Looking at the progress he and his colleagues at the General Electric Global Research Center in Niskayuna, NY, are making, computer scientist Rusty Irving believes that before the decade is out, jet engines and diesel locomotives will diagnose and report mechanical problems without human intervention. Sound far-fetched? Consider that these sensor-riddled machines already transmit performance data via satellite to maintenance crews, who look for subtle changes in pressures, temperatures and fuel flows that could mushroom into mechanical breakdowns. “The next level of this technology is to push this intelligence to the unit itself, let it say, Hey, I’m not feeling well. I think my problem is this; please take a look,’” says Irving.
This desire-to bring enhanced intelligence to transportation systems through information technologies-drives much of today’s research and development in the world of planes, trains and automobiles. Cheaper sensors, faster microprocessors, new generations of wireless communications and ever improving software-whether speech recognition for hands-free car phones or complex virtual-traffic models for creating congestion forecasts-make this infotech makeover of transportation feasible; the pressures on existing infrastructure make it necessary.
In some cases, the transition has already begun. Take highway “toll tags.” Velcroed to the windshields of many cars, they broadcast account numbers when hit with blasts of radio energy at tollbooths, and they also help count cars along the highway. Near the borders of some states, other tag readers help inspectors identify passing trucks and check their safety records. And in a quintessentially American adaptation, some southern California fast-food joints are even beginning to use them to debit purchases from drive-in customers’ accounts.
To piggyback on tag technology, Paul North, a program manager at the Johns Hopkins University Applied Physics Laboratory in Laurel, MD, is developing Web-accessible databases containing truck safety records that can be updated in near real time by federal and state officials. When a truck pulls into a roadside inspection station-or even before it arrives-these databases can be accessed using its tag number. “The idea is to make data available in a very timely, efficient and consistent way, not only to the inspectors, but to the trucking companies and their insurance companies,” North says. These Web-based databases should be implemented in the next two to five years, even at the most remote weigh stations, North says.
Toll tags linked to databases are only one example of how wireless technologies are transforming transportation. A growing number of drivers are using “telematics”-the marriage of wireless communications, computing and satellite-based Global Positioning System technologies-to navigate, find local services, check e-mail and even obtain remote diagnoses of car trouble through embedded sensors. Some two million drivers already use such systems-most commonly, General Motors’ OnStar service.
Even as these technologies get deployed today, a host of auto suppliers and startups are competing to provide future generations of telematics services with faster data connections, better voice recognition and wider market appeal. Delphi of Troy, MI, has a prototype that uses faster wireless-communications protocols and speech recognition software and could slide into standard automotive radio slots. The device bristles with telematics features such as hands-free voice dialing, Internet browsing and satellite radio. Far from an exotic luxury add-on, “this is the telematics that can get into every car,” says Bob Schumacher, general director of Delphi’s mobile-multimedia business group.
There will soon be more infotech under car hoods, too. Carmakers worldwide are competing to incorporate sophisticated, computer-controlled electronic actuators into everything from electromechanical valves to steering and braking devices that could replace traditional hydraulics and mechanical linkages. Systems from companies like Toyota Motor, Honda, BMW and DaimlerChrysler should be among the first to reach showrooms within the next several years.
In the aeronautics industry, a communications overhaul could help optimize flight patterns without compromising safety. This would mean gradually supplanting old-fashioned radio links, which only allow voice communication, with robust data links delivering torrents of information-from weather data to the locations and identities of nearby planes-so that cockpit computers can chart the best routes and provide collision avoidance instructions. While forerunners of such systems are today used only in radar-lacking regions like Alaska, in the long run “there will definitely be a move to more satellite-based air traffic control,” says John Hansman, professor of aeronautics at MIT and a TR100 judge. “There is already a gradual transition to satellite-based systems over oceans and unpopulated regions of the world, like western China.” Still, he says, implementation of such systems for commercial air traffic in developed areas will require overcoming “a lot of nitty-gritty technical and political problems.”
Andrew Barrows, president of Palo Alto, CA-based Nav3D, is one of the people trying to overcome those problems. He is developing a cockpit display technology that merges satellite location data with stored information about the earth’s surface to produce a virtual view of the ground even in bad weather. For him, the “nitty-gritty” work means further pushing the software to establish the extraordinary accuracy and near perfect reliability required to obtain Federal Aviation Administration certification.
Information technologies are also improving the way the air-, rail- and roadways on which we travel are managed. Kara Kockelman, a civil engineer at the University of Texas at Austin, builds computer simulations using data on where people live, how they move around and even their shopping habits to develop new transportation solutions-which corridor would benefit most from a new light-rail system, for example. “Improving highway traffic systems, optimizing rail schedules and even improving manufacturing processes comes down to being able to consider all the options and choose the best one”-a job best done through such modeling, she says.
|Austin, 2067: Projected shifts in where people work (top row) and where they live (bottom row) in segments of greater Austin, TX, in 2007, 2037 and 2067 (left to right) help researchers at the University of Texas at Austin plan optimal regional transportation systems. (Images courtesy of Kara Kockelman)|
Sophisticated traffic simulation models can make more accurate predictions by using data on how drivers behave when behind the wheel, according to Der-Horng Lee, a civil engineer in Singapore. He is developing algorithms that recognize how different drivers react differently to news of a traffic jam: some stay passively on the road, others take the long way around, others cut through city streets. If such behaviors are accurately reproduced using computer-generated scenarios, Lee says, planners can respond to real jams by, for example, adjusting traffic light sequences on side streets.
For some of these technological possibilities, the only remaining obstacles are human. “As the complexity and information flow in these systems increase, the human will become-or in some cases already has become-the limiting factor in the design and operation of the system,” Hansman says. Which may explain why GE’s Irving is striving to take us out of the equation. Even as he works on self-diagnosis for jets and trains, Irving envisions machines performing self-repair on the fly-literally. “This is really out-there stuff, but this is the long-range vision,” he says. And if the collective computer-centric predictions of researchers like Irving pan out? Getting from here to there will become easier and safer.
A quick guide to the most important AI law you’ve never heard of
The European Union is planning new legislation aimed at curbing the worst harms associated with artificial intelligence.
It will soon be easy for self-driving cars to hide in plain sight. We shouldn’t let them.
If they ever hit our roads for real, other drivers need to know exactly what they are.
This is the first image of the black hole at the center of our galaxy
The stunning image was made possible by linking eight existing radio observatories across the globe.
The gene-edited pig heart given to a dying patient was infected with a pig virus
The first transplant of a genetically-modified pig heart into a human may have ended prematurely because of a well-known—and avoidable—risk.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.