Аннотация:The report is devoted to solving the problem of maneuver typedetermination for aircraft flight recordings using machine learning methods. Thepurpose of the study was to obtain real flight data with labeled maneuvers for use indeveloping motion cueing algorithms. These algorithms are supposed to be used intraining simulators.During the flight synchronized inertial and video data were recorded. It was decidedto use machine learning methods to label these records because the manual markupprocess takes a lot of time, requires pilots’ consultations, and is very difficult if there isno video of the horizon overboard the aircraft. Markup during the flight also is not anoption since this will either interfere with the pilot and can lead to an accident, or theremust be another person in the cabin who is capable of marking in real time, which isquite expensive.Both video and inertial data were used to manually label the used data, but onlypreprocessed inertial data was used to train the classifiers. The article compares 10different types of classifiers which determine the type of maneuver being performed frominertial data recorded during 10 seconds of flight.