Working Group E

Agenda
	Working Groups

Working Group E	Intelligent Systems and Machine Learning for Space Ground Systems
	Monday – February 27, 2023 12:30 PM-3:00 PM PT
Description	Adaptive, reliable automation and intelligent decision making are essential for the success of our space ground systems. One big challenge is migrating capabilities out of the lab to supporting critical mission operations. In the past, these approaches were often misunderstood, misapplied, too complex or costly to sustain, or insufficient for mission needs. Applied intelligent systems and machine learning technologies have begun to address this challenge through self-evolving, efficient, and value-focused capabilities. In addition, un-realized opportunities exist for applying established, or rapidly emerging technologies, solutions and architectures to the area of ground system space control and mission processing. This year of the GSAW “Intelligent Systems / Machine Learning for Space Ground Systems” working group will explore deeper the themes of: Where do intelligent systems and machine learning currently exist in space ground systems? What underlying parts of the space ground systems, enterprise and operations are suited to intelligent systems and machine learning? What emerging capabilities and technologies are being developed in the community? What are real-world impediments for adoption in operations? What capability and technology gaps exist and might seed further research and investment? How have program dealt with end user trust and acceptance of planned intelligent and machine learning systems for space ground systems? What is the nature of the intersection of reliable autonomy and trusted AI, and how is this related to machine learning operations (MLOPs)? Format of the Working Group Part 1: Use Case Presentations. This portion will include a brief recap of definitions, concepts and misconceptions, and summary material from the previous year’s GSAW Intelligent System – Machine Learning Working Group. This will be followed by a series of thought-provoking presentations and case studies on where intelligent systems and machine learning are currently adopted in space ground systems for the focus areas listed above. The purpose will be to seed the Working Group with real-world, current examples of successes, failures, challenges, opportunities. Part 2: Panel Discussion. A group of panelists will address topics covered in the survey and answer questions from the audience. Desired Outcomes The primary goal of the Working Group is to compile the results from the presentations, survey and panel discussion, to be out-briefed at end of the GSAW Workshop. The outcome should inform adopters, as well as informing stakeholders who are able to support research and development and help integrate intelligent systems into operations. A secondary goal is to establish an enduring community for long-term collaboration.
Leads	Nick Perlongo and James Andrew Gilbertson, The Aerospace Corporation
Biographies	Dr. Nick Perlongo is currently a section manager in the CSG Artificial Intelligence, Analytics & Innovation department at The Aerospace Corporation. He and his team provide software development solution architecting, including ML/AI, to advance missions in civil space. Prior to joining Aerospace, Dr. Perlongo’s research included development of a comprehensive physics-based models to simulate geospace weather dynamics. He also worked as a data scientist at an advertising technology startup company. Dr. Perlongo holds a Ph.D. in Atmospheric and Space Science and a B.S.E in Earth Systems Science & Engineering from the University of Michigan. James Andrew Gilbertson is a Senior Engineering Specialist in the Data Science and Artificial Intelligence Department at The Aerospace Corporation. He has more than ten years of experience with ground systems concept exploration, prototyping, and acquisition support experience. More recently he his research has focused on applying machine learning to launch telemetry anomaly detection, cybersecurity, and object detection. Additionally, his work has explored machine learning engineering, MLOps, and integrating analytic capabilities in larger systems. Mr. Gilbertson has a master’s in Computer Science from Georgia Tech and a B.S in Computer Engineering and Mathematics from Virginia Tech.