Training neuroemulators using Multicriteria Extended Kalman Filter and pseudoregularization for Model Reference Adaptive Neurocontrol

Problems of identification and control of nonlinear dynamic plants using Model Reference Adaptive Neurocontrol are considered. Pseudoregularization problem for effective training of neuroemulators is stated, Multicriteria Extended Kalman Filter method for solving this problem is proposed. Results of numerical experiments of training neuroemulators and neurocontrollers are presented.