Marine Radar Moving Target Tracking Technology

Target tracking is very important in radar data processing.The research of target tracking has been paid great attention,especially the maneuvering target tracking.Theα-βfilter,α-β-γfilter and Kalman filter are commonly used in the area of radar, navigation and automation.This thesis studies the tracking technique of radar targets and the performance of various filters in the tracking process.The purpose and method of simulating moving targets are introduced first.In the research,simulating the accurate and reliable target movement status and real-time data information of targets are very important for the radar target tracking test.Secondly, detailed analysis of the basic principles of target tracking,including target motion model setup and the basic methods of tracking filtering is given,emphasizing in variable gainα-βfilter,α-β-γfilter,constant gainα-βfilter and constant gainα-β-γfilter.Because that the inertia of the ship is very large,the ship's navigation process can be regarded as constant speed linear motion during a certain period of time.In this case theα-βfiltering algorithm can be used to track the targets.But for a long period of time,change of course and speed of the ship is avoidless.So theα-βfilter has its limitations.Theα-β-γfiltering algorithm introduces the acceleration filter coefficient,which can detect the maneuvering of the ship and track the targets in constant acceleration movement.The variable gain filtering algorithm can change the filter coefficients with time in order to get the best performance,but the process of calculating the filter coefficients is complicated.So the use of constant gain filtering algorithm with fixed filter coefficients reduces the complexity of the computation.Based on the analysis of filters mentioned above,we researched and designed several radar moving target tracking algorithms.The computer software of tracking algorithm is developed with Visual C++.The performances of the tracking algorithms are tested with moving targets generated by the moving target simulation program. Experimental results are given and comparisons between different tracking algorithms are also made.