By Jim Pearce, Carolyn Krause, and Bill Cabage
Patricia Hu and Jennifer Young, both of ORNL's Center for Transportation Analysis, paint a foreboding picture of life on America's crowded highways. Their work shows that rush hour has become nearly an all-day affair,
stretching from dawn until dusk, with only a brief midmorning lull.Their report, based on a U.S. Department of Transportation survey of more than 21,000 households across the country, suggests a number of reasons for this trend. More drivers are driving more cars more often than ever before. More women are working outside the home, and more women are getting drivers' licenses. Teenagers are driving more too—nearly twice as much as they did in the 1960s. Even retirement-aged folks are getting in on the act, driving 40% more than they did 25 years ago.
And society itself is changing: Referring to urban sprawl, Hu says, "You can't really compare driving habits people had 30 years ago with the way we drive today. Thirty years ago, people could walk to the store, to school, even to work. Most of us can't do that anymore."
Carmakers and policymakers have been taking these trends into account for some time. So have scientists, who see technology as the source of solutions to the issues the travel boom has brought about—issues such as increases in traffic congestion, energy consumption, and emissions of pollutants that may threaten personal health and the global climate's stability.
When materials researcher Geoff Wood considers the range of transportation research going on in Oak Ridge, he sees practically unlimited potential. "We have tremendous opportunities in transportation research," he says, "particularly in the area of passenger cars, if we make the most of our resources." Lighter, stronger, and more durable materials, improved passenger safety systems, more efficient engines, cleaner burning fuels, and state-of-the-art manufacturing technologies are just some of the areas where Oak Ridge scientists are teaming up with researchers from U.S. industry to develop vehicles for the next century.
One hundred years ago, only trees and maybe a few buildings would have occluded
this sunset. Freeway horizons are now a common sight to commuters. The environmental effects of the increase in vehicle travel is the main driver of a
strong and diverse research effort at ORNL.
The transportation revolution that began almost a century ago with Ford's clattering Model T (and paved highways to drive it on) continues to transform our lives. In 1905, there were only 200,000 cars and trucks in the entire world; today there are 200 million automobiles in the United States alone, and the number grows daily. Although it has undoubtedly changed the world for the better, the transportation revolution also carries some heavy baggage—problems that threaten present and future generations' quality of life. For one thing, as vehicles consume more than 60% of our petroleum, they are responsible for more than 30% of the greenhouse gases produced in the United States. Gases emitted by cars and trucks can make the air we breathe unhealthy, may damage the ozone layer that protects us from ultraviolet light, and may eventually cause climate changes that could upset our economy and seriously restrict our standard of living. Regulated pollutants--nitrogen oxides, carbon monoxide, hydrocarbons, and particulates are known health hazards. There is also the possibility—or what many call the reality--that we will simply exhaust ready supplies of the petroleum fuels that power us down the pike.
To help solve these problems and help keep the revolution rolling without derailing the economies it has bolstered, ORNL has joined forces with other laboratories, corporations, and universities. In fact, transportation research in Oak Ridge is a $100-million-a-year business, bringing together world-class scientists with specialties from advanced materials to communications technologies to supercomputing. The goal is to keep people in the driver's seat of a transportation system that will carry us, as well as our children and grandchildren, through the next century.
ORNL's Energy Division has compiled, analyzed, and published transportation data since the mid-1970s, soon after the first gas lines brought on by the 1973 oil embargo shocked Americans into new energy realities. Because the "energy crisis" had such widespread effects on the U.S. population and economy, the U.S. government took an active role in addressing the problems in energy-related areas. That government role has coincided with ORNL's growing involvement with energy and environmental research over the past two decades, including studies of the effects of the increasing amount of carbon dioxide in the atmosphere, to which crowded highways contribute their share.
The chief goals of smart cars and ITS, which until recently was referred to as the Intelligent Vehicle Highway System, are safer and more efficient means of "getting there." Three goals drive the PNGV: One is to improve manufacturing technology, which includes using supercomputers to reduce the cost of developing and fabricating new products. The second is to provide near-term improvements to current vehicles, such as improving and reducing the cost of catalytic converters and developing a lean-burn engine—one that uses more air and less fuel. Third is the magic formula: achieve three times the fuel economy without giving up comfort or performance, at an affordable price.
ORNL's contributions to the Clean Car Initiative are highlighted in color.
Today Laboratory researchers are attacking the transportation problem on many fronts through partnerships with others in academia, industry, and government. The Partnership for a New Generation of Vehicles involves several federal agencies, including DOE, that have teamed up with the "Big Three" automakers—General Motors, Ford, and Chrysler—to rethink what vehicles will need to be and do in the next century. This partnership is developing technologies to make a new generation of highly efficient, safe, reliable, and environmentally friendly cars. The ultimate goal of this initiative is an affordable midsize car that gets 80 miles to the gallon and emits significantly less pollution.
ORNL researchers are using massively parallel supercomputers to improve the design, manufacture, and fuel combustion efficiency of future vehicles. They are using supercomputers to simulate car crashes and assess the effect lighter vehicle materials may have on a vehicle's ability to absorb energy and maintain passenger safety. They are providing flywheel and power electronics technology to an industry team developing a "hybrid vehicle" powered by gasoline and electricity. They are helping automotive suppliers use advanced manufacturing techniques. They are identifying fast-growing trees and grasses that can be converted to alternative fuels such as ethanol.
As more and more traffic clogs our highways, ORNL is working to unlock gridlock by helping to develop smart-car technologies. Part of the solution is evolution—adapting our current transportation system to the demands of a new century. ORNL researchers are developing on-board traveler information displays, collecting data for the design of crash avoidance systems, and evaluating "smart" cars that converse with central control computers about road conditions, safety concerns, and the fastest way to get from point A to point B. These developments won't just help Joe Commuter get to work on time—they'll also help travelers avoid traffic congestion in places like Atlanta, Georgia, during the 1996 Summer Olympic Games.
Most vehicles use gasoline or diesel fuel, which are made from petroleum. The United States now imports close to half (45%) of its petroleum. Reducing its dependence on foreign oil would help the United States improve its trade imbalance and ensure secure energy supplies. From a global point of view, petroleum is a fossil energy source that may be depleted before the end of the next century because less developed countries are buying and using increasing numbers of vehicles as their economies grow. Thus, the need for fuel-efficient vehicles takes on new importance to slow the increase in emissions and delay the depletion of the world's oil supply.
Bob Honea, director of the Center for Cooperative Transportation Research, says, "There is a potentially tremendous demand for vehicles, particularly in Asia and the developing countries, that could overwhelm the world's oil reserves. Current estimates by the petroleum industry itself predict that the world's oil supplies will be gone in 45 to 55 years. To head off that scenario, we need a vehicle that is both affordable and safe and either extremely fuel efficient or fueled by an alternative energy source." (Currently, more than 60,000 vehicles in the United States operate on compressed natural gas even though only 1200 natural gas filling stations are available. By 1997, DOE plans to put 250,000 alternative fuel vehicles in government fleets. These trends are driven by the 1990 Clean Air Act and the 1992 Energy Policy Act.)
Currently, the most favored approach to automotive efficiency is a hybrid vehicle powered by both fuel and electricity. Cars and trucks would be equipped with high-speed flywheels, batteries, or ultracapacitors that store energy accumulated while cruising downhill. Electric motors working in reverse (as generators) would recover energy lost in braking. Flywheels would supplement a small, fuel-powered engine for accelerating the vehicle. Because they spin at thousands of revolutions per second, flywheels store a considerable amount of energy—a reserve that can be called on when a little extra push is needed for passing or climbing a steep hill.
Mark Valk, a University of Tennessee senior mechanical engineering student,
prepares a 1995 Chrysler Neon for a test in the Large Scale Climate Simulator at ORNL's Buildings Technology Center. This hybrid electric vehicle placed
first in its category in the 1995 Hybrid Electric Vehicle Challenge held in
June 1995 in Auburn Hills, Michigan.
The hybrid strategy results from the realization by PNGV planners that reaching the goals set forward for the clean car—including the 80 mpg milestone—will require major advances in a number of technologies. Several of these technologie—gas turbines, flywheels, advanced batteries, and lightweight materials, for instance—have been actively pursued at ORNL over the years, many in programs not directly related to commercial transportation. The hybrid's appetite for technology instead of fuel may prove to be very well suited to ORNL's multidisciplinary scope of research.
ORNL's Buildings Technology Center was recently used for a test of a 1995 Chrysler Neon that was converted to a hybrid electric vehicle by Professor Jeffrey Hodgson and his mechanical engineering students at the University of Tennessee at Knoxville. ORNL tested how fast the car's heat pump warmed the passenger cabin during simulated winter conditions. The car placed first in its category in the 1995 Hybrid Electric Vehicle Challenge held in June 1995 in Auburn Hills, Michigan. The car uses computer controls to coordinate the parallel operation of the engine, which is fueled by natural gas, and the electric motor (powered by a rechargeable battery) to drive the wheels. The success of the competition suggests that future drivers may someday go to service stations not only to fill 'er up but also to plug 'er in.
One way to increase fuel efficiency in vehicles and reduce exhaust emissions is to build them from lightweight materials, such as aluminum, magnesium, and polymer matrix composites (PMCs)—high-strength plastics that are similar to fiberglass. Auto components targeted by ORNL studies are steering mechanisms, suspension systems, valves, structural components, and body parts such as hoods, roofs, and doors. The goal is to fabricate cars that are 40% lighter than today's cars by using steel substitutes; a 10% reduction in a vehicle's mass increases fuel economy by up to 6%.
Lightweight composite automotive body structures of the future will have to
withstand the effects of low-energy impacts,
such as tool drops and roadway
gravel kickups, without significant damage. Here, Rick Battiste (left) and Jim
Corum subject instrumented composite plaques to pendulum drop impacts
(representing tool drops) and projectile impacts from a gas gun (representing
roadway kickups).
You can't simply substitute for a material like steel by making the same component out of aluminum," says Phil Sklad of the Metals and Ceramics (M&C) Division. "You have to compensate and make design changes to take advantage of aluminum's unique properties so that the component works as well as the steel counterpart."
One concern is passenger safety in vehicles made of lightweight materials. These vehicles must be designed so that lightweight materials will absorb energy as well as or better than steel during collisions. Another concern is affordability. The costs of the materials and the processes for manufacturing vehicle components must be competitive with steel.
To help meet these goals, ORNL is involved in several cooperative research and development agreements (CRADAs) with the U.S. Automotive Materials Partnership. One CRADA involves a study of the use of adhesives, rather than fasteners and welds, to join materials such as aluminum to steel or PMCs to aluminum. "Because adhesives are a different joining technique, they will affect how energy is absorbed by the vehicle under different impacts," Sklad says. "Computer models of a car's energy-absorption ability must take into account the role of adhesives."
Many automobile parts are made by forging—forming a hot metal into desirable shapes by compressing it. To reduce manufacturing costs, another CRADA in which ORNL is involved is examining casting, which is less expensive than forging. Casting involves pouring a liquid metal into a mold and letting it solidify into the desired shape.
"The molded part must have certain properties," Sklad says. "These properties are related to certain microstructures which can be attained if casting conditions are properly controlled. Casting conditions include temperature of the liquid, pouring speed, mold design, and solidification rate. If these conditions are not right, the product will contain defects such as pores that can lead to cracking. We are trying to determine how parameters in the casting process should be controlled to achieve the desired microstructure and properties."
Jim Corum of ORNL's Engineering Technology Division is testing the durability of polymer matrix composites that are candidates for structural components of cars. PMCs are matted fibers held together by a resin. The fibers provide strength to the material, but they can break, weakening the component. "What happens to the material if an auto worker drops a wrench on it or if a stone is kicked up into it by a moving car or if oil or windshield wiper fluid spills on it?" Sklad asks. "Corum is characterizing PMC samples, exposing them in test rigs to impacts and various auto fluids, and characterizing them again to see how much they degrade." The High Temperature Materials Laboratory, where much transportation-related research is conducted, was built soon after work began on ceramic materials for automotive gas turbines and low heat rejection diesels in 1983. This jewel in ORNL's crown was made possible by DOE's renewed interest in developing highly efficient and cleaner-running engines for automobiles. These engines—gas turbines and diesel engines for cars and trucks—would operate at temperatures high enough to weaken metals used in internal combustion engines; thus, research was needed on how to form reliable engine parts from ceramics, ceramic-metal composites, and high-temperature alloys—materials that can withstand the higher temperatures needed to extract more energy from fuel and release less pollution. Thus, DOE supported efforts at ORNL and elsewhere to guide industry in developing reliable ceramic and nickel-based components for high-temperature engines and to develop low-cost methods of manufacturing these engine parts.
One of these promising materials is silicon nitride, the preferred material for components of gas turbines because it is strong, hard, and resistant to wear, oxidation, decomposition, and thermal shock at very high temperatures. Researchers in ORNL's M&C Division have contributed to the industrial development of silicon nitride ceramics that can be shaped into reliable engine parts at high temperatures. Their characterization and analysis of the microstructure of various ceramics have steered industrial firms into selecting the most promising types of silicon nitride and modifying their chemistry to make them fabricable and reliable. Silicon nitride parts are being used in high-wear parts and valve train components in large diesel engines. Researchers in the HTML are also proving that ceramics can be inexpensively milled and machined.
How is ORNL contributing to the development of a clean car in the literal sense—one that emits virtually no pollutants and greenhouse gases? For the long term, ORNL's work in developing reliable, low-cost ceramic parts should hasten the manufacture of vehicles by the automobile industry that make use of technologies like gas turbines. For the short term, Ron Graves and his group in the Engineering Technology Division have been working with GM, Ford, and Chrysler engineers on a project to clean up current and near-term automobile engines by developing more effective catalysts that remove pollutants from the exhaust. New catalysts would be needed for the "lean burn" engine, a fuel-sipping design that runs at cooler temperatures. The catalysts currently used require high temperatures; with a lean-burn engine, they wouldn't pass emissions tests. Research toward more advanced catalysts and other projects like it can begin to protect the environment now, while cleaner, more-efficient propulsion alternatives are being developed.
During the heat-treat process and subsequent cooling or quenching, metal parts, especially the teeth of gears, are subject to distortion. If it's severe enough, the part must be reworked, and that's expensive. In the 1980s, a team of researchers led by Gerard Ludtka, then in Y-12's Development Division and now in the M&C Division, developed a method of predicting distortion caused by heat treating by measuring the material and cooling parameters that govern the process. They applied the "Quench Simulator" model to uranium alloy pieces; ORNL and independent lab tests verified the accuracy of the model's predictions. That work was cited in 1995 when Ludtka received the E. O. Lawrence Award for materials science.
The cooling parameters yielded by the Quench Simulator are of obvious interest to the auto industry. The heat-treat distortion CRADA researchers will measure the properties of steels used for automotive gears and incorporate the information into the Quench Simulator model. According to Jim Park of ORNL's Computational Physics and Engineering Division, designers will start out knowing the materials they want to use and then rely on the model to fabricate the part in a way that allows for the heat-treat distortions. "In many cases," Park says, "they might start out with a part that appears to be out of specifications, but conforms after heat treating." Participants in the CRADA include General Motors, Ford Motor Company, the Torrington Company, the U.S. Army, and Illinois Institute of Technology as well as ORNL, Sandia National Laboratories, Los Alamos National Laboratory, and Lawrence Livermore National Laboratory. The CRADA is managed by the National Center for Manufacturing Sciences in Ann Arbor, Michigan.
In another development that promises to pay dividends in the factory, a U.S. car manufacturer (General Motors) has found that ORNL's modified nickel aluminide is more resistant to cracking than the material now being used for heat-treating trays in furnaces. According to Vinod Sikka in the M&C Division, nickel aluminide is not affected by exposure to oxygen and carbon, and thus is more durable in those high-temperature environments. These trays hold automobile parts—valves, ball bearings, and gears—that are hardened (to make them resistant to wear) by heating them in a carbon atmosphere. Longer-lasting nickel aluminide furnace components could save the company a considerable sum of money.
For materials researcher Geoff Wood, one of the keys to reaching ORNL's potential in transportation research is the Oak Ridge Centers for Manufacturing Technology, which could help clear a middle ground between the drawing board and the marketplace, enabling industry to explore the potential of new products and technologies more completely.
The challenge facing Wood and his colleagues isn't just how to make a car out of a half dozen or so pieces of polymer-matrix composite components. It's how to turn them out fast enough to match the industry's production line rate. That means figuring out how to make a part out of fibers and liquid resin and get it onto the assembly line in less than 5 minutes.
"The four basic steps in this process,"says Wood, "are loading the mold with fibers, injecting the resin, allowing the panel to harden, and removing it from the mold. The curing and removing steps aren't a big concern at this point. The areas we need to work on are creating and loading the fiber preform and injecting the resin.
"The first step is packing the fibers into the preform. Right now, this takes on the order of 5 minutes. Once the mold is loaded, we inject the resin—that's what holds all the fibers together. It's important that the resin completely saturates all of the fibers in the preform because the bonding between the fibers and the resin provides the strength of the vehicle. Pumping the resin into a thin mold that is tightly packed with fibers can take up to about 5 minutes. We're working on ways of reducing that to a half minute or so."
Of course, 5 minutes here and 5 minutes there begins to add up, particularly for a group aiming to put a part on the assembly line in under 5 minutes. However, says Wood, "These two processes are usually done in parallel—someone is filling molds with fiber on one machine while someone else is injecting them with resin on another. We'll meet the industry's goal—it's just a matter of time. We're down to 8 or 10 minutes for the whole process now."
In the past, Wood notes, transportation research projects were sponsored by a number of different groups that didn't talk to each other much. Now the U.S. automotive industry is becoming more focused and beginning to speak with one voice. "All the major components of any transportation system—materials, propulsion, and fuels—are here. Now that we're all working toward a common goal, progress can be made much more quickly."
"The automakers identified material performance modeling and process modeling as critical needs to allow them to determine, without many years of testing, how to design and produce cars with new, different materials," Ziegler said. In other words, if a car door is stamped out of a new alloy, will it spring back into a shape within specifications? Trial and error production is too costly for the car business.
"All of this must be known four to five years prior to new model introduction, and little or no risk can be taken in light of the severe consequences of failure," Ziegler said. ORNL's computer modeling could give the industry helpful information on crash effects, weight minimization, tool design and limitations, durability, and finished costs. "We are helping in ways that may be too costly or risky for them to do," he said.
In 1992 ORNL hosted an industry workshop that identifed a wide range of material and process interests in the industries. Those processes focused on forming, joining, alloy development, and finishing technology for high-strength steel, aluminum, and magnesium. Because of the automobile industry's raw material and production costs, ultralight vehicles are years away. New materials, fabrication processes, and knowledge are required to make it a reality. ORNL's best contributions may be through its materials development and computer modeling capabilities.
In addition to predicting how new materials will adapt to manufacturing, ORNL's supercomputers—primarily the Intel Paragon XP/S 150—are also being used for tasks such as modeling combustion processes in automobiles. Perhaps the most visually stimulating work has been produced when the Paragon is used to predict how cars behave when they slam into each other or into other objects.
Describing work being done by Srdan Simunovic of the M&C Division, Thomas Zacharia leafs through computer-generated images of a Ford Taurus colliding with a telephone pole. "You see these areas here and here," he says, pointing to crushed areas between the car's engine and front bumper. "These are crumple zones. Each time you crumple something you absorb some of the force of the impact, so less force is transferred to the passengers."
Not too long ago, solving design and manufacturing problems was a matter of trial and error. Car designs were tested by crashing specially built and instrumented cars into obstacles and studying the results. This was not only expensive—about a million dollars per crash—but it was also time-consuming. It took months to thoroughly test a design—inevitably increasing the time it took to get new products to market. In the last few years, however, supercomputers have shouldered much of this load by providing detailed computer models of collisions. The idea is to crunch numbers, not cars, to get the answer. This alternative has turned out to be considerably cheaper than crash-testing, but, until recently, it still took 6 weeks to simulate a single crash on a conventional Cray supercomputer.
Instead of using a single microprocessor, or "brain," to solve a problem one step at a time, parallel supercomputers use hundreds or even thousands of smaller brains to break a problem into pieces and solve them all at once.
"We're one of the few groups in the country putting together high-performance parallel computers with manufacturing sector applications," says Zacharia, who originally headed up ORNL's effort to apply supercomputing to automotive design. Using the Intel Paragon XP/S 150 supercomputer, a half-dozen crash scenarios can be modeled at once, in hours instead of days or weeks, helping designers to explore a range of alternatives to optimize the use of lightweight materials.
"Modeling on the parallel computer not only promises to combine the advantages of new lightweight materials with the safety of today's vehicles," says Zacharia, "but it should also knock weeks off overall vehicle development time." This head start helps get new designs off the drawing board and on the market sooner and for less money—and gives U.S. auto manufacturers a jump on the competition.
As described at the beginning of this article, ORNL Center for Transportation Analysis researchers Patricia Hu and Jennifer Young gather statistics that tell of a day-long urban traffic jam where pedestrians have become about as obsolete as the horse. They published their analysis of findings from a U.S. Department of Transportation (DOT) survey in the 1990 Nationwide Personal Transportation Survey. DOT collected data for a 1995 study, Hu says, and she's pretty confident that the number of cars and drivers on the road will show a continued increase.
Hu and Young's report brings to the surface a number of interesting developments in the activities of American drivers, and in American culture itself. The rate of increase in the number of U.S. licensed drivers from 1983 to 1990, for instance, was double the rate of population growth. The high rate is attributed to more women in the work force and more women who drive. The surge in traffic between the morning and afternoon rush hours, Hu says, is partly the result of an aging population—out in the noonday sun to avoid the rush hours. These same retirees have driven all their lives and have seen American culture become centered largely around the automobile. They've seen the corner grocery and hardware stores virtually disappear, replaced by superstores almost inaccessible to the foot traveler.
In the face of a never-ending rush hour, rising gasoline taxes, and a growing hole in the planet's ozone layer, you'd think the public would be clamoring for greater access to transportation alternatives, such as buses and trains. "We asked people about that," says Hu. "They'd still rather use their cars—it's more convenient."
Traffic is already unacceptably clogged in many U.S. cities, although devoting land and resources to building new highways is increasingly difficult. Adding one lane of interstate highway can cost $30 million per mile in an urban area and $10 million in the country. As a result of the increased number of vehicles on the highway, roads and bridges are deteriorating fast, energy and time are wasted, more carbon dioxide and pollutants are needlessly discharged to the air, and more accidents occur, causing property damage, injuries, and deaths. Mishaps kill 40,000 Americans a year; property losses cost $350 billion a year. Overall, a dilapidated highway system jammed with traffic could ultimately stall economic growth and curb U.S. competitiveness.
Will telecommuting—working at home and communicating with the office by computer—cut down on traffic congestion? David Greene, Ed Hillsman, and Amy Wolfe in ORNL's Energy Division developed a computer model to study this question.
"We found that telecommuters will drive fewer vehicle miles, reducing fuel consumption and reducing congestion," Greene says. "Remaining traffic would move more efficiently, further cutting fuel consumption. However, reduced congestion might induce drivers who normally avoid heavy traffic to use highways more during the day, increasing fuel consumption. Because telecommuting allows people to live and work farther apart, increased urban sprawl could result, increasing driving distances and fuel consumption. On balance, telecommuting appears to provide significant reductions in fuel consumption, thus decreasing emissions of carbon dioxide."
Telecommunications technology, particularly the broadband information highway, could further reduce fuel consumption, Greene says. "People will drive less if they rely on the information highway to do their shopping and banking, to pay their bills, to send their letters, and to provide education and medical advice."
State and local governments, national laboratories, universities, the transportation industries, and companies that were defense contractors during the Cold War have teamed to work on constructing the Intelligent Transportation System (ITS) under the leadership of the U.S. Department of Transportation. Smart cars and smart highways are now technologically feasible because of recent advances in computing, communication, display, and sensing technologies.
"One of ORNL's greatest successes has been our work with the ITS," Honea says. "Our involvement has grown from a few hundred thousand dollars worth of research to nearly $20 million in 1995. This is an area of research that isn't going away. It is estimated that, in the long term, the technology that will make for safer, more-efficient highways will be a $300 billion-dollar investment, most of which will be shouldered by the private sector.
"In addition to making our highways safer," Honea says, "ORNL is a partner in improving the safety of the cars we drive. We are doing research in car crash avoidance that will contribute to the design of crash avoidance systems."
When ORNL researchers first got involved with the ITS program, the only thing they had to worry about was the simulation of traffic flow on highway networks. Today the ITS program looks at virtually all aspects of applying advanced technology to transportation systems, including safety, navigation, congestion, and environmental impacts. "We have broadened our horizons somewhat," program director Ajay Rathi notes wryly.
Building on their experience with modeling traffic flow, ORNL researchers' ITS traffic research is spread over 20 different projects. Their studies range from mathematically modeling an ITS system to developing the electronics and other technologies that will make intelligent vehicles and highways a reality.
"Probably our most important project right now is the Real-Time Dynamic Traffic Assignment Model," Rathi says. "Eventually, the system will consist of a network of traffic sensors around a metropolitan area, for example, that communicates with a central traffic computer. Electronic information from on-the-road sensors will be supplemented with information on what traffic is usually like this time of day and instantaneous updates from drivers calling in on cellular phones. Vehicles will be equipped with an on-board navigator that will garner up-to-the minute traffic information from the central system on alternative routes, the time a certain route is likely to take, and `forecasts' of expected traffic conditions. This effort will involve hundreds of miles of roads and thousands of vehicles. It may well stretch the limits of computing and mathematical modeling." Over the next 5 years, ORNL's job is to demonstrate the model under real-world operational conditions in Atlanta and beyond to provide real-time route guidance to drivers.
Considering all of the electronic equipment that intelligent vehicles will
require, will there be any trunk space left? Thanks to the microchip and
miniature technology, ORNL researchers can conceal electronics in the rear-view
mirror.
Another project of Rathi's group is working with the Federal Emergency Management Agency to develop software to model emergency evacuations. This simulation is designed to help planners cope with the overwhelming number of variables that come into play during an evacuation.
"Ideally, people either leave town in their cars or stay where they are and put on protective equipment," says Rathi. "But things aren't usually that simple. Variables such as reaction time, number of people being evacuated, the types of vehicles they're driving, the area of the evacuation, destinations of the evacuated, human behavior, traffic flow, and other parameters are being considered in the simulation."
The Cognitive Systems and Human Factors Group of the Intelligent Systems Section within ORNL's Computer Science and Mathematics Division is developing an Advanced Traveler Information System. The group, led by Bill Knee, takes a broader view of transportation safety. Using technology originally developed by the U.S. military for helicopter navigation, the system alerts drivers to potential traffic hazards and slowdowns on the way to their destination and suggests the quickest ways around them.
Mobile displays will also be installed in 200 test vehicles and will offer five levels of information, ranging from simple warnings, such as "construction ahead," to an electronic road map showing the location of the vehicle, areas of traffic congestion en route to its destination, and suggested alternative routes. "The navigation display can really be valuable to the driver," says Richard Carter, a staff psychologist in the group. "It gives you an up-to-date, blow-by-blow description of what's going on down the road, so you don't end up stuck in traffic."
Carter has high hopes for the system. "We are using the most sophisticated technology available to reduce traffic congestion and move people to their destinations," he says. "All of the displays will communicate with a traffic control center, which will process a constant stream of data from road-based traffic sensors around Atlanta and from the Global Positioning System satellite overhead." The result will be unprecedented: an up-to-the-minute traffic advisory, custom-tailored to the traveler's location and destination.
"It's the least we can do," says Carter. "The world is coming to visit."
Recently, the Cognitive Systems and Human Factors Group arrived at functional requirements for an In-Vehicle Signing system to be developed for the Federal Highway Administration. An In-Vehicle Signing system will bring information from roadway signs, signals, and pavement markings into the vehicle and present it to the driver. By capitalizing on advances in computing, communications, and display technologies and by applying principles of human factors psychology, In-Vehicle Signing will be designed to tailor the display of sign messages to the individual driver and to the current driving situation.
The In-Vehicle Signing system has three functions important to the driver. One is an information filter to ensure that only messages important to the driver are displayed (such as directions for a particular route number). The second is a display that takes into consideration lighting, driver preferences, and other simultaneous messages; it also monitors whether a driver is responding to messages (stop signs, for instance) and adjusts the display accordingly. The third is the timing function, which determines the onset of a sign display to give the driver and vehicle enough time and distance to respond; it displays relevant sign messages for the entire time that they apply (such as speed limits) and for only the time that they apply (such as a traffic light).
Work on an In-Vehicle Signing prototype is scheduled to begin in 1996. Eventually, this system will play a major role in controlling the vehicle, whether it is being driven by a human driver or piloted by an automated highway system.
Steve Allison, David Howell, and Gary Capps, all of ORNL's Engineering Technology Division, are working with Supercond Technology, Inc., in Atlanta to develop a traffic monitoring system, again for the Summer Olympic Games in 1996. The goal of this CRADA is to improve the traffic flow during the games in Atlanta, which is noted for its traffic even in normal times. ORNL researchers will conduct tests to guide the development of smart structures made of sensors embedded in road materials. The sensors must be able to endure extreme temperatures and other harsh road conditions and be compatible with asphalt and concrete. Staff at computer control centers will use smart structures incorporated in major highways to monitor the pace of traffic and identify slowdowns.
Frank Barickman uses a voltmeter to check a laser device on a vehicle outfitted
to collect data on the behavior of drivers in various situations. The
information could aid the development of crash-avoidance technologies.
Richard Carter programs a test vehicle with a laptop computer. ORNL researchers
are working with Scientific Atlanta to equip automobiles with data acquisition
systems to obtain information on driver behavior.
Carter'steam, assisted by Scientific Atlanta and other partners, is developing an outfitted car called DASCAR (data acquisition system for crash avoidance research) for acquiring information on driver behavior and responses during accidents and near misses. The work is being performed for the National Highway Transportation Safety Administration (NHTSA). This advanced data acquisition system provides an up-close-and-personal look at drivers' responses to traffic problems courtesy of a fleet of test vehicles outfitted with high-speed computers, video cameras, and motion sensors. All of this gadgetry gives each vehicle's on-board computer a precise picture of the traffic around it, and a satellite hookup enables the vehicle to know exactly where it is by tapping into information from the Global Positioning System satellites. This tracking system is so accurate that it can determine the vehicle's position along the road and even within its own lane. Computers also take note of other driving-related factors, such as the driver's heart rate and brain functions, as well as braking, steering, and whether the radio and wipers are on or off.
This wealth of data will enable researchers to determine exactly what happens before and during accidents or near misses. "These specially equipped vehicles," says Carter, the principal investigator, "will be used both on a test track and on the open road to test the effects on driver performance of various conditions ranging from the weather to traffic congestion."
DASCAR, which is noted for being the first CRADA associated with the ITS program, has been successfully demonstrated to NHTSA and other DOT officials. It was so well received that it was also paraded before members of Congress and the national press corps. This year a fleet of 15 vehicles is scheduled to make its debut—first on the track and then on the road. Carter hopes these tests will shed some much-needed light on the 1-second window before a crash or a near miss. "Most vehicle crashes," he says, "could be prevented if the correct action is taken within a half second to a second before the collision."
Using the data provided by the smart car project, Carter envisions the development of an intelligent cruise control that could react to changes in the speed and distance of other vehicles—keeping both driver and car out of harm's way. "The ITS project lays the groundwork for a truly intelligent vehicle," he says. "If we can use this system to determine what information a driver needs to escape collisions, the next step will be to develop a computer-controlled vehicle that can take control of its own movement when an accident is imminent and avoid it."
"Ultimately," says Bronzini, "having the data made available by the government ought to reduce industry's cost for this kind of information. Customers will eventually take the data we're providing and modify them to suit their requirements." Complete and accurate data on the nation's transportation resources are essential for developing policy and increasing the efficiency of industy and commerce through improved traffic flow. Over the past few years, Bronzini says, ORNL researchers have been assessing the quality and coverage of transportation data and have begun the task of filling in the gaps.
One longstanding gap was found in information on the flow of goods, or commodities. So ORNL is helping the U.S. Department of Transportation fill the void with the agency's national commodity flow survey. Targeting 200,000 establishments that ship out goods 12.5 million times a year, the survey asks, "What are you shipping?", "How are you shipping it?", "How much are you shipping?", and "Where are you shipping it to?" The initial results from this survey were published in July 1995. Proper use of this information should lead to improved efficiency in traffic and commodity flow.
Another gap is being filled by the Department of Transportation's American Travel Survey, in which ORNL is participating. This computer-aided telephone survey targets trips of 100 miles or more. It tracks travelers' modes of transportation, the distance traveled on each leg of the trip, and the total length of the trip.
"Similarly," Bronzini says, "we are also helping to lay the groundwork for a new National Highway System by combining the resources of disparate databases, such as GIS databases and the National Highway Planning Network. Our task is to determine the highways that serve a national purpose. We are working with the Federal Highway Administration to build a massive database, covering some 400,000 miles of U.S. roads. This kind of information has a number of applications, including helping policymakers determine where to spend scarce dollars to improve or maintain the nation's heaviest-used highways."
It is hoped that use of these data systems will result in improved efficiency in construction and operation of highways to reduce traffic congestion, energy consumption, and greenhouse gas emissions.
Fortunately, a revolution in computing, communication, display, and
sensing technologies can make mid-course corrections in the transportation
revolution. It is now possible to put a new spin on the transportation wheel.
Through development of clean, efficient cars and smart vehicles and highways,
the transportation revolution that threatens to derail economies can get them
back on track. ORNL is contributing to the development and testing of these
technologies. As long as there is a demand for better transportation
technologies, ORNL's transportation research should continue to roll.
[ Next article | Search | Mail | Review Home Page | ORNL Home Page ]