TRANSPORTATION RESEARCH
`
`RECORD
`
`No. 1408
`
`Highway Operations, Capacity, and
`Traffic Control
`
`Intelligent Vehicle
`Highway Systems
`
`A peer-reviewed publication of the Transportation Research Board
`
`TRANSPORTATION RESEARCH BOARD
`NATIONAL RESEARCH COUNCIL
`
`NATIONAL ACADEMY PRESS
`WASHINGTON, D.C. 1993
`
`Google Ex. 1018
`
`

`

`Transportation Research Record 1408
`ISSN 0361-1981
`ISBN 0-309-05555-5
`Price $28.00
`
`Subscriber Category
`IV A highway operations, capacity, and traffic control
`
`TRB Publications Staff
`Director of Reports and Editorial Services: Nancy A. Ackerman
`Associate Editor/Supervisor: Luanne Crayton
`Associate Editors: Naomi Kassabian, Alison G. Tobias
`Assistant Editors: Susan E. G. Brown, Norman Solomon
`Production Coordinator: Sharada Gilkey
`Graphics Coordinator: Terri Wayne
`Office Manager: Phyllis D. Barber
`Senior Production Assistant: Betty L. Hawkins
`
`Printed in the United States of America
`
`Sponsorship of Transportation Research Record 1408
`
`GROUP I-TRANSPORTATION SYSTEMS PLANNING AND
`ADMINISTRATION
`Chairman: Sally Hill Cooper, Virginia Department of
`Transportation
`
`Transportation Forecasting, Data, and Economics Section
`Chairman: Mary Lynn Tischer, Virginia Department of -
`Transportation
`
`Committee on Transportation Supply Analysis
`Chairman: Hani S. Mahmassani, University of Texas at Austin
`David E. Boyce, Yupo Chan, Carlos F. Daganzo, Mark S. Daskin,
`Michel Gendreau, Theodore S. Glickman, Ali E. Haghani,
`Randolph W. Hall, Rudi Hamerslag, Bruce N. Janson, Haris N.
`Koutsopoulos, Chryssi Malandraki, Eric J. Miller, Anna Nagurney,
`Earl R. Ruiter, K. Nabil A. Safwat, Mark A. Turnquist
`
`GROUP 3-0PERATION, SAFETY, AND MAINTENANCE OF
`TRANSPORTATION FACILITIES
`Chairman: Jerome W. Hall, University of New Mexico
`
`Facilities and Operations Section
`Chairman: Jack L. Kay, JHK & Associates
`
`Committee on Communications
`Chairman: T. Russell Shields, Navigation Technologies, Inc.
`Walter A. Albers, Jr., Kenneth C. Allen, James W. Bourg,
`E. Ryerson Case, Kan Chen, Min I. Chung, Robert L. French,
`Charles J. Glass, L. F. Gomes, Robert L. Gordon, Jan E.
`He/laker, Gabriel Heti, Randall Jones, Kevin Kelley, Gerard J.
`Kerwin, Allan M. Kirson, Joseph A. LoVecchio, Wesley S. C.
`Lum, Roger D. Madden, Said Majdi, Frank J. Mammano, Patrick
`F. McGowan, B. F. Mitchell, Corwin D. Moore, Jr., Heinz
`Sodeikat, Richard E. Stark, Philip J. Tarnoff
`
`Committee on Traffic Signal Systems
`Chairman: Herman E. Haenel, Advanced Traffic Engineering
`Secretary: Alberto J. Santiago, Federal Highway Administration
`A. Graham Bullen, E. Ryerson Case, Edmond Chin-Ping Chang,
`David J. Clowes, Robert A. De Santo, Donald W. Dey, Gary
`Duncan, Nathan H. Gartner, Robert David Henry, Dallas W.
`Hildebrand, Paul P. Jovanis, Les Kelman, Alfred H. Kosik, Joseph
`K. Lam, Feng-Bor Lin, David C. Powell, Raymond S. Pusey,
`Dennis I. Robertson, Lionel M. Rodgers, Stephen Edwin Rowe,
`Tom L. Stout, Miroslav Supitar, James A. Thompson, Charles E.
`Wallace
`
`GROUP 5-INTERGROUP RESOURCES AND ISSUES
`Chairman: Patricia F. Waller, University of Michigan
`
`Committee on High-Occupancy Vehicle Systems
`Chairman: Donald G. Cape/le, Parsons Brinckerhoff
`Secretary: Dennis L. Christiansen, Texas A&M University
`David E. Barnhart, John W. Billheimer, John Bonsall, Donald J.
`Emerson, Charles Fuhs, Alan T. Gonseth, Leslie N. Jacobson,
`William A. Kennedy, Theodore C. Knappen, James R. Lightbody,
`Timothy J. Lomax, Adolf May, Jr., Jonathan David McDade, C. J.
`O'Connell, R. L. Pierce, Lew W. Pratsch, Morris J. Rothenberg,
`Sheldon G. Strickland, Gary K. Trietsch, Katherine F. Turnbull,
`Carole B. Valentine, Jon Williams
`
`Committee on Intelligent Vehicle Highway Systems
`Chairman: Daniel Brand, Charles River Associates, Inc.
`Sadler Bridges, Melvyn Ches/ow, Patrick J. Conroy, Randolph M.
`Doi, Eugene Farber, Ronald J. Fisher, Robert L. French, William
`J. Harris, Jr., Thomas A. Horan, Eva Lerner-Lam, Hani S.
`Mahmassani, Joel E. Markowitz, Stephen Edwin Rowe, Donald A.
`Savitt, Lyle Saxton, T. Russell Shields, Steven E. Shladover,
`William M. Spreitzer, Richard J. Weiland
`
`James A. Scott and Richard A. Cunard, Transportation Research
`Board staff
`
`Sponsorship is indicated by a footnote at the end of each paper.
`The organizational units, officers, and members are as of
`December 31, 1992.
`
`Google Ex. 1018
`
`

`

`Transportation Research Record 1408
`
`Contents
`
`Foreword
`
`Intelligent Vehicle Highway System Benefits Assessment Framework
`Daniel Brand
`
`Comparison of Advanced Traffic Management and Traveler
`Information System Architectures for Intelligent Vehicle
`Highway Systems
`Melvyn D. Cheslow and S. Gregory Hatcher
`
`Intelligent Vehicle Highway System Safety: Specification and
`.Hazard Analysis of a System with Vehicle-Borne Intelligence
`A. Hitchcock
`
`Investigations into Achievable Capacities and Stream Stability with
`Coordinated Intelligent Vehicles
`B. S. Y. Rao, P. Varaiya, and F. Eskafi
`
`Flow Benefits of Autonomous Intelligent Cruise Control in Mixed
`Manual and Automated Traffic
`B. S. Y. Rao and P. Varaiya
`
`Automatic Speed Monitor: An Intelligent Vehicle Highway System
`Safe-Speed System for Advance Warning or Hazardous Speed
`Monitoring
`Alan R. Kaub and Thomas Rawls
`
`Engineering Feasibility of Roadway Electrification in a
`High-Occupancy-Vehicle Facility
`T. Chira-Chavala and Edward H. Lechner
`
`v
`
`1
`
`8
`
`18
`
`27
`
`36
`
`44
`
`47
`
`Google Ex. 1018
`
`

`

`Phased Implementation of Lateral Guidance Systems in
`High-Occupancy-Vehicle Lanes
`T. Chira-Chavala and Wei-Bin Zhang
`
`Integrated Approach to Vehicle Routing and Congestion Prediction for
`Real-Time Driver Guidance
`Isam Kaysi, Moshe Ben-Akiva, and Haris Koutsopoulos
`
`Experimental Analysis and Modeling of Sequential Route
`Choice Under an Advanced Traveler Information System in a
`Simplistic Traffic Network
`Kenneth M. Vaughn, Mohamed A. Abdel-Aty, Ryuichi Kitamura, Paul P.
`Jovanis, Hai Yang, Neal E. A. Kroll, Robert B. Post, and Brian Oppy
`
`Network Performance Under System Optimal and User
`Equilibrium Dynamic Assignments: Implications for Advanced
`Traveler Information Systems
`Hani S. Mahmassani and Srinivas Peeta
`
`Time-Dependent, Shortest-Path Algorithm for Real-Time Intelligent
`Vehicle Highway System Applications
`Athanasios K. Ziliaskopoulos and Hani S. Mahmassani
`
`Communications Architecture for Early Implementation of Intelligent
`Vehicle Highway Systems
`D. J. Chadwick, V. M. Patel, and L. G. Saxton
`
`Integration of Machine Vision and Adaptive Control in the Fast-Trac
`Intelligent Vehicle Highway System Program
`Panos G. Michalopoulos, Richard D. Jacobson, Craig A. Anderson, and
`James C. Barbaresso
`
`Software for Advanced Traffic Controller~
`'
`Darcy Bullock and Chris Hendrickson
`
`56
`
`66
`
`75
`
`83
`
`94
`
`101
`
`108
`
`116
`
`Google Ex. 1018
`
`

`

`Foreword
`
`The papers in this Record represent results of research studies focusing on intelligent vehicle
`highway systems (IVHS).
`Brand discusses predictive models to evaluate IVHS improvements, including the formu(cid:173)
`lation of inputs that make it possible to anticipate the important consequences of IVHS and
`therefore carry out benefit-cost analysis of new investments as well as collect the appropriate
`data for planning and evaluating operational field tests.
`Cheslow and Hatcher identify and evaluate five alternative architectures for advanced traffic
`management (ATM) and advance traveler information systems (A TIS). These activities were
`created to focus on several key architecture issues.
`Hitchcock discusses the normal operation of one system of automated freeways as described
`by Hsu. This system is being used as a basis for many system engineering studies within
`Partners for Advanced Transit and Highways (PA TH). The system minimizes the degree to
`which the infrastructure is involved in minute-to-minute maneuvers. In it, each vehicle, as
`it enters, is given a route including lane choices to the destination. As described, however,
`no account is given of procedures on entry or exit or of possible faults. The PATH safety
`program demanded a second example of the process of full specification and fault tree analysis
`to determine if this process was generally applicable.
`Rao et aL discuss the application of simulation SmartPath, which models the passage of
`individual vehicles along the highway. The simulator allows the researchers to examine tran(cid:173)
`sient behavior of the traffic stream under various conditions. Three different strategies are
`discussed for allowing vehicles to enter and leave automated lanes and measuring the max(cid:173)
`imum flow rates that are attained. The authors conclude that although maximum theoretical
`capacity cannot be attained, through prudent design of entrance and express strategies ex(cid:173)
`tremely high throughput can be sustained.
`Rao and Varaiya examine the potential flow increases when only a proportion of vehicles
`on a highway are equipped with autonomous intelligent cruise control (AICC). The authors
`use a simulator that models interactions between vehicles to give detailed information on
`achievable capacity and traffic stream stability. The authors conclude that capacity gains from
`AICC are likely to be small.
`Kaub and Rawls present a single and inexpensive IVHS speed monitoring concept that
`can be easily adapted to existing vehicles and roadways. The concept relies on the speed(cid:173)
`distance-time relationship and on an on-board impulse detector and constant times to calculate
`the travel time or posted speed of the roadway.
`Chira-Chavala and Lechner discuss the preliminary engineering feasibility of early de-·
`ployment of a roadway-powered electric vehicle in El Monte Busway (a 3 + high-occupancy(cid:173)
`vehicle facility in Los Angeles). The evaluation consists of determinations of the scale of
`electrification, the location to be electrified, the mode of operation, the level of energy
`transfer, and consumption of energy.
`Chira-Chavala and Zhang discuss the phased implementation of advanced lateral guidance
`systems in high-occupancy vehicle lanes with exclusive right-of-way. Steering assistance in(cid:173)
`formation systems, partially automated lane-keeping systems, and fully automated lateral
`control systems are described.
`Kaysi et al. propose a system structure for real-time traveler information systems consisting
`of a surveillance module, a congestion prediction module, and a control and routing module.
`The focus is on the approaches that may be used for congestion prediction and the strategies
`that may form the basis for routing.
`Vaughn et al. discuss the result of an experiment to collect sequential route choice data
`under the influence of ATIS. The experiment collected information on drivers' pretrip route
`choice behavior at three levels of information accuracy: 60, 75, and 90 percent. The results
`
`v
`
`Google Ex. 1018
`
`

`

`vi
`
`of the experiment indicate that drivers can rapidly identify the accuracy level of information
`that is provided and that they adjust their behavior accordingly.
`Mahmassani and Peeta present a comparative assessment of network cost and performance
`under time-dependent system optimal (SO) and user equilibrium (UE) assignment patterns,
`with reference to the effectiveness of ATIS. Both SO and UE solutions are found using a
`new simulation-based algorithm for the time-dependent assignment problem. The results of
`their work affirm the validity of a meaningful demarcation between system optimal and user
`equilibrium assignments in urban traffic networks and provide useful insights for macroscopic
`network-level relations among traffic descriptors.
`Ziliaskopoulos and Mahmassani discuss an algorithm that calculates the time-dependent
`shortest paths from all modes in a network to a given destination mode for every time step
`over a given time horizon in a network with time-dependent arc costs. The motivation for
`this study was the need to compute time-dependent shortest paths in a real-time environment
`in connection with IVHS. The suitability of the proposed algorithm for such applications is
`demonstrated in this study.
`Chadwick et al. propose a communication architecture for IVHS called the subsidiary
`communications authority traffic information channel (STIC) on the basis of the widely
`available FM radio broadcast services infrastructure by making use of FM subcarrier tech(cid:173)
`nology. According to the authors, the STIC has a higher data transmission capacity than any
`other existing FM subcarrier broadcast system, and it has the potential to meet the one-way
`outbound (broadcast) data transmission capacity needs of IVHS for the next few years.
`Michalopoulos et al. indicate that machine vision is one of the most promising technologies
`for developing and deploying ATMS and A TIS applications. The paper describes the first
`and largest application of video detection and its integration with adaptive control in a network
`of 28 intersections. The broader project (FAST-TRAC) is also summarized.
`Bullock and Hendrickson discuss why improved traffic controllers will be essential for many
`proposed IVHS applications. The authors introduce a computer language called Traffic Con(cid:173)
`trol Blocks (TCBLKS) that could provide the foundation for constructing real-time traffic
`engineering software.
`
`Google Ex. 1018
`
`

`

`66
`
`TRANSPORTATION RESEARCH RECORD 1408
`
`Integrated Approach to Vehicle Routing
`and Congestion Prediction for Real-Time
`Driver Guidance
`
`lsAM KAYSI, MosHE BEN-AKIVA, AND HARIS KouTSOPOULOS
`
`the modeling requirements of its constituent elements are
`analyzed.
`
`INTEGRATED FRAMEWORK FOR
`IMPLEMENTATION OF A TIS
`
`The actual benefits realized from traveler information systems
`depend heavily on the quality of the traffic information pro(cid:173)
`vided to drivers (J). This section describes the framework
`being proposed to provide drivers with guidance that they can
`have confidence in and that, as a result of improved infor(cid:173)
`mation, can eliminate the occurrence of adverse impacts [see
`Ben-Akiva et al. (2)]. The discussion that follows describes
`the system structure and information flow embodied in the
`proposed framework. Later sections present the principles
`behind the proposed framework and describe the control and
`routing (CAR) and congestion prediction (COP) modules.
`
`System Structure
`
`A dynamic network modeling approach is critical to the ef(cid:173)
`fectiveness of real-time traveler information systems. Such a
`modeling approach is needed to accurately assess network
`performance as well as to forecast traffic conditions that may
`exist in the near future to develop real-time diversion strat(cid:173)
`egies to alleviate both recurring and nonrecurring congestion
`conditions.
`A proper framework for A TIS implementation should be
`able to integrate the functional needs referred to above into
`an operational system. Figure 1 illustrates the system structure
`and information flow of the framework within which real-time
`ATIS should be implemented (3). The functions performed
`by each element in the system are briefly described:
`
`m consists of traffic sensors de(cid:173)
`• The surveillance syst
`ployed on the various network elements (for example, de(cid:173)
`tectors in the pavement, video cameras, possibly other optical
`recording equipment). The surveillance system may also in(cid:173)
`clude roadside readers that gather information about vehicles
`that are passing selected points on the network over time.
`Thus, equipped vehicles themselves may act as elements of
`the surveillance system by providing information on travel
`times on specific sections of the network. The collected data
`may consist of information on flows, speeds, travel times, the
`numbers of queued and' moving vehicles on each link, and
`
`1e
`
`The generation and dissemination of driver guidance that can be
`used for real-time diversion of traffic are expected to be imple(cid:173)
`mented through the use of real-time traveler information systems.
`To implement these functions, a system structure consisting of a
`surveillance module, a congestion prediction module, and a con(cid:173)
`trol and routing (CAR) module is proposed, with the focus on
`the approaches that may be used for congestion prediction and
`the strategies that may form the basis for routing. It is argued
`that a congestion prediction capability is critical for the effec(cid:173)
`tiveness of an on-line traveler information system. Such a capa(cid:173)
`bility is required to accurately forecast traffic conditions that may
`exist in the near future. The use of a dynamic traffic assignment
`model for congestion prediction is suggested. Such a model con(cid:173)
`sists of dynamic driver behavior and network performance mod(cid:173)
`ules as well as origin-destination updating capability. Alterna(cid:173)
`tively, statistical time-series methods may be necessary to generate
`predictions of future traffic conditions. The advantages and dif(cid:173)
`ficulties of adopting either approach are discussed. The predicted
`congestion information is passed to the CAR module to develop
`diversion strategies to alleviate both recurring and nonrecurring
`congestion. The role of routing strategies and update frequency
`in determining guidance effectiveness is discussed.
`
`During the coming decades, efficient operation of existing
`road networks is expected to be achieved through dynamic
`traffic management schemes that make use of available and
`anticipated advanced technologies. Within this context, in(cid:173)
`telligent vehicle highway systems (IVHS) are currently being
`developed. These IVHS systems envision the linking of road
`infrastructure, vehicles, and drivers using advanced commu(cid:173)
`nication technology, computers, information display equip(cid:173)
`ment, and traffic control systems.
`In this context, advanced traveler information systems
`(A TIS) that are based on modern information technology may
`play an important role in reducing traffic congestion and im(cid:173)
`proving traffic flows and safety. It is expected that A TIS will
`reduce delays caused by both incident and recurrent conges(cid:173)
`tion by providing information to motorists about alternative
`paths to their destinations using a combination of roadside
`signals and onboard systems. Such schemes will aim at optim(cid:173)
`izing driver route selection and making this selection respon(cid:173)
`sive to real-time road and traffic conditions. In this paper a
`framework for the operation of A TIS is presented and
`
`I. Kaysi, Department of Civil Engineering, American University of
`Beirut, P.O. Box 11-0236, Beirut, Lebanon. M. Ben-Akiva and H.
`Koutsopoulos, Departmt:nt of Civil Engineering, Massachusetts In(cid:173)
`stitute of Technology, Cambridge, Mass. 02139.
`
`Google Ex. 1018
`
`

`

`Kaysi et al.
`
`67
`
`incident detection. The actual information gathered will vary
`from system to system, depending on the particular compo(cid:173)
`nents included in the system and the coverage of the network
`by the surveillance system.
`•The COP element has the responsibility for providing the
`control and routing module with the information that is needed
`to implement routing and guidance strategies. Among the
`principles adopted in this paper are the need for COP to
`provide CAR with projected traffic conditions and the fact
`that COP should be performed by a dynamic traffic assign(cid:173)
`ment (DTA) model that will take into account in its projec(cid:173)
`tions the driver response.
`• The CAR element generates guidance advice in response
`to information provided by the surveillance system and by the
`congestion prediction model. The fact that CAR should main(cid:173)
`tain projection or guidance consistency constitutes another
`principle adopted in this paper.
`
`Information Flow
`
`The flow of information from one element in the system to
`another is described as follows (refer to Figure 1):
`
`•Infrastructure data: Infrastructure data, an umbrella term,
`includes all network attributes that generally are invariant
`with time or that change slowly with time. The attributes
`include the network topology, the geometric attributes of all
`network elements, the control devices that are installed in the
`network, any channelization or other type of lane control,
`circulation restrictions such as one-way streets and prohibited
`turns, and so on. This information is required by the COP
`and CAR.
`• Historical origin-destination (0-D) data: The historical
`0-D data generally consist of 0-D information obtained by
`surveys or inferred from traffic counts by assignment models.
`
`Historical
`0-D Data
`
`Infrastructure
`Data
`
`Surveillance System
`
`Volumes, Speeds, Incidents
`
`Route Information
`& Guidance
`
`Control and Routing
`
`Actual Traffic Conditions f - - - - - '
`
`___.. Flow of Information c==J Information
`________ .,. Flow of Information
`under Some Scenarios ~System Element
`
`FIGURE 1 Proposed framework.
`
`This historical information is, for the most part, slowly chang(cid:173)
`ing over time but should be updated periodically.
`•The surveillance system: The information provided by the
`surveillance system consists mainly of direct measurements of
`volume, speed, occupancy, and the presence of incidents.
`Eventually the information may also include travel time data
`from equipped vehicles.
`•Updated 0-D data: The most recent information from
`the surveillance system and route information (if available)
`can be combined with historical 0-D data to provide updated
`three-dimensional (3-D) 0-D matrixes for the subsequent time
`periods.
`•COP: Congestion prediction provides estimates of traffic
`conditions on the network. The updated 0-D data are used
`by some COP scenarios including the proposed DTA. Be(cid:173)
`cause route choice modeling and the provision of guidance
`are sensitive to the destinations associated with flows, 0-D
`data are required to implement the DTA associated with the
`proposed framework.
`• CAR: The traffic conditions identified by the COP are
`transmitted to the CAR. The routing strategies need this in(cid:173)
`formation to develop an optimal response to the developing
`traffic environment.
`•Guidance: The outputs of the CAR generally take the
`form of route guidance information. Within the proposed
`framework, guidance data are transmitted to the COP as an
`input.
`
`PRINCIPLES UNDERLYING THE PROPOSED
`FRAMEWORK
`
`The following are the major principles underlying the pro- _
`posed framework:
`
`Principle 1: COP Should Provide CAR with Projected
`Traffic Conditions
`
`The travel times used for routing purposes by ATIS may be
`based on historical, current, or predicted traffic conditions.
`Although the use of historical data may provide a basis for
`static guidance and navigation, its use alone as a basis. for
`CAR decisions is not expected to be of any value for adaptive
`routing. The main reason behind this is that historical traffic
`data are a bad indicator of evolving, day-specific traffic con(cid:173)
`ditions, especially in situations in which traffic patterns display
`a significant amount of day-to-day variability. An analysis of
`the performance of various real-time routing strategies by
`Koutsopoulos and Xu ( 4) confirmed this intuition and indi(cid:173)
`cated that the use of historical data as a basis for real-time
`routing advice is significantly inferior to the use of, for ex(cid:173)
`ample, current or predictive information (5).
`French (6), Catling and McQueen (7), and Rillings and
`Betsold ( 8) provide reviews of many existing demonstration
`projects. In many such projects guidance passed to drivers
`consists of information regarding current traffic conditions
`(Smart Corridor, AMTICS, and RACS). Some researchers
`assert that routing strategies may be formulated on the basis
`of a control-theoretic approach that requires information on
`current traffic conditions only. For example, Papageorgiou
`
`Google Ex. 1018
`
`

`

`68
`
`TRANSPORTATION RESEARCH RECORD 1408
`
`and Messmer (9) use feedback control methods to split traffic
`between an 0-D pair among different routes. They claim that
`their methodology has low sensitivity with respect to unknown
`future demand levels and compliance rates that are assumed
`to be exogenous "disturbances." However, the authors warn
`that their feedback concept, which is based on observations
`of current traffic conditions, may not achieve its goal of es(cid:173)
`tablishing dynamic user optimum conditions if strong oscil(cid:173)
`lations in the demand levels occur or if the network perfor(cid:173)
`mance displays strong nonlinearities. in case of severe
`congestion. These remarks by the authors provide further
`evidence that using current traffic conditions as a basis for
`guidance will not succeed if current traffic conditions are not
`good predictors of future conditions.
`One of the major principles embodied in the framework
`being proposed in this paper is that such routing strategies
`have to be formulated on the basis of a forecast or projection
`of future traffic conditions on the network (2) rather than on
`instantaneous traffic conditions. The rationale behind using
`predictive information is that drivers' travel decisions are af(cid:173)
`fected by future traffic conditions expected to be in effect
`when they reach downstream sections of the network on their
`way to their destinations. The ALI-SCOUT system uses pro(cid:173)
`jected travel times in setting the guidance on the basis of a
`similar rationale. Therefore, the most useful type of guidance
`that can be provided to a driver faced with travel decisions
`would be based on a projection of traffic conditions. In ad(cid:173)
`dition, guidance based on traffic information that is predicted
`using an advanced COP module is most capable of improving
`the travel time reliability of drivers because its look-ahead
`capability helps them avoid long future delays. This issue is
`discussed in more detail under the second principle.
`
`Principle 2: A DT A Model Should Be Used for COP
`
`In all existing traveler information systems and demonstration
`projects, the guidance passed to drivers is based either on cur(cid:173)
`rent traffic conditions (Smart Corridor, Travtek, AMTICS,
`and RACS) or on simple predictions of future traffic condi(cid:173)
`tions (ALI-SCOUT).
`The travel time prediction methodology used by the ALI(cid:173)
`SCOUT system, for example, constructs a projection ratio of
`the historical travel time on a specific link to the current travel
`time, as reported by equipped vehicles (JO). This ratio is used
`to predict travel times for vehicles using that link during all
`future time intervals. Because only a few vehicles are equipped
`in Berlin, many links are not used by equipped vehicles during
`particular time intervals; therefore, no estimate of the pro(cid:173)
`jection ratio would be available. Consequently, the ratio is
`modified to reflect current conditions on neighboring links as
`well as conditions in preceding time intervals. Koutsopoulos
`and Xu ( 4) note that a problem with this methodology is the
`fact that the projection ratio is used to predict travel times
`for all future time intervals, thus implicitly assuming that trends
`currently observed remain constant for the entire prediction
`interval. To remedy this particular problem, Koutsopoulos
`and Xu suggest the use of information discounting. However,
`the methodology remains heuristic in nature and suffers from
`other omissions, which are discussed next.
`
`In all these systems and projects, as well as in all analyses
`being conducted by researchers related to A TIS (9 ,11), there
`has been no consideration whatsoever of the response of mo(cid:173)
`torists to route guidance in setting such guidance. Such an
`omission entails a major shortcoming in that the potential
`concentration of traffic on the recommended routes and the
`overreaction of drivers in their response to guidance infor(cid:173)
`mation are ignored. This problem is expected to become more
`severe as the number of guided vehicles increases. Guidance
`validity and, as a result, driver compliance would be adversely
`affected in such schemes.
`To overcome this problem it is required that the guidance
`be based on an advanced COP module that makes its pre(cid:173)
`dictions of future congestion in the network on the basis of
`
`• Current traffic conditions (consideration of initial con(cid:173)
`ditions),
`•Predicted 0-D demand levels (sensitivity to future de(cid:173)
`mand patterns),
`• Guidance being provided and anticipated driver response
`to guidance,
`• Traffic control actions to be implemented, and
`• Reduction in capacity as a result of incidents that have
`been detected.
`
`Principle 3: CAR Should Maintain Guidance/
`Prediction Consistency
`
`On the basis of the earlier discussion, it becomes clear that
`none of the guidance systems in existence attempt to antici(cid:173)
`pate the impact of guidance being provided. The same holds
`true for analyses conducted by researchers in relation to A TIS.
`A major principle underlying the proposed framework is that
`consistency has to be maintained between the guidance being
`provided to drivers and the predicted traffic conditions. That
`is, the information system has to check that the guidance being
`provided will prove to be optimal to guided drivers on the
`basis of a prediction of the future traffic conditions. This
`system would result in guidance information that is consistent
`with anticipated driver behavior and network conditions, would
`ensure the validity of the guidance information, and would
`encourage its use by more drivers. In addition, consistency
`has to be maintained between the traffic conditions as per(cid:173)
`ceived by the COP and the actual traffic conditions so that
`the COP remains attuned to the real world.
`
`CAR STRATEGIES
`
`When specifying the CAR module, three major issues have
`to be addressed:
`
`1. What logic should the CAR module adopt to provide
`drivers with guidance advice? Should information or route
`directives be provided to drivers? Should route directives be
`based on shortest-path guidance or on route distributive guid(cid:173)
`ance? How should route distributive guidance be imple(cid:173)
`mented? How can the CAR logic ensure guidance and pro(cid:173)
`jection consistency?
`
`Google Ex. 1018
`
`

`

`Kaysi et al.
`
`69
`
`2. Given that guidance provided at specific locations of the
`network can be updated periodically, what is the impact of
`temporal update frequency on the effectiveness of routing
`strategies?
`3. If drivers receive guidance updates at various locations
`as they move through the network toward their destinations
`(as in ALI-SCOUT), what is the impact of spatial update
`frequency on the effectiveness of routing strategies?
`
`A discussion of each of these issues follows.
`
`CAR Logic
`
`The most basic distinction between various logic types used
`as part of CAR relates to whether CAR provides drivers with
`information or route directives.
`
`Information
`
`If information is passed to drivers then CAR would constitute
`a simple link whereby the traffic data output from COP is
`interpreted, relayed to drivers, and presented in an under(cid:173)
`standable form. This represents the case in many demonstra(cid:173)
`tion projects, such as AMTICS, RACS, and Smart Corri(cid:173)
`dor, and has served as the basis for A TIS analyses conducted
`by several researchers (see, for example, Mahmassani and
`Jayakrishman (11)].
`
`that such logic may lead to strong perturbations of traffic flow,
`especially if a large fraction of vehicles are equipped. There(cid:173)
`fore, they suggested an alternative logic based on a "smooth
`regulator" using a more advanced feedback law that leads to
`route-distributive guidance and the specification of optimal
`rates of splitting traffic among alternate routes.
`
`Difficulties in Implementing Route Distributive Guidance
`Depending on the specific technology used for providing driv(cid:173)
`ers with route directives, it may not be technically possible
`to distribute traffic over a number of routes at the same instant
`of time. For example, if variable message signs are used, the
`best that can be accomplished is to change the sign within the
`guidance interval so that the time average of the route direc(cid:173)
`tives would correspond to the fractions we wish to achieve
`for each route. On the other hand, if in-vehicle units are used,
`it is technically possible to provide different. drivers with dif(cid:173)
`ferent route directives at the same time to split drivers among
`routes according to the optimal fractions. However, this proc(cid:173)
`ess may