Multiple-time-scale high-gain observer and loop transfer recovery

A method to design a multiple-time-scale high-gain observer is presented. The design procedure consists of two strategies: system decoupling and high-gain feedback. A decoupled system is suitable to construct a multiple-time-scale system. The design method is very easy because each gain matrice is defined for a reduced-order subsystem. It is also shown that the control system with the high-gain observer attains loop transfer recovery.<<ETX>>