Brdys and Kulawski reported the use of recently developed adaptive control
techniques based on neural networks to the induction
motor control. Although the stability of this control strategy
can be guaranteed, the complicated network structure
and strict constraints are required in the control process. In
previous work [9, lo], an intelligent uncertainty observer
was designed to estimate the bound of lumped uncertainty.
The disadvantages of these structures are the complex
network structure and inference mechanism. The aim of
this study is to design a control scheme that possesses the
learning ability of the intelligent control and the simple
control structure comparable with the traditional PID
controller.