Connected Vehicle Data Research Articles

Adaptive and multi-path progression signal control under connected vehicle environment

Through wireless communications, enriched information from connected vehicles (CVs) can describe traffic information near an intersection and would supplement a data source for an effective signal control. This paper proposes an adaptive traffic signal control system in a CV environment. At the intersection level, an adaptive control model is developed to assign optimal green times by minimizing the total vehicle delay. When the CV penetration rate is low, a method depending on limited CV data is presented to estimate the vehicle arrival information. At the corridor level, a real-time optimization model is formulated to design the dynamic progression plan for critical paths (i.e. paths with high flows). These two optimization models are solved by the dynamic programming technique. A real-world arterial is modeled in VISSIM to evaluate the effectiveness and efficiency of the proposed traffic signal control system. Simulations with various CV penetration rates and demand levels are conducted to compare the proposed system with fixed coordination and adaptive signal control systems. Results indicate that the proposed system outperforms both base systems by reducing 15.67% and 13.81% of the average delay, respectively, when the CV penetration rate is relatively high. Moreover, the proposed system can improve the arterial performance under all tested scenarios with various CV penetration rates and demand levels. Since the tested demand scenarios can reflect the real traffic fluctuations, it can be proved that the proposed control system could be applied in the field to boost the overall traffic efficiency along the arterial.

Connected vehicle-based road safety information system (CROSS): Framework and evaluation

Valuable high-resolution data representing the maneuvering of both individual subject vehicles and adjacent vehicles are available in the era of the connected vehicle systems, which is also referred to as cooperative intelligent transportation systems (C-ITS). C-ITS can share useful traffic information between connected vehicles (CV) and between vehicles and infrastructure in support of vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) wireless communications. An excellent feature of a C-ITS pre-deployment project in Korean freeways that CVs are equipped with an in-vehicle forward collision warning system. This technical support provides a useful opportunity to evaluate crash risks more objectively and scientifically based on the analysis of vehicle interactions, which motivates our study. The purpose of this study is to develop a road safety information system based on the analysis of CV data. The proposed system estimates individual vehicle crash risks based on the crash potential index (CPI) and further utilizes them to develop a methodology for assessing road safety risks on freeways. High CPIs were observed in toll plaza area, recurrent congestion sections, and on and off-ramp areas. An encouraging result showed that the relationship between the estimated CPI and the actual crash frequencies was statistically meaningful. In addition, the impact of the CV market penetration rate (MPR) on the feasibility of the proposed road risk monitoring method was explored by microscopic traffic simulation experiments using VISSIM. A safety evaluation equivalent to 100 % MPR was obtainable with 30 % MPR. The outcomes of this study are expected to be utilized as fundamental to support the development of novel road risk monitoring systems in C-ITS environments.

Evaluation of the Impact of Queue Trucks with Navigation Alerts Using Connected Vehicle Data

Back of queue crashes on Interstates are a major concern for all state transportation departments. In 2020, Indiana DOT begin deploying queue warning trucks with message boards, flashers and digital alerts that could be transmitted to navigation systems such as Waze. This study reports on the deployment and impact evaluation of digital alerts on motorist’s assistance patrols and 19 Queue trucks in Indiana. The motorist assistance patrol evaluation is provided qualitatively. A novel analysis of queue warning trucks equipped with digital alerts was conducted during the months of May-July in 2021 using connected vehicle data. This new data set reports locations of anonymous hard-braking events from connected vehicles on the Interstate. Hard-braking events were tabulated for when queueing occurred with and without the presence of a queue warning truck. Approximately 370 hours of queueing with queue trucks present and 58 hours of queueing without queue trucks present were evaluated. Hard-braking events were found to decrease approximately 80% when queue warning trucks were used to alert motorists of impending queues.

Algorithm-driven Sensing Devices and Connected Vehicle Data in Smart Transportation Networks

Algorithm-driven Sensing Devices and Connected Vehicle Data in Smart Transportation Networks

Analysis of Electric and Hybrid Vehicle Usage in Proximity to Charging Infrastructure in Indiana

This paper explores the movement of connected vehicles in Indiana for vehicles classified by the NHTSA Product Information Catalog Vehicle listing as being either electric (EV) or hybrid electric (HV). Analysis of trajectories from July 12-18, 2021 for the state of Indiana observed nearly 33,300 trips and 267,000 vehicle miles travelled (VMT) for the combination of EV and HV. Approximately 53% of the VMT occurred in just 10 counties. For just EVs, there were 9814 unique trips and 64,700 Electric Vehicle Miles Traveled (EVMTs) in total. A further categorization of this revealed that 18% of these EVMTs were on Interstate roadways and 82% on non-interstate roads. Proximity analysis of existing DC Fast charging stations in relation to interstate roadways revealed multiple charging deserts that would be most benefited by additional charging capacity. Eleven roadway sections among the 9 interstates were found to have a gap in available DC fast chargers of 50 miles or more. Although the connected vehicle data set analyzed did not include all EV’s the methodology presented in this paper provides a technique that can be scaled as additional EV connected vehicle data becomes available to agencies. Furthermore, it emphasizes the need for transportation agencies and automotive vendors to strengthen their data sharing partnerships to help accelerate the adoption of EV and reduce consumer range anxiety with EV. Graphics are included that illustrate examples of counties that are both overserved and underserved by charging infrastructure.

Longitudinal Performance Assessment of Traffic Signal System Impacted by Long-Term Interstate Construction Diversion Using Connected Vehicle Data

Local arterials can be significantly impacted by diversions from adjacent work zones. These diversions often occur on unofficial detour routes due to guidance received on personal navigation devices. Often, these routes do not have sufficient sensing or communication equipment to obtain infrastructure-based traffic signal performance measures, so other data sources are required to identify locations being significantly affected by diversions. This paper examines the network impact caused by the start of an 18-month closure of the I-65/70 interchange (North Split), which usually serves approximately 214,000 vehicles per day in Indianapolis, IN. In anticipation of some proportion of the public diverting from official detour routes to local streets, a connected vehicle monitoring program was established to provide daily performances measures for over 100 intersections in the area without the need for vehicle sensing equipment. This study reports on 13 of the most impacted signals on an alternative arterial to identify locations and time of day where operations are most degraded, so that decision makers have quantitative information to make informed adjustments to the system. Individual vehicle movements at the studied locations are analyzed to estimate changes in volume, split failures, downstream blockage, arrivals on green, and travel times. Over 130,000 trajectories were analyzed in an 11-week period. Weekly afternoon peak period volumes increased by approximately 455%, split failures increased 3%, downstream blockage increased 10%, arrivals on green decreased 16%, and travel time increase 74%. The analysis performed in this paper will serve as a framework for any agency that wants to assess traffic signal performance at hundreds of locations with little or no existing sensing or communication infrastructure to prioritize tactical retiming and/or longer-term infrastructure investments.

Autonomous Driving Algorithms and Behaviors, Sensing and Computing Technologies, and Connected Vehicle Data in Smart Transportation Networks

Autonomous Driving Algorithms and Behaviors, Sensing and Computing Technologies, and Connected Vehicle Data in Smart Transportation Networks

Diverging Diamond Interchange Performance Measures Using Connected Vehicle Data

Since the first Diverging Diamond Interchange (DDI) implementation in 2009, most of the performance studies developed for this type of interchange have been based on simulations and historical crash data, with a small number of studies using Automated Traffic Signal Performance Measures (ATSPM). Simulation models require considerable effort to collect volumes and to model actual controller operations. Safety studies based on historical crashes usually require from 3 to 5 years of data collection. ATSPMs rely on sensing equipment. This study describes the use of connected vehicle trajectory data to analyze the performance of a DDI located in the metropolitan area of Fort Wayne, IN. An extension of the Purdue Probe Diagram (PPD) is proposed to assess the levels of delay, progression, and saturation. Further, an additional PPD variation is presented that provides a convenient visualization to qualitatively understand progression patterns and to evaluate queue length for spillback in the critical interior crossover. Over 7000 trajectories and 130,000 GPS points were analyzed between the 7th and the 11th of June 2021 from 5:00 AM to 10:00 PM to estimate the DDI’s arrivals on green, level of service, split failures, and downstream blockage. Although this technique was demonstrated for weekdays, the ubiquity of connected vehicle data makes it very easy to adapt these techniques to analysis during special events, winter storms, and weekends. Furthermore, the methodologies presented in this paper can be applied by any agency wanting to assess the performance of any DDI in their jurisdiction.

Estimation of Connected Vehicle Penetration on US Roads in Indiana, Ohio, and Pennsylvania

Connected vehicle data is an important assessment tool for agencies to evaluate the performance of freeways and arterials, provided there is sufficient penetration to provide statistically robust performance measures. A common concern by agencies interested in using crowd sourced probe data is the penetration rate across different types of roads, different hours of the day, and different regions. This paper describes and demonstrates a methodology that uses data from state highway performance monitoring systems in Indiana, Ohio and Pennsylvania. The study analyzes 54 locations over the 3 states for select Wednesdays and Saturdays in 2020 and 2021. Overall, across all locations and dates, the median penetration was approximately 4.5%. The median penetration for August 2020 for Indiana, Ohio, and Pennsylvania was 4.6%, 4.3%, and 4.0%, respectively. The median penetration for those same states in August 2020 on interstates and non-interstates was 3.9% and 4.6%, respectively. Additionally, the study conducted a longitudinal evaluation of Indiana penetration for selected months between January 2020 and June 2021. Indiana penetration increased modestly between December 2020 and June 2021, perhaps due to the post-COVID rebound of passenger vehicle traffic. This paper concludes by recommending that the techniques described in this paper be scaled to other states so that traffic engineers can make informed decisions on the use and limitations of connected vehicle data for various use cases.

Using Anonymous Connected Vehicle Data to Evaluate Impact of Speed Feedback Displays, Speed Limit Signs and Roadway Features on Interstate Work Zones Speeds

Annually, there are over 120,000 crashes in work zones in the United States. High speeds in construction zones are a well-documented risk factor that increases the frequency and severity of crashes. This study used connected vehicle data to evaluate the spatial and temporal impact that regulatory signs, speed feedback displays, and construction site geometry had on vehicle speed. Over 27,000 unique trips over 2 weeks on a 15-mile interstate construction work zone near Lebanon, IN were analyzed. Spatial analysis over a 0.2-mi segment before and after the posted speed limit signs showed that the regulatory signs had no statistical impact on reducing speeds. A before/after analysis was also conducted to study the impact of radar-based speed feedback that displays the motorists’ speed on a sign below a regulatory speed limit sign. Results showed a maximum drop in median speeds of approximately 5 mph. Speeds greater than 15 mph above the speed limit dropped by 10% - 15%. The reduction in speeds began approximately 1000 feet ahead of the sign and results were found to be statistically significant. The analysis also revealed that larger speed drops inside the work zone were due to geometric constraints that required additional driver workloads, especially during shoulder width changes and lane shifts. The results from this study will be helpful for agencies to understand driver behavior in the work zones and to identify proper speed limit compliance techniques that significantly reduce driver speeds in and around work zones.

Deep Transfer Learning Based Intersection Trajectory Movement Classification for Big Connected Vehicle Data

Trajectory movement labelling is an important pre-stage for predicting connected vehicle (CV) movement at intersections. Drivers’ movement prediction and warning at intersections ensure advanced transportation safety and researchers use machine learning-based data-driven approaches to implement these technologies. However, prediction of drivers’ movements at intersections requires labelling the train and test dataset accurately with different vehicle movements at intersections to evaluate the performance of the prediction model by comparing the actual and predicted intersection movements. Moreover, due to GPS detection error or missing co-operative awareness messages (CAM), the data resides with many abnormal trajectories which are unable to be matched with regular straight or any turning movements. Especially for big data with million trajectories, it is tedious to label the movements manually. To solve this problem, we have created an automated trajectory movement classification technique using a dual approach of map matching technique and deep transfer learning modelling. Data of connected vehicle trajectory information is taken from the Ipswich Connected Vehicle Pilot (ICVP) Project, which is one of the largest connected vehicle pilots within a naturalistic driving environment in Australia. Map matching approach is performed as initial labelling by analysing the origin and destination of the vehicle CAM messages at intersections and then was converted as image datasets of 19202 samples. The map matching error and abnormal trajectories are identified by visual inspection. With properly labelled 9496 training images, 10 transfer learning models are built and tested through the remaining 9706 testing images. The maximum testing accuracy (99.73%) is achieved from the Densenet169 model, and the result shows satisfactory accuracy for individual classes: straight (99.85%), turn left (99.59), turn right (99.25), u-turn (100%), abnormal (98.63%). This model becomes a routine tool that is used daily to automatically classify thousands of trajectory movements of the C-ITS data in the ICVP project.

Predictive Control Algorithms, Real-World Connected Vehicle Data, and Smart Mobility Technologies in Intelligent Transportation Planning and Engineering

Predictive Control Algorithms, Real-World Connected Vehicle Data, and Smart Mobility Technologies in Intelligent Transportation Planning and Engineering

Real-World Connected Vehicle Data, Deep Learning-based Sensing Technologies, and Decision-Making Self-Driving Car Control Algorithms in Autonomous Mobility Systems

Real-World Connected Vehicle Data, Deep Learning-based Sensing Technologies, and Decision-Making Self-Driving Car Control Algorithms in Autonomous Mobility Systems

Safety evaluation of connected and automated vehicles in mixed traffic with conventional vehicles at intersections

Connected and Automated Vehicles (CAVs) can potentially improve the performance of the transportation system by reducing human errors. This paper investigates the safety impact of CAVs in a mixed traffic with conventional vehicles at intersections. Analyzing real-world AV crashes in California revealed that rear-end crashes at intersections are the dominant crash type. Therefore, to enhance our understanding of the future interactions between human-driven vehicles with CAVs at intersections, a simulation framework was developed to model the mixed traffic environment of Automated Vehicles (AV), cooperative AVs, and conventional human-driven vehicles. In order to model AVs driving behavior, Adaptive Cruise Control (ACC) and cooperative ACC (CACC) models are utilized. Particularly, this study explores system improvements due to automation and connectivity across varying CAV market penetration scenarios. ACC and CACC car following models are used to mimic the behavior of AVs and cooperative AVs. Real-world connected vehicle data are utilized to modify and tune the acceleration/deceleration regimes of the Wiedemann model. Next, the driving volatility concept capturing variability in vehicle speeds was utilized to calibrate the simulation to represent the safety performance of a real-world environment. Two surrogate safety measures are used to evaluate the safety performance of a representative intersection under different market penetration rate of CAVs: the number of longitudinal conflicts and driving volatility. At low levels of ACC market penetration, the safety improvements were found to be marginal, but safety improved substantially with more than 40% ACC penetration. Additional safety improvements can be achieved more quickly through the addition of cooperation and connectivity through CACC. Furthermore, ACC/CACC vehicles were found to improve mobility performance in terms of average speed and travel time at intersections.

Quasi-real time estimation of turning movement spillover events based on partial connected vehicle data

Turning movement spillover (TMS) is the result of a turning bay section (TBS) not being able to accommodate all arriving vehicles, so that the turning-vehicle queue spills back and blocks other vehicles turning in different directions. We are not aware of any TMS estimation method that can remedy this situation or support relevant applications in real time. This research proposes a quasi-real time algorithm for estimating TMS, which includes triggering movement as well as duration estimation. The proposed method is based on data for connected vehicles (CVs), including their trajectories and their desired turning directions. In addition, a model that uses partial trajectory data is proposed. For each assumed TMS, a “simplified trajectory” is developed by the construction of a piece-wise linear curve. To minimize any deviation of the simplified trajectory from observation, a TMS estimation can be made. This proposed method is effective and computationally efficient when tested against dynamic demand in two mainstream signal phase settings, with varied sample sizes. Even though data for a higher number of vehicle samples is generally favorable, the proposed model still makes a good estimate when only one trajectory is available.

Review of Usage of Real-World Connected Vehicle Data

GPS loggers and cameras aboard connected vehicles can produce vast amounts of data. Analysts can mine such data to decipher patterns in vehicle trajectories and driver–vehicle interactions. Ability to process such large-scale data in real time can inform strategies to reduce crashes, improve traffic flow, enhance system operational efficiencies, and reduce environmental impacts. However, connected vehicle technologies are in the very early phases of deployment. Therefore, related datasets are extremely scarce, and the utility of such emerging datasets is largely unknown. This paper provides a comprehensive review of studies that used large-scale connected vehicle data from the United States Department of Transportation Connected Vehicle Safety Pilot Model Deployment program. It is the first and only such dataset available to the public. The data contains real-world information about the operation of connected vehicles that organizations are testing. The paper provides a summary of the available datasets and their organization, and the overall structure and other characteristics of the data captured during pilot deployments. Usage of the data is then classified into three categories: driving pattern identification, development of surrogate safety measures, and improvements in the operation of signalized intersections. Finally, some limitations experienced with the existing datasets are identified.

Estimation of Traffic Stream Density Using Connected Vehicle Data: Linear and Nonlinear Filtering Approaches.

The paper presents a nonlinear filtering approach to estimate the traffic stream density on signalized approaches based solely on connected vehicle (CV) data. Specifically, a particle filter (PF) is developed to produce reliable traffic density estimates using CV travel-time measurements. Traffic flow continuity is used to derive the state equation, whereas the measurement equation is derived from the hydrodynamic traffic flow relationship. Subsequently, the PF filtering approach is compared to linear estimation approaches; namely, a Kalman filter (KF) and an adaptive KF (AKF). Simulated data are used to evaluate the performance of the three estimation techniques on a signalized approach experiencing oversaturated conditions. Results demonstrate that the three techniques produce accurate estimates—with the KF, surprisingly, being the most accurate of the three techniques. A sensitivity of the estimation techniques to various factors including the CV level of market penetration, the initial conditions, and the number of particles in the PF is also presented. As expected, the study demonstrates that the accuracy of the PF estimation increases as the number of particles increases. Furthermore, the accuracy of the density estimate increases as the level of CV market penetration increases. The results indicate that the KF is least sensitive to the initial vehicle count estimate, while the PF is most sensitive to the initial condition. In conclusion, the study demonstrates that a simple linear estimation approach is best suited for the proposed application.

Efficient mapping of crash risk at intersections with connected vehicle data and deep learning models

Traditional methods for identifying crash-prone roadways are mainly based on historical crash data. It usually requires more than three years to collect a sufficient amount of dataset for road safety assessment. However, the emerging connected vehicles (CVs) technology generates rich instantaneous information, which can be used to identify dangerous road sections proactively. Information about the identified crash-prone intersections can be shared with the surrounding vehicles via CVs communication technology to promote cautious driving behaviors; in the longer term, such information will guide the implementation of countermeasures to prevent potential crashes. This study proposed a deep-learning based method to predict the risk level at intersections based on CVs data from the Michigan Safety Pilot program and historical traffic and intersection crash data in areas around Ann Arbor, Michigan, USA. One month of data by CVs at intersections were used for analyses, which accounts for about 3%–12% of overall trips. The risk levels of 774 intersections (i.e., low, medium and high risk) are determined by the annual crash rates. Feature extraction process is applied to both CV’s data and traffic data at each intersection and 24 features are extracted. Two black-box deep-learning models, multi-layer perceptron (MLP) and convolutional neural network (CNN) are trained with the extracted features. A number of hyperparameters that affect prediction performance are fine-tuned using Bayesian optimization algorithm for each model. The performance of the two deep learning models, which are black-box models, were also compared with a decision tree model, a white-box type of simple machine learning model. The results showed that the accuracies of deep learning (DL) models were slightly better (both over 90 %) than the decision tree model (about 87 %). This indicated that the DL models were capable of uncover the inherent complexity from the dataset and therefore provided higher accuracy than the traditional machine learning model. CNN model achieves slightly higher accuracy (93.8 %) and is recommended as the classifier to predict the risk level at intersections in practice. The interpretability analysis of the CNN model is conducted to confirm the validity of the model. This study shows that combination of CVs data (V2V and V2I) and deep learning networks (i.e. MLP and CNN used in this paper) is promising to determine crash risks at intersections with high time efficiency and at low CV penetration rates, which help to deploy countermeasures to reduce the crash rates and resolve traffic safety problems.

Using Connected Vehicle Data to Reassess Dilemma Zone Performance of Heavy Vehicles

The rate of fatalities at signalized intersections involving heavy vehicles is nearly five times higher than for passenger vehicles in the US. Previous studies in the US have found that heavy vehicles are twice as likely to violate a red light compared with passenger vehicles. Current technologies leverage setback detection to extend green time for a particular phase and are based upon typical deceleration rates for passenger cars. Furthermore, dilemma zone detectors are not effective when the max out time expires and forces the onset of yellow. This study proposes the use of connected vehicle (CV) technology to trigger force gap out (FGO) before a vehicle is expected to arrive within the dilemma zone limit at max out time. The method leverages position data from basic safety messages (BSMs) to map-match virtual waypoints located up to 1,050 ft in advance of the stop bar. For a 55 mph approach, field tests determined that using a 6 ft waypoint radius at 50 ft spacings would be sufficient to match 95% of BSM data within a 5% lag threshold of 0.59 s. The study estimates that FGOs reduce dilemma zone incursions by 34% for one approach and had no impact for the other. For both approaches, the total dilemma zone incursions decreased from 310 to 225. Although virtual waypoints were used for evaluating FGO, the study concludes by recommending that trajectory-based processing logic be incorporated into controllers for more robust support of dilemma zone and other emerging CV applications.

Data Infrastructure for Connected Vehicle Applications

Ann Arbor Connected Vehicle Test Environment (AACVTE) is the world’s largest operational, real-world deployment of connected vehicles (CVs) and connected infrastructure, with over 2,500 vehicles and 74 infrastructure sites, including intersections, midblocks, and highway ramps. The AACVTE generates a massive amount of data on a scale not seen in the traditional transportation systems, which provides a unique opportunity for developing a wide range of connected vehicle (CV) applications. This paper introduces a data infrastructure that processes the CV data and provides interfaces to support real-time or near real-time CV applications. There are three major components of the data infrastructure: data receiving, data pre-processing, and visualization including the performance measurements generation. The data processing algorithms include signal phasing and timing (SPaT) data compression, lane phase mapping identification, trajectory data map matching, and global positioning system (GPS) coordinates conversion. Simple performance measures are derived from the processed data, including the time–space diagram, vehicle delay, and observed queue length. Finally, a web-based interface is designed to visualize the data. A list of potential CV applications including traffic state estimation, traffic control, and safety, which can be built on this connected data infrastructure is discussed.