An automated training model for selecting maritime traffic monitoring and tracking model types using AIS data with missing information

This project will utilize Automatic Identification System (AIS) datasets to design scalable Automated Maritime Traffic Monitoring and Analysis (AMTMA) applications and tools and work with two Computational data enabled science and engineering (CDS&E) Ph.D. students to produce two dissertations in this direction. Critical applications such as the detection of anomalies, offshore and onshore attacks and data intrusions, require fast mechanisms for Artificial Intelligence (Al) analysis of thousands of events per second, as well as efficient techniques for the analysis of massive historical AIS data with missing information. There have been major developments of Big Data Analysis Frameworks for analyzing the AIS historical data, but their applications and scalable analysis techniques to the AMTMA domain remains poorly understood and difficult to benchmark due to the frequency of missing information in the often collected AIS datasets. This project introduces a Least-squares regression model with missing data and a Principal Component model for sparse functional data using AIS data that will aid in monitoring maritime traffic and directly assist in averting accidents, tracking vessels, and support in avoidance of dangerous environments. The elements of a cleaned AIS dataset with missing information are often presented as curves (trajectories) rather than single points. Functional principal components can be used to describe models of variation of such curves. If one has complete measurements for each vessel trajectory or, as is more common, one has the dataset in a spreadsheet format collected at the same time points (time stamps) for all trajectories, then many standard data analytics techniques may be applied. However, vessel trajectory data as appeared in the AIS dataset is collected at irregular and sparse set of time stamps which can differ widely across individual vessels. This project will present a technique for handling this more difficult case using a reduced rank mixed effects framework. This project also explores a Least-square regression model with missing data to develop a Regression Learner app to automatically train a selection of different models on the AIS data. An automated training model will be developed to quickly try a selection of model types, and then explore promising models interactively.

Language

• English

Project

• Status: Active
• Funding: $82500
• Contract Numbers:

  69A3552348331

• Sponsor Organizations:

  Office of the Assistant Secretary for Research and Technology

  University Transportation Centers Program
  Department of Transportation
  Washington, DC  United States  20590
• Managing Organizations:

  Maritime Transportation Research and Education Center (MarTREC)

  University of Arkansas
  4190 Bell Engineering Center
  Fayetteville, AR  United States  72701
• Performing Organizations:

  Jackson State University, Jackson

  Department of Civil and Environmental Engineering
  Jackson, MS  United States  39217-0168
• Principal Investigators:

  Kwembe, Tor

• Start Date: 20240801
• Expected Completion Date: 20250830
• Actual Completion Date: 0
• USDOT Program: University Transportation Centers

Subject/Index Terms

• TRT Terms: Artificial intelligence; Automatic vehicle identification; Data analysis; Identification systems; Traffic surveillance; Water traffic
• Subject Areas: Data and Information Technology; Marine Transportation;

Filing Info

• Accession Number: 01924919
• Record Type: Research project
• Source Agency: Maritime Transportation Research and Education Center (MarTREC)
• Contract Numbers: 69A3552348331
• Files: UTC, RIP
• Created Date: Jul 23 2024 4:29PM