety; second is congestion; and the third is environmental problems.
Each year in the United States, more than 40,000 people are killed and 5 million people are injured in automobile crashes. Because human error is a leading factor in nine out of 10 crashes and because AHS promises to significantly reduce the element of human error, AHS offers a great potential for saving lives and avoiding injuries.
AHS can reduce congestion and increase mobility in several ways, but primarily, by being able to safely reduce the distance between vehicles, AHS "can double or triple the capacity of our roadways at today's legal speeds and make trips faster and trip times more reliable by avoiding the backups due to stop-and-go traffic and congestion," said Jim Rillings, former NAHSC program manager. Congestion is another leading factor in automobile crashes; so, reducing congestion will also have safety advantages.
Vehicles traveling in a tight, automated platoon with about half a vehicle-length inteival have a dramatic reduction in aerodynamic drag that results in a 20-percent to 25-percent improvement in fuel economy and emissions reduction. AHS will also have great economic advantages. Today's vehicles are about as crash-worthy as it is possihle to make them within reasonable cost. Therefore, the automobile companies, as well as the federal government, are now turning to crash avoidance as a way of avoiding injuries and death and also as wav of saving economic losses due to crashes, which amount to approximately $150 billion per year. The economic losses due to highway congestion are in the neighhorhood of $50 billion per year. Adding those up, a sizable amount of money is lost each year due to motor vehicle crashes and congestion," Rillings said.
Different approaches to AHS were showcased in seven different "scenarios" during the demo. Cutting-edge technologies to provide adaptive cruise control, collision warning, obstacle avoidance, lane departure warning, and lateral and longitudinal control (steering and interval) were used to show variations on an AHS of the future.
The 1,350 passengers who rode in the Demo '97 vehicles were the first people to experience such a diversity of AHS technologies in a real-time, real-world setting. In a survey conducted after the rides, 98 percent of the riders said they believe AHS technologies can help to improve highway safety, and 87 percent believe the technologies can help to provide relief from urhan highway congestion.
All together, more than 3,500 people attended the demo and exposition, and the demonstration vehicles drove more than 16,000 kilometers in the automated mode. This was the biggest AHS event in history with hundreds of foreign observers from around the world, principally from Europe, Japan, and Korea.
And because of the widespread media coverage -- in the largest and most influential newspapers in the country, through the Associated Press to more than 700 other newspapers, and on major network television, cable, and radio news programs -- most of the country was aware of the demonstration. In addition, media representatives from many other countries, including Brazil, Canada, Denmark, France, Germany, Great Britain, Japan, and Switzerland, covered the event, and this international coverage played a strong role in establishing the United States as a leader in AHS development.
The demo showed that an automated system can work in a real-world environment. But it also showed that there is more than one way to get the job done. Some of the demonstrated technologies perform the same functions but with different advantages and disadvantages. For example, several different approaches to lateral control were demonstrated using cameras, magnets, radar-reflective tape, and lasers.
The camera or vision-based system uses a small television camera that is approximately 3 centimeters cubed and costs about $150. The camera works in concert with an onboard computer, sophisticated software, and vehicle control actuators to guide the vehicle to stay within the lane markings. A major advantage is that this system works without any structural changes to the highway and can be operated easily in mixed (automated and nonautomated) traffic.
As a matter of fact, this system has been tested extensively by researchers at Carnegie Mellon University. One of those researchers is Todd Jochim, who has been commuting to and from the university in normal traffic on the roads around Pittsburgh, Pa., in a 1996 Pontiac Bonneville equipped with adaptive cruise control and a vision-based system for lane-keeping. Jochim "drove" this vehicle to San Diego for the demonstration, and the car drove autonomously -- with Jochin monitoring and ready to take over if necessary -- for 98.2 percent of the way.
 |
| Todd Jochim (standing) of Carnegie Mellon University explains the events of the free-agent, multiplatform scenario, included an automatic lane change, to Secreatary of Transportation Rodney Slater (right forward, closest to display monitor) and NAHSC program manager Jim Rillings (right center). |
The magnet-based system relies on magnets embedded along the center of the lane 1.2 meters apart to define the roadway. The car tracks the magnets with less than 7.5 centimeters of error. Although this requires doing something to the road, it is very cost-effective to spend about $10,000 per lane mile to buy and install the magnets compared to spending from $1 million to $100 million per mile to build new lanes. And the life of the magnets is 30 to 50 years -- longer than the best pavements.
|
|
| In the platoon scenario, the automatic steering control system uses magnetic markers buried along the center of the road 1.2 meters apart to define the roadway. |
The major advantages of the magnet-based system are that it is very reliable in all weather conditions even when the roadway is covered with snow and that it is more precise than the vision-based system. The vision-based system is essentially as precise and reliable as a human driver. Also, by alternating the polarities of the magnets, they can be used to transmit roadway information -- such as the location, direction, and sharpness of an upcoming curve; milepost locations; and location of the next roadway exit -- to the vehicle. The cars in the platoon scenario were guided by the magnets. In the platoon demonstration, eight cars traveled in close coordination under fully automated longitudinal and lateral control. The cars maintained a fixed spacing of 6.5 meters between themselves at all speeds up to full highway speed. The spacing was maintained with an accuracy within 10 centimeters during cruising and 20 centimeters during maneuvers like acceleration and deceleration.
The platoon moves as a unit, accelerating, decelerating, changing lanes, and avoiding obstacles like a "highway train" with the vehicles linked electronically. The cars "talk" to each other comparing information 50 times each second.
I was lucky during my ride in the platoon scenario. We had a "fault," and I was able to witness the safety features built into the system. If any vehicle in the platoon misses just 10 consecutive "packages" of information -- and this occurs within one-fifth of a second -- the platoon automatically goes into the fault management mode, and the vehicles extend their interval to about 15 meters.
The preciseness of the electronic systems is mind-boggling -- the "wow factor" with a capital W -- but it is something with which you can easily and quickly become comfortable.
"I thought I'd be nervous because I hate tailgating, but because you're traveling at a uniform speed under automated control, it's very comfortable," said Cherie Hearne of Berkeley, Calif. "The first five minutes are very interesting and exciting. Then you get used to it, and it's no big deal," said Kin Yen from the University of California, Davis.
As Aleks Gollu of the California PATH (Partners for Advanced Transit and Highways) Program stated, "The purpose of the demonstration is not to give an exciting ride but to provide a comfortable ride."
What's Next
In discussions with the general public about AHS, the first question of many people is, "When will this system be in place?" This is a question that most AHS experts don't want to answer except in broad, general terms because AHS will happen in steps as technologies are made available in a piecemeal fashion. How fast will this happen? NAHSC's "deployment roadmap" predicts an end of operational testing about 2010. But that depends on the willingness of automobile manufacturers to offer these technology applications, the acceptability of these features to drivers, and the willingness of car buyers to spend more money for them. As the public accepts each step, it will lead to additional steps.
"Adaptive cruise control will help people accept a vehicle that controls the throttle and brakes," said Dr. Robert Schumacher, director of advanced development engineering for Delco Electronics Corp. "Cars used to function mechanically; now, more and more functions are electronic ... controlled by software and computers."
"Today, we are already seeing electronic throttle controls, and most vehicles have electronic brake control -- that is, electronically controlled ABS (anti-lock brake systems). Elsewhere in the world, we're beginning to see electronically controlled steering ... So, very quickly, all of the actuators required for automation are on the ve- hicle," Rillings said.
But what about the drivers who don't trust the system or who simply enjoy driving?
"I really didn't want to let go of the steering wheel when they said, 'Let go,' but after seeing what the system could do, I became very comfortable with the safety of the system. It was a scary feeling at first, but now it's OK," said Max Bonton, the operator of the year for the Houston Metro and the driver of one of the Houston Metro buses used in the demonstration.
"It's really exciting for the first 15 seconds; then, it gets really dull. It's like driving with a chauffeur -- an electronic chauffeur," said Rillings.
Even when an automated high-way system is developed, it won't be on all lanes or on all highways. People who like to drive can continue to manually operate their cars. Piecemeal introduction of the technology also helps to make price increases more tolerable.
"A rule of thumb in the auto industry is that any feature on an automobile that costs more than $800 to $1000 is really hard to sell, and you get very low market penetration.... That's the kind of price target we're going to have to meet if this technology is ever to realistically be de- ployed. It will be, after all, a consumer product," Rillings said.
For all of its success, the demonstration, showing a range of technologies, was just an intermediate ohjective of NAHSC. The consortium is about one-third of the way into a seven-year program to develop a prototype automated highway system.
"After the demo, our major focus will be on concept development work. That effort has been underway for two years but is now becoming more concrete," Rillings said. "We will investigate key attributes of these concepts and begin making selections. Essentially, we'll be making design decisions on those concepts, developing and testing the high-risk portions of those designs and eventually focusing on a single AHS system architecture in two years. In early 1999, we should have essentially the system design. Then we will begin to prototype that design."
By 2002, the consortium will build an AHS prototype. That will be followed by about eight years of operational testing and development of an integrated, fully automated system in ordinary traffic.
However, plans can -- and probably will -- change.
"Guidance (from Congress) in NEXTEA (the reauthorization of the Intermodal Surface Transportation Act of 1991) will be critical. We may have several individual capabilities instead of a prototype. Plans are bound to change or evolve as the ISTEA era closes and the NEXTEA era begins," said Dick Bishop, who was the federal AHS project manager until October 31. But nevertheless, AHS is off to a grand start.
"I am impressed by the successful development and demonstration of the Automated Highway System concept and technology and encouraged by the enthusiastic reception by the transportation community and general public at San Diego," said Bob Ferlis, program manager for vehicle-highway automation reasearch and development and Bishop's replacement at the Federal Highway Administration. "I am optimistic that the team will continue to step up to the challenges as this research program evolves into a key element of USDOT's (the U.S. Department of Transportation's) new Intelligent Vehicle Initiative." (See "The Intelli- gent Vehicle Initiative: Advancing 'Human-Centered' Smart Vehicles" in Public Roads, September/October 1997.)
"This demonstration is just the beginning," said U.S. Secretary of Transportation Rodney E. Slater.
"Our aim must be to continue to provide Americans the opportunity to benefit from these and other advanced technologies.... Even now the investment we made in the automated highway system is providing dividends. The technologies developed by the consortium are real and can be made available in cars we purchase early in the next decade. And at the heart of these technologies is ITS (intelligent transportation systems), the federal government's commitment to intelligent transportation and intelligent vehicles.... To date, we have invested $1.2 billion in the ITS program. In 1998, we have proposed investing another $250 million with $53 million for this research alone. Clearly, our investment is paying dividends for the nation." Reference
1. Jerry Werner. "The U.S. Automated Highway System at a Crossroads: A Discussion With Project Manager Dr. James Rillings," ITS Online, August 1997.
Bob Bryant is the editor of Public Roads. He is employed by Avalon Integrated Services Corp, as the project manager/supervisor of an editorial support team in Federal Highway Administration's Office of Research and Development.
 |
TFHRC Home | FHWA
Home | Feedback
United States Department of Transportation - Federal Highway Administration |
