Tracking a swift target using the interacting fuzzy multi-modelalgorithm

摘要

This paper focuses on the generation of an intelligent trackermodule equipped with a wavelet based neural network that learnspredictions from past experience. The perception of actual targetmanoeuvre and prediction of its future states are achieved in this workby “projecting” actual observations into decision spaces oflocal fuzzy predictions based on independent prototypical trajectorytypes: linear, parabolic and sinusoidal. Decentralized trackingdecisions are thus generated which are further evaluated by a learningprediction module and are fused before being sent to the guidance module