PPM: A conceptual framework for an AUV propulsion plant manager

The Propulsion Plant Manager (PPM) represents a conceptual design for the development of an expert system to manage the propulsion plant of an Autonomous Underwater Vehicle (AUV). Data from propulsion plant sensors are collected on a real-time basis and monitored by PPM. Data trends are analyzed and projections made. Abnormal propulsion plant conditions as represented by the data or by data trends are identified and form the basis for diagnosing and resolving anomalous conditions. The PPM program uses the techniques of Artificial Intelligence (AI) to represent and process the knowledge of an expert in the domain of AUV propulsion plants to accomplish the functions described above. The requirements of AUV design dictate the needs for continuous monitoring and frequent analysis of a multitude of real-time data, along with providing a rapid response to the diagnosed abnormalities, in order to ensure that the AUV can accomplish its mission in a reliable manner without human interaction. Many of the basic concepts developed in VM (Ventilation Manager) for representing time-related knowledge are explored and some are adapted to PPM. The VM program is designed to interpret on-line physiological data on patients in an intensive care unit and used to assist the physician in managing the mechanical breathing assistance received by patients. The unique demands of the AUV domain can be met by an extension of the concepts used in VM.