A fusion method that performs better than best sensor
Nageswara S. V. Rao · University of North Texas Digital Library (University of North Texas) · 1998
In a multiple sensor system, sensor S i , i = 1; 2 : : : ; N , outputs Y (i) 2 [0; 1], according to an unknown probability distribution P Y (i) jX , in response to input X 2 [0; 1]. We choose a fuser, that combines the outputs of sensors, from a function class F = ff : [0; 1] N 7! [0; 1]g by minimizing empirical error based on a sample. If F satisfies a simple isolation property, we show that the fuser performs at least as well as the best sensor in a probably approximately correct sense. Several well-known fusers such as linear combinations, special potential functions, and certain feedforward piecewise-linear networks satisfy the isolation property. Keywords: Sensor fusion, fusion rule estimation, empirical estimation 1 Introduction In multiple sensor systems, it is generally known that a good fuser outperforms the best sensor, and on the other hand, an inappropriate fuser can perform worse than the worst sensor. If the error distributions of the sensors are precisely known, a...