Multisensor Data Fusion Fuzzy Similarity-based of Several Kalman Filters

Abstract:
The main objective of this work is the development of an intelligent multisensor
integration and fusion model that uses fuzzy similarity-based data fusion of several
Kalman filters outputs. First, the estimation of sensors outputs are calculated using a set
of Kalman filters with pre-estimated measurement noise. Using fuzzy set theory, the
fuzzy similarity between the predicted data is extracted to determine the importance
weight of each sensor. Weights assigned to different sensors measurement data to reflect
the confidence in the sensor's behavior and performance and to realize the multi-sensor
data fusion. According to the algorithm theory, its application software is developed
using MATLAB. This work has wide applications especially in the development of
radar target tracking, smart structural health monitoring systems, biomedical imaging,
and robotics control. The applied example proves that the algorithm can give priority to
the high-reliability and stability sensors. Moreover, it reflects the efficiency and
feasibility to real-time data processing and monitoring.