Research On Target Tracking Of Radar/Ir Multi-Guidance Using Information Fusion

As one of the key technologies for multi-guidance, information fusion has a great significance to increase target detection, identity and tracking ability of the guidance system, as well as the survival ability of the system. The target tracking methods based on information fusion of Radar/Infra-red (IR) multi-guidance are studied in this thesis.Target tracking plays an important role in military area, the system can implement effective beat only if it gets credible and accurate tracking information. For different levels of target maneuvering, corresponding motion models and filtering algorithms are studied. When the targets maneuver slightly, we can get good result using Extended Kalman Filter (EKF). For large maneuvering targets, we can adopt the variable dimension (VD) filter or Interacting Multiple Model (IMM). The numeral simulation results indicate that the IMM can get better tracking performance.Distributed fusion of Radar/IR is studied in the thesis. Tracking performance is bad to anti-jamming and unstable in Cartesian coordinate using EKF because of the bearing-only problem existing in the IR sensor. A distributed fusion model of Radar/IR with partly feedback is presented, as well as the filer and track fusion algorithms. The simulation results indicate that the precision of the local fusion is better using the model proposed, at the same time; the anti-jamming ability of the system is strengthened compare to the model no feedback.A new distributed fusion method of Radar/IR based on separation and combination of the measurements is proposed after analyzing the influence of rate measurement. The rate information separated from the radar measurements composes a fake vector of IR. It really makes the information from different sensors complementary. The results indicate that the method not only dose a great improvement to local tracker's performance but also improves the global tracking precision.