Applying the Extended Kalman Filter to Nonlinear Regression Models.

摘要

In using an extended Kalman filter to estimate the parameters of a nonlinear regression model, the order in which the measurements are processed can be important. The filter cannot always be expected to produce a satisfactory global fit when processing the measurements in the casual order in which they occur. Interestingly, if one can accept off-line processing, it may be possible to obtain such a fit when the measurements are processed in a random order.