Evolutionary programming Kalman filter

From MaRDI portal





The authors consider discrete-time linear and nonlinear systems with uncertain parameters. The uncertainty is specified by restricting the parameters to intervals rather than single values. The statistical assumptions remain the same as in the classical Kalman filter approach. Optimal estimates of the state vector and the error covariance matrix are sought through an evolutionary programming algorithm. Both interval and nominal estimates are provided by the algorithm. Simulations of typical linear and non-linear systems are reported. They suggest that the new method is more accurate and less conservative than the existing interval Kalman filtering.





Describes a project that uses

Uses Software






This page was built for publication: Evolutionary programming Kalman filter

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5946287)