"X-2000 Anomaly Detection Language" denotes a developmental computing language, and the software that establishes and utilizes the language, for fusing two diagnostic computer programs, one implementing a numerical analysis method, the other implementing a symbolic analysis method into a unified eventbased decision analysis software system for real-time detection of events (e.g., failures) in a spacecraft, aircraft, or other complex engineering system. The numerical analysis method is performed by beacon- based exception analysis for multimissions (BEAMs), which has been discussed in several previous NASA Tech Briefs articles. The symbolic analysis method is, more specifically, an artificial-intelligence method of the knowledge-based, inference engine type, and its implementation is exemplified by the Spacecraft Health Inference Engine (SHINE) software. The goal in developing the capability to fuse numerical and symbolic diagnostic components is to increase the depth of analysis beyond that previously attainable, thereby increasing the degree of confidence in the computed results. In practical terms, the sought improvement is to enable detection of all or most events, with no or few false alarms.
This program was written by Mark James of Caltech for NASA's Jet Propulsion Laboratory. For further information, access the Technical Support Package (TSP) free online at www.techbriefs.com/tsp under the Software category.
This software is available for commercial licensing. Please contact Karina Edmonds of the California Institute of Technology at (626) 395-2322. Refer to NPO-42512.
This Brief includes a Technical Support Package (TSP).
Fusing Symbolic and Numerical Diagnostic Computations
(reference NPO-42512) is currently available for download from the TSP library.
Don't have an account?
Overview
The document from NASA's Jet Propulsion Laboratory (JPL) outlines advancements in diagnostic technologies aimed at enhancing the monitoring, analysis, and diagnosis of flight systems. It focuses on the fusion of symbolic and numerical diagnostic computations, specifically through the integration of tools like BEAM (a state estimator/predictor) and SHINE (a real-time expert reasoner). This innovative approach allows for real-time analysis of complex systems, achieving a level of depth and confidence previously unattainable.
The key feature of this technology is its ability to provide near-zero false alarms while ensuring 100% detection of anomalies. By simultaneously fusing sensor data, software results, and commands, the system can abstract system physics and identify information invariants, making it highly sensitive to system degradation and changes. This capability allows for the isolation of changes in both time and space, pinpointing specific sensors that may be affected.
The document emphasizes the scalability of these techniques, which can be applied to systems with sensor networks ranging from 10 to 10,000 sensors. The successful application of these technologies has been demonstrated in various JPL missions, including Voyager II, Galileo, Magellan, Cassini, and the Extreme Ultraviolet Explorer (EUVE). The technologies have proven to outperform traditional diagnostic methods and human operators, particularly in responding to unknown fault modes and providing prognostic capabilities.
The integration of these advanced diagnostic tools is particularly significant for the development of reusable launch vehicles (RLVs), as it promises to reduce mission costs, enhance safety, and decrease launch turnaround times. The document serves as a technical support package under NASA's Commercial Technology Program, aiming to disseminate aerospace-related developments with broader technological, scientific, or commercial applications.
For further inquiries or assistance regarding this technology, the document provides contact information for the Innovative Technology Assets Management at JPL. Overall, the fusion of symbolic and numerical diagnostics represents a significant leap forward in aerospace technology, with the potential to transform how flight systems are monitored and maintained.
