Performance of neural data associator

The paper presents the performance of neural data association based on a mean field Hopfield network. The authors create a new energy function for measurement data association (MDA) that consists of assigning radar plots to predicted track positions which plays a key role in all track-while-scan systems. The network presented in the paper in combination with the new energy function can minimise a global cost, which is a function of the distances between the plots in a given scan of data and the predicted track positions. The data association capacities of the neural network have been studied in different environments, and the results are presented. The authors also give the results of tracking trials based on neural data association