TERRA TECH & Yaroslavl State University to Develop Artificial Intelligence Technologies Based on Satellite Data

ST. PETERSBURG, RUSSIA (Roscosmos PR) — TERRA TECH, a subsidiary of the Russian Space Systems Holding (RKS, part of the Roscosmos State Corporation), and Yaroslavl State University named after P.G. Demidov at the St. Petersburg Economic Forum agreed to unite efforts to develop and market high-tech software solutions based on space imagery data using artificial intelligence technologies.

The signing of the agreement is intended to strengthen the competences of the parties and create new groundwork in the development of modern geoinformation solutions using deep machine learning and neural networks. Cloud geoservices are in demand on the market today, providing information and analytical products based on Earth remote sensing data from space to both government agencies and commercial consumers, including individuals. Consumer requirements for the delivery time of space data and ready-made analytics necessitate the connection of artificial intelligence technologies for automation, increasing the speed and volume of generated information.

TERRA TECH General Director Milana Elerdova: “Creation of innovative digital solutions based on geodata is a modern necessity. At TERRA TECH today, a center of competence for processing satellite images using artificial intelligence technologies has been created, an excellent team of specialists has gathered. For us, a technological partnership with a university is an opportunity to expand existing competencies, to form a personnel and product reserve for the future. Together with our partners, we will improve algorithms based on neural networks and machine learning to improve the accuracy of object recognition based on Earth remote sensing data.”

Rector of Yaroslavl State University Alexander Rusakov: “It is very important for us to have a technological partner from among the enterprises of the Roscosmos State Corporation.  Partnership with TERRA TECH allows our employees and students to test their scientific developments in the field of computer vision and artificial intelligence technologies in relation to real, serious problems and to contribute to the implementation of significant federal projects. The tasks of recognizing objects of the earth’s surface in images are in demand in various fields, from forestry to monitoring the state of landfills. We are especially proud that our common team is developing unique solutions for the use of neural network algorithms on Russian satellite imagery data, and we hope that this joint work will benefit not only the Yaroslavl region, but the entire country.”

Today, artificial intelligence subsystems using neural networks help to process images from domestic and foreign Earth remote sensing satellites. In automatic and automated mode, neural networks assess forest changes, identify illegal dumps and quarries, disturbed lands, identify buildings and structures, etc.