Skip to main content

Anuj Sharma

Iowa State University, Civil, Construction And Environmental Engineering, Faculty Member

Followers

36

Following

1

Public Views

Interests

Uploads

Papers by Anuj Sharma

Massively parallelizable approach for evaluating signalized arterial performance using probe-based data

Journal of Intelligent Transportation Systems, May 3, 2022

Managing this performance, considering the dynamic nature of demand requires updating traffic sig... more Managing this performance, considering the dynamic nature of demand requires updating traffic signal timings through various strategies. Agency resources for these activities are commonly scarce and are primarily by public complaints. This paper provides a data-driven prioritization approach for traffic signal re-timing on a corridor. In order to remove any dependence on available detection, probe-based data are used for assessing the performance measures. Probe-based data are derived from in-vehicle global positioning system observations, eliminating the need for installing on-field traffic infrastructure. The paper provides a workflow to measure and compare different segments on arterial corridors in terms of probe-based signal performance measures that capture different aspects of signal operations. The proposed method can serve as a tool to guide agencies looking to alter their signal control. The methodology identifies a group of dynamic days followed by evaluation of travel rate based upon non-dynamic days. Dynamic days represent the variability of traffic on a segment. Performance measures on non-dynamic days include Median Travel Rate, Within-Day Variability of travel rate, between-days variability of travel rate Minimum Travel Rate Dispersion, and two variables which include Overall Travel Rate Variabilities. Consequently, a corridor having high number of dynamic segments along with poor performance during normal days would be a candidate for adaptive control. A case study was conducted on 11 corridors within Des Moines, Iowa where Merle Hay Road and University Avenue were identified suitable for adaptive control.

Automated Approach for Computer Vision-based Vehicle Movement Classification at Traffic Intersections

ArXiv, 2021

Movement specific vehicle classification and counting at traffic intersections is a crucial compo... more Movement specific vehicle classification and counting at traffic intersections is a crucial component for various traffic management activities. In this context, with recent advancements in computer-vision based techniques, cameras have emerged as a reliable data source for extracting vehicular trajectories from traffic scenes. However, classifying these trajectories by movement type is quite challenging as characteristics of motion trajectories obtained this way vary depending on camera calibrations. Although some existing methods have addressed such classification tasks with decent accuracies, the performance of these methods significantly relied on manual specification of several regions of interest. In this study, we proposed an automated classification method for movement specific classification (such as right-turn, left-turn and through movements) of vision-based vehicle trajectories. Our classification framework identifies different movement patterns observed in a traffic sce...

The 2018 NVIDIA AI City Challenge

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2018

The NVIDIA AI City Challenge has been created to accelerate intelligent video analysis that helps... more The NVIDIA AI City Challenge has been created to accelerate intelligent video analysis that helps make cities smarter and safer. With millions of traffic video cameras acting as sensors around the world, there is a significant opportunity for real-time and batch analysis of these videos to provide actionable insights. These insights will benefit a wide variety of agencies, from traffic control to public safety. The second edition of the NVIDIA AI City Challenge, being organized as a CVPR workshop, provided a forum to more than 70 academic and industrial research teams to compete and solve real-world problems using traffic camera video data. The Challenge was launched with three tracks-speed estimation, anomaly detection, and vehicle re-identification. Each track was chosen in consultation with traffic and public safety officials based on the value of potential solutions. With the largest available dataset for such tasks, and ground truth for each track, the Challenge enabled 22 teams to evaluate their solutions. Given how complex these tasks are, the results are encouraging and reflect increased value addition year over year for the Challenge.

Leading Pedestrian Interval Implementation as a Marginal Costs and Benefits Problem

To improve the safety of people walking at particular signalized intersections, traffic signal en... more To improve the safety of people walking at particular signalized intersections, traffic signal engineers may implement leading pedestrian intervals (LPI) to provide pedestrians with a walk signal for a few seconds prior to the parallel vehicular green indication. Previous research using before-after studies and simple economic analyses shows that LIPs are low cost tools that can reduce vehicle-pedestrian conflicts and crashes at some signalized intersections. Despite this evidence, there is a little guidance for municipalities on when to implement LPIs. this paper develops a marginal costs and benefits framework using quantitative metrics, extending the concept of traffic conflicts and marginal safety-delay tradeoffs to analyze the appropriateness of implementing an LPI at specific signalized intersections. The guidance provided by this method helps quantify the probability of a conflict happening, and provides direction on whether or not to implement an LPI at a given location base...

Region-based deformable fully convolutional networks for multi-class object detection at signalized traffic intersections: NVIDIA AICity challenge 2017 Track 1

2017 IEEE SmartWorld, Ubiquitous Intelligence & Computing, Advanced & Trusted Computed, Scalable Computing & Communications, Cloud & Big Data Computing, Internet of People and Smart City Innovation (SmartWorld/SCALCOM/UIC/ATC/CBDCom/IOP/SCI), 2017

Multi-class object detection is critical for intelligent traffic monitoring applications in smart... more Multi-class object detection is critical for intelligent traffic monitoring applications in smart cities as well as connected autonomous vehicles. Although, numerous research works evaluate the performance of image processing algorithms for on-vehicle cameras, the body of research evaluating performance of image processing of stationary cameras located near intersections is limited. In this research, we use region-based deformable fully convolutional networks to detect 14 different object classes within images from traffic surveillance cameras. The object classes include vehicles, pedestrians, bicyclists and traffic signals. The goal of the NVIDIA AICity challenge is to provide accurate localization and correct classification of objects for stationary cameras mounted near signalized traffic intersections. Our proposed method scores mean average precision of 0.41, 0.37, and 0.34 for aic480, aic1080, and aic540 challenge datasets.

Deep Reinforcement Learning for Adaptive Traffic Signal Control

Volume 3, Rapid Fire Interactive Presentations: Advances in Control Systems; Advances in Robotics and Mechatronics; Automotive and Transportation Systems; Motion Planning and Trajectory Tracking; Soft Mechatronic Actuators and Sensors; Unmanned Ground and Aerial Vehicles, 2019

Many existing traffic signal controllers are either simple adaptive controllers based on sensors ... more Many existing traffic signal controllers are either simple adaptive controllers based on sensors placed around traffic intersections, or optimized by traffic engineers on a fixed schedule. Optimizing traffic controllers is time consuming and usually require experienced traffic engineers. Recent research has demonstrated the potential of using deep reinforcement learning (DRL) in this context. However, most of the studies do not consider realistic settings that could seamlessly transition into deployment. In this paper, we propose a DRL-based adaptive traffic signal control framework that explicitly considers realistic traffic scenarios, sensors, and physical constraints. In this framework, we also propose a novel reward function that shows significantly improved traffic performance compared to the typical baseline pre-timed and fully-actuated traffic signals controllers. The framework is implemented and validated on a simulation platform emulating real-life traffic scenarios and sen...

Traffic sensor health monitoring using spatiotemporal graphical modeling

International Journal of Prognostics and Health Management, 2020

Accurate traffic sensor data is essential for traffic operation management systems and acquisitio... more Accurate traffic sensor data is essential for traffic operation management systems and acquisition of real-time traffic surveillance data depends heavily on the reliability of the traffic sensors (e.g., wide range detector, automatic traffic recorder). Therefore, detecting the health status of the sensors in a traffic sensor network is critical for the departments of transportation as well as other public and private entities, especially in the circumstances where real-time decision is required. With the purpose of efficiently determining the sensor health status and identifying the failed sensor(s) in a timely manner, this paper proposes a graphical modeling approach called spatiotemporal pattern network (STPN). Traffic speed and volume measurement sensors are used in this paper to formulate and analyze the proposed sensor health monitoring system and historical time-series data from a network of traffic sensors on the Interstate 35 (I-35) within the state of Iowa is used for valid...

Driver Behavior Analysis Using Lane Departure Detection Under Challenging Conditions

ArXiv, 2019

In this paper, we present a novel model to detect lane regions and extract lane departure events ... more In this paper, we present a novel model to detect lane regions and extract lane departure events (changes and incursions) from challenging, lower-resolution videos recorded with mobile cameras. Our algorithm used a Mask-RCNN based lane detection model as pre-processor. Recently, deep learning-based models provide state-of-the-art technology for object detection combined with segmentation. Among the several deep learning architectures, convolutional neural networks (CNNs) outperformed other machine learning models, especially for region proposal and object detection tasks. Recent development in object detection has been driven by the success of region proposal methods and region-based CNNs (R-CNNs). Our algorithm utilizes lane segmentation mask for detection and Fix-lag Kalman filter for tracking, rather than the usual approach of detecting lane lines from single video frames. The algorithm permits detection of driver lane departures into left or right lanes from continuous lane dete...

Effect of Narrower Lane Width on Traffic Operations for Urban Midblock Segments

The narrower lane width would reduce travel speed on high speed roadways, but for urban streets, ... more The narrower lane width would reduce travel speed on high speed roadways, but for urban streets, the past studies have been inconclusive in establishing consistent relationship between lane width and vehicle speed. This paper presents an analysis of the effects of narrower lane width on travel speed and vehicle lane violation for urban midblock segments between signalized intersections. Data collected from 14 midblock segments in Nebraska were used for analysis. General regression techniques were used for modeling. The study found that 10 ft lane width is associated with higher travel speed on midblock segments with a speed limit of 25 mph or 35 mph. For speed limits of 40 mph and 45 mph, the 10 ft lane width appeared to reduce the travel speeds, which might be used for traffic calming. The midblock segments with 11 ft lane width would have higher travel speed compared to those with 12 ft lane width, but the differences in reducing lane violation in terms of encroachments on adjacen...

Leveraging high-resolution traffic data to understand the impacts of congestion on safety

Since vehicle crashes in urban area may potentially cause higher societal costs than those in rur... more

Active Traffic Management Case Study: Phase 1

This study developed a systematic approach for using data from multiple sources to provide active... more This study developed a systematic approach for using data from multiple sources to provide active traffic management solutions. The feasibility of two active traffic management solutions is analyzed in this report: ramp-metering and real-time crash risk estimation and prediction. Using a combined dataset containing traffic, weather, and crash data, this study assessed crash likelihood on urban freeways and evaluated the economic feasibility of providing a ramp metering solution. A case study of freeway segments in Omaha, Nebraska, was conducted. The impact of rain, snow, congestion, and other factors on crash risk was analyzed using a binary probit model, and one of the major findings from the sensitivity analysis was that a one-mile-per-hour increase in speed is associated with a 7.5% decrease in crash risk. FREEVAL was used to analyze the economic feasibility of the ramp metering implementation strategy. A case study of a 6.3 mile segment on I-80 near downtown Omaha showed that, a...

Active traffic management case study: phase 1 : final report

This study developed a systematic approach for using data from multiple sources to provide active... more This study developed a systematic approach for using data from multiple sources to provide active traffic management solutions. The feasibility of two active traffic management solutions is analyzed in this report: ramp-metering and real-time crash risk estimation and prediction. Using a combined dataset containing traffic, weather, and crash data, this study assessed crash likelihood on urban freeways and evaluated the economic feasibility of providing a ramp metering solution. A case study of freeway segments in Omaha, Nebraska, was conducted. The impact of rain, snow, congestion, and other factors on crash risk was analyzed using a binary probit model, and one of the major findings from the sensitivity analysis was that a one-mile-per-hour increase in speed is associated with a 7.5% decrease in crash risk. FREEVAL was used to analyze the economic feasibility of the ramp metering implementation strategy. A case study of a 6.3 mile segment on I-80 near downtown Omaha showed that, after applying ramp metering, travel time decreased from 9.3 minutes to 8.1 minutes and crash risk decreased by 37.5% during the rush hours. The benefits of reducing travel time and crash cost can easily offset the cost of implementing ramp metering for this road section. The results from the real-time crash risk prediction models developed for the studied road section are promising. A sensitivity analysis was conducted on different models and different temporal and spatial windows to estimate/predict crash risk. An adaptive boosting (AdaBoost) model using a 10 minute historical window of speeds obtained from 0.25 miles downstream and 0.75 miles upstream was found to be the most accurate estimator of crash risk. 17. Key Words active traffic management, crash risk, ramp metering 18. Distribution Statement 19. Security Classif. (of this report) Unclassified 20. Security Classif.

Freeway Traffic Incident Detection from Cameras: A Semi-Supervised Learning Approach

2018 21st International Conference on Intelligent Transportation Systems (ITSC), 2018

Early detection of incidents is a key step to reduce incident related congestion. State Departmen... more Early detection of incidents is a key step to reduce incident related congestion. State Department of Transportation (DoTs) usually install a large number of Close Circuit Television (CCTV) cameras in freeways for traffic surveillance. In this study, we used semi-supervised techniques to detect traffic incident trajectories from the cameras. Vehicle trajectories are identified from the cameras using state-of-the-art deep learning based You Look Only Once (YOLOv3) classifier and Simple Online Realtime Tracking (SORT) is used for vehicle tracking. Our proposed approach for trajectory classification is based on semi-supervised parameter estimation using maximumlikelihood (ML) estimation. The ML based Contrastive Pessimistic Likelihood Estimation (CPLE) attempts to identify incident trajectories from the normal trajectories. We compared the performance of CPLE algorithm to traditional semisupervised techniques Self Learning and Label Spreading, and also to the classification based on the corresponding supervised algorithm. Results show that approximately 14% improvement in trajectory classification can be achieved using the proposed approach.

A Rubric-Driven Evaluation of Open Data Portals and Their Data in Transportation

Journal of Big Data Analytics in Transportation, 2020

The open data movement has been gaining momentum in the Transportation industry in recent years, ... more The open data movement has been gaining momentum in the Transportation industry in recent years, with multiple State Departments of Transportation (DOT) launching their own repository of datasets. These actions offer the potential to improve both transportation safety and mobility. That said, an examination of the quality of data, ease of use, and availability of metadata across agencies revealed a significant variation among them and presented a serious concern to realizing the full potential of these principled efforts. This study highlights the imminent need for agency support towards assessing data warehouse quality and in establishing an index to measure performance. The approach includes a data portal evaluation rubric (DPER) that captures the essence of the National Open Data Policy. Its metrics are devised from practices implemented across several agencies to ensure the effective use of open data. The proposed DPER was applied to 43 transportation-related open data portals at the state (39) and national level (4). The DPER examines the quality, the openness of data, and the relevance of published content to the Transportation industry. The evaluation found that the portal hosted by the state of New York scored the highest due to its user-friendly interface with interactive visualization tools, relevant data content, detailed information about the data, and the useful API (Application Program Interface) references for application developers. The absolute scramble by agencies to fight the life-impacting events of 2020 emphasizes both the need and urgency to provide standardized, accessible, and accurate data for a ubiquitous audience.

Milkweed ( Asclepias syriaca ) plant detection using mobile cameras

Ecosphere, 2020

Milkweed (Asclepias spp.) are host plants of monarch butterflies (Danaus plexippus). It is import... more Milkweed (Asclepias spp.) are host plants of monarch butterflies (Danaus plexippus). It is important to detect milkweed plant locations to assess the status and trends of monarch habitat in support of monarch conservation programs. In this paper, we describe autonomous detection of milkweed plants using cameras mounted to vehicles. For detection, we used both aggregated channel features (ACF) for running the detectors on embedded computing platforms with central processing unit and faster regionbased convolutional neural network (Faster R-CNN) with a ResNet architecture-based detector that is suitable for graphics processing unit optimized processing. The ACF-based model produced 0.89 mean average precision (mAP) on the training dataset and 0.29 mAP on the test dataset, whereas the ResNet-based Faster R-CNN model provided 0.98 mAP on training and 0.44 mAP on the test dataset. The detections were used to calculate approximate densities of milkweed plants in geo-referenced locations based on global positioning system point correspondences of recorded images. Probability-of-count distributions are compared for the actual milkweed plant locations near roadsides. This is one of the first examples of using automated milkweed plant detection and density mapping using a vehicle-mounted camera.

Data-driven parallelizable traffic incident detection using spatio-temporally denoised robust thresholds

Transportation Research Part C: Emerging Technologies, 2019

Automatic incident detection (AID) is crucial for reducing non-recurrent congestion caused by tra... more Automatic incident detection (AID) is crucial for reducing non-recurrent congestion caused by traffic incidents. In this paper we propose a data-driven AID framework that can leverage large-scale historical traffic data along with the inherent topology of the traffic networks to obtain robust traffic patterns. Such traffic patterns can be compared with the real-time traffic data to detect traffic incidents in the road network. Our AID framework consists of two basic steps for traffic pattern estimation. First, we estimate robust univariate speed threshold using historical traffic information from individual sensors. This step can be parallelized using MapReduce framework thereby making it feasible to implement the framework over large networks. Our study shows that such robust thresholds can improve incident detection performance significantly compared to traditional threshold determination. Second, we leverage the knowledge of the topology of the road network to construct threshold heatmaps and perform image denoising to obtain spatio-temporally denoised thresholds. We used two image denoising techniques, bilateral filtering and total variation for this purpose. Our study shows that overall AID performance can be improved significantly using bilateral filter denoising compared to the noisy thresholds or thresholds obtained using total variation denoising.

Exploring spatio-temporal effects in traffic crash trend analysis

Analytic Methods in Accident Research, 2017

Unobserved heterogeneity produced by spatial and temporal correlations of crashes often needs to ... more Unobserved heterogeneity produced by spatial and temporal correlations of crashes often needs to be captured in crash frequency modeling. Although many studies have included either spatial or temporal effects in crash frequency modeling, only a limited number of studies have considered both. This study addresses the limitations of existing studies by exploring multiple models that best fit the spatial and temporal correlations. In this study, we used Bayesian spatio-temporal models to investigate regional crash frequency trends, and explored the effects of omitting spatial or temporal trends in spatio-temporal correlated data. The fast Bayesian inference approach, integrated nested Laplace approximation, was used to estimate parameters. It was found that fatal crashes showed decreasing trends in all Iowa counties from 2006 to 2015, but the decreasing rates varied by counties. Among all the covariates investigated, only vehicle miles traveled (VMT) was significant. None of the socioeconomic or weather indicators were found to be significant in the presence of VMT. Both spatial and temporal effects were found to be important, and they were responsible for both over dispersion and zero inflation in the crash data. In addition, spatial effects played a more important role than did temporal effects in the studied dataset, but temporal component selection was still important in spatio-temporal modeling.

Evaluating the Reliability, Coverage, and Added Value of Crowdsourced Traffic Incident Reports from Waze

Transportation Research Record: Journal of the Transportation Research Board, 2018

Traffic managers strive to have the most accurate information on road conditions, normally by usi... more Traffic managers strive to have the most accurate information on road conditions, normally by using sensors and cameras, to act effectively in response to incidents. The prevalence of crowdsourced traffic information that has become available to traffic managers brings hope and yet raises important questions about the proper strategy for allocating resources to monitoring methods. Although many researchers have indicated the potential value in crowdsourced data, it is crucial to quantitatively explore its validity and coverage as a new source of data. This research studied crowdsourced data from a smartphone navigation application called Waze to identify the characteristics of this social sensor and provide a comparison with some of the common sources of data in traffic management. Moreover, this work quantifies the potential additional coverage that Waze can provide to existing sources of the advanced traffic management system (ATMS). One year of Waze data was compared with the rec...

Automated Vehicle Recognition with Deep Convolutional Neural Networks

Transportation Research Record: Journal of the Transportation Research Board, 2017

In recent years there has been growing interest in the use of nonintrusive systems such as radar ... more In recent years there has been growing interest in the use of nonintrusive systems such as radar and infrared systems for vehicle recognition. State-of-the-art nonintrusive systems can report up to eight classes of vehicle types. Video-based systems, which arguably are the most popular nonintrusive detection systems, can report only very coarse classification levels (up to four classes), even with the best-performing vision systems. The present study developed a vision system that can report finer vehicle classifications according to FHWA’s scheme and is also comparable to other nonintrusive recognition systems. The proposed system decoupled object recognition into two main tasks: localization and classification. It began with localization by generating class-independent region proposals for each video frame, then it used deep convolutional neural networks to extract feature descriptors for each proposed region, and, finally, the system scored and classified the proposed regions by ...

Real-time energy-efficient traffic control via convex optimization

Transportation Research Part C: Emerging Technologies, 2018

This article proposes a macroscopic traffic control strategy to reduce fuel consumption of vehicl... more This article proposes a macroscopic traffic control strategy to reduce fuel consumption of vehicles on highways. By implementing Greenshields fundamental diagram, the solution to Moskowitz equations is expressed as linear functions with respect to vehicle inflow and outflow, which leads to generation of a linear traffic flow model. In addition, we build a quadratic cost function in terms of vehicle volume to estimate fuel consumption rate based on COPERT model. A convex quadratic optimization problem is then formulated to generate energy-efficient traffic control decisions in real-time. Simulation results demonstrate significant reduction of fuel consumption on testing highway sections under peak traffic demands of busy hours.