A5025201295- Traffic Prediction and Management Techniques
- Traffic control and management
- Autonomous Vehicle Technology and Safety
- Video Surveillance and Tracking Methods
- Transportation Planning and Optimization
- Advanced Neural Network Applications
- Anomaly Detection Techniques and Applications
- Time Series Analysis and Forecasting
- Simulation Techniques and Applications
- Data Management and Algorithms
- COVID-19 epidemiological studies
- Robotics and Sensor-Based Localization
- Medical Imaging and Analysis
- Advanced Image and Video Retrieval Techniques
- Fire effects on ecosystems
- Impact of Light on Environment and Health
- Simulation and Modeling Applications
- Target Tracking and Data Fusion in Sensor Networks
- Manufacturing Process and Optimization
- Vehicular Ad Hoc Networks (VANETs)
- Seismic Imaging and Inversion Techniques
- Automated Road and Building Extraction
- Remote Sensing and LiDAR Applications
- Traffic and Road Safety
- Image Processing and 3D Reconstruction
Integrated Software (United States)
2019-2025
Vanderbilt University
2019-2025
In this article, we assess the string stability of seven 2018 model year <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">adaptive cruise control</i> (ACC) equipped vehicles that are widely available in US market. Seven distinct vehicle models from two different makes analyzed using data collected more than 1,200 miles driving car-following experiments with ACC engaged by follower vehicle. The resulting dataset is used to identify parameters a...
This work introduces a multi-camera tracking dataset consisting of 234 hours video data recorded concurrently from overlapping HD cameras covering 4.2 mile stretch 8-10 lane interstate highway near Nashville, TN. Video is in cooperation with Tennessee State Department Transportation and its policies. The during period high traffic density 500+ objects typically visible within the scene typical object longevities 3-15 minutes. GPS trajectories 270 vehicle passes through are manually corrected...
This review offers a comprehensive overview of current traffic modeling, estimation, and control methods, along with resulting field experiments. It highlights key developments future directions in leveraging technological advancements to improve management safety. The focus is on macroscopic, microscopic, micro-macro models, as well state-of-the-art techniques estimation methods for deploying vehicles
This work presents a novel video dataset recorded from overlapping highway traffic cameras along an urban interstate, enabling multi-camera 3D object tracking in monitoring context. Data is released 3 scenes containing at least 16 each, totaling 57 minutes length. 877,000 bounding boxes and corresponding tracklets are fully accurately annotated for each camera field of view combined into spatially temporally continuous set vehicle trajectories scene. Lastly, existing algorithms to benchmark...
Vehicle trajectory data has received increasing research attention over the past decades. With technological sensing improvements such as high-resolution video cameras, in-vehicle radars and lidars, abundant individual contextual traffic is now available. However, though quantity massive, it by itself of limited utility for because noise systematic errors, thus necessitates proper processing to ensure quality. We draw particular extracting vehicle from cameras monitoring are becoming...
Despite the high utility of traffic volume and turning movement data, such data is still hard to come by for vast majority roadways intersections in nearly every city. Edge computing devices offer a promising tool recording if lightweight algorithms can be designed run real-time with relatively modest computational complexity. To that end, this work presents Vehicle Turning-Movement Counting using Localization-based Tracking (LBT-Count). This method fast because it never performs detection...
This paper introduces a toolbox for generating virtual trajectories, which we call VT-tools v1.0, to address challenges in analyzing large but imperfect trajectory datasets. v1.0 is able generate trajectories from raw datasets that are typically challenging process due their size. We also provide set of these resulting I–24 MOTION INCEPTION data and demonstrate the practical utility as-sessing speed variability travel times across different lanes within dataset. The opens future research on...
We introduce the I-24 Mobility Technology Interstate Observation Network (MOTION), a transportation cyber-physical systems testbed under development in Tennessee. It consists of six-mile freeway segment instrumented with 400 4K resolution cameras, processed by real-time compute system to enable continuous performance monitoring traffic. The is being developed support next generation connected and autonomous vehicle technologies advanced traffic management. When complete, will be longest...
The dissipation of stop-and-go waves attracted recent attention as a traffic management problem, which can be efficiently addressed by automated driving. As part the 100 vehicles experiment named MegaVanderTest, feedback controls were used to induce strong via velocity smoothing. More precisely, single vehicle driving differently in one four lanes I-24 Nashville area was able regularize profile reducing oscillations time and differences among vehicles. Quantitative measures this effect...
End-to-end production of vehicle tracking data from video in real-time and with high accuracy remains a challenging problem due to the computational cost object detection on each frame. In this work we present Tracking Crop-based Detection, method for speeding constrained contexts (with stable cameras relatively-predictable motion) such as traffic monitoring. We leverage context provide strong prior locations, which use 1.) boost speed by detecting objects only regions corresponding priors...
Social distancing has become a pressing and challenging issue during the Covid-19 pandemic. In smart cities context, it becomes possible to measure inter-personal distance using networked cameras computer vision analysis. We deploy pipeline based on Retinanet that identifies pedestrians in streaming video frames, then converts their positions GPS coordinates for calculation further This processing is applied nine camera streams at three locations from around Vanderbilt University. collect 70...
Analyzing stop-and-go waves at the scale of miles and hours data is an emerging challenge in traffic research. In past, datasets were limited could be easily analyzed by hand or with rudimentary methods to identify a very set present within data. This paper introduces automatic scalable wave identification method capable capturing generation, propagation, dissipation, as well bifurcation merging, which have previously been observed only rarely. Using concise simple critical-speed based...
Stop-and-go waves are a fundamental phenomenon in freeway traffic flow, contributing to inefficiencies, crashes, and emissions. Recent advancements high-fidelity sensor technologies have improved the ability capture detailed dynamics, yet such systems remain scarce costly. In contrast, conventional sensors widely deployed but suffer from relatively coarse-grain data resolution, potentially impeding accurate analysis of stop-and-go waves. This article explores whether generative AI models can...
Adaptive cruise control (ACC) is the first wave of vehicle automation that will reach mainstream. It has been shown in [3] a small fraction vehicles traffic (e.g., 5%) can change emergent properties flow, for example by dissipating phantom jams. Substantial theoretical and experimental underpinnings platooning were established from USDOT Automated Highway System effort [1]. However, it not yet clear whether ACC are currently commercially available dampen or amplify
The Interstate-24 MObility Technology Interstate Observation Network (I-24 MOTION) is a new instrument for traffic science located near Nashville, Tennessee. I-24 MOTION consists of 276 pole-mounted high-resolution cameras that provide seamless coverage approximately 4.2 miles I-24, 4-5 lane (each direction) freeway with frequently observed congestion. are connected via fiber optic network to compute facility where vehicle trajectories extracted from the video imagery using computer vision...
Monocular 3D object detection is a challenging task because depth information difficult to obtain from 2D images. A subset of viewpoint-agnostic monocular methods also do not explicitly leverage scene homography or geometry during training, meaning that model trained thusly can detect objects in images arbitrary viewpoints. Such works predict the projections bounding boxes on image plane estimate location boxes, but these are rectangular so calculation IoU between projected polygons...
This article introduces a new virtual trajectory dataset derived from the I-24 MOTION INCEPTION v1.0.0 to address challenges in analyzing large but noisy datasets. Building on concept of trajectories, we provide Python implementation generate trajectories raw datasets that are typically challenging process due their size. We demonstrate practical utility these assessing speed variability and travel times across different lanes within dataset. The opens future research traffic waves impact energy.
End-to-end production of object tracklets from high resolution video in real-time and with accuracy remains a challenging problem due to the cost detection on each frame. In this work we present Localization-based Tracking (LBT), an extension any tracker that follows tracking by or joint paradigms. focuses only regions likely contain objects boost speed avoid matching errors. We evaluate LBT as two example trackers (KIOU SORT) UA-DETRAC MOT20 datasets. LBT-extended outperform all other...