Upcoming Event: Center for Autonomy Seminar

Satellite Chasers: Intelligent Adversaries on the Edge of Space

Gregory Falco, Cornell University

Abstract

As space becomes increasingly crowded and contested, developing robust autonomous capabilities for spacecraft operating in complex multi-agent scenarios has emerged as a critical priority. Existing autonomous methods in space applications primarily leverage optimization-based path planning or pre-planned orbital maneuvers. However, these conventional approaches have proven insufficient for adversarial situations in which one satellite actively pursues another. In this talk, we explore recent research advances addressing these applied challenges, revealing fundamental autonomous systems research problems. We begin by examining critical algorithmic issues encountered in six-degree-of-freedom sensor fusion and the rigorous hardware-in-the-loop testing on actual flight hardware necessary to validate autonomous technologies. Subsequently, we introduce Divergent Adversarial Reinforcement Learning (DARL), an innovative two-stage Multi-Agent Reinforcement Learning (MARL) framework designed specifically to enable spacecraft to autonomously develop effective evasion strategies against coordinated adversarial engagements. The discussion will conclude with reflections on open research questions in autonomy and control, highlighting how the unique constraints of space serve as a compelling testbed for advancing multi-agent reinforcement learning, adversarial decision-making, and spacecraft autonomy.

Biography

Prof. Gregory Falco has been at the forefront of space system and critical infrastructure security in both industry and academia for the past decade. As a tenure-track Assistant Professor at Cornell University, he is the Director of the Aerospace ADVERSARY (Autonomy, Defense and Vulnerability Exploitation for Resilient, Secure and Assured Risk/Yield) Lab. The ADVERSARY designs and develops future aerospace technology enabling secure, resilient and assured autonomous infrastructure. Falco is the founding chair of the IEEE International Technical Standard for Space System Cybersecurity. He was the former co-founder and CEO of the blockchain-based industrial control security company NeuroMesh Inc., which was acquired by Meta in 2022. He has been listed in Forbes 30 Under 30 for his inventions and contributions to critical infrastructure cyber security, is a Fulbright Scholar and is the recipient of the DARPA RISER and DARPA’s Young Faculty Award for work on building a zero-trust collaboration ecosystem for space systems. Prof. Falco serves as a member of the Department of Homeland Security’s Space Systems Critical Infrastructure Working Group and has been awarded contracts relating to cyber-physical system security for AFRL, US Space Force, NASA and DARPA. Prior to joining Cornell University, Falco was an Assistant Professor at Johns Hopkins University, jointly appointed between the Whiting School of Engineering and the Applied Physics Laboratory. Falco completed his PhD at MIT’s Computer Science and Artificial Intelligence Laboratory, Master’s degree at Columbia University and Bachelor’s degree at Cornell University. Prior to academia, Falco co-founded and led Accenture’s Smart Cities business.