An Application Research of Unbiased Converted Measurement Kalman Filter

摘要

The unbiased converted measurement Kalman filter (UCMKF) keeps the advantage of Kalman filter and has been widely used for target tracking. In the practical application, it is discovered that when the UCMKF is directly used in the east-north-up coordinates, the filter performance degrades or even diverges in some motion models or scenes with moving radars. In this paper, the cause of the degradation of filter performance in that situation is analyzed, and the application of unbiased converted measurement Kalman filter (UCMKF) algorithm is extended to Earth Centered Earth Fixed (ECEF) coordinate to filter random errors. Theoretical analyses and simulation results show that the extended algorithm is of better compatibility for moving radars, and owns a good filtering performance when the radar is moving with high-velocity.