Development and Field Testing of an Automatic Turning Movements Identification System

The goal of this research is will assist the Ohio Department of Transportation (ODOT) develop and evaluate a real-time system, which can automatically collect the Turning Movements Information at signalized intersections using signal control information and video detection data. To achieve this goal, the researcher, will be working in a partnership with Akron Metropolitan Area Transportation Study and the City of Akron's Traffic Engineering Division. Currently, Turning Movements Information is collected manually with handheld devices in the field, which is tedious and labor intensive involving high cost. Previous efforts on this problem relied on a mathematical model by solving an origin and destination (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 with using supplementary volume data from the local detectors; yet previous studies showed that the results from the O-D method are not accurate. Therefore, driven by the need to identify vehicle turning movements automatically in different geometric and traffic control conditions, an Automatic Turning Movement Identification System (ATMIS) has been developed in the transportation Laboratory of The University of Akron. By interfacing with existing signal controller and detection devices, the system was tested in a laboratory environment using video from the field. The results from the preliminary lab experiment showed that the methodology is very promising and it can potentially be expanded and enhanced for field applications.