Author: Jialei Mao

Publisher:

Published: 2009

Total Pages: 75

ISBN-13:

"It is important for many traffic operations, such as real-time adaptive signal control, dynamic traffic assignment and traffic demand estimation, to obtain the vehicle turning movement information at a signalized intersection. However, it is laborious and time consuming to count the turning movements manually. Previous efforts on this problem relied on a mathematical approach by solving an O-D matrix in which the turning movements represent distributions of the arriving flow at each intersection approach. However, such a matrix cannot be mathematically solved without using supplementary volume data from the local detectors; previous studies showed the results from the O-D method are not accurate. Identifying vehicle turning movements from detector information is a more direct method. Limited studies using this method have been found for intersections without shared lanes. Most intersections allow shared lane operations, thus this method is not practical without further improvements. Driven by the need to identify vehicles turning movement automatically regardless of the geometry and operation of the intersections, this research studied and developed a system called Automatic Turning Movement Identification System (ATMIS). ATMIS utilizes intelligent detection matching algorithm to identify vehicle turning movements from the detections collected from the field in real-time. This algorithm has the capability to compensate the error caused by the faults detections and shared lanes. The results from laboratory experiments and field tests are very encouraging. Future work is also discussed in the thesis. While another related study, intersection delay measurement, is also included in this thesis. Delay is regarded as one of the most important Measures of Effectiveness (MOE) for signalized intersections. Different methods to measure control delay are recommended by many researchers but most of them are costly, time consuming and labor intensive. Some key components of delay measurement, such as intersection delay and delay for different turning movements, are not addressed in the previous studies. Driven by the need to develop an automatic delay estimation system, a method is presented to improve or extend the current methods to estimate intersection travel time based on the detection information derived from ATMIS. Yet, the detection information is not perfect from ATMIS, we still can estimate intersection delay accurately with the help of data filters. The mechanism and result of proposed method are discussed in this thesis. Further work of delay measurement is also presented."--Abstract.