This paper describes a new method of sensor failure detection, isolation, and accommodation using a neural network approach. In a propulsion system such as the Space Shuttle Main Engine, the dynamics are usually very complicated and sometimes not well known. However, the number of variables measured is usually much higher than the order of the system. This built-in redundancy of the sensors can be utilized to detect and correct sensor failure problems.

No abstract

Software for categorizing a cloud of more than 300,000 three-dimensional (3-D) surface data points, captured from a human subject, is presented. The software is part of an incremental approach that progressively refines the identification of human anthropometric landmarks. The first phase of identification is to orient and segment the human body data points. A step-by-step method for these tasks is presented. One of the algorithms, a discrete point cusp detector, plays a fundamental role in separating the data cloud. The theory and operation of the algorithm is explained. The discrete cusp detector can be used to separate any two cylindrical objects, in (3-D) data, whose boundaries touch. The software has been tested on over a hundred different body scan data sets and shown to be robust.

No abstract