Home
The Generic Software Architecture for Prognostics (GSAP) is a framework for applying prognostics. It makes applying prognostics more efficient by implementing many of the common elements across prognostic applications. The standard interface of the GSAP framework enables adaptability of prognostic algorithms and models to systems of interest.
The GSAP framework is used through the creation of communicators, prognosers, or models (the deployment layer). The elements of the deployment layer plug into the framework and use the tools of the support layer.
-
Models: Models are utilized to represent the behavior of a component. Prognostics is commonly performed by using a model that describes both the healthy and damaged behavior of the components. The ModelBasedPrognoser class then uses these models to perform prognostics.
-
Prognosers: Prognosers, the core of the GSAP system, contain the fundamental logic for performing prognostics. Prognosers take in live sensor data, configuration and future loading parameters, and output state estimation and prognostic results.
-
Communicators: Communicators are used to communicate data with the outside world. These function as interfaces with various data sources and sinks by receiving data which will be used by prognosers or communicate the results with operators. Some examples could be a playback agent that reads from a file, a GUI for displaying prognostic results, an automated report generator, or a client that connects into a network messaging system.
Each of these components is configured through the use of configuration files. This allows for a GSAP deployment to be configured to a new configuration or system without any software changes.
GSAP was created by the NASA Ames Research Center Diagnostics and Prognostics Group.
This wiki aims to provide complete documentation and examples of how to implement GSAP. The ultimate goal of GSAP is to create a consistent and compatible framework that the user can easily incorporate into their desired system to perform Prognostics.