Search

The Rocky Mountain Naturalistic Driving Study (RMNDS) was, at the time of data collection, the first attempt to use the NDS methodology to conduct research on the effects of cannabis on driving performance. The resulting dataset comprises over 14,000 trips made by 23 participants who self-reported medium to heavy cannabis use and who also reported they had a history of driving under the influence of cannabis. A unique aspect of the study was the collection of quantitative and qualitative drug use data. Qualitative drug use data was collected via an online journal, while quantitative drug use data was collected using a Quantisal oral fluid collection device prior to one driving trip each week. Samples were sent to a toxicology laboratory for analysis, which produced quantifiable test results for the National Institute on Drug Abuse 5 drug panel, including delta-9-THC. Out of the 14,000+ trips, there were a total of 178 verified drug test results along with 1,549 drug use journal entries. While the study proved successful for collecting naturalistic driving data and both objective and subjective drug use data, the difficulty comes in linking the drug use data to the driving data in order to identify periods of driving that may be impacted by the consumption of cannabis and/or other drugs. Further analysis of the RMNDS data, including identification of trips linked to drug use, would provide invaluable information about the impact of cannabis and/or other drugs on driving performance.

Automatic traffic monitoring is becoming an important investment for transportation specialists, especially as the overall volume of traffic continues to increase, as do crashes at intersections. Infrastructure cameras can be a good source of information for automatic monitoring of traffic situations at intersections. However, efficient computer vision methods that can process the video data effectively are required for this endeavor. Intersection cameras often record video data that are of low quality and low frame rate, making them challenging to use. In this project, we have demonstrated how traffic cameras can be used to automatically track roadway agents, find their kinematic behavior, and devise a safety measurement strategy, leveraging recent advancements in computer vision and deep learning. In this process, we have specifically focused on the Virginia Beach area and used publicly available traffic data to demonstrate our results. We developed a full computer vision pipeline that trains a custom object detector specifically using traffic data. We also used an optical flow method and a graph neural network to improve the accuracy of object tracking. The tracked objects from the image frames were further used as a point source and mapped to their GPS locations. Finally, the speed of each object was calculated to understand the traffic dynamics and determine possible crash predictors. This information can be used to quickly alert traffic control operators to a specific intersection that likely needs their attention so that crashes can be mitigated.

Intersection collisions account for 40% of all crashes on U.S. roadways. It is estimated that 165,000 accidents, which result in approximately 800 fatalities annually, are due to vehicles that pass through intersections during red signal phases. Although infrastructure-based red-light violation countermeasures have been deployed, intersections remain a top location for vehicle crashes. The Virginia Department of Transportation and its research arm, the Virginia Transportation Research Council, partnered with the Virginia Tech Transportation Institute to create the Virginia Connected Corridors (VCC), a connected vehicle test bed located in Fairfax and Blacksburg, Virginia, that enables the development and assessment of early-stage connected and automated vehicle applications. Recently, new systems have been deployed that transmit position correction messages to support lane-level accuracy, enabling development of signal awareness applications such as red-light violation warning. This project enhances the current capabilities of VCC platforms by developing new signal awareness safety and mobility features. Additionally, this project investigated the technical and human factors constraints associated with user interfaces for notifying and alerting drivers to pertinent intersection-related information to curb unsafe driving behaviors at signalized intersections.

Intersection collisions account for 40% of all crashes on U.S. roadways. It is estimated that 165,000 accidents, which result in approximately 800 fatalities annually, are due to vehicles that pass through intersections during red signal phases. Although infrastructure-based red-light violation countermeasures have been deployed, intersections remain a top location for vehicle crashes. The Virginia Department of Transportation and its research arm, the Virginia Transportation Research Council, partnered with the Virginia Tech Transportation Institute to create the Virginia Connected Corridors (VCC), a connected vehicle test bed located in Fairfax and Blacksburg, Virginia, that enables the development and assessment of early-stage connected and automated vehicle applications. Recently, new systems have been deployed that transmit position correction messages to support lane-level accuracy, enabling development of signal awareness applications such as red-light violation warning. This project enhances the current capabilities of VCC platforms by developing new signal awareness safety and mobility features. Additionally, this project investigated the technical and human factors constraints associated with user interfaces for notifying and alerting drivers to pertinent intersection-related information to curb unsafe driving behaviors at signalized intersections.

69A3551747115/Project VTTI-00-021, Intersection collisions account for 40% of all crashes on U.S. roadways. It is estimated that 165,000 accidents, which result in approximately 800 fatalities annually, are due to vehicles that pass through intersections during red signal phases. Although infrastructure-based red-light violation countermeasures have been deployed, intersections remain a top location for vehicle crashes. The Virginia Department of Transportation and its research arm, the Virginia Transportation Research Council, partnered with the Virginia Tech Transportation Institute to create the Virginia Connected Corridors (VCC), a connected vehicle test bed located in Fairfax and Blacksburg, Virginia, that enables the development and assessment of early-stage connected and automated vehicle applications. Recently, new systems have been deployed that transmit position correction messages to support lane-level accuracy, enabling development of signal awareness applications such as red-light violation warning. This project enhances the current capabilities of VCC platforms by developing new signal awareness safety and mobility features. Additionally, this project investigated the technical and human factors constraints associated with user interfaces for notifying and alerting drivers to pertinent intersection-related information to curb unsafe driving behaviors at signalized intersections.

DTFH6114H00002, The Advanced Messaging Concept Development (AMCD) Project objective was to evaluate the ability of connected vehicles to generate, and infrastructure to collect, Basic Safety messages (BSM), Probe Data Message (PDM), and Basic Mobility Message (BMM) alternatives using cellular and DSRC communications while employing message control strategies in real-world driving conditions for non-safety-critical applications. These three message schemes represented potential alternatives for transferring data from equipped vehicles to the infrastructure. Such data transfer is intended to enable a broad array of Vehicle-to-Infrastructure (V2I) applications which may be used to improve operations. AMCD implemented the three messaging schemes on the Virginia Connected Corridor with the aim of validating efficacy and characterizing their associated behavior. The schemes were exercised in live traffic using ten instrumented vehicles and an emulated traffic operation center interface where the experimenters manipulated the various message control parameters while measuring the resulting message traffic. Results support the feasibility of advanced V2I messaging and demonstrate potential advantages of a flexible, multi-threaded message scheme. A number of recommendations are provided with the aim of further improving V2I messaging capabilities and leading to implementation of advanced messaging with infrastructure applications to assess the operational value in deployment.

The market penetration levels needed to realize the full safety, economic, and environmental benefits of connected vehicle (CV) systems will not be met for some time. During the transition, it would be beneficial if data on non-CVs could be measured and included within the real-time CV data stream. Conceptually, a connected vehicle with advanced sensors, such as radar, could measure the dynamics of adjacent vehicles and, in addition to broadcasting its own Basic Safety Message (BSM), broadcast a pseudo BSM representing the non-connected vehicles. This project investigated the use of radar sensors to compute the position, speed, and heading of a non-connected vehicle (non-CV) for packaging into a pseudo BSM. An algorithm was developed to estimate the speed, position, and heading of a nearby non-CV via speed, Global Positioning System (GPS) coordinates, and radar data from the CV. Field tests were conducted with two vehicles on the Virginia Smart Road and on public roads in the New River Valley of Virginia. The field tests were designed to cover a variety of vehicle formations, traffic densities, velocities, and roadway environments. The final results showed that 67.9% of the position estimates were within 3 m of the measured position along the x-axis (longitudinal) and within 1.5 m of the measured position along the y-axis (lateral). Heading and speed estimates were generally excellent. Although the estimated position accuracy was lower than desired, the data that were collected and analyzed were sufficient to suggest ways to improve the system, such as fusing the radar data with camera-based vision data or using a more accurate GPS.

DUNS: 0031370150000, EIN: 54-6001805, This project identified vehicle-to-infrastructure (V2I) communication system limitations on the Northern Virginia Connected Vehicle Test Bed. Real-world historical data were analyzed to determine wireless Dedicated Short Range Communication (DSRC) coverage gaps and overlaps. In addition, a simulated scalability test was run to determine the effects of network congestion on the system. The results from the real-world historical data showed that significant loss of signal occurred due to obstructions commonly found in complex highway systems, including overpasses and underpasses, elevated concrete roadways, and foliage. Consequently, care must be taken to minimize loss of signal when selecting an installation site for roadside equipment (RSEs). The deployment of multiple RSEs or repeaters may be necessary to maximize coverage in localized dead zones. The results from the scalability test showed that the current network architecture is not able to handle a large deployment of connected vehicles (CV). If a large scale of CV were to be deployed, an assessment of the current network design needs to be investigated to account for the number of vehicles and subsequent flow of data expected in the operational area.

69A3551747115/[ Project 04-121], Automatic traffic monitoring is becoming an important investment for transportation specialists, especially as the overall volume of traffic continues to increase, as do crashes at intersections. Infrastructure cameras can be a good source of information for automatic monitoring of traffic situations at intersections. However, efficient computer vision methods that can process the video data effectively are required for this endeavor. Intersection cameras often record video data that are of low quality and low frame rate, making them challenging to use. In this project, we have demonstrated how traffic cameras can be used to automatically track roadway agents, find their kinematic behavior, and devise a safety measurement strategy, leveraging recent advancements in computer vision and deep learning. In this process, we have specifically focused on the Virginia Beach area and used publicly available traffic data to demonstrate our results.