Research & Projects

Graduate Research Assistant, Transportation Research Institute – Old Dominion University, Norfolk VA


Truck-trailer classification using side-fire LIDAR data

Ongoing Projects:


Project : Analyzing LIDAR data in precipitation conditions

Completed Projects:


Project : Classification of Truck Trailers Based on Side-Fire LIDAR Data

  • A custom algorithm has been developed to extract and analyze LIDAR cloud point data
  • Support Vector Machine (SVM) machine learning algorithm has been used
  • A heuristic model developed to distinguish refrigerated trailers from non-refrigerated trailers from cloud point data


Project : Evaluation Of Strategies To Reduce Truck Turnarounds At The Hampton Roads Bridge-Tunnel (HRBT)

  • Analysis of past overheight violations
  • Conduct a survey with truck drivers stopped for overheight violations
  • Deploy a scanner to measure truck profiles and analyze data
  • Analyze the potential impacts of the new OH system on traffic operations


Project 1: Real-Time System State Prediction & Optimal Rebalancing Strategies for Public Bike Sharing Systems

  • Developed PostgreSQL and NoSQL database server for collecting real time Twitter, Hotel, Weather, and Bike Data

Project 2: Estimation Of Free-Flow Speed Distributions In Large Networks From Probe Data

  • Mobile GPS data has been used to estimate free-flow speed of freeways in Hampton Roads Area
  • From a methodological standpoint, the general linear model developed for the estimate FFS as a function of roadway characteristics.

Project 3: Investigating Relationship between Driving Patterns and Traffic Safety using Smartphones Based Mobile Sensor Data

  • Mobile sensor data has been used to identify unsafe driving patterns and quantified the relationship between these driving patterns and traffic crash incidences
  • Recent advancements in the count modeling literature including Heterogeneous Dispersion (HD) and Generalized Ordered Response (GOR) modeling methods are better suited for analyzing crash frequency data compared to standard Poisson and Negative Binomial (NB) models.


Project 1: Classification Algorithms for Detecting Vehicle Stops from Smartphone Accelerometer Data

  • Machine learning algorithms developed to extract useful traffic information from crowdsourced data
  • Developed algorithm detect stops from large accelerometer data produced by smartphones onboard vehicles in the traffic stream

Project 2: Combining Different Data Sources to Predict Origin-Destinations and Flow Patterns for Trucks in Large Networks

  • Developing vehicle re-identification algorithms that can integrate travel time information with pavement loop data for matching trucks between two sites
  • Assessing the accuracy of the re-identification algorithms as a function of the reliability of the travel time information


Project: Developing GoGreen Android application for vehicle and driver data collection

  • PostgreSQL database server has been developed for offline data analysis
  • The application can collect data from smartphone inertial sensors and on board diagnosis (OBDII)

Graduate Research Assistant, Virginia Modeling Analysis and Simulation Center (VMASC)


Project 1: Retrieving Africa related funding opportunities from the web, Virginia Modeling Analysis and Simulation Center (VMASC)

  • Explored and retrieved Africa funding opportunities from the web
  • Put opportunities in a SQL database for further analysis

Project 2: Identify the body of knowledge of modeling and simulation (M&S), Virginia Modeling Analysis and Simulation Center (VMASC)

  • Explored and retrieved M&S related publications from the web
  • Analyzed the retrieved publications with the Mahout machine learning framework
  • Put in a NoSQL (MongoDB) database in a structured format.


Project 1: Interactive Terrain Visualization, Old Dominion University

  • Understood raw NED data
  • Generated a 3D terrain using AutoCAD MAYA based on the NED data
  • Understood and applied 3D rendering techniques
  • Generated triangular and polygonal meshes
  • Applied text mapping process to the terrain
  • Created an interactive scene with a 3D character based on the 3D terrain
  • Built a character with some behavioral functions on the terrain (walk, run, etc.)

Project 2: Sea level rise decision model, Continuously Running Simulation Environment CoRSE, Virginia Modeling Analysis and Simulation Center (VMASC).

  • Created a Human social culture behavior environment for the support of decision making under short-and long-term emergency and urban-planning scenarios.
  • The model builds upon the infrastructure data taxonomy created by the U.S. Department of Homeland Security (DHS)
  • The data collection algorithm programmed to automatically retrieve real values from authorized government web sited such as Census and CIA
  • Modeled and simulated in Any Logic simulation software.


Project 1: Cave Automatic Virtual Environment Renovation, MSVE Department, Old Dominion University.

  • Disassembly of the old C.A.V.E
  • Cleaned and rearranged the space in the facility room
  • Prepared the raw materials used in assembly of the new C.A.V.E
  • C.A.V.E frame and projector screen assembly

Project 2: Simulation Study on Supply Chain, Old Dominion University

  • Conducted meeting with Analyst and Director of Supply Chain of Ferguson Enterprises, INC
  • Collected data on inventory levels
  • Modeled and optimized C9D inventory levels at each distribution center in Arena 12.0
  • Conducted statistical analysis on simulation results


Project: De-noising Using Wavelet-Bayesian Method

  • Implemented a wavelet Bayesian to de-noise both 1 and 2 dimensional signals
  • Coefficients at a number of frequency sub-bands contaminated by noises or artifacts, are corrected by thresholding technique
  • The preliminary results show that the method is efficient and promising to this type of problem