Panos G. Michalopoulos

Vehicle detection video through image processing: the Autoscope system

Recent worldwide developments concerning the use of vehicle detection for wireless large-scale data collection and implementation of advanced traffic control and management schemes such as vehicle guidance/navigation are briefly discussed, and the system developed in the US, Autoscope, is described. Early work on Autoscope, leading to its present advantages over other emerging devices, and recent progress, such as preproduction line prototype development, field testing, and plans for extensive field validation and verification, are examined. The aforementioned plans include two large demonstration projects recently initiated in Minneapolis. In the first project, Autoscope will be used for incident detection over a section of Interstate 394. The second project involves implementation of the machine vision system at a signalized intersection. >

PRACTICAL PROCEDURE FOR CALIBRATING MICROSCOPIC TRAFFIC SIMULATION MODELS

Since the use of simulation is becoming widespread in traffic engineering practice, questions about the accuracy and reliability of its results need to be addressed convincingly. A major criticism related to this issue is proper calibration of the simulation parameters as well as validation, which is often not done or dealt with in an ad hoc fashion. A complete, systematic, and general calibration methodology is presented for obtaining the accuracy needed in high-performance situations. A technique for automating a significant part of the calibration process through an optimization process is also presented. The methodology is general and is implemented on a selected simulator to demonstrate its applicability. The results of the implementation in two freeway sections of reasonable size and complexity, in which detailed data were collected and compared with simulated results, demonstrate the effectiveness of the manual calibration methodology. For instance, through calibration the average volume correlation coefficient on 21 detecting stations improved from 0.78 to 0.96. Comparable results were obtained with the automated calibration procedure with significant time savings and reduced effort.

Continuum modelling of traffic dynamics for congested freeways

This paper investigates advanced continuum models for describing traffic dynamics at congested flows. First, the existing high-order models are reviwed and discussed, and a new formulation is proposed which does not require the use of an equilibrium speed-density relationship. (A model is of high-order when it includes both mass and momentum conservation.) Traffic friction at interrupted flows and changing geometries is also addressed through the use of a viscosity term. The model is then discretized through two alternative numerical schemes to examine the effectiveness of the implementation. Stability analysis is performed to set up numerical limits of the model parameters. Finally, in two test cases with congested and uncongested field data, the new model is compared with the existing models; this comparison suggests that the proposed mode is more accurate and computationally more efficient.

Modelling and analysis of traffic queue dynamics at signalized intersections

Abstract The dynamics of formation and dissipation of queues at isolated signalized intersections are investigated by analyzing the vehicle conservation equation along the street. Solution of the conservation equation is obtained by the method of characteristics for general initial and boundary conditions, and the effect of the control variables (cycle length, green and red intervals) and system parameters (arrival rates, capacity) on the length of the queue is examined. The theoretical results are obtained by examining the dynamics of the shock waves which are formed because of the intermitted service of traffic by the signal. These results include analytical criteria for the development of saturated or unsaturated situations, analytical expressions for the maximum queue length and limits on the control variables so that the queues remain bounded. The basic findings of this study can be employed for the solution of the traffic signal optimal control problem.

Analysis of interrupted traffic flow by finite difference methods

Three numerical techniques for macroscopic analysis of traffic dynamics at signalized links (isolated or coordinated) are presented. The techniques are based on finite differences in time and space and assist in implementing continuum models at real situations. The general methodology presented allows treatment of any arrival and departure pattern, inclusion of sinks or sources, employment of any desired equilibrium flow-density relationship and arbitrary specified initial conditions. Implementation of the proposed method to a signalized intersection and a coordinated link suggests satisfactory agreement with field data and notable agreement with analytical results, respectively. Comparisons made under progressively realistic assumptions demonstrate substantial improvements in model performance as the complexity of the assumptions increases

Multilane traffic flow dynamics: Some macroscopic considerations

The subject of macroscopic modelling and analysis of multilane homodirectional freeway flow is discussed in this paper. Two existing models are extended and treated numerically so that their simplifying assumptions are relaxed. Further, two new formulations are developed; the first is two dimensional with respect to space (i.e. in addition to the street length it includes the street width explicitly) while the second is one dimensional high order dynamic (i.e. it incorporates a momentum equation in order to take into acount acceleration and inertia effects). All modelling alternatives are implemented into a few exemplary situations representing uninterrupted and interrupted flow conditions. Finally, comparisons with aggregate results obtained from microscopic simulation are presented.

Improved high-order model for freeway traffic flow

An improved high-order continuum model is developed based on hyperbolic conservation laws with relaxation, linearized stability analysis, and more realistic considerations of traffic flow. The improved high-order model allows smooth traveling wave solutions as well as contact shocks (different densities moving at the same speed), is able to describe the amplification of small disturbances on heavy traffic, and allows fluctuations of speed around the equilibrium values. Furthermore, unlike existing high-order models, it does not result in negative speeds at the tail of congested regions and disturbance propagation speeds greater than the traffic flow velocity because the improved model has a zero characteristic speed and a nonnegative characteristic speed that is equal to the traffic flow velocity. The relaxation time is a function of density and, in the equilibrium limit, the improved high-order model is consistent with the simple continuum model. The improved high-order model is compared with the simple continuum model. Exemplary test results suggest that the improved high-order model is intuitively correct. Comparison of numerical results with field data suggests that the improved high-order model yields lower error levels than the simple continuum model.

Real-time detection of crash-prone conditions at freeway high-crash locations

Because of growing concern over traffic safety and rising congestion costs, recent research efforts have been redirected from the traditional reactive traffic management (crash detection and clearance) toward on-line proactive solutions for crash prevention. Such a solution for high-crash areas is explored by the identification of the most relevant real-time traffic metrics and the incorporation of them in a model to estimate crash likelihood. Unlike earlier attempts, this model is based on a unique detection and surveillance infrastructure deployed on the freeway section that has the highest crash rate in Minnesota. State-of-the-art infrastructure allowed the video capture of 110 live crashes, crash-related traffic events, and contributing factors while measuring traffic variables (e.g., individual vehicle speeds and headways) over each lane in several places in the study area. This crash-rich database was combined with visual observations and analyzed extensively to identify the most relevant real-time ...

EVALUATION OF RAMP CONTROL EFFECTIVENESS IN TWO TWIN CITIES FREEWAYS

Recent public opposition threatened to eliminate ramp control as a traffic management option in the Twin Cities of Minneapolis and St. Paul, which have one of the most extensive ramp control systems in the nation. In response to this, the Minnesota Department of Transportation had to produce tangible independent evidence that ramp metering is effective, to avoid turning off the meters. Simulation is the most widely accepted technique for achieving the stated objectives without turning the metering system off, and therefore it was used in this study. Two freeway sections were selected for detailed testing, and the results along with the methodology are presented here. The results confirm that ramp metering is effective on the ramp and freeway system (not just the freeway), but they also revealed excessive delays on certain ramps that appear to support the concerns raised by the users. Real-life issues related to the simulation implementation process (data collection and filtering, calibration, and interpreting and summarizing results) are also presented. Through the course of this work, simulation reliability was established by defining a successful calibration and validation methodology and by identifying, in the process, certain operational problems related to the deployed surveillance and control system that had been unknown. Finally, a general methodology was developed for evaluation that can easily be adapted to any user-specified control strategy or used to improve an already existing one without field disruptions.

An application of shock wave theory to traffic signal control

A real time control policy minimizing total intersection delays subject to queue length constraints at an isolated signalized intersection is developed in this paper. The policy is derived from a new traffic model which describes the simultaneous evolution of queue lengths of two conflicting traffic streams, controlled by a traffic light, in both time and space. The model is based on the examination of shock waves generated upstream of the stop lines by the intermittent service of traffic at the signal. The proposed policy was tested against the existing pre-timed control policy at a high volume intersection and it was found superior, especially when demands increase well above the saturation level.