Autonomous Space Training for Robust Operations and Edge-Driven Growth in Experimentation (ASTROEDGE)

The ASTROEDGE project presents a transformative approach to autonomous space command and control (C2) capabilities by integrating artificial intelligence (AI) and machine learning algorithms with diverse, high-quality space situational awareness data. Focusing on the development of a digital twin training environment, ASTROEDGE aims to simulate a wide array of space scenarios for robust AI algorithm training and evaluation. This environment leverages electro-optical, infrared, and radio frequency data from real and synthetic sources to simulate space object cataloging and proximity operations, thereby addressing the challenge of varied data quality and ensuring the algorithms' performance across different scenarios. The project also demonstrates the importance of on-orbit adaptation and trust in AI policies by incorporating advanced learning approaches and runtime assurance systems. With a foundation built on extensive experience in DoD programs,Āexpertise in space digital twin modeling, and AI spacecraft control, ASTROEDGE aims to set aĀnew standard for developing and deploying autonomous systems in space operations, promoting space domain superiority and operational efficiency.