Neural Network Data Fusion for Maneuvering Targets Tracking

A neural network data fusion approach for radar/infrared dual mode guidance was presented.By taking a neural network consisting of one neuron for each component of the measurement vector as the fusion center, the infrared measurements were fused first to keep synchronous with the radar measurements processed by Kalman filter. The processed data were then transmitted to the central neural network where a fused estimation of target was formed. The goal of this pager was to propose a method for computing the fused estimation in the absence of covariance information. Simulation result was demonstrated the effectiveness of this method.