About

Sanjay Ranka is a Distinguished Professor in the Department of Computer & Information Science & Engineering at the University of Florida. His research focuses on high-performance computing and big data science, with an emphasis on developing efficient computational methods and data analysis techniques to model scientific phenomena. His practical applications include improving healthcare quality and reducing traffic accidents. Ranka has a background that includes serving as the Chief Technology Officer at Paramark, where he developed a real-time optimization service called PILOT, which served over 10 million decisions daily with high uptime, and was recognized as a top internet technology company before its acquisition. He has held faculty positions at Syracuse University, been an academic visitor at IBM, and a summer researcher at Hitachi America Limited. Ranka has co-authored a book, four monographs, and over 300 journal and conference articles. His work has received multiple awards, including best paper recognitions and the 2020 Research Impact Award from the IEEE Technical Committee on Cloud Computing. He is a fellow of IEEE and AAAS, and serves as an editor for several prominent journals. His research interests include data mining, informatics, grid computing, digital health, embedded systems, and energy-aware computing.

Research topics

• Computer Science
• Artificial Intelligence
• Medicine
• Psychology
• Machine Learning
• Computer vision
• Physical medicine and rehabilitation
• Mathematics
• Physical therapy
• Engineering
• Human–computer interaction
• Gerontology
• Clinical psychology
• Data Mining
• Neuroscience
• Statistics
• Internal medicine
• Psychiatry
• Computer network
• Automotive engineering
• Geography
• Real-time computing

Selected publications

• Measuring Braking Behavior Using Vehicle Tracking and Camera-to-Satellite Homography Rectification

  arXiv (Cornell University) · 2026-01-24

  preprintOpen access

  This paper presents an open-source software application for analyzing traffic camera footage, focusing on vehicle behavior and braking events at signalized urban highways. The core innovation is a robust ground-plane homography estimation that links fixed traffic camera views to satellite orthoimagery. This process rectifies the camera's oblique perspective, ensuring that pixel distances accurately represent real-world distances. This enables the acquisition of features such as vehicle trajectory, speed, deceleration, and braking severity without the need for camera calibration. The pipeline employs the MAGSAC++ estimator to build the homography, converting YOLO11 object detections into a rectified top-down coordinate system. All detection and trajectory data are stored in a ClickHouse database for subsequent analysis. A real-world case study at two signalized intersections in Key West, Florida, showcased the system's capabilities. Across two days of daytime footage, braking activity at the higher-volume intersection peaked around 4 PM at approximately 57.5 events per hour, while the second intersection peaked around 10 AM at roughly 15.5 events per hour. The spatial analysis revealed that most braking events initiated upstream, with mild and moderate braking mostly occurring 30 to 45+ meters away from the stop bar and severe braking distributed throughout, but particularly concentrated in lanes with higher interaction and merging activity. The findings highlight the significant potential of this centralized safety information system to support connected vehicles, facilitating proactive traffic management, crash mitigation, and data-driven roadway design and safety analysis.

  PublisherDOI

• Measuring Braking Behavior Using Vehicle Tracking and Camera-to-Satellite Homography Rectification

  ArXiv.org · 2026-01-24

  articleOpen access

  This paper presents an open-source software application for analyzing traffic camera footage, focusing on vehicle behavior and braking events at signalized urban highways. The core innovation is a robust ground-plane homography estimation that links fixed traffic camera views to satellite orthoimagery. This process rectifies the camera's oblique perspective, ensuring that pixel distances accurately represent real-world distances. This enables the acquisition of features such as vehicle trajectory, speed, deceleration, and braking severity without the need for camera calibration. The pipeline employs the MAGSAC++ estimator to build the homography, converting YOLO11 object detections into a rectified top-down coordinate system. All detection and trajectory data are stored in a ClickHouse database for subsequent analysis. A real-world case study at two signalized intersections in Key West, Florida, showcased the system's capabilities. Across two days of daytime footage, braking activity at the higher-volume intersection peaked around 4 PM at approximately 57.5 events per hour, while the second intersection peaked around 10 AM at roughly 15.5 events per hour. The spatial analysis revealed that most braking events initiated upstream, with mild and moderate braking mostly occurring 30 to 45+ meters away from the stop bar and severe braking distributed throughout, but particularly concentrated in lanes with higher interaction and merging activity. The findings highlight the significant potential of this centralized safety information system to support connected vehicles, facilitating proactive traffic management, crash mitigation, and data-driven roadway design and safety analysis.

  PublisherOA PDF

• BigSUMO: A Scalable Framework for Big Data Traffic Analytics and Parallel Simulation

  arXiv (Cornell University) · 2026-01-05

  preprintOpen access

  With growing urbanization worldwide, efficient management of traffic infrastructure is critical for transportation agencies and city planners. It is essential to have tools that help analyze large volumes of stored traffic data and make effective interventions. To address this need, we present ``BigSUMO", an end-to-end, scalable, open-source framework for analytics, interruption detection, and parallel traffic simulation. Our system ingests high-resolution loop detector and signal state data, along with sparse probe trajectory data. It first performs descriptive analytics and detects potential interruptions. It then uses the SUMO microsimulator for prescriptive analytics, testing hundreds of what-if scenarios to optimize traffic performance. The modular design allows integration of different algorithms for data processing and outlier detection. Built using open-source software and libraries, the pipeline is cost-effective, scalable, and easy to deploy. We hope BigSUMO will be a valuable aid in developing smart city mobility solutions.

  PublisherDOI

• BigSUMO: A Scalable Framework for Big Data Traffic Analytics and Parallel Simulation

  ArXiv.org · 2026-01-05

  articleOpen access

  With growing urbanization worldwide, efficient management of traffic infrastructure is critical for transportation agencies and city planners. It is essential to have tools that help analyze large volumes of stored traffic data and make effective interventions. To address this need, we present ``BigSUMO", an end-to-end, scalable, open-source framework for analytics, interruption detection, and parallel traffic simulation. Our system ingests high-resolution loop detector and signal state data, along with sparse probe trajectory data. It first performs descriptive analytics and detects potential interruptions. It then uses the SUMO microsimulator for prescriptive analytics, testing hundreds of what-if scenarios to optimize traffic performance. The modular design allows integration of different algorithms for data processing and outlier detection. Built using open-source software and libraries, the pipeline is cost-effective, scalable, and easy to deploy. We hope BigSUMO will be a valuable aid in developing smart city mobility solutions.

  PublisherOA PDF

• Enactor: From Traffic Simulators to Surrogate World Models

  arXiv (Cornell University) · 2026-03-18

  preprintOpen accessSenior author

  Traffic microsimulators are widely used to evaluate road network performance under various ``what-if" conditions. However, the behavior models controlling the actions of the actors are overly simplistic and fails to capture realistic actor-actor interactions. Deep learning-based methods have been applied to model vehicles and pedestrians as ``agents" responding to their surrounding ``environment" (including lanes, signals, and neighboring agents). Although effective in learning actor-actor interaction, these approaches fail to generate physically consistent trajectories over long time periods, and they do not explicitly address the complex dynamics that arise at traffic intersections which is a critical location in urban networks. Inspired by the World Model paradigm, we have developed an actor centric generative model using transformer-based architecture that is able to capture the actor-actor interaction, at the same time understanding the geometry to the traffic intersection to generate physically grounded trajectories that are based on learned behavior. Moreover, we test the model in a live ``simulation-in-the-loop" setting, where we generate the initial conditions of the actors using SUMO and then let the model control the dynamics of the actors. We let the simulation run for 40000 timesteps (4000 seconds), testing the performance of the model on long timerange and evaluating the trajectories on traffic engineering related metrics. Experimental results demonstrate that the proposed framework effectively captures complex actor-actor interactions and generates long-horizon, physically consistent trajectories, while requiring significantly fewer training samples than traditional agent-centric generative approaches. Our model is able to outperform the baseline in traffic related as well as aggregate metrics where our model beats the baseline by more than 10x on the KL-Divergence.

  PublisherDOI

• Enactor: From Traffic Simulators to Surrogate World Models

  ArXiv.org · 2026-03-18

  articleOpen accessSenior author

  Traffic microsimulators are widely used to evaluate road network performance under various ``what-if" conditions. However, the behavior models controlling the actions of the actors are overly simplistic and fails to capture realistic actor-actor interactions. Deep learning-based methods have been applied to model vehicles and pedestrians as ``agents" responding to their surrounding ``environment" (including lanes, signals, and neighboring agents). Although effective in learning actor-actor interaction, these approaches fail to generate physically consistent trajectories over long time periods, and they do not explicitly address the complex dynamics that arise at traffic intersections which is a critical location in urban networks. Inspired by the World Model paradigm, we have developed an actor centric generative model using transformer-based architecture that is able to capture the actor-actor interaction, at the same time understanding the geometry to the traffic intersection to generate physically grounded trajectories that are based on learned behavior. Moreover, we test the model in a live ``simulation-in-the-loop" setting, where we generate the initial conditions of the actors using SUMO and then let the model control the dynamics of the actors. We let the simulation run for 40000 timesteps (4000 seconds), testing the performance of the model on long timerange and evaluating the trajectories on traffic engineering related metrics. Experimental results demonstrate that the proposed framework effectively captures complex actor-actor interactions and generates long-horizon, physically consistent trajectories, while requiring significantly fewer training samples than traditional agent-centric generative approaches. Our model is able to outperform the baseline in traffic related as well as aggregate metrics where our model beats the baseline by more than 10x on the KL-Divergence.

  PublisherOA PDF

• BigSUMO: A Scalable Framework for Big Data Traffic Analytics and Parallel Simulation

  2025-12-10

  article

  With growing urbanization worldwide, efficient management of traffic infrastructure is critical for transportation agencies and city planners. It is essential to have tools that help analyze large volumes of stored traffic data and make effective interventions. To address this need, we present “BigSUMO”, an end-to-end, scalable, open-source framework for analytics, interruption detection, and parallel traffic simulation. Our system ingests high-resolution loop detector and signal state data, along with sparse probe trajectory data. It first performs descriptive analytics and detects potential interruptions. It then uses the SUMO microsimulator for prescriptive analytics, testing hundreds of what-if scenarios to optimize traffic performance. The modular design allows integration of different algorithms for data processing and outlier detection. Built using open-source software and libraries, the pipeline is cost-effective, scalable, and easy to deploy. We hope BigSUMO will be a valuable aid in developing smart city mobility solutions.

  PublisherDOI

• TGDT: A Temporal Graph-based Digital Twin for Urban Traffic Corridors

  2025-11-18 · 1 citations

  articleSenior author
  PublisherDOI

• A Comprehensive Survey on Multi-Layer Graph Embedding Methods

  Vietnam Journal of Computer Science · 2025-07-23

  articleOpen accessSenior author

  The use of graphs enables the systematic modeling, analysis, and optimization of complex systems in various real-world domains. When multiple types of relationships or interactions exist among entities, whether homogeneous or heterogeneous, graphs can be structured into multiple layers to model context-specific interdependencies and more effectively capture the complexity of these interactions. This survey introduces a novel and comprehensive taxonomy that categorizes the diverse spectrum of multi-layer graph embedding methods into three main groups: algorithmic, machine learning, and deep learning approaches. This survey aims to serve as a guide for the research community in navigating the graph embedding methods for multi-layer graphs by providing a structured summary, analysis, and comparison within and across different categories that highlight their respective strengths, limitations, and suitability for various application domains. Furthermore, we examine key factors that influence the selection of appropriate methods, including graph structure, inherent properties, application domain, learning paradigm, and computational constraints. Finally, we outline several promising research directions to advance this rapidly evolving field.

  PublisherDOI

• Graph Attention Network for Lane-Wise and Topology-Invariant Intersection Traffic Simulation

  IEEE Transactions on Intelligent Transportation Systems · 2025-03-10 · 5 citations

  articleSenior author

  Traffic congestion poses significant economic, environmental, and social challenges. High-resolution loop detector data and signal state records from Automated Traffic Signal Performance Measures (ATSPM) offer new opportunities for traffic signal optimization at intersections. However, additional factors such as geometry, traffic volumes, Turning-Movement Counts (TMCs), and human driving behaviors complicate this task. Existing simulators (e.g., SUMO, Vissim) are computationally intensive, while machine learning models often lack lane-specific traffic flow estimation. To address these issues, we propose two computationally efficient Attentional Graph Auto-Encoder frameworks as “Digital Twins” for urban traffic intersections. Leveraging graph representations and Graph Attention Networks (GAT), our models capture lane-level traffic flow dynamics at entry and exit points while remaining agnostic to intersection topology and lane configurations. Trained on over 40,000 hours of realistic traffic simulations with affordable GPU parallelization, our framework produces fine-grained traffic flow time series. This output supports critical applications such as estimating Measures of Effectiveness (MOEs), scaling to urban freeway corridors, and integrating with signal optimization frameworks for improved traffic management.

  PublisherDOI

Recent grants

• MCDA: Collaborative Research: A Multi-Element and Multi-Objective Optimization Approach for Allocating tasks to Multi-Core Processors

  NSF · $270k · 2009–2013

• Sparse Direct Methods on High-Performance Heterogeneous Architectures

  NSF · $310k · 2011–2015

• ITR: Collaborative Research: A Data Mining and Exploration Middleware for Grid and Distributed Computing

  NSF · $535k · 2003–2008

• CSR: Medium: Collaborative Research: SparseKaffe: high-performance, auto-tuned, energy-aware algorithms for sparse direct methods on modern heterogeneous architectures

  NSF · $395k · 2015–2019

• SCC: Video Based Machine Learning for Smart Traffic Analysis and Management

  NSF · $2.0M · 2019–2024

Frequent coauthors

• Anand Rangarajan

  140 shared

• Tania Banerjee

  82 shared

• Sartaj Sahni

  University of Florida

  74 shared

• Yashaswi Karnati

  41 shared

• Chilukuri K. Mohan

  Virtual High School

  37 shared

• Todd M. Manini

  University of Florida

  35 shared

• Rahul Sengupta

  University of Florida

  31 shared

• Pan He

  Auburn University

  29 shared

Labs

• Adaptive Learning & Optimization LabPI

Education

• Ph.D., Computer Science

  University of California, Santa Barbara

  1990

• M.S., Computer Science

  University of California, Santa Barbara

  1986

• B.S., Computer Science and Engineering

  Indian Institute of Technology, Kanpur

  1983

Awards & honors

• 2020 Research Impact Award from IEEE Technical Committee on…
• Best Paper Award at ICN 2007
• Best Student Paper Award at ACM-BCB 2010
• Best Paper Award at BICOB 2014
• Best Student Paper Runner-up Award at IGARSS 2015