Publication Number: FHWA-HRT-08-004

Exploratory Advanced Research: A Journey Toward New Solutions

by David Kuehn and Ariam Asmerom

Breakthrough opportunities receive Federal funding to address long-term transportation challenges.

The research participant in this vehicle at the Texas Transportation Institute is wearing eye-tracker goggles as part of a project to understand the visual requirements for human drivers and autonomous vehicles. The project is one of those in FHWA's program for exporatory advanced research.

Expansion of transportation research knowledge and continued innovation are key factors in the U.S.’s economic well-being and continued prosperity. Research can be defined by its characteristics: long-term, fundamental, and high-risk research is considered “advanced research,” while shorter term, incremental research that answers a specific need is considered “applied research.” An example of applied research from 100 years ago might be developing better springs for a smoother carriage ride in a horse and buggy duo, whereas advanced research from back then might entail developing the breakthroughs needed to replace the horse and buggy with the automobile. Advanced research moves forward in “leaps” rather than incremental “steps.”

Developing solutions to current and future transportation issues will take both types of research, although in the real world, competing interests and a constrained fiscal environment often can limit an agency’s ability to carry out high-risk advanced research. A recent report by the National Science Board, Science and Engineering Indicators 2008, indicates that the years 2004 to 2007 marked the first multiyear decline in Federal obligations for academic research of both kinds since 1982. The report calls for the Federal Government to improve funding for advanced research, and for industry, government, and academia to seek greater intellectual interchange.

Even before the National Science Board published the report, the Federal Highway Administration (FHWA) and the Transportation Research Board (TRB) both agreed that FHWA was in a unique position to pursue innovative advanced research opportunities. In 2001, TRB commissioned a study and released a report titled The Federal Role in Highway Research and Technology (Special Report 261). Authored by the TRB Research and Technology Coordinating Committee, the report identifies the need for FHWA to increase its focus on “fundamental, long-term research aimed at achieving breakthroughs in the understanding of transportation-related phenomena.”

FHWA’s own assessment also supported the need for this type  of advanced research, and in 2005 President George W. Bush signed the Safe, Accountable, Flexible, Efficient Transportation Equity Act: A Legacy for Users (SAFETEA-LU) into law, authorizing $14 million per year for the Exploratory Advanced Research (EAR) Program. Strong support from FHWA leadership established the program with strategic corporate objectives centered on supporting long-term research with high-payoff potential, collaboration, and stakeholder input.

In 2007, FHWA’s EAR Program committed more than $10 million  to support 11 high-risk, high-payoff projects. FHWA selected the projects based on scientific or technical merit, evidence of research partnership, relevance and importance to the agency’s mission, and funding availability. The award of nine of the 11 projects signifies an important step in FHWA’s efforts to affirm the unique role of the Federal Government in supporting long-term, breakthrough solutions to the Nation’s transportation challenges. (FHWA anticipates awards for the other two projects in 2008.)

Shaping the “Next- Generation” Research Agenda

Development of a corporate agenda for exploratory advanced research began in early 2005 and is an ongoing process. In developing the first EAR agenda, FHWA referenced recommendations from stakeholder forums, priority research areas identified in the authorizing legislation, the U.S. Department of Transportation’s (USDOT) Transportation  Research, Development and Technology Strategic Plan 2006–2010, and internally solicited recommendations. For details of the process, see “The Evolution of Advanced Research” in the May/June 2006 issue of Public Roads.

In January 2007, FHWA released the first open solicitation for research proposals. The EAR Program defined the working research agenda in terms of the following general areas of interest:

Highway Safety

• Increased understanding of driver visibility
• Innovative technologies to detect and warn pedestrians and drivers of potential conflicts
• Development of an integrated safety system
• Improved understanding of the relationship between vehicle dynamics and roadway friction capacity

Planning and the Environment

• Environmental and land use planning, including complex data collection and use, assuming current trends continue
• Accurate models for predicting future trends, assuming some degree of trend reversal

Innovative Transportation  Policy Solutions

• Analytical approaches to multimodal investment
• Emerging issues in innovative finance
• Alternative futures to freight mobility
• Integration of transportation data from multiple sources
• Travel demand response to travel pricing
• Geospatial data acquisition for performance monitoring of the highway system
• Advanced approaches for identifying fuel tax evasion practices

Highway Infrastructure

• Next generation infrastructure materials
• Breakthroughs in construction technology
• Systems-based analytical modeling and design methods
• Innovative infrastructure management

Innovative Operations Solutions

• Mobility applications for vehicle-infrastructure integration
• Heuristic-based solutions for traffic management systems
• Development of an origin-destination model
• Pavement materials that facilitate radio communication
• Enhanced monitoring of highway condition and performance

The Awardees

The EAR Program received close to 400 proposals from universities and private companies in response to the 2007 solicitation. Of these, the program invited approximately 70  to submit full proposals for further consideration. Various panels of subject matter experts, including FHWA staff and peers from State departments of transportation (DOTs), the American Association  of State Highway and Transportation Officials, and TRB, convened to evaluate the technical merit of each proposal. Of the 70 full proposals, the panels selected 22 that were superior and responsive to the stated evaluation criteria. Because the total estimated cost of the 22 recommended proposals exceeded $20 million in Federal funds, a panel of FHWA officials was charged with selecting those proposals that best addressed the strategic goals and priorities of FHWA and USDOT.

The panel also looked at the relationships between the 22 proposals and considered the potential for synergistic or complementary projects that could work together to achieve strategic agency and departmental objectives. After completion of the technical reviews, the final selection panel voted unanimously to select 11 projects for a total investment of $10.1 million.

Based on an almost 50/50 split resulting from various cost-share agreements, the combined total investment for all 11 projects is an estimated $19.5 million. As of January 31, 2008, FHWA had awarded contracts or cooperative agreements to nine of the winning proposals, and those contracts and agreements now are active. The following is a look at the research topics from the awards.

The data acquisition antenna interface shown here will eliminate the need for operators to collect data for observing network traffic, predicting demand, and setting traffic signal phase durations.

Congestion Relief Through Smart Technology and Innovative Safety Research

For almost two decades, highway agencies have recognized intelligent transportation systems(ITS) as a “movement” in the war against gridlock and other impediments to improved mobility and safety. Initiatives implementing vehicle infrastructure integration (VII), History Lessons From the National Bridge Inventory, and electronic freight management are just some of the promising ways that ITS is affecting transportation. VII as used here refers to the ability of vehicles to exchange information among themselves and with roadside components via wireless communications technology.

California PATH is using these two sedans for vehicle infrastructure integration experiments to test driver usage of cooperative adaptive cruise control at reduced time gaps and to transmit individualized speed advisory information to drivers inside their vehicles and also directly to the automated speed control systems on the vehicles.

To complement USDOT’s long-established ITS program, FHWA selected several EAR proposals that addressed existing research gaps  in safety and congestion relief. The selected topics are expected to lead to breakthroughs in smart traffic signals, application of VII and advanced sensor technology, and autonomous vehicle control.

Next Generation Smart Traffic Signals. The next generation smart signals research effort will explore the possibilities of replacing existing traffic signal systems with self-adaptive systems that significantly reduce the typically high costs of installation and maintenance of traffic signals. Self-adaptive signal systems have proved to increase roadway capacity, and if widely implemented, could transform traffic operations significantly.

The first phase of the study will eliminate the need for operators to collect data necessary for system operation. Instead, the researchers will use leading-edge sensor technology to observe network traffic, predict demand, and set phase durations to optimize a given objective. Further system expansion through vehicle-infrastructure integration ultimately could provide a communication network that greatly enhances corridor management and capacity. The University of Arizona’s Advanced Traffic and Logistics Algorithms and Systems (ATLAS) Center will lead this project in partnership with local government and private sector collaborators.

Auburn University’s instrumented vehicle on the National Center for Asphalt Technology (NCAT) test track at Auburn is being used to test lanelevel navigation technologies as part of an FHWA funded project.

Mobility Applications for VII. The EAR Program awarded research funding to California Partners for Advanced Transit and Highways (PATH), a statewide program housed at the University of California, Berkeley, to develop and test VII-enabled technology with promising innovative concepts across several dimensions.

Successful results would capitalize on data collected on freeways using a high-resolution probe vehicle enabled by VII technology. This technology will make it possible to transmit individualized speed advisory information to drivers inside their vehicles, and also directly to the automated speed control systems on the vehicles.

Using vehicle-to-vehicle and infrastructure-to-vehicle data communication, engineers could enhance the performance of adaptive cruise control far beyond the current state of the art. Finally, fully automated control of heavy trucks would represent an important step in the longer term deployment of automated highway systems.

The impacts of such progress would be significant. The speed advisory and control systems would enable traffic shock waves to dissipate, resulting in optimized throughput capacity of highway lanes and reduced traffic congestion. Truck platooning (using wireless communications to closely coordinate acceleration and braking) could double the capacity of truck-only lanes if  all trucks using those lanes were equipped with the needed sensor, communications, and speed control systems. Also, reduction in fuel consumption by 10 to 20 percent would be likely.

This work will extend far beyond the current focus of the VII program, where efforts are centered on nearer term concepts that would enable the initial deployment of VII technology. Although many VII ideas are based on the use of aggregated data accumulated over a long period of time, the proposed research will rely on individualized, real-time data.

This autonomous car developed by Austin Robot Technology (a team of technologists) and Professor Peter Stone’s research group at the University of Texas at Austin participated in the semifinals of the 2007 Defense Advanced Research Projects Agency’s (DARPA) Urban Challenge and will be used in the Autonomous Vehicle Intersection Control project.

Intelligent Multisensors for Vehicles. According to the FHWA Office of Safety, close to 50 percent of all traffic fatalities are due to lane departures. Therefore, efforts to advance driver assistance systems that prevent unintended lane departures have a high-payoff potential. Auburn University will perform research to develop an accurate, robust, and reliable positioning system capable of providing high-accuracy, high-update-rate, lane-level measurements for future vehicle navigation and control systems.

The most important element of any lane-keeping assistance system is the reliability of lane position information. Currently available lane departure warning (LDW) systems rely on cameras to track lane markings. These systems suffer from technical limitations, particularly in areas where lane markings are missing or difficult to detect. At the same time, the reliability of vehicle positioning is becoming more important as map-supported safety systems gain popularity. Thus, FHWA supports research that would improve both lane detection systems and vehicle positioning to achieve near 100 percent availability.

The proposed work will examine the safety improvements possible by fusing sensor signals from several systems expected to be available in passenger vehicles in the near future. The key innovative concept is found in the idea of using the strengths of individual sensor modalities to make up for the weaknesses of others.

This urban scene shows pedestrians and vehicles with color boxes that indicate the detected and recognized pedestrians using a visionbased layered framework.

Autonomous Vehicle Intersection Control. Self-guided, self-controlled vehicles have been technologically feasible for some time, and FHWA has tested them in a research setting.

Such vehicles likely will become economically feasible within the next 5 to 10 years. The many advantages of this technology — for example, enabling minors and the elderly to be independently mobile — make these vehicles a highly desirable and perhaps inevitable eventuality.

Researchers at The University of Texas at Austin propose a study that considers the impact of autonomous vehicles on urban traffic infrastructure, specifically at intersections. By taking advantage of the precise controllability of autonomous vehicles, researchers hypothesize that the vehicles can be coordinated at intersections by reserving just the amount of space and time needed to ensure safe passage through any given intersection.

The university researchers will test this hypothesis in scenarios consisting of all autonomous vehicles and in scenarios with mixed autonomous and human-driven vehicles. The researchers also will investigate coordination along multiple intersections to maximize corridor throughput. The proposed research focuses specifically on intersection throughput, but the cumulative impacts realized from related research could have a profound impact on future mobility concerns.

Researchers are using this robotic vehicle in a project to develop a hybrid human/computer model for describing the visual parameters needed by human drivers.

Object Recognition for Pedestrian Sensing. Can a vehicle-based pedestrian detection system be developed that detects moving or stationary pedestrians with a high accuracy and low false-alarm rate? Sarnoff Corporation and Autoliv Inc. propose to prove that such a system is achievable by using a vision-based layered processing framework. This innovative safety concept proposes an inherently low-cost system that is capable of being widely implemented.

The current state-of-the-art technology uses techniques that can detect large- and medium-sized objects accurately during daylight. Although these approaches are encouraging, they suffer from a large number of inadvertent warnings because object detection is performed without a proper understanding of scene context. Researchers believe that improved performance at lower cost would achieve widespread acceptance and greatly reduce the frequency of incidents involving automobiles and pedestrians.

The performance of slip-form concrete pavements on major highway projects like this one depends on limiting or eliminating random, unplanned cracking—the purpose of the High-Performance SRCC for Crack-Free Pavements project.

The three key innovations proposed are: (1) a layered approach that integrates multiple cues (shape, appearance, depth, and motion) to detect, recognize, and track objects; (2) simultaneous multiobject classification; and (3) the use of scene context in improving object recognition. The study proposes to demonstrate detection of pedestrians in a vehicle’s path at a rate of 98 percent, with a false positive rate of approximately one per hour. The system will include a collision prediction model that can be used in a collision warning system.

If successful, the proposed innovations could improve significantly the performance of automotive safety systems that currently are under study by other Federal agencies.

Researchers are using pad foot roller compactors like the one shown here in a project to improve quality control during earthwork construction.

Understanding Driver Requirements for Visual Information. A collaborative effort by Science Applications International Corporation™ (SAIC), the National Institute of Standards and Technology (NIST), and the Texas Transportation Institute proposes to exploit the synergy between two current approaches to advance the understanding of the visual requirements for both human drivers and autonomous vehicles. The research objective is to determine the quantity and quality of visual information needed, under both driving modes, to navigate the road safely.

The study’s point of departure is a methodology developed by NIST’s Intelligent Systems Division for the control of autonomous onroad vehicles known as the Four Dimensional/Real-Time Control System (4D/RCS). This project proposes to combine psychology, neurobiology, and engineering in new ways to bring structure and order to the complex process of dynamic visual perception involved in negotiating curves, for example, on a roadway at night.

The end product of this effort will be a hybrid human/computer model that describes the visual parameters needed by human drivers to make appropriate driving decisions under  a limited set of circumstances. In the long term, manipulation of this product to model robotic driving would provide added benefits in anticipation of future roadways designed for a mix of human- and computer-driven vehicles.

Next-Generation Research To Plan, Build, and Maintain Future Highways

The next set of research efforts addresses cementious composites, a device for real-time measurement of soil properties, and infrastructure management through systems-based monitoring.

High-Performance, Stress-Relaxing Cementitious Composites (SRCC) for Crack-Free Pavements.The Texas A&M Research Foundation at the Texas A&M University is attempting a crosscutting and bold approach to address the issue of cracking in concrete pavements and structures. The research study proposes new concepts to prevent or reduce cracks, which are a major problem in portland cement concrete. Concrete is one of the most used construction materials, and methods to reduce cracking can have important applications in current concrete mixes as well as innovative mixes that incorporate new materials used in pavements and bridges.

If successful, the project can point the way to a new use of nano- to microscale inclusions in the concrete to enable it to relax a little under stress, thus reducing harmful cracking in concrete pavements. The project should be complete by early 2010, with milestones that involve developing predictive models in 2008 and conducting experimental validation of crack-resistant concrete in 2009. The researchers expect to generate a proposed design methodology for SRCC based on the models and experimental evidence developed under the project.

Development of a Soil Stiffness Measuring Device for Pad Foot Roller Compactors. The aim of this research is to develop a measurement approach to determine mechanistic soil properties continuously, in real time, during the compaction of fine- and mixed-grain soils with a pad foot roller compactor. The researchers will integrate the measurement system with global positioning system- based mapping software to provide roller operators with a continuous record of soil stiffness and modulus for quality control. The new measurement approach also will provide inspectors with a record of soil stiffness for quality assurance.

The Colorado School of Mines will pursue the study with participation from industry (roller manufacturers). If successful, this effort will represent a significant leap forward in quality control during earthwork construction, accelerated construction, and direct measurement of mechanistic soil parameters.

Systems-Based Monitoring Approaches for Improved In-frastructure Management Under Certainty. Academia, industry, and government partners are collaborating on a project focused  on the effective management and maintenance of bridge infrastructure. Researchers from the University of Central Florida and Lehigh University will use novel sensing technologies, deterioration models, and decisionmaking tools, leading  to safer structures and more cost- effective maintenance techniques. A successful outcome would produce a much-needed advance over current practice, where many decisions for infrastructure management are based on visual inspections and experiment-based engineering approaches.

The researchers expect that the proposed framework will serve not only for a particular structure, but also for a network of bridges. The new techniques will provide an integrative monitoring platform with necessary data for future design and maintenance, using analysis and prediction methods. Finally, this study will develop effective means for handling large amounts of intermittent or real-time data collected from field applications on bridges. The researchers will convert this data into useful knowledge by using advanced system-based analysis methods, which bridge owners then can use for decisionmaking.

The results of this research will represent an important step in forming the next generation of complete bridge infrastructure monitoring, an issue of special concern in light of the August 2007 bridge collapse in Minneapolis, MN. This project is funded under FHWA’s EAR Program. University of Central Florida Professor F. Necati Catbas, Ph.D., is the principal investigator for this project.

University of Central Florida and Lehigh University researchers will use novel sensing technologies, deterioration models, and decisionmaking tools to produce an advance over current infrastructure management practice. This graphic illustrates a bridge health monitoring system for decisionmaking. In this illustration, the bridge is a movable bridge, and both structural and mechanical components of the bridge are monitored. A data acquisition system (DAS) collects the measurements from the sensors distributed at critical regions and components of the movable bridge. The idea of this monitoring is that the sensor data along with video images are collected, processed, and transmitted to the bridge engineers for timely and objective decisionmaking.

Innovative Research, Breakthrough Results

FHWA established the EAR Program with ambitious goals and a long- term vision. A range of well-focused investments, along with strong, long-term measures and expert evaluation, will ensure that the program moves successfully toward major research payoffs.

With the diversity of advanced research projects now underway — in terms of research focus areas, kinds of organizations conducting the work, and types of partnerships — the EAR Program has a strong foundation to affect the future of highway research and ultimately the quality of the Nation’s transportation system. Part of what led to a strongly diversified portfolio of advanced research projects has been coordination at the national level, stakeholder involvement, and a corporate and industry focus. Moving forward will require careful management of annual outcomes and charting a course toward longer term research results.

David Kuehn became the first program manager for the FHWA EAR Program. Previously he served as a community planner for the FHWA Office of Planning and as a planner in local government. Kuehn entered Federal service as a presidential management fellow after completing a master’s of public administration from the University of Southern California. He also holds a B.A. from the University of California, Irvine.

Ariam Asmerom, P.E., was a transportation specialist and coordinator of FHWA’s advanced research initiative in the Office of Corporate Research, Technology, and Innovation Management. She recently resigned from FHWA with 16 years of experience in transportation planning and engineering, working in both the public and private sectors. Asmerom has a bachelor’s of science in civil engineering from the University of Virginia.

For more information, see www.fhwa.dot.gov/advancedresearch or contact David Kuehn at 202–493–3414 or david.kuehn@fhwa.dot.gov.