Showing papers in "Transportation Research Part C-emerging Technologies in 1995"
TL;DR: It is postulated that a more rigorous approach to matters such as comparison with other techniques and also the methodology used to design the neural networks would help a clearer picture to emerge as to best practice and future research directions.
Abstract: This paper attempts to summarise the findings of a large number of research papers concerning the application of neural networks to transportation. A brief introduction to neural networks is included, for the benefit of readers unfamiliar with the techniques. Because the subject is so young, some of the papers appear only in conference proceedings or other less formal publications. I make no apology for this; I felt it was important to cover as much of the contemporary work as was possible. The paper surveys both the application areas found to be fruitful and the range of neural network paradigms which have been used. Not surprisingly, multilayer feedforward networks such as backpropagation have so far been by far the most popular, but there are signs of a growing diversity; practitioners using neural networks are urged to seek out the less well known paradigms and experiment with them themselves. A particular weakness noted in much of the work is the informal approach taken to detailed analysis of the results of the research. It is postulated that a more rigorous approach to matters such as comparison with other techniques and also the methodology used to design the neural networks would help a clearer picture to emerge as to best practice and future research directions.
403 citations
TL;DR: This algorithm constitutes one of the three major components of an AI-based hybrid solution approach to solving the transit network design problem and implements efficient search techniques using AI tools, algorithmic procedures developed by others, and modules for tools implemented in conventional languages.
Abstract: In this paper we present a Lisp-implemented route generation algorithm (RGA) for the design of transit networks. Along with an analysis procedure and an improvement algorithm, this algorithm constitutes one of the three major components of an AI-based hybrid solution approach to solving the transit network design problem. Such a hybrid approach incorporates the knowledge and expertise of transit network planners and implements efficient search techniques using AI tools, algorithmic procedures developed by others, and modules for tools implemented in conventional languages. RGA is a design algorithm that 1. (a) is heavily guided by the demand matrix, 2. (b) allows the designer's knowledge to be implemented so as to reduce the search space, and 3. (c) generates different sets of routes corresponding to different trade-offs among conflicting objectives (user and operator costs). We explain in detail the major components of RGA, illustrate it on data generated for the transit network of the city of Austin, TX, and report on the numerical experiments conducted to test the performance of RGA.
275 citations
TL;DR: In this paper, the authors present a rolling horizon framework for addressing the real-time traffic assignment problem, where an ATIS/ATMS controller is assumed to have O-D desires up to the current time interval, and short-term and medium-term forecasts of future O -D desires.
Abstract: Existing dynamic traffic assignment formulations predominantly assume the timedependent O-D trip matrix and the time-dependent network configuration to be known a priori for the entire planning horizon. However, there is also a need to provide real-time path information to network users under ATIS/ATMS when unpredicted variations in O-D desires and/or network characteristics (e.g. capacity reduction on certain links due to incidents) occur. This paper presents a rolling horizon framework for addressing the real-time traffic assignment problem, where an ATIS/ATMS controller is assumed to have O-D desires up to the current time interval, and short-term and mediumerm forecasts of future O-D desires. The assignment problem is solved in quasi-real time for a near-term future duration (or stage) to determine an optimal path assignment scheme for users entering the network in real-time for the short-term roll period. The resulting model is intricate due to the intertemporal dependencies characterizing this problem. Two formulations are discussed based on whether a capability to reroute vehicles en route exists. A rolling horizon solution procedure amenable to a quasi-real time implementation of a multiple user classes (MUC) time-dependent traffic assignment solution algorithm developed previously by the authors is described. Implementation issues are discussed from the perspective of ATIS/ATMS applications.
217 citations
TL;DR: The finding that it is possible to relate standard GPS signal quality indicators to increased errors in speed and position provides an enhanced degree of confidence in the use of the GPS system for real-time traffic observations.
Abstract: Much of the research and development work in intelligent vehicle-highway systems (IVHS) relies on the availability of methods for locating and monitoring vehicles (e.g. “probe vehicles”) in real time across a road network. This paper considers the use of the global positioning system (GPS) as one method for obtaining information on the position, speed and direction of travel of vehicles. It reports the results of a series of field studies, in which real-time GPS data were compared to data collected by an instrumented vehicle, under a range of physical and traffic conditions. The field studies and consequent data analysis provide a picture of the reliability and usefulness of GPS data for traffic monitoring purposes, and hence the possibilities for the use of GPS in IVHS projects. The use of GPS receivers tailored for mobile applications, and able to provide direct observations of vehicle speed and travel direction, coupled with database management using geographic information systems (GIS) software, was found to provide a reliable and efficient system for vehicle monitoring. Field data collection under “ideal” GPS conditions indicated that accurate speed and position data were readily obtained from the GPS. Under less favourable conditions (e.g. in downtown networks), data accuracy decreased but useful information could still be obtained. In addition, the conditions and situations under which GPS data errors could be expected were noted. The finding that it is possible to relate standard GPS signal quality indicators to increased errors in speed and position provides an enhanced degree of confidence in the use of the GPS system for real-time traffic observations.
192 citations
TL;DR: Three types of neural network models, namely the multi-layer feedforward (MLF), the self-organizing feature map (SOFM) and adaptive resonance theory 2 (ART2), were developed to classify traffic surveillance data obtained from loop detectors, with the objective of using the classified output to detect lane-blocking freeway incidents.
Abstract: A major source of urban freeway delay in the US is non-recurring congestion caused by incidents The automated detection of incidents is an important function of a freeway traffic management center A number of incident detection algorithms, using inductive loop data as input, have been developed over the past several decades, and a few of them are being deployed at urban freeway systems in major cities These algorithms have shown varying degrees of success in their detection performance In this paper, we present a new incident detection technique based on artificial neural networks (ANNs) Three types of neural network models, namely the multi-layer feedforward (MLF), the self-organizing feature map (SOFM) and adaptive resonance theory 2 (ART2), were developed to classify traffic surveillance data obtained from loop detectors, with the objective of using the classified output to detect lane-blocking freeway incidents The models were developed with simulation data from a study site and tested with both simulation and field data at the same site The MLF was found to have the highest potential, among the three ANNs, to achieve a better incident detection performance The MLF was also tested with limited field data collected from three other freeway locations to explore its transferability Our results and analyzes with data from the study site as well as the three test sites have shown that the MLF consistently detected most of the lane-blocking incidents and typically gave a false alarm rate lower than the California, McMaster and Minnesota algorithms currently in use
160 citations
TL;DR: In this article, a unified approach to the design of integrated control strategies for traffic corridors of arbitrary topology including both motorways and signal-controlled urban roads is presented based on suitable application of the store-and-forward modeling philosophy that leads to the formulation of a linear optimal control problem involving a number of possible control actions, such as ramp metering, signal control, motorway-to-motorway control, route guidance, and VMS control.
Abstract: The paper presents a unified approach to the design of integrated control strategies for traffic corridors of arbitrary topology including both motorways and signal-controlled urban roads The presented approach is based on suitable application of the store-and-forward modeling philosophy that leads to the formulation of a linear optimal-control problem involving a number of possible control actions, such as ramp metering, signal control, motorway-to-motorway control, route guidance, and VMS control The control objective is minimisation of a common criterion, such as the total delay or the total time spent in the network The formulated optimal-control problem may be resolved in real time using suitable algorithms to provide traffic-responsive queue management, particularly under saturated traffic conditions The presentation of numerical results and case studies is to follow in subsequent publications
160 citations
TL;DR: In this article, a behavioral model of transit path choice is presented that frames the choice in terms of a decision whether to board a departing vehicle, and this path choice model accommodates network travel times that are both stochastic and time-dependent, two elements that have been neglected in previous studies.
Abstract: This paper considers information systems in public transit in which the passenger receives information in real time regarding projected vehicle travel times. Such information systems may have value to passengers in situations where they may choose among different origin-to-destination paths. To provide a preliminary assessment of these systems, an analytic framework is presented to evaluate path choices and travel time benefits resulting from real-time information. A behavioral model of transit path choice is presented that frames the choice in terms of a decision whether to board a departing vehicle. Furthermore, this path choice model accommodates network travel times that are both stochastic and time-dependent, two elements that have been neglected in previous studies but are critical to evaluating real-time information systems. The path choice model is extended to demonstrate how real-time information may be incorporated by the passenger in making a path choice decision. This analytic framework is applied to a case study corridor at the Massachusetts Bay Transportation Authority (MBTA), using a computer simulation to model vehicle movements and passenger path choices in the corridor. The results suggest that real-time information yields only very modest improvements in passenger service measures such as the origin-to-destination travel times and the variability of trip times. Based on this analysis, the quantitative benefits of real-time information for transit passenger path choices appear to be questionable.
149 citations
TL;DR: This paper presents a dynamic traffic assignment model with traffic-flow relationships based on a bi-level optimization framework that prevents violations of the first-in-first-out (FIFO) condition using constraints on the distances moved by vehicles during each time step.
Abstract: Conventional traffic assignment methods assume that the origin-destination (OD) demand is uniformly distributed over time to estimate the traffic pattern. This assumption does not hold for modeling peak periods of congestion in which the OD demand is time varying. In this paper, we present a dynamic traffic assignment model with traffic-flow relationships based on a bi-level optimization framework. Our assignment variable is the number of vehicles present on a link during a time step, rather than traffic flow, which is used in static assignment. Using the modified Greenshields speed-density relationship, we derive a link-cost function that is monotonically nondecreasing and convex with respect to density. To capture traffic dynamics, we use short time-steps. The model prevents violations of the first-in-first-out (FIFO) condition using constraints on the distances moved by vehicles during each time step. A solution algorithm which resembles a Stackelberg leader-follower problem is presented, and numerical results from networks of different sizes demonstrate that the proposed model performs satisfactorily.
133 citations
TL;DR: Autonomous Dial-A-Ride Transit (ADART) as discussed by the authors employs fully-automated order-entry and routing-and-scheduling systems that reside exclusively on board the vehicle.
Abstract: This paper introduces a modernized version of many-to-few dial-a-ride called autonomous dial-a-ride transit (ADART), which employs fully-automated order-entry and routing-and-scheduling systems that reside exclusively on board the vehicle. Here, “fully automated” means that under normal operation, the customer is the only human involved in the entire process of requesting a ride, assigning trips, scheduling arrivals and routing the vehicle. There are no telephone operators to receive calls, nor any central dispatchers to assign trips to vehicles, nor any human planning a route. The vehicles' computers assign trip demands and plan routes optimally among themselves, and the drivers' only job is to obey instructions from their vehicle's computer. Consequently, an ADART vehicle fleet covers a large service area without any centralized supervision. In effect, the vehicles behave like a swarm of ants accomplishing their chore without anyone in charge.
107 citations
TL;DR: The results indicate that if drivers behave according to boundedly rational principles without being provided with information in a road network with non-recurrent congestion, the road network will not be used efficiently in terms of travel time.
Abstract: This paper analyses the potential of advanced traveller information systems (ATIS) in a road network in which incidents are generated in a random fashion. A simulation model is applied in which the traffic flows are the aggregation of drivers' decisions. These decisions, in turn, are modelled using boundedly rational principles. The experiments performed focus on the relationship between the network wide performance, the level of market penetration, the quality of the information, and the en route switching propensity. The results indicate that if drivers behave according to boundedly rational principles without being provided with information in a road network with non-recurrent congestion, the road network will not be used efficiently in terms of travel time. In these circumstances, ATIS is useful. However, the commercial viability of ATIS might be frustrated by the quickly diminishing additional benefits to equipped drivers. Further, the complexity of the implications of ATIS is stressed by the strong interaction between, on the one hand, the level of market penetration, the quality of the information and the en route switching propensity and, on the other hand, the network wide performance.
103 citations
TL;DR: It was generally found that data from a single link provided almost equally good incident detection as data obtained from pairs of links, thereby supporting incident detection on any link that has a current data independent of data availability from other links.
Abstract: This paper describes incident detection algorithms for urban arterial streets using two distinct data sources: fixed traffic detectors and probe vehicles. The data sources are used independently to obtain two distinct algorithms. This approach is undertaken to increase the overall coverage of incident detection capabilities as early implementation will result in relatively few cases when data is available from both fixed detectors and probe vehicles on the same link and during the same time period. The algorithms were developed using simulation data for the ADVANCE ITS operational test; they will subsequently be recalibrated with field data collected during the ADVANCE demonstration project. Discriminant analysis was used to estimate a variety of models based on different traffic flow measures from each data source. Various functions of fixed detector measures (volume and occupancy) and probe vehicle travel times were considered for inclusion in the fixed detector and probe vehicle algorithms, respectively. The most effective variables for detecting incidents were volume divided by occupancy (which is related to average speed) for fixed detectors and average speed for probe vehicles. In both cases, traffic measures for the incident link were most useful for incidents located in the downstream portion of the link and for the next upstream link for incidents located at the upstream end or in the middle portion of the link. Further, it was generally found that data from a single link provided almost equally good incident detection as data obtained from pairs of links. This led to the development of an algorithm that uses data from a single detector or link, thereby supporting incident detection on any link that has a current data independent of data availability from other links. The performance of the algorithms was evaluated using detection rates and false alarm rates, which were found to be in the same range for both the algorithms. The fixed detector algorithm showed better detection ability, but its use is limited by the number of detectorized links in the network, while the performance of the probe vehicle algorithm was dependent on the number of reports available per time period.
TL;DR: Recommendations for simulator design characteristics that increase the reliability of the data collected are made and enhancements are suggested so that current simulators can be used for the collection of data related to access and acquisition of ATIS products as well.
Abstract: Understanding traveler response to potential ATIS services is critical for designing such services and evaluating their effectiveness. Extensive data is required for developing the models necessary to provide this understanding. In this paper we examine one source of such data: traveler simulators. We make a distinction between travel simulators, used to study the travelers response to information acquisition, and driving simulators, which are elaborate tools used mainly for human factors research. Traveler simulators have the potential to provide a wealth of data collected relatively inexpensively under controlled conditions. However the data may suffer from biases introduced because of the laboratory nature of travel simulators. We examine various existing simulators and comment on their advantages and disadvantages. We make recommendations for simulator design characteristics that increase the reliability of the data collected and suggest enhancements so that current simulators can be used for the collection of data related to access and acquisition of ATIS products as well. We conclude the paper with recommendations for future research in the area.
TL;DR: A general approach for real time traffic management support using knowledge based models is described, and it is concluded that such an approach is feasible, and is compatible with existing state of the art traffic control systems.
Abstract: This paper describes a general approach for real time traffic management support using knowledge based models. Recognizing that human intervention is usually required to apply the current automatic traffic control systems, it is argued that there is a need for an additional intelligent layer to help operators to understand traffic problems and to make the best choice of strategic control actions that modify the assumption framework of the existing systems. The need for an open architecture is stated, in order to allow users to modify decision criteria according to their experience, given that no skills are available yet to deal with real time strategy decision making. An architecture of knowledge is described that is oriented towards traffic management strategic advice applied in the TRYS system developed by the authors. This system has been installed for urban motorway control in several Spanish cities. Finally, an example of knowledge-based modeling, using TRYS, is presented in a case study where both the TRYS model and its operation are described. It is concluded that such an approach is feasible, and is compatible with existing state of the art traffic control systems.
TL;DR: The proposed models, although lacking in mathematical elegance, are capable of providing the acceptable prediction accuracy at 3 time-steps ahead and are sufficiently detailed for both responsive signal control and intersection operations.
Abstract: To capture the complex nature of intersection queue dynamics, this study has explored the use of neural network models with data from extensive simulation experiments. The proposed models, although lacking in mathematical elegance, are capable of providing the acceptable prediction accuracy (more than 90%) at 3 time-steps ahead. As each time-step is as short as 3 s, the resulting information on queue evolution is sufficiently detailed for both responsive signal control and intersection operations. To accommodate the differences in available surveillance systems, this study has also investigated the most suitable neural network structure for each proposed queue model with extensive exploratory analyses.
TL;DR: In this paper, the authors conducted three nationwide surveys designed to obtain user information requirements for an advanced traveler information system (ATIS) and the ATIS portion of commercial vehicle operations (CVO).
Abstract: This paper discusses three nationwide surveys designed to obtain user information requirements for an advanced traveler information system (ATIS) and the ATIS portion of commercial vehicle operations (CVO). A description of the survey methodology for targeting the three populations—private vehicle drivers, commercial vehicle drivers, and commercial vehicle operators—is provided, as well as the implications for the design of ATIS based on driving behavior and preference. Analysis of the data from the 1,610 returned surveys revealed that commercial drivers and private drivers valued road and traffic information as the most important ATIS service, whereas commercial vehicle operators (or dispatchers) valued the ability to have two-way communication as most important. This study also revealed that dispatchers were willing to pay significantly more for an ATIS than drivers. Finally, the data indicated that, of the four main nationwide geographical areas considered in the survey, no differences in response to the survey questions were observed. Implications of the results for the design of ATIS are discussed.
TL;DR: Model estimation results show that travelers' socio-economics, habitual travel patterns, commute congestion levels and attitudes toward in-vehicle technologies are significant determinants of travelers' importance ratings and the preferred distance ahead of in-Vehicle system information.
Abstract: A sample of travelers' preferences toward in-vehicle traffic information systems was undertaken and appropriate statistical models were estimated. Specifically, ordered logit and regression analyses were conducted to quantify travelers' ratings of the importance of in-vehicle system attributes, and the distance ahead that they prefer to be notified of various types of information provided by in-vehicle systems. Model estimation results show that travelers' socio-economics, habitual travel patterns, commute congestion levels and attitudes toward in-vehicle technologies are significant determinants of travelers' importance ratings and the preferred distance ahead of in-vehicle system information. These model results provide important information for both marketing and design of in-vehicle information systems.
TL;DR: System optimum and user equilibrium dynamic assignments on an 18-arc test network are compared in terms of total travel times and schedule delays at different levels of traffic congestion and provide important implications for the success of the intelligent transportation systems (ITS) in reducing traffic congestion.
Abstract: One way to estimate the potential benefits of new traffic control and management systems is to compare the total cost incurred in equilibrium with the system optimized total cost. To do this, we formulate the dynamic traffic assignment models with schedule delays under the system optimum and user equilibrium principles and solve them using numerical methods. System optimum and user equilibrium dynamic assignments on an 18-arc test network are then compared in terms of total travel times and schedule delays at different levels of traffic congestion. This comparison provides important implications for the success of the intelligent transportation systems (ITS) in reducing traffic congestion.
TL;DR: In this paper, the authors analyzed the responses from two nationwide surveys designed to obtain user information requirements for the design of advanced traveler information systems (ATIS) and commercial vehicle operations (CVO) with respect to commercial system operators (dispatchers) and Commercial vehicle drivers.
Abstract: This paper analyzes the responses from two nationwide surveys designed to obtain user information requirements for the design of advanced traveler information systems (ATIS) and commercial vehicle operations (CVO) with respect to commercial system operators (dispatchers) and commercial vehicle drivers. A total of 673 returned surveys (348 dispatcher surveys and 325 commercial driver surveys), were used in the analysis. Mathematical models were developed, using a binomial logit to predict whether the commercial driver or dispatcher would use an intelligent transportation system, and an ordered probit to estimate the importance of information (i.e. route and navigation, roadside services, personal communication and road and traffic information) to be provided by in-vehicle information systems. The results of this study provide guidelines for the design of information systems and help define informational requirements for users of ATIS/CVO.
TL;DR: The modeling of the basic elements of SmartPath is discussed, which describes the organization of these elements in the simulation and the assumptions underlying the design, such as availability of certain communication infrastructure and appropriate sensors.
Abstract: SmartPath is simulation package for an automated highway system (AHS). The program may be used to understand how an AHS would perform under various control policies in terms of highway capacity, traffic flow, and other performance measures of interest to transportation system planners and engineers. SmartPath also can be used to test, simulate, and evaluate the performance of the designs of different modules and instrumentations like engine models, sensors, and communications. The package consists of two separate modules: simulation and animation. The animation program runs on Silicon Graphics workstations and the simulation runs on Sun Spare or Silicon Graphics workstations. The animation program produces a three-dimensional color animation of AHS traffic. SmartPath is a microsimulation: the system elements and the control policies are each individually modeled. The control policies are, for the most part, parametrically specified, so users can study the performance variations by changing the specifications. In this paper, we discuss the modeling of the basic elements of SmartPath. We describe the organization of these elements in the simulation and the assumptions underlying the design, such as availability of certain communication infrastructure and appropriate sensors. We also summarize SmartPath's computational performance.
Journal Article•
TL;DR: In this article, the authors explored the use of neural network models with data from extensive simulation experiments to capture the complex nature of intersection queue dynamics, and the proposed models, although lacking in mathematical elegance, are capable of providing the acceptable prediction accuracy (more than 90 percent) at 3 time-steps ahead.
Abstract: To capture the complex nature of intersection queue dynamics, this study has explored the use of neural network models with data from extensive simulation experiments. The proposed models, although lacking in mathematical elegance, are capable of providing the acceptable prediction accuracy (more than 90 percent) at 3 time-steps ahead. As each time-step is as short as 3 s, the resulting information on queue evolution is sufficiently detailed for both responsive signal control and intersection operations. To accommodate the differences in available surveillance systems, this study has also investigated the most suitable neural network structure for each proposed queue model with extensive exploratory analyses.
TL;DR: Issues that are important in selecting the optimal neural network model including the number of hidden layers, their units, learning rule, tiling characteristics of the input image and the output representation of the network—are addressed in this paper.
Abstract: Current vision-based vehicle detection systems use image-processing algorithms to monitor the presence of vehicles on the roads. Recent research has shown that an artificial feedforward neural network can be trained to provide similar capabilities. A properly trained and configured network should be able to recognize the presence of vehicles in the images it has never been exposed to. This paper discusses the development of a feedforward-backpropagation neural network-based vehicle detection system that recognizes and tracks vehicles with satisfactory reliability and efficiency. Various issues that are important in selecting the optimal neural network model—like the architecture of the network including the number of hidden layers, their units, learning rule, tiling characteristics of the input image and the output representation of the network—are addressed in this paper. This paper also analyzes how the neural network internally learns the mapping knowledge of the input-output training pairs. The final section describes an output post processor that produces the traditional pulse and presence signals.
TL;DR: A framework, based on the following four elements, is presented for guiding changes in transportation system design, which is meant to apply to all forms of transportation and material handling.
Abstract: Development of new technologies for vehicle control and wireless communication will make automated transportation increasingly common over the next decade As these technologies evolve, fundamental changes will occur in transportation system design Within this paper, a framework, based on the following four elements, is presented for guiding these changes 1 (a) A vehicle/guideway/terminal/ coordinator classification of transportation entities, and a sensor/intelligence/memory/transmitter/ actuator classification of devices; 2 (b) Functional residence of actuators; 3 (c) Communication media and “illocutionary point” of messages; 4 (d) Degree of coordination, residence of control functions and topology Examples applications are provided for intelligent transportation systems However, the framework is meant to apply to all forms of transportation and material handling
TL;DR: The paper takes a systems view and identifies the safety issues regarding the movement coordination of the large number of vehicles on an Automated Highway System (AHS).
Abstract: The task of traffic control for automated highway systems (AHS) is drastically different from and much more complex than its conventional counterpart. This paper proposes a conceptual framework for designing a traffic control scheme. It adopts a top-down approach to defining major design steps starting with high-level feature definition. Since all AHS control features materialize through vehicle movements and there exists an infinite number of possible vehicle movements, specifying these movements and verifying that they indeed suffice for the desired features could be extremely complicated. One approach to simplifying such tasks is to define a small number of permissible moves as “building blocks” and define all permissible movements in terms of these moves. With the desired features defined, the top-down approach then identifies and defines moves and related planning and movement functions that are required for supporting the desired features. Central to traffic control is planning, including system flow planning and vehicle movement planning. The former tries to optimize the macroscopic flow of aggregate traffic in the AHS while the latter plans for the microscopic movement of individual vehicles. Making the actual movements according to the vehicle movement plans requires the most detailed data about the immediate neighborhood affecting or affected by the movements. To ensure safety, initiation/continuation/ abort conditions for all permissible vehicle moves must be clearly and safely defined at this level of detail. Since vehicle movement plans are generated by various controllers in the AHS, the planned vehicle moves have the potential of conflicting and interfering with one another. Based on key attributes of a move and the concept of initiation/continuation/abort conditions, we are able to define rigorously the concepts of conflict and interference among different moves. Such conflicts must be recognized and resolved in time for safety; they should also be prevented at the planning stage. Such interferences should be minimized for efficiency. A “running” example defining the traffic control scheme of a simplified AHS operating scenario is provided for illustration.
TL;DR: In this article, a system of automatic control for car traffic is proposed based on the similarity with physical thermodynamic systems and providing effective correlations between moving cars in flow, the control force on a car is defined through an artificial force function, depending on the distance between the adjacent cars but also possessing a non-conservative velocity dependence.
Abstract: A system of automatic control for car traffic is proposed based on the similarity with physical thermodynamical systems and providing effective correlations between moving cars in flow. The control force on a car is defined through an “artificial force” function, depending on the distance between the adjacent cars but also possessing a nonconservative velocity dependence. The latter assures existence of stable equilibrium states of the system under control in various traffic situations. A practical realization mode for such correlative control is outlined. Several simple examples of the system action are considered, demonstrating its capacities in establishing a secure and fluid mode of traffic flow.
TL;DR: The work described here arose from the application of a novel deductive DBMS called PFL (Persistent Functional Language) to the storage and manipulation of road accident data.
Abstract: Transport data are typical of many application areas in that they arise from a variety of sources and are used in various ways. Furthermore, important information that is required in practical applications of transport databases is often not stored explicitly, but rather has to be deduced from some that is. It is therefore natural to consider the application in this field of deductive database management systems (DBMSs). These extend traditional DBMSs by permitting the definition of inference rules, default rules, complex data structures and integrity constraints, each of which can be used to provide facilities of substantial practical value. The work described here arose from the application of a novel deductive DBMS called PFL (Persistent Functional Language) to the storage and manipulation of road accident data. Although the primary entity type in such a database is that of an accident, queries are typically based upon the concept of a site. Because site information is not stored explicitly in the database, it must be deduced from auxiliary information that provides some indication of location. The combination of large amounts of data and computationally intensive queries presents extraordinary demands for the database system and has led to the development of new software techniques of high efficiency in both computation and data manipulation.