XANALYTIX SYSTEMS LLC — Department of Defense STTR Phase I: AF22A-T003

Description

The main objective of the solicitation is to assess the viability of a system to perform multi-perspective information fusion in an autonomous and distributed satellite architecture.  To address this objective, CARDS (Collaborative Autonomy and Resiliency for Distributed Satellites) will decompose the problem in answering three critical questions 1) What is the potential Information Gain (IG) of an AOI when fusing heterogeneous sensors from different agents (satellites)?  Given space and time, what is the value of assigning resources to that area at particular periods.  Answer:  Maximizing the effective utilization of a set of satellites in an ISR mission requires knowing when and where the information collected will provide the highest value for mission success.  Our approach allows for satellites with different levels of computational power.  The proposed approach satisfies this requirement in that it can be used both with intelligence-on-the-edge and legacy systems with limited computational capabilities and in a delayed/disconnected, intermittently-connected, low-bandwidth environment. 2) What is the suitability of a particular satellite or group of satellites (force composition) to address a specific potential IG over space and time?  Answer: Composing optimal forces for military tasking is a non-trivial problem when accounting for variables such as situational awareness, unit weapon/sensor capabilities, tasking priorities, and adversary behavior.  To facilitate decision-making in this sphere, we need to find the suitability of a particular asset to perform a task.  In space, the problem becomes more difficult since the orbit of satellites is predetermined and very costly (we will assume virtually impossible) to change.  This objective will investigate a scalable tree structure to determine the fitness of space objects to perform tasks over space and time 3) What is the scheduling assignment of tasks for satellites to maximize the potential IG over multiple missions in multiple AOIs based on required time windows and priorities?  Answer: CARDS will develop a multi-perspective approach based on a decentralized or hierarchical framework in which each satellite will have a “perspective” of what tasks other satellites might be performing based on the last synchronization and plausible futures.  To deal with the heterogeneous computational power, multiple algorithm types will work together to coordinate and collaborate when communications are possible.  The approach allows a commander to have multiple criteria to achieve objectives in both centralized and decentralized environments.  When the decision-maker is disconnected from coordination with other owners, they will utilize current situational awareness to maintain insight into what the assets that are not under their control have been assigned to do.  When coordination is reestablished, they will re-align their perspective of the mission to match the ground truth.