Detection of Manoeuvre for the Needs of Estimating the Target's State Vector in the Process of Radar Tracking

摘要

The article presents two concepts of estimating the target's state vector in the process of radar tracking, by means of the neural filter based on GRNN network and the results of initial research on the method of manoeuvre detection for the needs of such concepts. Both concepts assume using networks of different parameters for a tracked target proceeding with established movement and different for tracking a target performing manoeuvres. This is becoming an essential problem - a method of detecting a manoeuvre, as information about the moment of changing network parameters.