TY - GEN
T1 - A new type of fuzzy logic system for adaptive modelling and control
AU - Zhang, J.
AU - Knoll, A.
AU - Le, K. V.
N1 - Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 1997.
PY - 1997
Y1 - 1997
N2 - We present a new type of fuzzy controller constructed with the B-spline model and its applications in modelling and control. Unlike the other normalised parameterised set functions for defining fuzzy sets, B-spline basis functions do not necessarily span from membership value 0 to 1, but possess the property “partition of unity”. These B-spline basis functions are automatically determined after the input space is partitioned. By using “product” as fuzzy conjunction, “centroid” as defuzzification, “fuzzy singletons” for modelling output variables and adding marginal linguistic terms, fuzzy controllers can be constructed which have advantages like smoothness, automatic design and intuitive interpretation of controller parameters. Furthermore, both theoretical analysis and experimental results show the rapid convergence for tasks of data approximation and unsupervised learning with this type of fuzzy controller.
AB - We present a new type of fuzzy controller constructed with the B-spline model and its applications in modelling and control. Unlike the other normalised parameterised set functions for defining fuzzy sets, B-spline basis functions do not necessarily span from membership value 0 to 1, but possess the property “partition of unity”. These B-spline basis functions are automatically determined after the input space is partitioned. By using “product” as fuzzy conjunction, “centroid” as defuzzification, “fuzzy singletons” for modelling output variables and adding marginal linguistic terms, fuzzy controllers can be constructed which have advantages like smoothness, automatic design and intuitive interpretation of controller parameters. Furthermore, both theoretical analysis and experimental results show the rapid convergence for tasks of data approximation and unsupervised learning with this type of fuzzy controller.
UR - http://www.scopus.com/inward/record.url?scp=84947708487&partnerID=8YFLogxK
U2 - 10.1007/3-540-62868-1_129
DO - 10.1007/3-540-62868-1_129
M3 - Conference contribution
AN - SCOPUS:84947708487
SN - 3540628681
SN - 9783540628682
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 363
EP - 380
BT - Computational Intelligence
A2 - Reusch, Bernd
PB - Springer Verlag
T2 - 5th Fuzzy Days International Conference on Computational Intelligence, CI 1997
Y2 - 28 April 1997 through 30 April 1997
ER -