Nasas integrated vehicle management project is developing technologies for detecting, diagnosing, predicting, and mitigating adverse events during the flight of an aircraft, regardless of the…

Nasas integrated vehicle management project is developing technologies for detecting, diagnosing, predicting, and mitigating adverse events during the flight of an aircraft, regardless of the subsystem(s) from which the adverse event arises. To support this capability IT is essential that a multidisciplinary vehicle-level reasoning system (vlrs) be established for integrating large heterogeneous (i.e., continuous and discrete signal) asynchronous data streams from multiple subsystems in order to detect a potential adverse event, diagnose its cause, predict the effect of that event on the remaining useful life of the vehicle, and then take appropriate steps to mitigate the event as required. The vlrs may take advantage of component, subsystem, and vehicle level models which would represent connectivity and potential causal chains of failure. Similarly, physics-based models of damage propagation for certain subsystems (such as the airframe or actuators) may be appropriate for inclusion in the vlrs. Vlrs technologies will produce a significant amount of data that can be used to enable conditioned-based maintenance (CBM) for aircraft. CBM continues to be one of the key motivations for the development of health management technology in a wide variety of aeronautical applications, and includes tools and technologies that enable automated, need-based maintenance (as opposed to schedule-based maintenance) using data and information derived from in-situ sensors and external test and measurement equipment and maintenance logs. With CBM, maintenance is only to be performed when there is an anticipated need. This order is the result of proposal 09-IVHM1-09-0001 submitted by honeywell international INC. For award under the NASA headquarters aeronautics research mission directorate (armd) NASA research announcement (NRA) NNH09ZEA001N2 entitled research opportunities in aeronautics 2009, amendment b.4, project integrated vehicle health management (ivhm), topic vehicle-level reasoning system / evaluation of multidisciplinary ivhm technologies, tools, and techniques. This defines a 3-year effort leading to the design, development, and demonstration of a vlrs for protecting systems and components against failures and malfunctions. Year one is the base period, years two and three are option years and may be exercised based on NASA s down selection review of year 1 technical progress and on the availability of funds. The objective of the proposed vehicle-level reasoner is to achieve NASA s ivhm goal for detecting, diagnosing, and predicting adverse events during the flight of an aircraft. The vipr effort advances the state-of-the-art aircraft-level reasoner (honeywell s aircraft diagnostics and monitoring system, adms) by addressing the following challenges identified by NASA s ivhm team. Detecting adverse events caused by fast progression and intermittent faults arising from the propulsion and aircraft actuator subsystems that currently cannot be detected using built-in test results. Such detection will be made possible through an open interface that allows vipr to take advantage of any enriched diagnostic and prognostic monitors in the components and subsystems. Detecting adverse events caused by cascading faults and fast progression faults arising from avionics and software subsystems that cannot be detected by passive aggregation of available data. This goal is achieved by incorporating intelligent data collection functionality that allows vipr to take an active role in the diagnostic process by querying the appropriate subsystem. Fusing a wide variety of heterogeneous, partial, and uncertain conclusions arising from multiple subsystems by managing the propagation of uncertainty through the reasoning process. This fusion is achieved by incorporating probabilistic elements within honeywell s b-777 diagnostic reasoner algorithm. Understanding the trade space between user requirements for cost, complexity, and safety IMP