The New Jersey Institute of Technology's
Electronic Theses & Dissertations Project

Title:	A conceptual framework for using feedback control within adaptive traffic control systems
Author:	Chhonkar, Renu
View Online:	njit-etd2004-021 (xiv, 154 pages ~ 6.6 MB pdf)
Department:	Executive Committee for the Interdisciplinary Program in Transportation
Degree:	Doctor of Philosophy
Program:	Transportation
Document Type:	Dissertation
Advisory Committee:	Daniel, Janice Rhoda (Committee chair) Bladikas, Athanassios K. (Committee member) Spasovic, Lazar (Committee member) Chien, I-Jy Steven (Committee member) Liu, RongFang (Committee member)
Date:	2004-01
Keywords:	Traffic control Feedback control Conceptual framework Oversaturation
Availability:	Unrestricted
Abstract:	Existing adaptive traffic control strategies lack an effective evaluation procedure to check the performance of the control plan after implementation. In the absence of an effective evaluation procedure, errors introduced in the system such as inaccurate estimates of arrival flows, are carried forward in time and reduce the efficiency of the traffic flow algorithms as they assess prevalent traffic conditions. It is evident that the feed-forward nature of these systems cannot accurately update the estimated quantities, especially during oversaturated conditions. This research is an attempt to develop a conceptual framework for the application of feedback control within the basic operation of existing adaptive traffic control systems to enhance their performance. The framework is applied to three existing adaptive traffic control strategies (SCOOT, SCATS, and OPAC) to enable better demand estimations and queue management during oversaturated condition. A numerical example is provided to test the performance of an arterial in a feedback environment. The example involves the design and simulation test of Proportional (P) and Proportional-Integral (P1) controllers and their adaptability to adequately control the arterial. A sensitivity analysis is further performed to justify the use of a feedback control system on arterials and to choose the type of controller best suited under given demand conditions. The simulation results indicated that for the studied arterial, the P1 controller can handle demand estimation and queuing better than P controllers. It was determined that a well designed feedback control system with a PI controller can effectively overcome some of the deficiencies of existing adaptive traffic control systems.