Search Results

This paper provides several examples of silicon “micromachined” semiconductor sensors with which the authors are involved for aerospace condition monitoring. This and related work in MEMS (Micro Electro Mechanical Systems) has the potential to revolutionize condition monitoring in aerospace condition and “health monitoring” by (1) moving “smart” electronics out to the sensor chip itself and (2) combining a vast quantity and types of, not only electronic, but micromechanical sensing schemes into the silicon chip . Precisely formed cantilevers, gears, valves, microplumbing and even micro motors of the cross-section of a human hair can be fabricated on a single silicon microchip. Silicon is an excellent mechanical material with a yield strength several times that of stainless steel. Also silicon has excellent thermal properties , whereas compatible silicon dioxide (which we typically use in connection with silicon microelectronics patterning) is virtually a thermal insulator.

In this work, the goal of enhanced reliability through redundancy is explored. Two levels of fusion have been defined: the first is a fusion of sensors, redundant in both number and type, and the second is a statistical fusion of the resulting data at a software level. An intermediate preprocessing level is required to connect both fusions. The various types of sensors which are included are bulk micromachined flow, pressure and hydrogen sensors and a thin film poly-crystalline silicon temperature sensor. Individual sensors have been fabricated and packaged in arrays. Associated preprocessing has been designed to be able to handle all of the signals coming from each sensor and prepare them for statistical analysis. Data fusion algorithms have been written and tested.

The High Pressure Oxidizer Turbine (HPOT) discharge temperature of the Space Shuttle Main Engine (SSME) was estimated using Radial Basis Function Neural Networks (RBFNN) during the startup transient. Estimation was performed for both nominal engine operation and during simulated input sensor failures. The K-means clustering algorithm was used on the data to determine the location of the basis function centers. The performance of the RBFNN is compared with that of a feedforward neural network trained with the Quickprop learning algorithm.

Building and operating dependable systems is fundamental to many critical applications, such as designing integrated hardware and software systems for vehicles or satellites. Dependable systems techniques, methods, and tools are developed and used by researchers and practitioners working in widely varying disciplines. In order to provide a unifying framework for the successful dissemination and sharing of dependability results, the development of a dependable systems knowledge base is underway.1 Two database support subsystems are under development: one that manages the storage and retrieval of document information, as well as communicating between the user interface layer and the physical database layer, and another that manages the lexicon of dependability terminology for the user interface layer. The system will provide access to information in a sophisticated, intelligent manner that enables a human user to function more effectively in learning and decision-making capacities.

Looking forward to an autonomous Unmanned Combat Aerial Vehicle (UCAV) for future applications, it becomes apparent that on-board intelligent controllers will be necessary for these advanced systems. LETHA (Learning Enhanced Tactical Handling Algorithm) was created to develop intelligent managers for these advanced unmanned craft through the novel means of a genetic cascading fuzzy system. In this approach, a genetic algorithm creates rule bases and optimizes membership functions for multiple fuzzy logic systems, whose inputs and outputs feed into one another alongside crisp data. A simulation space referred to as HADES (Hoplological Autonomous Defend and Engage Simulation) was created in which LETHA can train the UCAVs intelligent controllers.