Performance evaluation of multi sensor data fusion techniques in tracking and thermal systems by modeling and simulation

Abstract

The performance evaluation of parameter estimation algorithms for target tracking and thermal systems is the main issue
of this article. A single measurement data from a system may not be sufficient to estimate the parameter accurately.
Therefore multiple sensor data observed from the systems are fused to improve the parameter estimation by redundant
and additional data available. In this paper the most preferred state estimation method known as Kalman filter is applied
for the four test systems by fusing two sensors using three algorithms namely Measurement fusion (MF), State vector
fusion (SVF) and Gain fusion (GF). The main purpose of this work is to implement the multi sensor data fusion
algorithms in target tracking systems and thermal systems by using system models with simulated data and to evaluate all
of their performances by calculating the various estimation errors namely Percentage Fit Error (PFE), Mean Absolute
Error (MAE) and Mean Square Error (MSE) using MATLAB.