A Data Fusion Optimization Method Based on GA and PSO

In this paper state fusion and measurement fusion are introduced as the two fields of data fusion. As a more optimal choice, measurement fusion has been focused within a couple of usual methods based on augmented observation vector and weighted sum of observations and a third recently proposed method based on the modified form of Kalman filter. The main contribution of this paper was based on the optimization of the third method, using GA and PSO. Meanwhile, the sensor gains were defined as variables to be optimized. The results showed a great improvement in terms of leading to smaller error covariance matrix values which are second to none in all recent researches.