The project will apply robotic positioning techniques to improve the accuracy of the satellite’s orbit. The project focuses on determining the orbit of a satellite based on sensor data and a dynamic model of the system. Traditional Kalman Filter-based methods will be extended with machine learning models.
Technologies: Machine learning, Extended Kalman filter (EKF), Orbit determination
The project has programmed a working application based on the Extended Kalman Filter algorithm. Current development work focuses on using the machine learning model as part of the algorithm, and on improving the accuracy of the algorithm. The company will benefit from accurate orbit detection, as knowing the exact position of satellites is critical for some satellite applications. The University will develop and test the proposed solution. The company can use the method to increase the positioning accuracy of its mobile robots.
ReOrbit
ReOrbit is a company founded in 2020 that provides software-based satellites and ready-to-use space systems.