Since the early 1980s, short-term traffic forecasting has been an integral part of most Intelligent Transportation Systems (ITS) research and applications; most effort has gone into developing methodologies that can be used to model traffic characteristics and produce anticipated traffic conditions. Existing literature is voluminous, and has largely used single point data from motorways and has employed univariate mathematical models to predict traffic volumes or travel times. Recent developments in technology and the widespread use of powerful computers and mathematical models allow researchers an unprecedented opportunity to expand horizons and direct work in 10 challenging, yet relatively under researched, directions. It is these existing challenges that we review in this paper and offer suggestions for future work.

Short-term traffic forecasting: Where we are and where we’re going

The prevalence of smartphones presents a unique opportunity to develop a system that can have a significant impact on reducing the annual 400,000 fatalities from pedestrian traffic accidents. This system gives 360 degrees, extended range, NLOS view where both the driver and the pedestrian are warned of a possible collision. This, the first of its kind, system was developed from a two year collaborative research effort between Honda and Qualcomm to leverage DSRC so vehicles can communicate with smartphones to preempt a possible collision between a pedestrian (with a smartphone) and an approaching vehicle. This paper describes the pedestrian and vehicle-based algorithms and gives an overview of how this system warns both the driver and the pedestrian so they can take evasive action and prevent a collision. We present the results from our field tests where we demonstrate several pedestrian safety scenarios and present the over-the-air performance data collected in the field tests. Finally, we discuss remaining challenges and present possible approaches to reducing false positives, minimizing spectrum and channel congestion and improving security and localization.

Cars Talk to Phones: A DSRC Based Vehicle-Pedestrian Safety System

Probe vehicle data are increasingly becoming more attractive for real-time system state estimation in transportation networks. This paper presents analytical models for the real-time estimation of queue lengths at traffic signals using the fundamental information (i.e., location and time) that probe vehicles provide. For a single queue with Poisson arrivals, analytical models are developed to evaluate how error changes in queue length estimation as the percentage of probe vehicles in the traffic stream varies. When the overflow queue is ignored, a closed-form solution is obtained for the variance of the estimation error. For the more general case with the overflow queue, a formulation for the error variance is presented, which requires the marginal probability distribution of the overflow queue as the input. In addition, an approximate model is presented for the latter case, which yields results that are comparable with the exact solution. Overall, the formulations presented here can be used to assess the error in queue length estimation from probe data without conducting simulation runs for various scenarios of probe vehicle market-penetration rates and congestion levels.

Analytical Evaluation of the Error in Queue Length Estimation at Traffic Signals From Probe Vehicle Data

As mobile traffic sensor technology gets more attention, mathematical models are being developed that utilize this new data type in various intelligent transportation systems applications. This study introduces simple analytical estimation models for queue lengths from tracked or probe vehicles at traffic signals using stochastic modeling approach. Developed models estimate cycle-to-cycle queue lengths by using primary parameters such as arrival rate, probe vehicle proportions, and signal phase durations. Valuable probability distributions and moment generating functions for probe information types are formulated. Fully analytical closed-form expressions are given for the case ignoring the overflow queue and approximation models are presented for the overflow case. Derived models are compared with the results from VISSIM-microscopic simulation. Analytical steady-state and cycle-to-cycle estimation errors are also derived. Numerical examples are shown for the errors of these estimators that change with probe vehicle market penetration levels, arrival rates, and volume-to-capacity ratios.

Simple analytical models for estimating the queue lengths from probe vehicles at traffic signals

Assessing the Potential Impacts of Connected Vehicles: Mobility, Environmental, and Safety Perspectives

This paper presents an effort to develop a virtual testbed for assessing probe vehicle data generation by IntelliDrive vehicles within a microscopic traffic-simulation environment. Simulation capabilities are implemented through the development of a portable plug-in module using the application programming interface of the Paramics microscopic traffic simulation. This module simulates the generation of snapshots by individual vehicles, the uploading of these snapshots to roadside units, and some probe vehicle data postprocessing. While some temporary simplifying assumptions are made, the simulation generally follows operational concepts described in the Society of Automotive Engineers (SAE) J2735 Surface Vehicle Standard. Application of the model is demonstrated by simulating IntelliDrive probe data collection over the U.S. Department of Transportation (USDOT)'s Michigan Proof-of-Concept testbed. Simulation results show the sensitivity of probe data collection to communication range, market penetration, number of active roadside communication units (RSEs), interval between snapshots, and snapshot buffer size. Impacts on link travel time estimates are also presented. These results clearly demonstrate the utility of the simulator in conducting evaluations and sensitivity analyses for scenarios that would be difficult to execute in existing testbeds.

Virtual Testbed for Assessing Probe Vehicle Data in IntelliDrive Systems

This paper assesses various issues associated with the use of IntelliDrive probe vehicle data generated according to existing protocols. These evaluations are conducted using a virtual IntelliDrive probe vehicle data generator implemented within the Paramics microscopic traffic simulation model. Issues investigated included the ability to monitor queue conditions, the potential for data-sampling bias, effects of privacy rules on collected data, the ability to track vehicles over short distances, the magnitude of data latency induced by the protocols, and the ability to use probe vehicle data to estimate link travel times. The evaluations indicate various improvement needs: (1) generating snapshots at fixed intervals rather than at speed-based intervals, (2) promoting the use of short snapshot intervals, (3) allowing vehicles to include the snapshots they generate while stopped, (4) enabling vehicles to generate link exit snapshots, (5) allowing vehicles to keep uploading new snapshots generated while within range of a roadside communication unit, and (6) revising privacy protocols to ensure that adequate short vehicle tracks can be obtained. The impacts of current privacy rules on the ability to track vehicles over short distances are also outlined.

Evaluation of Usability of IntelliDrive Probe Vehicle Data for Transportation Systems Performance Analysis

This report covers several key subjects related to the generation of IntelliDriveSM probe vehicle data and use of this data in application of interest to state departments of transportation and local public transportation agencies. The evaluations conducted as part of this project are primarily based on the probe vehicle data collection system that was deployed by the U.S. Department of Transportation (USDOT) around Novi, Michigan, in 2008 for its Vehicle‐Infrastructure Integration (VII) Proof‐of‐ Concept (POC) test program. This system was designed around the use of the 5.9‐GHz Dedicated Short Range Communication (DSRC) wireless protocol to enable vehicles to communicate with Roadside Equipment (RSE). The generation of snapshots further followed the protocols defined within the SAE J2735 DSRC Message Set standard. Following a general introduction in Chapter 1, Chapter 2 briefly reviews the protocols that were used to generate and retrieve probe vehicle snapshots, while Chapter 3 presents a general evaluation of the POC test data that were accumulated during the 2008 test program. This is followed by a presentation in Chapter 4 of the evaluation framework of the current project. This presentation includes an overview of the envisioned DUAP system and descriptions of project stakeholders, potential data sources, supporting technologies, applications of interests, and potential operational constraints. Chapter 5 then presents a general description of the Paramics IntelliDriveSM virtual simulator that is used to conduct some of the subsequent evaluations. While the initial POC test program aimed to evaluate data collection capabilities across a range of application, this program was significantly shortened due to various technical issues. This resulted in incomplete data collection and partial application designs that were insufficient to complete the initial project deliverables associated without relying on simulation. Chapter 6 then examines the effects of snapshot generation protocols and privacy policies on data latency, data quality, and the ability to track vehicles over short distances. Chapter 7 follows with a mapping of application data needs and general descriptions of processes required to convert raw probe data into useful information, while Chapter 8 evaluates how basic traffic flow performance measures (flow rates, flow density, travel times, speed profiles, queue parameters) can be estimated from probe data in systems featuring full and partial proportions of probe vehicles. Chapter 9 further develops a concept of operations for an enhanced traffic monitoring system incorporating probe vehicle and other data sources, while Chapter 10 investigates various issues that must be considered when developing application deployment plans. Chapters 11, 12 and 13 finally present a summary of primary findings, lessons learned and recommendations for future work.

VII Data Use Analysis and Processing (DUAP) Final Project Report (Phase II)

This paper presents the initial efforts to develop a virtual VII simulation testbed within the Paramics microscopic traffic simulation environment. This simulator is being developed to help mitigate limitations of real-world testbeds. VII simulation capabilities are implemented through the development of a portable plug-in module using Paramics’ Applications Programming Interface (API). This module simulates the generation of VII snapshots by individual vehicles and the uploading of these snapshots to roadside units. While temporary simplifying assumptions are made, the simulation generally follows operational concepts developed by the U.S. Department of Transportation. Application of the model is demonstrated by simulating VII probe data collection over the Michigan tested recently used by the U.S. Department of Transportation for VII Proof-of-Concept testing. Results of these simulations demonstrate the utility of the simulator in conducting evaluations and sensitivity analyses in scenarios hard to execute within the current testbeds, complementing real-world testing activities. Simulation results showed the sensitivity of VII probe data collection to communication range, market penetration, number of active RSEs, interval between snapshots, and snapshot buffer size.

VII Test Bed Simulation Framework for Vehicle-to-Infrastructure Communication and Application Evaluations

Trip origin and destination (O‐D) data play a crucial role in various transportation activities. This information not only includes the starting and end points of a trip, but also information that can be obtained through the ability to track vehicles across a network. Collected information may include the location and speed of a vehicle every second, a record of the links entered during a journey, the time a vehicle has entered each link, the time taken to travel a link, etc. Despite some recognized benefits, there are still concerns, primarily related to privacy, about whether or not probe vehicles should be allowed to collect O‐D trip information. Within this context, this report examines various issues associated with the potential collection of O‐D data by IntelliDrive probe vehicles. Research findings presented in the report include: an overview of current uses of O‐D data in activities related to transportation planning and system operations; a description of the methods currently used to collect or generate O‐D trip data; a description of issues associated with current O‐D data collection and generation methods, as well as how O‐D trip data are applied in transportation applications; a discussion on how O‐D information collected from IntelliDrive probe vehicles can enhance existing applications and enable the development of new applications; a discussion on the potential benefit‐cost ratio offered by the inclusion of O‐D trip information in IntelliDrive data sets; a discussion of privacy issues regarding O‐D data collection, including a review of current concerns, emerging policies regarding driver privacy in IntelliDrive systems, methods currently promoted to ensure driver privacy, impacts of promoted privacy methods on IntelliDrive applications, and methods available to mitigate privacy concerns; and a simulation study with the Paramics microscopic traffic simulation model demonstrating the benefits to individual vehicles and network operations of using O‐D data reported by IntelliDrive vehicles to provide dynamic route guidance around an incident.

Ralph Robinson

Papers

Virtual Testbed for Assessing Probe Vehicle Data in IntelliDrive Systems

Evaluation of Usability of IntelliDrive Probe Vehicle Data for Transportation Systems Performance Analysis

VII Data Use Analysis and Processing (DUAP) Final Project Report (Phase II)

VII Test Bed Simulation Framework for Vehicle-to-Infrastructure Communication and Application Evaluations

Sweden/Michigan naturalistic field operational test - phase 1 : benefits of origin and destination information in IntelliDrive data sets