Computing Beyond Earth: AI and Distributed Intelligence for Space Systems

Abstract: 
What are the emerging challenges in bringing advanced AI, data-driven intelligence, and distributed computing into the space domain? This seminar aims to explore these questions by bringing together researchers from Politecnico di Milano, industry leaders such as Thales Alenia Space Italia, and Motus ml, a Politecnico di Milano spin-off specializing in data-driven and edge-AI solutions. The session introduces the industrial context and innovation ecosystem of Thales Alenia Space Italia, highlighting the expanding role of artificial intelligence in streamlining processes and enabling next-generation mission capabilities. Building on this perspective, the seminar delves into recent advancements in data analysis pipelines, knowledge-graph–based reasoning, and edge and distributed AI, technologies increasingly central to spaceborne and space-related systems. Attention is given to opportunities arising from distributed architectures, federated learning paradigms, micro-model deployment, and the growing convergence between cloud and space infrastructures. The goal of the seminar is to spark informed discussion, highlight promising research directions, and lay the foundations for potential collaborations with the MIT research community at the intersection of data science, AI, and future space systems engineering.

Speakers: 
Andrea Proia is a senior professional at Thales Alenia Space Italia, where he leads Performance, Workload, and Data-Driven Management across a multi-plant industrial organization. His work focuses on introducing advanced analytics and AI-enabled methodologies to enhance operational efficiency and support strategic decision-making. He has extensive experience in process and data harmonization within complex and international environments, guiding teams through initiatives that improved production governance, reduced cost variances, and strengthened risk and opportunity management. His responsibilities span cost prediction and planning, project review cycles, product line oversight, and the development of structured strategic partnerships. An AI enthusiast and active contributor to the scientific community, he leads a multi-year innovation program in collaboration with Politecnico di Milano and its spin-off Motus ml, advancing data-driven transformation and AI governance within the company. 

Francesco Corallo is the Artificial Intelligence Officer at Thales Alenia Space Italia, working within the Chief Technical Officer organization. With a background in space engineering and an early specialization in Artificial Intelligence, he has developed strong expertise at the intersection of spacecraft operations and data-driven methodologies. He began his career as a Spacecraft Operations Engineer, contributing to major Earth observation missions, where he played a key role in the development of advanced analytics and machine learning solutions for spacecraft operations. His work includes anomaly detection on time-series data and engineering tools powered by large language models. In his current role, he leads Artificial Intelligence initiatives across the company, delivering production-ready applications that enhance productivity, automation, and operational efficiency, while bridging engineering expertise with state-of-the-art Machine Learning. 

Giacomo Ziffer is a PhD Candidate in Information Technology at Politecnico di Milano and the CEO of Motus ml, a PoliMi spin-off specializing in data-driven solutions for real-time data streams and edge AI. His research focuses on advanced AI methods for dynamic data, addressing challenges such as abrupt fluctuations and complex temporal dependencies. He holds a Double Degree in Data Science from the EIT Digital Master School and has developed expertise in industrial Data Science, real-time time-series analytics, and adaptive learning techniques. He co-founded Motus ml in 2023, a company undergoing rapid consolidation and working with major industrial partners for over two years. Motus ml develops Machine Learning solutions designed for continuous learning from streaming data and efficient deployment in distributed and edge environments. The company is also part of the ESA Business Incubation Centre (ESA BIC), supporting its growth in the space-tech domain.