Error Model-converted Measurement and Error Model-modified Extended Kalman Filters for Target Tracking

摘要

Two-filter schemes have been evaluated to handle the polar measurements using error model (for bias correction and measurement noise covariance computation) for target-tracking application. It is assumed that a good reference source of target information is available. Schemes based on error model converted measurement Kalman filter (ECMKF) and error model modified extended-Kalman filter (EMEKF) algorithms are presented. Also some comparison with CMKF (debiased) is given. It is inferred that EMEKF gives better performance compared to other filters. Features of CMKF (debiased), ECMKF, and EMEKF are highlighted. Also the sensitivity study on the performance of EMEKF is carried out wrt to processing order of radar measurement channels.