STELLAR SCIENCE LTD. CO. — Department of Defense SBIR Phase I: AF221-0027

Description

Stellar Science proposes a distributed multi-target manager as a proliferated low-Earth orbit (pLEO) expert system for autonomous target-track management, called Pythia. Each manager in the Pythia distributed architecture is comprised of a multi-target track manager as well as a sensor task manager, and each manager continually collaborates with neighboring managers in a distributed manner to converge on global track-state estimates and task allocation consensus. The proposed distributed estimation algorithm, the sample greedy gossip information consensus filter (SGG-ICF), synchronizes tracks between multiple track managers in a distributed architecture while lowering communication burdens. This distributed estimation algorithm is extended to multi-target tracking via data association methods, such as joint probabilistic data association (JPDA). A task manager utilizes the set of synchronized tracks to hypothesize the information value gained for potential future sensor tasking based on information theoretic metrics on moving target-track distributions. These metrics are shared with neighboring task managers to perform distributed task allocation through a bidding process, specifically the consensus-based bundle algorithm (CBBA). The resulting consensus tasks represent an allocation that maximizes collaborative information value. Both the distributed estimation and distributed task allocation processes are only performed in emergent, temporary, and spatially confined local neighborhoods in the constellation with current and near-future field-of-regard (FOR) overlap among multiple sensing platforms. In this way, the computational requirements and communication burdens of these distributed processes are reduced. Stellar Science proposes to integrate the Pythia expert system for pLEO autonomous target track management into an existing government-owned modeling, simulation, and analysis (MS&A) tool, the Advanced Framework for Simulation, Integration, and Modeling (AFSIM). The MS&A framework will enable users to test expert systems for autonomous target track management against scriptable, multi-domain, physics-based scenarios. In this way, the integrated expert system may represent a digital twin of any proposed sensing constellation, ideal for expert system verification and validation (V&V) and for trade-space analysis of planned architectures as well as inclusion in tabletop exercises, simulated wargames, or other operational-like environments. Furthermore, a realistic implementation of system constraints (e.g., communication, attitude management, processing latencies, etc.) will enable an analytical analysis of the challenges and impacts of deploying the Pythia expert system on-orbit in low-to-medium size, weight, and power (SWaP) satellites.