(Friday 4/29) SAIL Seminar: AI in Space – Dr. Shima Ghasemi (ESA)

(Friday 4/29) SAIL Seminar: AI in Space – Dr. Shima Ghasemi (ESA)

Abstract:

One of the current challenges in Earth Observation is the successful application of ML approaches for the extraction of relevant information from a large variety of sensors. The application of latest development in AI provides the potential to more efficiently exploit the earth observation data. This talk focuses on research activity is going on between ESA, NASA and other national space agencies started in November 2019. This research activity has been evolved mainly around the exploitation of the (Hyperspectral) earth observation data, although it is a potentially useful application in other new missions. The aim of this talk to explain the ongoing research activity to exploit the earth observation data by means of artificial intelligence techniques. This can be carried out in two different ways: 1) by using optimisation methods to select the most relevant spectral bands for specific case studies (metaheuristic algorithms have already proven their ability in this tasks, according to the literature), and 2) by using Deep Learning to extract features from the whole collection of bands, which is more demanding from the computational point of view but does not discard any source of information. Hence a new era of using AI technology in earth observation has begun, a brief explanation of upcoming mission and usage of AI-on-board will be discussed as well as application of cutting edge AI practices for compression data in the orbit.
Speaker: Shima Ghasemi, Ph.D. is currently a researcher at European Space Agency (ESA), ESTEC, The Netherlands mainly focused on AI on board for a new earth observation mission. Her main activities are focused on looking into best practices for building AI unit in the orbit and earth observation applications. Her previous experiences relatively are: working as data & AI consultant at Microsoft and Shell oil & energy company. She has a PhD on earth observation systems with a topic related to using AI/ML for processing Polarimetric-Interferometric-Synthetic-Aperture-Radar data.
Categories: Latest Events