Determining manoeuvre detection threshold of GRNN filter in the process of tracking in marine navigational radars

The article presents the results of simulation research aimed at determining the manoeuvre detection threshold of a neuron filter, based on GRNN artificial neural network. The structure of a filter has been presented, from which the necessity of manoeuvre detection results and a detector has been described able to be used in this case; it is based on an analysis of course and speed increase in particular estimation steps. The research was performed by the Monte Carlo method for 100 independent simulations.