Showing papers in "Transportation Research Part C-emerging Technologies in 1993"

A neural network-based methodology for pavement crack detection and classification

Abstract: This paper presents a methodology for automating the processingof highway pavement video images using an integration of artificial neural network models with conventional image-processing techniques. The methodology developed is able to classify pavement surface cracking by the type, severity, and extent of cracks detected in video images. The approach is divided into five major steps: (1) image segmentation, which involves reduction of a raw gray-scale pavement image into a binary image, (2) feature extraction, (3) decomposition of the image into tiles and identification of tiles with cracking, (4) integration of the results from step (3) and classification of the type of cracking in each image, and (5) computation of the severities and extents of cracking detected in each image. In this methodology, artificial neural network models are used in automatic thresholding of the images in stage (1) and in the classification stages (3) and (4). The results obtained in each stage of the process are presented and discussed in this paper. The research results demonstrate the feasibility of this new approach for the detection, classification, and quantification of highway pavement surface cracking.

Optimal infrastructure management decisions under uncertainty

Abstract: The planning of maintenance and rehabilitation activities for transportation facilities uses information on facility condition from two sources: measurement and forecasting. Both of these sources are characterized by the presence of significant uncertainties, which have important life-cycle cost implications. State-of-the-art decision-making models ignore the uncertainty either in one or both sources of information. This paper presents a methodology (the Latent Markov Decision Process) that explicitly recognizes the presence of random measurement errors in the measurement of facility condition. The methodology can also be used to quantify the “value of more precise information,” which allows an agency to evaluate measurement technologies of different precisions and costs. A parametric study, which demonstrates such an evaluation in the case of highway pavements, is performed.

Simulation Of Freeway Incident Detection Using Artificial Neural Networks

Abstract: A major source of traffic delay in many large urban areas in the United States is non-recurring congestion caused by incidents. In the last several decades, a number of incident detection algorithms have been developed for freeway surveillance and control systems. However, conventional algorithms have generally met with mixed success in terms of performance criteria, such as detection rate, false alarm rate, and the mean time to detect incidents. The need for improved techniques is pressing, particularly with the advent of intelligent vehicle-highway system concepts. These systems will rely heavily on the ability to detect non-recurring traffic congestion automatically. In this paper, we hypothesize that spatial and temporal traffic patterns can be recognized and classified by an artificial neural network, and we present an investigation of such models for the automated detection of lane blocking incidents in a one-mile section of urban freeway. The artificial neural network was trained with data obtained from a microscopic freeway traffic simulation model that was specially calibrated for the actual freeway test section. The neural network first classifies the traffic state of the freeway section into either “incident-free” or “incident” conditions in every 30-second interval. The change in traffic state from incident-free to incident conditions is then used to trigger an incident alarm. Based on the results of an off-line test using simulated data, and comparisons with the well known California incident detection algorithm and the recently developed modified McMaster algorithm, the results suggest that neural network models have the potential to achieve significant improvements in incident-detection performance.

Non-recurrent congestion: How big is the problem? Are traveler information systems the solution?

Abstract: Recent research on highway congestion has calculated that over 50% of delays are non-recurrent (incident produced). A common inference seems to be that non-recurrent delay constitutes over 50% of the “congestion problem.” But this inference overlooks the fact that non-recurrent delays would not be nearly as large if highways were not already overloaded, and that as travelers respond to changes in non-recurrent delay, additional demand could significantly reduce the percentage gain. Within this paper, the issue of incident delay is examined from the alternative perspective of “effective capacity” (which will be equivalent to the expected capacity over time). When evaluated from this view, strategies aimed at alleviating peak-period, incident-caused congestion (such as Incident Management, IM, and Advanced-Traveler-Information-Systems, ATIS) have only a marginal long-term effect on the average delay of congested highways. The conclusion is that neither ATIS nor IM can be relied on as the solution to peak-period congestion. It is also unrealistic to consider either ATIS or IM as an effective alternative to the conventional strategy of adding lanes and building highways.

Behavioral issues in the design and evaluation of advanced traveler information systems

Abstract: Decisions about implementing Advanced Traveler Information Systems (ATIS) should be based on the individual and social benefits expected from such technologies, which will be strongly dependent on the ways travelers respond to these new information sources. This paper explores the behavioral issues important to understanding traveler reactions to ATIS; it discusses evaluation strategies, including stated preference methods and observation of revealed behavior in laboratory simulations and field tests with various degrees of control and complexity. Advantages and disadvantages of different approaches are reviewed, and the experimental design challenges of site selection, recruitment of test subjects, and measurement of behavior are explored.

Modeling the benefits of advanced traveler information systems in corridors with incidents

Abstract: This paper concerns the benefits from Advanced Traveler Information Systems (ATIS) in corridors under incident conditions. A deterministic queueing model is developed and applied to an idealized corridor composed of two routes. A user optimal strategy is implemented to disseminate real-time traffic information to vehicles equipped with ATIS as they approach the incident bottleneck. The sensitivity of route guidance benefits to relevant parameters such as the fraction of vehicles guided with ATIS is analyzed. The findings show that a few cases of queue evolution result when ATIS is used under incident conditions. Both the proportion of guided traffic and the incident duration play an important role in determining which case results. When an incident occurs, ATIS will divert all equipped vehicles to the alternate route until equilibrium is achieved. Equilibrium is achieved only if a sufficient number of guided vehicles comply with diversion instructions. Equilibrium is maintained by reducing the rate of diversion from one route to the other. The implication is that during equilibrium some guided travelers will be diverted to the alternate route, while others will remain on the route where the incident has occurred. It is found that the benefits to guided traffic decrease when the proportion of guided traffic exceeds a critical value that causes a queue on the alternate route. The system benefits also level off once the critical value is exceeded. Therefore, if the system management has the choice, there is no need to equip more than the critical fraction of vehicles with ATIS. There is a need to develop a methodology that can find practical estimates of the critical fraction for use in large-scale simulations of real-life networks.

The effect of motorist information on commuter behavior: classification of drivers into commuter groups

Abstract: This paper describes a methodology originally devised for analysis of travel-activity patterns and applies it to commuters' responses to the influence of traffic information upon commuting decisions. The method of cluster analysis was employed to identity commuter groups (from 3,893 motorists who responded to an on-road survey) with similar patterns of responses to the influence of traffic information. The resulting groups were defined as (a) route changers, willing to change route both on Interstate 5 and before leaving; (b) non-changers, unwilling to change departure time, route, or mode of transportation; (c) route and time changers, willing to change route and departure time; and (d) pre-trip changers, willing to change departure time, route, or mode before departure but unwilling to change en route. Knowledge of such groups and their behavioral characteristics is useful in designing advanced traveler information systems that seek to affect commuter behavior and increase the efficiency of current transportation facilities. This paper discusses the methodology used to derive the commuter groups and investigates their commuting behaviors, decision-making processes, and information needs.

The modelling of dynamic route guidance systems

Abstract: In attempting to simulate the operation of a dynamic route guidance system, the modelling task is concerned both with the operation of the control system and with the implications this has for modelling driver behaviour (whether or not the driver is receiving information from the controller) and network conditions. The aim of this paper is to provide an overview of the modelling issues that need to be considered when addressing such a problem, and that have been identified by various authors in reports on experimental/survey work and in discussion papers. We achieve this by presenting a structured survey of recent research into dynamic route guidance and highlighting issues that are critical to its effectiveness. It is our belief that the development of a model that adequately represents the performance of a dynamic route guidance system is of the utmost importance to the success of the system. It will not only provide a means for evaluating the potential benefits, but should also provide an essential insight into the most appropriate means for its implementation and improve our understanding of transportation networks.

Freeway incident detection through filtering

Abstract: The structure and performance of existing freeway incident detection methods are reviewed and a new method is proposed. The new method reduces the likelihood of false incident decisions that occur during short duration traffic disturbances by smoothing detector occupancies. Temporal smoothing is performed by the statistical mean or median of a data window moved over time. The algorithm considers the smoothed spatial occupancy difference between detector stations, and detects an incident when the difference changes significantly in a short time period. The proposed algorithm and a set of existing algorithms are tested with traffic and incident data from I-35W in Minneapolis. Test results, expressed in terms of detection rate at indicative false alarm rates, demonstrate the improved performance of the new algorithm over previous algorithms.

Some human factors design issues and recommendations for automobile navigation information systems

Abstract: This paper addresses some of the major human factors issues associated with automobile navigation information systems. Human factors objectives of navigation efficiency, ease, safety, and roadway use efficiency are specified. Major decision trade-offs for the display or navigation-related information and the operator control of navigation functions are addressed. When appropriate, design recommendations are made based on the current state of human factors knowledge. For every case, major underlying human factors concerns associated with various design alternatives are discussed.

Demands on driver resources associated with introducing advanced technology into the vehicle

Abstract: Efforts are underway to provide drivers with high-technology systems directed toward improving travel capability, safety, and comfort. In most cases, these systems cause increases in driver workload. This paper describes and ranks the driver resources that may be needed to service these in-vehicle systems, and briefly cites the literature relevant to the potential conflict between performing the primary task of driving and performing in-vehicle tasks associated with such systems.

An information discounting routing strategy for advanced traveler information systems

Abstract: Although there is a lot of enthusiasm and hope that in-vehicle information and route guidance systems will become an integral part of the solution to the traffic congestion problem, there are still a lot of technological and other problems that need to be addressed before such systems operate successfully. This paper examines design aspects of advanced traveler information systems (ATIS) such as frequency of information update, location of information nodes and, most importantly, approaches to estimate the travel times for use in routing decisions and suggestions. In particular, a routing strategy based on information discounting for travel time projection is developed. The importance of the above issues and their implications on the effectiveness of ATIS are demonstrated with a small network.

A neural network for image-based vehicle detection☆

Abstract: Vehicle detection on roadways is useful for a variety of traffic engineering applications from intersection signal control to transportation planning. Traditional detection methods have relied on mechanical or electrical devices placed on top of, or embedded in, pavements. These systems are relatively expensive to install, tend to be unreliable over time, and are limited in their capabilities. Considerable research has been conducted in the area of machine vision for Wide Area Vehicle Detections Systems (WADS). These systems have typically employed conventional image processing and pattern matching algorithms, and many installations have been sensitive to varying lighting conditions, camera perspective, and shadows. In addition, these systems have often required large amounts of computing resources. This paper reports on the development of a new image based vehicle detection system that is based on a simple back propagation/feedforward neural network for tracking vehicles. Application of this concept in a field system is discussed and preliminary results are presented. These results suggest that the neural network vehicle tracking model can be used to reliably detect vehicles. In addition, the training capability of the neural network detection model permits the system to adapt to variations in lighting and camera placement. This should lead to simplified installation and maintenance of WADS.

A dynamic system-optimum control model for commuting traffic corridors

Abstract: The need to implement effective traffic control in commuting corridors has long been recognized by transportation professionals. However, most existing focused either on optimizing freeway ramp metering rates or providing coordinated surface street signals, without taking account of the vital interaction between these two subsystems. As a result, it is not unusual that an effective control strategy for freeway operation may cause significant detrimental effects to the adjacent surface streets. On the other hand, the access to freeway ramps is often impeded by the formation of congestion or bottlenecks on surface streets due to the increasing peak-period traffic demand and ineffective signal operation. This paper presents a dynamic system-optimal control model (DSOCM) for commuting corridors which consist of both freeway and surface street segments. The proposed DSOCM considers the complex interactions among the freeway, surface street and diversion flows, and allows the system operators to compute the optimal time-dependent ramp metering rate and signal setting over the selected time horizon. Depending on the input reliability, DSOCM need not be executed at every control interval as long as the differences between the projected and actual traffic conditions are within the acceptable range. An effective and coordinated control operation for integrated traffic systems can then be achieved.

Adaptive demand-diversion prediction for integrated control of freeway corridors

Abstract: Integrated freeway corridor control, which includes efficient real-time management of freeway traffic diversion onto less congested arterials, is one of the most cost-effective ways to cope with freeway congestion. Because traffic diversion is influenced by ramp metering and intersection signal timing, the effectiveness of an integrated corridor control strategy draws on its ability to predict the diversion resulting from the control in real time. An adaptive demand-diversion predictor is developed that reflects the drivers' choice behavior in a rapidly changing traffic environment. The new method explicitly treats the time-variant effects of control on the traffic demand to be predicted by combining behavioral modeling with filtering. Behavioral demand-diversion models and an extended Kalman filter are developed, with the filter continuously updating the model parameters with the most recent prediction error. The method was applied in several freeway entrance ramps of the Minneapolis-St. Paul metropolitan area freeway system to predict the demand-diversion of traffic flow approaching the ramp area in real time. Following extensive testing and evaluation, the method was incorporated in a new demand-responsive control logic for the online control of freeway corridors.

Exploring real-time traffic simulation with massively parallel computing architecture

Abstract: The advent of parallel computing architectures presents an attainable opportunity for transportation professionals to simulate a large-scale traffic network with sufficiently fast response time for real-time operation. However, it necessitates a fundamental change in the modelling algorithm to take full advantage of parallel computing. Currently there are two general types of parallel processing architectures: (a) single instruction multiple data (SIMD) streams, and (b) multiple instruction multiple data streams (MIMD). This paper describes a model to simulate network traffic with the Connection Machine, a massively parallel SIMD computer. First we introduce the basic parallel computing architectures along with a list of commercially available parallel computers. It is followed by an in-depth presentation of the proposed simulation methodology with a massively parallel computer. The proposed traffic simulation model has an inherent path-processing capability to represent drivers' route choice behavior at the individual/vehicle level. Such a feature is critical to its integration with a real-time dynamic assignment model in IVHS applications. The proposed model has been implemented on the Connection Machine. Several simulation experiments were carried out which show that massively parallel computers provide a viable alternative for use in the real-time application. The results show that the CM-2 with 16,384 processors can simulate 32,000 vehicles for 30 minutes at a one-second interval within 3 1 2 minutes.

A primer on usability testing for developers of traveler information systems

Abstract: This paper reviews the requirements for conducting successful usability tests, particularly on traveler information and traffic management systems; offers a general approach for conducting such tests; and presents a case study of usability tests on an interactive traveler information system.

Validating expert system prototypes using the turing test

Abstract: High-performance, expert-level computer systems require that the expert system prototype be continously evaluated during its development. Expert system validation—that is, testing systems to ascertain whether they achieve acceptable performance levels—has with few exceptions been ad hoc, informal, and of dubious value. This paper describes a variant of the Turing Test technique that may be used to formalize the validation process. A microcomputer-based prototype expert system, the Hazardous Location Analyst (HLA), in the accident location analysis domain was used as a case study of the technique. Turing tests provide a blind method for multiple experts to assess expert system performance qualitatively, and provide a means of determining reasonable performance levels for the particular domain. The HLA has been modified based upon findings from the case study work done previously. The improved prototype was implemented in the City of Greensboro, North Carolina, and a set of 10 case studies are analyzed using the HLA, and by a Greensboro traffic engineer. The results were summarized in identical form and mailed to four traffic engineers outside of Greensboro. Their ratings of case study findings were summarized, and the results used to find a reasonable performance level for this application and to assess the performance of the HLA. The consistency of the expert ratings was also assessed. It was concluded that there exists excellent consistency among experts with regard to human performance, but not with regard to the HLA. This paper demonstrates that the Turing Test as a validation methodology provides an objective, quantitative way to assess system performance.

Driver recruitability for advanced traveler information system experiments

Abstract: A number of Advanced Traveler Information System (ATIS) field experiments are being undertaken to study the effectiveness of the ATIS concept in ameliorating traffic congestion and reducing delays. Many of these experiments require the participation of private drivers willing to allow in-vehicle navigation units to be installed in their vehicles over an extended period of time. A critical part of any ATIS field experiment is the selection or recruitment of private drivers to fulfill the multi-purpose participation needs of the ATIS experiment. To provide an informed basis for designing such a driver recruitment effort, it is important to understand the factors affecting driver recruitability or “willingness to participate.” This research presents the results of a quantitative analysis of driver recruitability conducted to aid in the design of recruitment procedures for ADVANCE (Advanced Driver and Vehicle Advisory Navigation Concept), the largest ATIS field experiment of its kind. The approach used a telephone survey to assess driver willingness to participate in the ADVANCE field experiment and to explore variations in that willingness among different drivers and across characteristics of the ADVANCE system and experimental design. The results indicate that the willingness to participate in the ADVANCE field test is greater for men, persons who hold executive or managerial occupations, individuals who drive extensively, persons who use electronic devices such as personal computers and car phones regularly, and persons who have positive beliefs regarding the usefulness of the ADVANCE concept. The result also suggest that drivers' willingness to participate is not strongly affected by monitoring/reporting requirements such as responding to surveys, mailing electronically stored records of system operation, and periodic service requirements. However, the willingness decreases considerably if drivers have to bear the financial responsibility for damage of the navigation equipment and any equipment-caused electrical failures to the car. Finally, the incentive of a lottery prize raises the level of participation willingness. These results have important implications for the recruitment effort, both in terms of recruiting drivers for participation in the demonstration and specifying the operational details of the field test.

Discrete time-adaptive traffic signal control: the calculation of expected delays and stops

Abstract: The random component of delays and stops can play a significant role in discrete time-adaptive traffic signal control when implemented at isolated intersections. However, the use of Markov chain techniques to calculate expected delay and stops for the latter part of the rolling horizon, where accurate estimates of the undelayed arrival times of individual vehicles at the stopline are not available from detector observations, requires prohibitively large amounts of computer storage and processing. This paper sets out four approximate approaches for the calculation of expected delay and three for the calculation of expected stops, all requiring substantially less storage and processing. These methods are then assessed against results yielded by the Markov chain approach. The assessment suggests that approximations that propagate the first and second moment of the queue-length distribution forward in time give the best results. This approach has the added advantage that it does not presuppose a particular form of vehicle arrival distribution.