Multi-Object Behavior Modeling of Space Systems

In response to the Department of Defense (DoD) Small Business Innovation Research (SBIR) 2023.3 Solicitation for the Space Force Phase II Program, topic F233-0014 “Multi-Body Behavior Modeling of Space Systems,” Advanced Space, LLC (Advanced Space) proposes to apply ML algorithms to create a containerized application that autonomously detects and assesses potential multi-vehicle threats and to remove the need for humans-in-the-loop until decision-quality information is referred to the end-user for final authorization. Critical to this application are automated intelligence agents that can assess real-time data intuitively, make informed and contextual decisions, and recommend the best course of action to a human operator who ultimately decides whether to pursue the recommended action. In other words, we aim to reduce the burden on the human-in-the-loop requirements throughout the entire data analysis process and strategically automate certain aspects of this process to increase tactical response capabilities.  The purpose of this tool is to fit within larger DoD strategic frameworks for Space Domain Awareness (SDA) and constellation Command and Control (C2). The individual components of the tool will be utilized in an integrated software application to containerize the software to be compatible with broader SDA and C2 pipelines. As part of the Phase II effort, Advanced Space will work with Redwire Space, Inc. (Redwire) to better understand the containerization and integration requirements for the software. Redwire has existing expertise in the SDA and C2 realm that will be leveraged alongside Advanced Space’s ML expertise to validate this technology's long-term objectives.  The goal of the proposed Multi-Object Behavioral Modeling effort is to create a software tool to anticipate and mitigate threats to satellites on a constellation scale to defend U.S. and allied capabilities in the event of hostile action against them. Using machine learning (ML) methodologies, the proposed system will define the parameters of potentially threatening maneuvers; perform trajectory analysis to recognize when hostile actions are being initiated; and refer suggested actions to the end-user for final authorization.