The recursive prediction error methods in state-space form have been efficiently used as parameter identifiers for linear systems, and especially Ljung's innovations filter using a Newton search direction has proved to be quite ideal. In this paper, the RPE method in state-space form is developed to the nonlinear case and extended to include the exact form of a nonlinearity, thus enabling structure preservation for certain classes of nonlinear systems. Both the discrete and the continuous-discrete versions of the algorithm in an innovations model are investigated, and a nonlinear simulation example shows a quite convincing performance of the filter as combined parameter and state estimator.
|Title of host publication||25th IEEE Conference on Decision and Control|
|Publication status||Published - 1986|
|Event||25th IEEE Conference on Decision and Control - Athens, Greece|
Duration: 10 Dec 1986 → 12 Dec 1986
Conference number: 25
|Conference||25th IEEE Conference on Decision and Control|
|Period||10/12/1986 → 12/12/1986|