Tracking filter design is discussed. It is argued that the basis of the present stochastic paradigm is questionable. White process\r\nnoise is not adequate as a model for target manoeuvring, stochastic least-square optimality is not relevant or required in practice,\r\nthe fact that requirements are necessary for design is ignored, and root mean square (RMS) errors are insufficient as performance\r\nmeasure. It is argued that there is no process noise and that the covariance of the assumed process noise contains the design\r\nparameters. Focus is on the basic tracking filter, the Kalman filter, which is convenient for clarity and simplicity, but the arguments\r\nand conclusions are relevant in general. For design the possibility of an observer transfer function approach is pointed out. The\r\nissues can also be considered as a consequence of the fact that there is a difference between estimation and design. The a-�Ÿ filter is\r\nused for illustration.
Loading....