U

Gateway-MAESTRO

Gateway Management through Adaptive, Autonomous, Fault Identification & Diagnosis, Reconfiguration/Replanning/Rescheduling Optimization (Gateway-MAESTRO)

Stottler Henke developed Gateway-MAESTRO for NASA. The Lunar Gateway is a lunar space station that will orbit the Moon in a highly elliptical Near-Rectilinear Halo Orbit (NRHO). The Gateway is envisioned as a multi-purpose spacecraft serving as a temporary habitat for astronauts, a science laboratory (for crewed and un-crewed experiments), a communications store/relay, a holding area for other spacecraft and devices, etc. The Gateway requires autonomous system and operations management, including the ability to follow commands (from ground stations and astronauts), fault/degradation detection and diagnosis; warnings, recovery; preventative/precautionary measures; and more. The Gateway includes many subsystems such as the Electrical Power System (EPS) and Communications. The EPS underlies much of the PPE, the HALO, and the Gateway as a whole. Stottler Henke developed Gateway-MAESTRO to manage the Gateway system and operations.

During normal operations, Gateway-MAESTRO will monitor onboard sensor values in order to automatically characterize Gateway components and detect failures. Based on that characterization, the system automatically predicts resource availability over time and automatically schedules the actions (i.e., determine what activities will occur and when, along with the modes of the associated equipment). Gateway-MAESTRO hybridizes model-based and ML methods emphasizing the benefits of each, while minimizing their disadvantages. This hybrid approach also uses the Stottler Henke developed Thermodynamic Reasoning Intelligent Anomaly Detection (TRIAD) algorithm which learns to classify nominal, known faults, and unexpected off-nominal situations based on telemetry data. 

During a failure scenario, Gateway-MAESTRO operations are as follows:

  • Detect the problem
  • Immediately safe the spacecraft to minimize damage
  • Diagnose the problem and determine the root cause
  • Determine potential feasible courses of action (COAs) given the failed components or set of possible failed components
  • Determine the impact and ramifications of each COA
  • Select the most appropriate COA
  • Generate the detailed schedule/sequence of actions to implement the COA
  • Adaptively execute the required actions

We tested Gateway-MAESTRO on a hardware simulation of the Gateway system referred to as a LabSat. During this testing phase, we validated that Gateway-MAESTRO could detect a wide array of organic and intentionally triggered faults in a real world system. In fact, Gateway-MAESTRO was able to detect that faults had occurred after a person unexpectedly bumped into the table with the LabSat on it, despite there being no prior knowledge that a fault like this was a possibility.