Target tracking with implementation problems

In this thesis, the problem of target tracking has been investigated by means of simulation for various scenarios. In these scenarios, the motion data for the targets was generated realistically by simulation. While the measurement data is synthesized from these data, it has been taken into consideration that the actual target tracking applications can cause problems, and the data has been added. This thesis focuses on missed measurements, clutter measurements, closely spaced targets, model mismatch, multiple target tracking and track management issues. The purpose of this thesis is to study the classical target tracking methods in realistic scenarios and develop target tracking systems that solve the problems in different scenarios. The Kalman filter and the Extended Kalman filter were used for estimation purposes. Probabilistic and joint probabilistic data association methods, standard and global nearest neighbor data association methods are used for data association. For the management of traces, track management approach known as M / N logic was used. In the thesis, target tracking applications for all the scenarios have been successfully developed. Errors for different methods are calculated arithmetically and geometrically. The methods were evaluated for their success and complexity, and the results were presented comparatively.