National General Aviation Flight Information Database (NGAFID) Data Releases

This page contains data releases and relevant information from the NGAFID. All information provided is completely de-identified to protect pilots, and released with approval from the necessary review boards and other entities involved. The goal of these releases is completely non-punitive, with the data only being used to examine next generation machine learning and data mining methods in order to improve the safety of general aviation.

The following contains flight data releases from the NGAFID. As we gain further approvals we will release larger data sets for the public.

Within the files, columns are space separated, and comment lines begin with a # character. The first line contains the headers for the columns.

  • [2014, October 20] First Release - Five Flights:
    This is the flight data that was used along with the Toolkit for Asynchronous Optimization to gather results for the following publications:
    • Travis Desell, Sophine Clachar, James Higgins and Brandon Wild. Evolving Neural Network Weights for Time-Series Prediction of General Aviation Flight Data. In the 13th International Conference on Parallel Problem Solving from Nature (PPSN 2014). Ljubljana, Slovenia. September 13-17, 2014. [pdf]
    • Travis Desell, Sophine Clachar, James Higgins and Brandon Wild. Evolving Deep Recurrent Neural Networks Using Ant Colony Optimization. In the 15th European Conference on Evolutionary Computation in Combinatorial Optimisation (Evo* 2015: EvoCOP). Copenhagen, Denmark. April 08-10, 2015. To Appear.