Unscented Kalman Filter and Its Implementation in Digital Image Correlation

Abstract

With the development of industrial technology and the improvement of automation level of industrial production, automatic recognition and prediction has gradually become another key issue in automated production. The main purpose of this paper is to elucidate the many advantages of the unscented Kalman filter (UKF) method and to demonstrate its full applicability to many identification and prediction processes in order to explore more possibilities for prediction and state estimation. This study focuses on the new application of unscented Kalman filter in recognition and prediction, explains and discusses its principles and functions, and explains its advantages and functions with an example of digital image correlation method. In the fourth part, an example of the application of unscented Kalman filter in digital image correlation is given, and its specific principle and advantages are explained. The function of unscented Kalman filter in digital image correlation is demonstrated to speed up the operation and avoid errors. Finally, it is concluded that the UKF can be well suited for various prediction and state estimation tasks.