Featuring developments in Federal highway policies, programs, and research and technology.
|This magazine is an archived publication and may contain dated technical, contact, and link information.|
|Federal Highway Administration > Publications > Public Roads > Vol. 61· No. 2 > The Intelligent Vehicle Initiative: Advancing "Human-Centered" Smart Vehicles|
The Intelligent Vehicle Initiative: Advancing "Human-Centered" Smart Vehicles
by Cheryl Little
In the wake of the computer and information revolutions, motor vehicles are undergoing the most dramatic changes in their capabilities and how they interact with drivers since the early years of the century.
In 1908, Henry Ford's Model T exemplified major breakthroughs in automotive design. Not only did its interchangeable parts inaugurate easy and economical mass production, but its "user-friendly" operation allowed almost anyone to drive.1 Nearly 90 years later, the motor vehicle is resembling less and less Ford's simple machine and quickly becoming a complex "mobile computer," capable of acting as a navigator, a safeguard, and even, a second driver. These new capabilities will not only change how we drive; intelligent vehicles could also enhance transportation services, save lives, and bolster the competitiveness of U.S. industries.
However, intelligent vehicles aren't quite here. Instead, the components that make vehicles smarter -- new information, safety, and automation technologies -- are arriving on the market as piecemeal accessories, offered either as optional equipment by new vehicle manufacturers or as speciality components by after-market suppliers. These technologies are being developed and marketed to increase driver safety, performance, and convenience. These individual technologies, however, have yet to be integrated to create a fully intelligent vehicle that works cooperatively with the driver.
In particular, the infusion of new and uncoordinated technologies in motor vehicles could potentially distract and overwhelm drivers. An individual could fill his or her vehicle with a smorgasbord of information, safety, and automation technologies (along with portable phone, computer, and fax machine) but suffer information indigestion as these uncoordinated technologies deliver excessive, competing, or contradictory messages and demands.
The automotive industry is already aware of and addressing potential problems associated with the uncoordinated influx of technology. But their progress is hampered by technical and economic obstacles, uncertain consumer interest, and insufficient standards and guidelines. Also, neither original vehicle manufacturers or government regulators (unless safety problems are clearly proven) have control over after-market products, especially their use in trucks and buses. However, without a "human-centered" design approach for the intelligent vehicle that attempts to integrate and coordinate various technologies, we may not only lose the opportunity to realize the benefits offered by new in-vehicle technologies, but we could inadvertently degrade driving safety and performance.
The challenges involved in developing a "human-centered" intelligent vehicle are not unlike those faced by a tailor designing a coat out of many, varied pieces of cloth. The tailor must first understand the proportions and needs of the wearer, then meticulously sew the pieces together to fashion a coat that fits and is also useful and desirable. Similarly, the designer of an intelligent vehicle must integrate disparate technologies and systems to create a coherent machine that complements the human driver.
The Intelligent Vehicle Initiative
Recognizing the importance of smart vehicles and the potential for unintended consequences if human factors are not placed at the center of their design, DOT launched the Intelligent Vehicle Initiative (IVI) in 1997. This initiative aims to accelerate the development, availability, and use of integrated in-vehicle systems that help drivers of cars, trucks, and buses operate more safely and effectively.2
IVI rises to the challenge posed a year earlier by the National Science and Technology Council (NSTC), an executive office of the president, to develop "human-centered" transportation systems in the United States. In the spirit of this national effort, IVI advocates the creation of smart vehicles that fully consider the driver's requirements, capabilities, and limitations. Driver-centered design, however, will mean more than the ergonomics of "knobs and dials." It will also require that designers adopt what the Japanese call kansei -- the infusion of human sensibility into the design process.3
IVI is a multiagency research and development (R&D) effort. DOT's Intelligent Transportation Systems (ITS) Joint Program Office provides a single budget for the ongoing vehicle-related ITS projects of the Federal Highway Administration (FHWA), the National Highway Traffic Safety Administration (NHTSA), and the Federal Transit Administration (FTA). These projects have collectively investigated the human factors, user acceptance, and technical development of individual driver information systems, advanced collision-avoidance and vehicle safety systems, and automated highway systems (AHS). By coordinating activities among DOT's agencies, IVI captures synergies in research and economizes its resources. IVI's role is to research and advance integrated concepts that have safety implications or benefits that will likely not be accounted for by the marketplace or are too far from commercialization to be of interest to most companies.
ITS research has already shown the benefits and feasibility of many of the technologies that will be contained within intelligent vehicles:
Aside from delivering safety and efficiency benefits for the traveling public, the federal government expects that indigenous development of intelligent vehicles could promote America's economic competitiveness.
"The nation that develops and integrates an architecture that provides a seamless interface to the driver will dominate the automobile industry for many years to come," stated NSTC's Human-Centered Transportation Committee.7 With this perspective, IVI is endeavoring to forge alliances with the private sector, particularly auto manufacturers and suppliers. These alliances are crucial since driver-centered intelligent vehicles cannot be produced and marketed without strong industry action and consumer interest.
The production of a safe, effective, and affordable intelligent vehicle, however, will be a long and difficult undertaking -- an undertaking in which IVI must overcome many technical obstacles, better comprehend the complexities and idiosyncrasies of human behavior, and form cooperative relationships with industry to establish a vision for "human-centered" design of intelligent vehicles.
What is a "Human-Centered" Intelligent Vehicle?
The 1980s television series "Knight Rider" featured an intelligent vehicle that could leap moderately tall buildings, drive itself at seemingly supersonic speeds, spy on bad guys, and had the diction and personality of an English butler. The car was not only smart, but smart-alecky. Although intelligent vehicles in the real world will not be able to fly over standing traffic, they will have formidable capabilities. As envisioned by IVI, smart vehicles will be able to give route directions, sense objects, warn drivers of impending collisions, automatically signal for help in emergencies, keep drivers alert, and may ultimately be able to take over driving.
The use of information- and computer-based technologies in motor vehicles, however, is not new. Widescale computerization of motor vehicles began in the 1980s with technologies designed to enhance vehicle operation and driver comfort. These technologies included electronic fuel injection to control engine performance, particularly to reduce vehicular emissions and improve fuel economy, antilock braking systems to help drivers retain control on slippery roads, and cruise control to relieve driver tedium during long stretches of driving. Whereas these technologies were primarily aimed at enhancing the capabilities of the vehicle, the most recent wave of in-vehicle technology, which is of most interest to IVI, are the intelligent transportation systems designed to enhance the capabilities of the driver. These systems include warning and information, driver assistance, and automation technologies.
Just as people possess different specialized abilities, in-vehicle ITS technologies endow vehicles with different types and levels of "intelligence" to complement the driver. Driver information systems expand the driver's knowledge of routes and locations. Warning systems, such as collision-avoidance technologies, enhance the driver's ability to sense what's going on in the surrounding environment. And driver assistance and automation technologies simulate a driver's thinking and physical actions to operate a vehicle temporarily during emergencies or for prolonged periods.
But while a smart vehicle will extend the driver's capabilities, it will also potentially expand the driver's traditional role. In particular, in the midst of new in-vehicle technologies, the human role expands from that of sensory-motor skill, writes Thomas Sheridan, a professor who heads the Human-Machine Systems Laboratory at the Massachusetts Institute of Technology (MIT), "to that of planner, programmer, monitor of the automation, diagnostician ..., learner and manager."8
Integration: The Key to Human-Centered Design
A key criteria of human-centered design is ensuring that a technology provides the intended benefits without engendering unintended adverse consequences. Driving is a potentially dangerous activity that requires attentive and alert drivers. Although technologies in the vehicle can enhance the driver's capabilities and comfort, they can also create potential distractions that transform even the best driver into a road hazard. The National Public Services Research Institute, for example, found that individuals with cellular phones in their cars had a 34-percent higher chance of having a collision.9
"The three tasks associated with cellular-phone use -- placing calls, simple conversations, and complex conversations -- all led to significant increases in the time to respond to highway traffic conditions and in the likelihood of failure to respond at all," noted the institute.10 The uncoordinated infusion of new ITS technologies in motor vehicles could have similar negative impacts on safety and driving performance unless they are integrated to work cooperatively.
Just as the human brain coordinates memory, senses, thinking, and physical reaction during driving, a vehicle needs a coordinating system to integrate the various capabilities of information, safety, and automation technologies. This coordinating system is the defining feature of a human-centered intelligent vehicle. Without it, the vehicle is simply a container for potentially overlapping or conflicting technologies.
Although DOT's ITS programs have developed preliminary human factors guidelines for individual systems, such as driver information,11 collision avoidance,12 and automated highway systems,13 these programs, through IVI, are beginning to explore fully how these systems should be integrated within a "driver-centered" framework. In particular, this framework must specifically address how information will be managed and displayed, how technologies can be improved to better serve drivers, and how drivers can learn to operate intelligent vehicles safely and effectively.
In an intelligent vehicle, drivers will have access to more information than they are traditionally accustomed. This includes information on road and weather conditions, route directions, vehicle diagnostics, anti-collision warnings, the driver's physiological status, etc. As a result, the vehicle itself must do some information filtering, a process called "data fusion." Data fusion, however, is a complicated task, which involves suppressing non-critical information -- such as "the nearest fast food restaurant is two blocks to the left" -- when a safety-critical message -- such as "brake immediately" -- is relayed to the driver.14 Data fusion must also consider how many messages a driver can receive within a given period without becoming overwhelmed or distracted.
Developing Effective and Affordable Technology
A key component of the intelligent vehicle, collision-avoidance technology, is not readily available to the buying public as standard or optional equipment. Many of the technologies used in crash-avoidance systems -- radar detection, head-up displays, location referencing -- have been used in the defense industry for decades; however, transferring these technologies from military applications to the driving environment has been difficult. First, the driving environment is complex; varying patterns of driver and traffic behavior, uneven road conditions, changing weather, and other variables must all be considered and addressed. Second, the technologies must be re-engineered for mass production and made for much less cost. One of the challenges of manufacturers is to produce affordable technologies that are also highly effective and reliable.
Streamlining the Driver-Vehicle Interface
The design of the driver-vehicle interface -- the place where the driver interacts physically and mentally with the vehicle -- is critical. In an article entitled "Safety in Our Hands," published in Traffic Technology International, authors Susan Scott and Eugene Farber astutely remark, "The end result of successful ITS market deployment should not and cannot be a dashboard resembling that of a commercial aircraft."15
"An important safety benefit is achieved merely by making manufacturers' systems consistent -- if the button for 'engage automated cruise control' is always on the same place on the dash and/or has the same icon on the button, it will be easier for customers to avoid dangerous mistakes (such as assuming the automated cruise control is engaged when it is not)," Scott and Farber go on to say.
DOT, as well as the Society of Automotive Engineers (SAE), have already begun to develop guidelines on the ergonomics of individual in-vehicle ITS technologies, particularly driver information and warning systems.
Presenting Information Effectively
A fundamental human factors research challenge is to determine how people best understand and process information. FHWA's human factors research has its roots in trying to design road signs that effectively communicate safety instructions and directions to drivers. These signs, however, can only present fixed information. As a result, a whole new area of language and symbols must be developed to communicate the dynamic information that can be provided by in-vehicle ITS technologies. For example, how should a driver information system communicate that traffic is stalled ahead because of construction? Microsoft and Apple, for example, employ "user-friendly" icons in their software that take advantage of the natural ways that people think about and manage information.
Designers of intelligent vehicles must also consider the media by which messages are conveyed to the driver, such as text via displays, speech via voice recognition systems, or aural "bells and whistles," such as the repeated "dinging" a car makes when keys are accidently left in the ignition. The U.S. Air Force Systems Command, when developing instruments for pilots, found that 80 percent of human knowledge is acquired through visual input. But visual communication is not always best in all situations. For example, NHTSA found in its studies of human reactions to collision warnings that drivers responded best to a voice warning of short discrete words.
Helping Drivers Understand the Technology
Drivers must also understand how to use technology, particularly automation technologies that may be invisible to them. For example, a survey by the AAA Foundation for Traffic Safety revealed that 49 percent of drivers misused or didn't know how to use antilock braking systems (ABS). ABS requires the driver to press the brake pedal firmly and hold it down without letting up; this runs counter to the traditional method of pumping brakes. Perhaps as a result, early research has shown no clear improvement in the number or severity of collisions of passenger vehicles equipped with ABS, and in some cases, ABS-equipped vehicles were more likely to be involved in certain types of accidents.16
For intelligent vehicles, these concerns can be addressed by careful design, such as allowing certain features to be engaged only when the car is parked or hands-free activation of in-vehicle services via voice recognition systems. However, many ITS technologies will likely require increased driver education and training.
When Might We See a Human-Centered Intelligent Vehicle?
The timing of "human-centered" intelligent vehicles' arrival on the market will depend on the evolution of technology, particularly the resolution of technical and cost constraints for some advanced concepts, such as collision-avoidance and automated systems, manufacturers' interest and production lead-times, and consumer demand. The federal role, as manifested by IVI, is to protect safety and enhance transportation services by helping both manufacturers and users recognize the importance of the integration of in-vehicle systems as well as the concept of "human-centeredness" in intelligent vehicle design.
How Might Intelligent Vehicles Evolve?
Over time, IVI expects that intelligent vehicles will advance in three primary ways: in the capabilities of in-vehicle systems, in the sophistication of the driver-vehicle interface, and in the ability of vehicles to communicate with each other and a smart infrastructure.
Capabilities of Individual In-Vehicle Systems Will Improve
Rudimentary information and warning systems are already appearing on the market, while driver assistance and automation technologies are positioned farther out on the horizon. The key technical challenge in developing driver information systems is to reduce production costs while providing dynamic route guidance abilities. Most current systems, for example, provide routes based on "static" maps irrespective of traffic considerations. Dynamic route guidance systems could determine optimum routes based on prevailing "real-time" conditions on the roadway, such as the location of bottlenecks caused by accidents or construction. The primary technical challenge in collision-avoidance system development is to reduce costs while also enhancing sensing capabilities to improve accuracy and reliability. Automated vehicles will share many of the sensing elements of collision-avoidance systems and must also possess an intelligence that can continuously assess the changing contexts and circumstances of the driving environment.
Driver-Vehicle Interface Will Become Increasingly Sophisticated
The integration of individual in-vehicle technologies will be reflected in coordinated and streamlined information displays and controls. Over time, the vehicle will become increasingly sophisticated in how it communicates information to and accepts commands from the driver. This sophistication will stem from the use of voice recognition, head-up displays, and vision-enhancement technologies that can relay information and commands and can also allow the driver to keep both hands on the steering wheel. Voice recognition software is already becoming increasingly common in Japanese in-vehicle navigation systems although it is still quite expensive for the average consumer.
Intelligent Vehicles Will Communicate With Other Vehicles and With a Smart Infrastructure
For intelligent vehicles to reach their maximum potential, they must be able to communicate with an intelligent transportation infrastructure and with other intelligent vehicles. For example, communication with a smart infrastructure would allow an intelligent vehicle to learn of incidents and then proactively suggest alternative routes in real time. Smart vehicles could also act as probes that could send information about travel conditions back to the infrastructure to create a richer base of knowledge about travel conditions on roads and highways. In addition, fully automated vehicles will likely rely to some extent on the guidance provided by an intelligent infrastructure and on communication with other smart vehicles. For example, in the recent AHS demonstration in San Diego, automated vehicles with magnetic sensors under their bumpers were guided by magnets implanted at 1.2 meter intervals just below the road surface.
Based on a review of the best publicly available information on technology development and potential cost and technical breakthroughs in these three areas, IVI foresees three "generations" of intelligent vehicles arriving on the market over the next 20 years.
A fully intelligent vehicle must work cooperatively with the driver. Uncoordinated technologies could deliver excessive, competing, or contradictory messages and demands that might distract and overwhelm the driver.
Over the next five to 10 years, we should see the first generation with advances in the capabilities of individual driver information and warning systems. These systems will become increasingly integrated with information coordinated through displays. Drivers will still maintain full control over their vehicles although collision-warning systems will provide limited automated assistance. In addition, vehicles would have a greater intelligence about road conditions in real time due to rudimentary communications with an intelligent infrastructure.
In about 10 to 15 years, the application of improvements in individual ITS systems will bring on a second generation with more and better intelligence in the vehicle. Although drivers will still maintain full control over their vehicles, collision-avoidance systems could take control temporarily during emergencies. In addition, more sophisticated voice recognition systems will be incorporated within the driver-vehicle interface. Vehicles will be able to communicate with each other to improve collision-avoidance capabilities. And communications with an intelligent infrastructure will be more interactive.
In about 20 years, in the third generation, we could see fully automated highway systems, cooperative systems of vehicles and infrastructure, and advances in the driver-vehicle interface, such as use of vision enhancements and head-up displays.
What Is the Industry's Interest in Intelligent Vehicles?
Intelligent vehicles hold major potential for industry.
"A breakthrough in navigation and autopiloting could revitalize the consumer's desire to spend discretionary income on motor vehicles, even in the most saturated markets," concludes a study by MIT on the future of the automobile industry.17
The trucking industry could also benefit from technologies that could keep commercial drivers safe and help them navigate the nation's interstate corridors more efficiently.
The critical issues for manufacturers are reducing the production costs of in-vehicle ITS technologies and systems integration, eliminating liability concerns, and better gauging consumers' potential demand for autonomous and integrated systems. Recognizing the importance of human factors in making their systems more attractive to consumers, the industry is already involved in domestic and international standards development efforts.
In addition, the private sector is becoming increasingly intrigued by the prospect of integrating in-vehicle technologies. As an initial step toward integration, the Society of Automotive Engineers' ITS Data Bus Committee is trying to develop open platforms to integrate navigation and in-vehicle communications equipment.18 Others are taking additional steps.
In early 1996, Delco Electronics unveiled a concept car, the SSC, which featured about two dozen safety, security, and communication technologies, some of them linked.19 For example, the vehicle's navigation, phone, and air-bag systems were all linked so that emergency services would be automatically notified if the air bag deployed during a crash.
"Through integration, we wanted to show that it really is possible to enhance quality and to make the car safer, more secure, and a true communications device," said Gary W. Dickinson, chief executive officer of Delco Electronics in an article profiling the SSC.20
In 1997, Intel and Microsoft began promoting similar personal computer software platforms for integrating in-vehicle technologies. The platforms are called Connected Car-PC and Auto PC, respectively. Both platforms would run on Microsoft's new open operating system, Windows CE. As reported by The Hansen Report on Automotive Electronics, this PC architecture would be able to coordinate multiple in-vehicle technologies and to "connect the driver to the Internet, e-mail, cell phone, fax/pager, as well as weather and traffic information." The computer would also be able to run entertainment and security systems or contact emergency services. Furthermore, the PC could allow electronic features to be operated by voice commands. Neither Microsoft nor Intel intend to produce a vehicle PC themselves, but they are instead trying to interest automotive suppliers and consumer electronics companies to take on vehicle PC development.21
Will Consumers Accept Intelligent Vehicles?
Perhaps the ultimate in vehicle IQ will be a vehicle's ability to drive itself. In August 1997, the National AHS Consortium demonstrated automated vehicle prototypes.
User acceptance will play a critical role in how intelligent vehicles will look and perform. Preliminary DOT research suggests that consumers are interested in advanced traveler information systems that provide accurate and timely information; are reliable, affordable, and easy to use; and improve personal safety and security.22 But the driving public is notoriously picky about what goes in their vehicles. In the 1980s, consumers in the United States almost universally rejected voice-alert systems in the vehicle, including those reminding passengers to buckle their seat belts.23 But consumers are also growing more accustomed to interacting with machines, which have become an intractable feature of everyday life in banking, communications, and entertainment.
From a market perspective, customization is an attractive feature, as exemplified by the thriving vehicle accessories market, which allows consumers to individualize "one size fits all" vehicles. Given the wide variety of information and technologies that could be placed in vehicles, it is not too improbable that in the future, drivers would be able to personalize the types of information provided by their vehicle. The ability to personalize would mean that the vehicle would know the driver's routine, recommend routes, and remember the context of those trips, such as ongoing construction on a much traveled exit ramp. In another fashion, an intelligent vehicle could allow the driver to customize how information is presented, in much the way that Microsoft or Apple allows a user to customize how he or she maneuvers through programs or the way an individual can touch a button in high-end luxury cars to position the seat, mirrors, and steering wheel for a particular driver. Paralleling the current discussions on air bags, however, are questions such as whether drivers would be able to disengage systems, particularly safety systems, they didn't like or found personally distracting.
Although DOT, as part of the national ITS program, has identified or will soon investigate user acceptance of individual in-vehicle systems, a challenge to the manufacturers and to IVI is to more precisely identify how public consumers would value the integration of these systems within an intelligent vehicle.
What is the Federal Role?
The federal role in the development of the human-centered intelligent vehicle is to cooperate with industry to develop technology, identify societal costs and benefits, build public confidence, promote standards and guidelines, encourage industry research and development, and promote U.S. industry.
The IVI research and development effort will include system definition and then development of intelligent vehicle concepts, evaluation of benefits and user acceptance, field operational testing, and demonstration and showcasing. IVI is structured to ensure that useable products are available for operational tests and demonstrations within three to four years.
The availability of the smart vehicle, however, will largely depend on industry and consumer interest. As a result, IVI will aggressively seek out and incorporate public and private stakeholder views to refine its vision of human-centered intelligent vehicles and to assess how its research and development efforts can assure that intelligent vehicles are technically, socially, institutionally, and economically beneficial.
Looking back on a century inundated by technology, the motor vehicle stands out as a singularly dynamic invention. In the next century, this dynamism will be driven by advances in information and computer technology. Our challenge is to ensure that new information, safety, and automation technologies are integrated to create human-centered intelligent vehicles that can advance safety, surface transportation efficiency, and economic competitiveness.
Technological change, however, is constant. Even as we envision one set of technologies, the envelope of possibilities expands. The latest innovations in information and computer technology, for example, are focusing on the development of neural links that will allow commands to be given with muscle impulses, eye movements, and brain waves, creating an almost symbiotic relationship between humans and machines.24 The vision of a human-centered intelligent vehicle, therefore, is not fixed but will continuously evolve in the wake of continuing technological breakthroughs. For DOT, however, putting people at the center of technology, as witnessed by the efforts of IVI, will remain paramount.
1. James P. Womack, Daniel T. Jones, and Daniel Roos. The Machine That Changed the World, Macmillian Publishing Co., New York, N.Y., 1990, p. 26.
2. Intelligent Vehicle Initiative: Strategic Planning Overview, Draft, April 29, 1997.
3. "The Human Centered Transportation System of the Future," presentation the National Science and Technology Council's Human Centered Transportation Safety Team at ITS America 7th Annual Meeting in Washington, D.C., June 1997.
4. Peter Haapaniemi. "Smart Vehicles Have Minds of Their Own," Safety + Health, November 1996, p. 64.
5. Implementation of the National Intelligent Transportation Systems Program, Report to Congress, ITS Joint Program Office, Federal Highway Administration, Washington, D.C., 1996.
7. NSTC Human Centered Transportation Safety Team, op.cit.
8. Thomas B. Sheridan. "Trains, Planes and Automobiles," The Reflector, May 1997, p. 9.
9. Sarah Wortham. "Are Cell Phones Dangerous on the Road?" Safety + Health, February 1997, p. 43.
10. Ibid., p. 44.
11. Preliminary Human Factors Design Guidelines for Driver Information Systems, Publication No. FHWA-RD-94-087, Federal Highway Administration, Washington, D.C., December 1995.
12. Preliminary Human Factors Guidelines for Crash Avoidance Warning Devices, National Highway Traffic Safety Administration, Washington, D.C., January 1996.
13. Preliminary Human Factors Guidelines for Automated Highway System Design, Publication No. FHWA-RD-95-053, Federal Highway Administration, Washington, D.C., 1995.
14. Susan Scott and Eugene Farber. "Safety in Our Hands," Traffic Technology International, February/March 1997, p. 61.
16. Mick Hans. "Learn the ABCs of Antilock Brakes," Safety + Health, March 1996, pp. 72-74.
17. Womack et al., op.cit.
18. Scott and Farber, op.cit.
19. Haapaniemi, op.cit., p. 67.
21. "Microsoft and Intel Pushing Vehicle PCs," The Hansen Report on Automotive Electronics, March 1997, pp. 1-2.
22. Department of Transportation, User Acceptance of ATIS Products and Services: A Report of Qualitative Research, January 1997.
23. "Speech Recognition Advances," The Hansen Report on Automotive Electronics, June 1997, p. 1.
24. Hugh S. Lusted and R. Benjamin Knapp. "Controlling Computers With Neural Signals," Scientific American, October 1996, pp. 82-87.
Cheryl Little is a policy analyst for the Volpe National Transportation Systems Center, where she has worked on technical and societal implications of ITS for five years.
Page Owner: Office of Corporate Research, Technology, and Innovation Management
Scheduled Update: Archive - No Update
Technical Issues: TFHRC.WebMaster@dot.gov