Smart robust interpolator

Abstract

A proposed smart robust interpolator that based on robust trained neural networks is presented and compared with other popular interpolation methods widely implemented in mathematical, industrial and manufacturing applications. Recently many interpolation methods have been developed, and examined. Most of them are based on looking for the optimal interpolation trajectories based on the well known data set. However, it is rare to build robust interpolator based on noisy data, and this is one of the most popular topics in industrial testing and measurement applications. The smart robust neural network (SRNN) interpolator reported in this paper provides a convenient and simple way to solve this problem and offers more accurate interpolation results based on given data set in the presence of outliers. This method can be implemented in many applications, such as manipulators measurements and calibrations, automations, unmanned air vehicles, Upward Velocity of Rockets, and semiconductor manufacturing processes.