Modeling and Simulation of a Spacecraft Payload Hardware Using Machine Learning Techniques

Abstract

Space systems are complex and consist of multiple subsystems. Research and development teams of such complex systems are usually distributed among various institutions and space agencies. This affects the quality of the On-board Software (OBSW) since testing it without having all required subsystems at the software development site can be troublesome. In this paper, we present a data-driven method which can be used to synthesize parts of a system or even an entire system as a black-box model. We exploit the data collected from the real hardware to derive a model using a Machine Learning (ML) algorithm. The proposed model can easily be distributed among development teams and is dedicated to emulate the system for testing the OBSW.