Improving the Estimation of the Tracking Error Signal in Real Time for IR Seeker Using Nonlinear Kalman Filters

摘要

In nonimaging IR seekers, the received target radiation on the IR detector is modulated via a reticle and produces the information signal (IS). The IS contains the tracking error signal (TES), which is proportional with the target position. TES is used in the control and optic section in the missile. The main task is to extract the TES from IS. The accuracy of TES extraction may be affected by several items such as noise from engines. In this paper, we used for the first time this field Square Root Unscented Kalman Filter (SRUKF) and Extended Kalman Filter (EKF) to estimate the TES from the IS for wagon wheel reticle.Due to the high computational complexity of these algorithms, their execution in real time is not an easy task especially if there is space limitation for hardware. By using the minicomputer, such as Raspberry Pi 3 Model B+ platform, the task can be done.The results showed that the SRUKF presented the best phase estimation for TES. The implementation by using Raspberry Pi was in real time because all algorithm executions for one period was less than 5ms, this time in our problem is less than strict timing window.