Data Fusion Method Based on Adaptive Kalman Filtering

Abstract

This paper discusses data fusion methods to combine the data from a rotary encoder and ultrasonic sensor. Both sensors are used in a micro-flow calibration system developed by the Research Center of Metrology LIPI. The methods studied are hierarchical data fusion and Kalman filtering. Three types of Kalman filters (KFs) are compared: the conventional Kalman filter and two adaptive Kalman filters. Moreover, a method to combine the uncertainty results from KF in hierarchical data fusion is proposed. The aim of this study is to find appropriate methods of data fusion that can be implemented in microflow calibration systems. Data from two experiment setups are used to compare the methods. The result indicates that one of the methods (with little adjustment) is more appropriate than the other.