Model-Driven Deep Graph Learning for Risk-Aware Anomaly Response

Future deep space habits will require resilient and autonomous operations that have a holistic understanding of the systems contained within. The HOME team stated that a fundamentally disruptive approach is required to safely perform human-exploration driven by a holistic perspective on systems across a wide range of domains. Our approach addresses the critical area of balancing human and autonomous system intervention by leveraging the capabilities inherent in model-based systems engineering (MBSE) to understand the complexity of a fault or anomaly to the mission and the safety of the astronauts. By having a systems perspective on the FDIR, a more cohesive story can be presented to astronauts, including potentiality of cascading failures, recommended courses of action, time to failure predictions, and other human factors in risk analysis. To address the need for holistic understanding, we leverage an ensemble approach that combines MBSE, graphical models, probabilistic graphical models, semantic learning, and decision theory to provide anomaly response and fault action impacts. Our innovative approach disaggregates the problem into three components: 1. Model-based systems engineering and semantic understanding that gathers all the design information, model parameters, interrelations from the systems' models and expands their context with semantic understanding of the taxonomies and classifications of components. 2. Online Guided Deep Structure Learning provides the fault propagation modeling of the entire system that is guided by a priori information about the system from the system model but learns the real characteristics of the sensors and their interdependencies through physics-informed online learning. 3. Risk-aware Anomaly Response is the final module that makes decisions based upon actual relationships and an understanding of the impact a decision will have across the entire system, the mission, and any safety concerns.

Assuming we are successful in Phase I, the outcome of this program will be a demonstrated model-based systems engineering approach to in situ graphical structure learning and risk-informed decision making that addresses a need across almost every industry from seabed to space. We see potential benefits for any system that involves complexity where the system needs to quickly ascertain the source of a problem and take risk-tolerable actions to mitigate to ensure system resiliency and timely response. The main benefit will be the increased trust and resiliency in the autonomous systems, because it has mission- and human-centric understanding of the decisions it makes. In the NASA domain, this mission- and safety-centric approach – where inappropriate actions can have disastrous consequences that cannot be recovered from, repaired, or replaced because of the difficulty of distances, costs, and time – takes a holistic view to the entire system and the interrelation of decisions on the entire system, not just a subsystem or component. It not only understands what the designers were intending in their design but learns in situ further relations that might have been overlooked by engineers, often across multiple companies and agencies, to provide further assurances on the success of the mission and the safety of astronauts. This capability goes beyond just smart habitats, but for any autonomous system in space, manned or unmanned, from satellites to deep-space autonomous robot explorers. Assuming we are successful in Phase I, the outcome of this program will be a demonstrated model-based systems engineering approach to in situ graphical structure learning and risk-informed decision making that addresses a need across almost every industry from seabed to space. We see potential benefits for any system that involves complexity where the system needs to quickly ascertain the source of a problem and take risk-tolerable actions to mitigate to ensure system resiliency and timely response. The main benefit will be the increased trust and resiliency in the autonomous systems, because it has mission- and human-centric understanding of the decisions it makes. In the Non-NASA domain, this mission- and safety-centric approach is applicable to any system where an autonomous agent is controlling response actions to faults, anomalies, or more. Other agencies working in space, would benefit from the capability for the system to make risk-informed decisions to the mission of their system, including its own survivability (safety). It is also beneficial to systems that cannot afford the time for a human-autonomy component, and the system needs to react quicker than human analysis will allow. This is especially true for many defense systems that are deployed in the battlefield. A soldier will need to prioritize their own mission (looking down the barrel of a rifle), then spending time examining and selecting response actions. Therefore, the solider needs to know it can depend on the system responding appropriately to keep the mission going and soldiers alive.

This is early/mid-stage (TRL 2) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.