Browsing by Author "Greenwood, Ian"

Now showing 1 - 5 of 5
  • Thumbnail Image
    Agent-Based Game Theory Modeling for Driverless Vehicles at Intersections
    Rakha, Hesham A.; Zohdy, Ismail H.; Kamalanathsharma, Raj Kishore (United States. Department of Transportation, 2013-02-19)
    This report presents three research efforts that were published in various journals. The first research effort presents a reactive-driving agent based algorithm for modeling driver left turn gap acceptance behavior at signalized intersections. This model considers the interaction between driver characteristics and vehicle physical capabilities. The model explicitly captures the vehicle constraints on driving behavior using a vehicle dynamics model. In addition, the model uses the driver's input and the psychological deliberation in accepting/rejecting a gap. The model is developed using a total of 301 accepted gaps and subsequently validated using 2,429 rejected gaps at the same site and also validated using 1,485 gap decisions (323 accepted and 1,162 rejected) at another site. The proposed model is considered as a mix between traditional and reactive methods for decision making and consists of three main components: input, data processing and output. The input component uses sensing information, vehicle and driver characteristics to process the data and estimate the critical gap value. Thereafter, the agent decides to either accept or reject the offered gap by comparing to a driver-specific critical gap (the offered gap should be greater than the critical gap for it to be accepted). The results demonstrate that the agent-based model is superior to the standard logistic regression model because it produces consistent performance for accepted and rejected gaps (correct predictions in 90% of the observations) and the model is easily transferable to different sites. The proposed modeling framework can be generalized to capture different vehicle types, roadway configurations, traffic movements, intersection characteristics, and weather effects on driver gap acceptance behavior. The findings of this research effort is considered as an essential stage for modeling autonomous/driverless vehicles The second effort develops a heuristic optimization algorithm for automated vehicles (equipped with cooperative adaptive cruise control CACC systems) at uncontrolled intersections using a game theory framework. The proposed system models the automated vehicles as reactive agents interacting and collaborating with the intersection controller (manager agent) to minimize the total delay. The system is evaluated using a case study considering two different intersection control scenarios: a four-way stop control and the proposed intersection controller framework. In both scenarios, four automated vehicles (a single vehicle per approach) were simulated using a Monte Carlo simulation that was repeated 1000 times. The results show that the proposed system reduces the total delay relative to a traditional stop control by 35 seconds on average, which corresponds to an approximately 70 percent reduction in the total delay. The third effort presents a new tool for optimizing the movements of autonomous/driverless vehicles through intersections: iCACC. The main concept of the proposed tool is to control vehicle trajectories using Cooperative Adaptive Cruise Control (CACC) systems to avoid collisions and minimize intersection delay. Simulations were executed to compare conventional signal control with iCACC considering two measures of effectiveness - delay and fuel consumption. Savings in delay and fuel consumption in the range of 91 and 82 percent relative to conventional signal control were demonstrated, respectively. It is anticipated that the findings of this report may contribute in the future of advanced vehicles control and connected vehicles applications.
  • Thumbnail Image
    Development of a Web Based Pavement Management Reporting System to Enhance the Virginia Department of Transportation Pavement Performance Monitoring Program
    Gerber, Aaron; Kanaan, Ahmad; Perrone, Eric; Chowdhury, Tanveer; Shekharan, Raja A. (2015-06-04)
    The Virginia Department of Transportation (VDOT) is the third largest state-maintained highway system in the nation. With such a significant taxpayer investment in road infrastructure, VDOT has embarked on a comprehensive pavement management process that includes the use of state of the art optimization analysis software to determine appropriate funding allocation of the pavement network. Since VDOT uses the results of the optimization analysis in determining their paving targets and for determining their funding allocations for the Districts, and because the final project selection is usually done at the District level, it is necessary to track how the scheduled Maintenance & Rehabilitation (M&R) projects being performed in the field align to the targets produced by the optimization analysis and it is also important to track the difference between the targeted, planned and actual performance of the network. This is not a simple process; this also requires an extensive coordination between the Central Office Pavement Management and the Districts. In this paper, the development of the VDOT Pavement Performance Monitoring Program is described. This Performance Monitoring Program consists of a set of programmed reports that link the VDOT PMS system and database with VDOT contracting system which gives the CO decision makers the ability to efficiently determine whether the funding allocations and performance targets are being met with the Districts project selection. This integration of the network level PMS analysis results with the Districts project selection allows VDOT to close the loop between project and network level pavement management.
  • Thumbnail Image
    From Nothing to Something… Creating a Road Maintenance Culture in Tonga
    Schlotjes, Megan R.; Bennett, Christopher R.; Faiz, Asif; Visser, Alex; Greenwood, Ian (2015-06-04)
    The Kingdom of Tonga is a small island country in the South Pacific, located some 1,500 km north-east of New Zealand. The road network, approximately 870 km consisting of about 640 km of public roads, is almost exclusively low volume roads, with only a few urban roads in the capital Nuku'alofa carrying over 1,000 vehicles per day. Over 90 % of the public roads are sealed. Due to a variety of structural and fiscal issues, road maintenance - both routine and periodic - effectively halted in the late 2000's. The Ministry of Works, nominally responsible for the maintenance of the road network, did not have the resources or equipment to maintain the roads so for about a decade there was effectively no routine or periodic maintenance. In 2010 the Government of Australia provided funds to Tonga through the World Bank with the objective of establishing a routine and periodic maintenance program, with a secondary objective of stimulating the economy by creating employment through road works. This paper describes the success story of how Tonga went from having no regular road maintenance, indeed, not even contractors to undertake road maintenance, to having a competitive road maintenance industry undertaking routine and periodic maintenance across the country.
  • Thumbnail Image
    Long-Term Pavement Performance Automated Faulting Measurement
    Agurla, Mahesh; Lin, Sean (2015-06-04)
    This study focused on identifying transverse joint locations on jointed plain concrete pavements (JPCP) using an automated joint detection algorithm and computing faulting at these locations using Long-Term Pavement Performance (LTPP) program profile data collected by the program’s high speed inertial profilers (HSIP). This study evaluated two existing American Association of State Highway and Transportation Officials (AASHTO) R 36-12 automated faulting measurement (AFM) models: ProVAL (Method-A) and Florida Department of Transportation (FDOT) PaveSuite (Method-B). A new LTPP AFM was developed using LTPP profile data. The LTPP AFM devised an automated algorithm to identify joint locations where faulting is also computed for each joint identified in order to replicate the manually collected faulting data using the Georgia Faultmeter (GFM), which has been used on LTPP test sections since the program’s inception. The study compared the LTPP manual faulting measurements collected using the GFM with the ProVAL AFM and the LTPP AFM using LTPP profile data. Similarly, the FDOT GFM measurements were compared to the FDOT PaveSuite AFM and the LTPP AFM using the same FDOT profile data. The initial results for six LTPP test sections show that the LTPP AFM can identify joint locations with a joint detection rate (JDR) ranging from 95 to 100 percent. ProVAL's JDR range is from 58 to 99 percent for the same six LTPP test sections. Similarly, for the one FDOT test section available, the LTPP AFM’s and FDOT PaveSuite's JDRs are approximately 96 percent. This study outlines the LTPP AFM algorithm, discusses the comparison of the three AFM results, and recommends future research needs in this area.
  • Thumbnail Image
    Microscopic Analysis of Traffic Flow in Inclement Weather
    Rakha, Hesham A.; Zohdy, Ismail H.; Park, Sangjun; Krechmer, Daniel (United States. Federal Highway Administration, 2010-12)
    This report documents the second part of the FHWA research study involving analysis of the microscopic impacts of adverse weather on traffic flow, but is a third phase of the research effort on the impacts of weather on traffic flow. The first phase of FHWA research involved macroscopic analysis, which focused on the impacts of adverse weather on aggregate traffic flow. The second phase of research analyzed the impacts of adverse weather on microscopic traffic behavior. This report documents the results of three research efforts (1) The impacts of icy roadway conditions on driver behavior at a microscopic level, using field measured car-following data,; (2) An investigation of the influence of weather precipitation and roadway surface condition on left-turn gap-acceptance behavior using traffic and weather data collected during the winter of 2009-2010 at a signalized intersection in Blacksburg, Virginia; and (3)The development and demonstration of methodologies for the use of weather-related adjustment factors in microsimulation models, including general approaches to construct simulation models accounting for the impact of precipitation. For the third effort, the general approach was applied to the calibration of the VISSIM and INTEGRATION simulation software.