An adaptive neuro-fuzzy architecture for intelligent control of a servo system and its experimental evaluation

Comparison of self-tuned Neuro-Fuzzy controllers on 2 DOF helicopter: an application

“…The NFC inputs are as follows; error, error derivative, and sum of errors. As stated in [19], the NFC response is faster comparing to classical methods but the result is relatively wavy. PI, Fuzzy tuned PID, Fuzzy tuned PI-PD and adaptive NFC, having two inputs which are error and its derivative, are compared on a DC motor [20].…”

“…In this method, the steady-state error is zero but noise is greater compared to classical methods. In 2010, a new adaptive NFC is proposed, having three inputs, which is implemented into a servo system [19]. The NFC inputs are as follows; error, error derivative, and sum of errors.…”

“…Within the scope of this study, the NFC controllers proposed in the open literature have been tested and tried to reach the best NFC controller combination. So, the 6 different NFC structures are tested as: (normalized error, output derivative [18]), (error, error derivative, summation of error [19]), (error, output derivative, summation of error), (error, output derivative, error integral), (normalized error, error derivative, summation of error), (normalized error, output derivative, error integral). As a result of these tests, (error, output derivative, the summation of error) inputs structure NFC is found to be better than other NFC combinations.…”

“…As a result of these tests, (error, output derivative, the summation of error) inputs structure NFC is found to be better than other NFC combinations. The proposed NFC input combination is tested on the nonlinear 2 Degree of Freedom (DOF) helicopter system and compared to other NFC combinations those given in open literature as (normalized error, output derivative [18]), (error, error derivative, the summation of error [19]). So, the performance of the proposed controller is tested for a nonlinear dynamical system.…”

Neural – Adaptive Control Based on Feedback Linearization for Electro Hydraulic Servo System

“…An intelligent CMAC, FNN neural controller that uses a feedback error learning approach appears in [4,5] which is highly complicated and [3] explain ways based on feedback linearization and back stepping However, it is not easy , nor is it simple to design such controllers. Other control methods will appear in [6][7][8][9][10][11][12][13].…”