top
This paper presents a new learning algorithm for the design of Mamdani- type or fully-linguistic fuzzy controllers based on available input-output data. It relies on the use of a previously introduced parametrized defuzzification strategy. The learning scheme is supported by an investigated property of the defuzzification method. In addition, the algorithm is tested by considering a typical non-linear function that has been adopted in a number of published research articles. The test stresses on data-fitting, function shape representation, noise insensitivity and generalization capability. The results are compared with those obtained using neuro-fuzzy and other fuzzy system design approaches.
Saade, Jean Jamil. "A defuzzification based new algorithm for the design of Mamdani-type fuzzy controllers." Mathware and Soft Computing 7.2-3 (2000): 159-173. <http://eudml.org/doc/39196>.
@article{Saade2000, abstract = {This paper presents a new learning algorithm for the design of Mamdani- type or fully-linguistic fuzzy controllers based on available input-output data. It relies on the use of a previously introduced parametrized defuzzification strategy. The learning scheme is supported by an investigated property of the defuzzification method. In addition, the algorithm is tested by considering a typical non-linear function that has been adopted in a number of published research articles. The test stresses on data-fitting, function shape representation, noise insensitivity and generalization capability. The results are compared with those obtained using neuro-fuzzy and other fuzzy system design approaches.}, author = {Saade, Jean Jamil}, journal = {Mathware and Soft Computing}, keywords = {Controladores difusos; Conjuntos difusos; Algoritmos de aprendizaje; learning algorithm; fuzzy controllers}, language = {eng}, number = {2-3}, pages = {159-173}, title = {A defuzzification based new algorithm for the design of Mamdani-type fuzzy controllers}, url = {http://eudml.org/doc/39196}, volume = {7}, year = {2000}, }
TY - JOUR AU - Saade, Jean Jamil TI - A defuzzification based new algorithm for the design of Mamdani-type fuzzy controllers JO - Mathware and Soft Computing PY - 2000 VL - 7 IS - 2-3 SP - 159 EP - 173 AB - This paper presents a new learning algorithm for the design of Mamdani- type or fully-linguistic fuzzy controllers based on available input-output data. It relies on the use of a previously introduced parametrized defuzzification strategy. The learning scheme is supported by an investigated property of the defuzzification method. In addition, the algorithm is tested by considering a typical non-linear function that has been adopted in a number of published research articles. The test stresses on data-fitting, function shape representation, noise insensitivity and generalization capability. The results are compared with those obtained using neuro-fuzzy and other fuzzy system design approaches. LA - eng KW - Controladores difusos; Conjuntos difusos; Algoritmos de aprendizaje; learning algorithm; fuzzy controllers UR - http://eudml.org/doc/39196 ER -