Traffic congestion is becoming a major problem as the number of vehicles on the road increases. To address the issue of congestion, there are many options such as building a proper infrastructure, and imposing restrictions on the number of automobiles per individual. However, these solutions are not efficient from an economic perspective. To overcome this disadvantage, congestion can be reduced by providing accurate travel time prediction so that users can always plan their trips prior to departure.
Several methods exist to predict travel time with different equipment such as license plate matching and probe vehicle technique to monitor the current traffic status. The equipment adds an additional overhead to the traffic congestion issue. While using external equipment to measure the traffic status from the freeway, small deviations are introduced; as the traffic increases, the deviation in the data increases. If the data from the source is not accurate, travel time prediction is not accurate either.
In this thesis, a robust model, Dynamic Travel Time Prediction (DTTP), is proposed for travel time prediction based on the data collected from the Global Positioning System (GPS) devices used in the vehicles and the data from the Vehicle Detection System (VDS) on the path. VDS is a system used for detecting vehicles and suitable for use in conjunction with other equipment for traffic monitoring and control. To provide accurate travel time prediction, the current detailed traffic status is obtained and updated periodically to maintain the accuracy of the prediction. In order to obtain the current traffic status, a GPS device installed in cars that provides information about the speed, distance, and current location of these vehicles is used, so there is no deviation in the collected data. To prevent the misuse of traffic information and maintain privacy of users, an existing system called Virtual Trip Lines (VTL) is used as an intermediary system to obtain the traffic related data. The information obtained from the car is merged with data collected from VDS. By combining the VDS and GPS dataset, the model uniquely minimizes the limitations of each dataset and enhances the prediction’s accuracy. The obtained current traffic status is then matched with the available historical traffic data (archived) using k-NN (k Nearest Neighbor) algorithm to obtain the travel time from the origin to the destination. The primary focus of this research is to develop a robust model for travel time prediction based on the data collected from GPS devices in the vehicles and the data from VDS on the path. After comparison with 6 different models, the DTTP model implies that using the GPS and VDS data together contributes to improved accuracy.
|Commitee:||He, Min, Wang, Sen|
|School:||California State University, Long Beach|
|Department:||Computer Engineering and Computer Science|
|School Location:||United States -- California|
|Source:||MAI 56/06M(E), Masters Abstracts International|
|Keywords:||Gps, Prediction, Travel, Vehicles|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be