An algorithm for automatic identification of asymmetric transits in the TESS database

Abstract Currently, the Transiting Exoplanet Survey Satellite (TESS) searches for Earth-size planets around nearby dwarf stars. To identify specific weak variations in the light curves of stars, sophisticated data processing methods and analysis of the light curve shapes should be developed and applied. We report some preliminary results of our project to find and identify minima in the light curves of stars collected by TESS and stored in the MAST (Mikulski Archive for Space Telescopes) database. We developed Python code to process the short-cadence (2-min) TESS PDCSAP (Pre-search Data Conditioning Simple Aperture Photometry) light curves. Our code allows us to create test samples to apply machine learning methods to classify minima in the light curves taking into account their morphological signatures. Our approach will be used to find and analyze some sporadic events in the observed light curves originating from transits of comet-like bodies.