Speaker
Description
In the last decade, satellite miniaturization has significantly expanded access to space, enabling scientific, technological, and educational missions. This situation introduces inherent limitations in terms of structural redundancy, onboard computational resources, and physical testing opportunities, factors that can compromise mission reliability and performance assurance. To address these challenges, this research proposes the development of a digital twin focused on the payload subsystem of a nanosatellite mission.
The project envisions a detailed modeling approach of the payload structure, prioritizing the inclusion of a wide range of variables to closely approximate real-world physical behavior. The framework will simulate critical operational scenarios that realistically represent the environmental and mechanical conditions experienced throughout the mission lifecycle. In parallel, machine learning algorithms will be implemented to enable predictive data analysis, aiming to anticipate anomalies and forecast system behavior under diverse operational scenarios, with the advantage of conducting these assessments without compromising the integrity of the physical payload.
The proposed validation process will be conducted in two stages: a pre-launch phase using ground-based testing platforms, and an in-orbit phase where the digital twin will continuously receive telemetry from the satellite. This real-time data integration will allow the model to adapt and refine its predictions, improving the fidelity and responsiveness of the digital representation. This line of research aims to consolidate the digital twin as a strategic tool for enhancing operational efficiency, reducing design, testing, and deployment costs, and increasing the reliability of future satellite missions.