An extended Kalman filter using self-organizing map neural network

This paper proposed a Kalman filter using self-organizing map neural network for the filtering problem of nonlinear systems. The system is approximated by the multiple models using self-organizing map neural network and the resulting model is subject to Kalman filter. The method has no such difficulties as classical extended Kalman filter may encounter, and compared with other nonlinear filtering methods, the on-line computation consumption is reduced. Some features of the method are discussed and an example is given to show the application of the method to the nonlinear system filtering problem.