Abstract—Self-Organizing controller is a very good tool for
providing a system with adaptive abilities. However there are
some problems that are drawbacks of this controller. Firstly,
although the process model needed for an SOC is an
incremental one, it is still difficult to find and calculate such
models in practice, and furthermore if an SOC is to be used,
there is always a need for the knowledge of time lags in the
process under control. The second problem is that in an SOC
the process of updating the rules, assumes that there is only one
sample in the past that is responsible for the current poor
performance, while in fact more than one sample may be
responsible and require correction. And the third problem is
that if a fuzzy controller is used in the lower level of an SOC, the
process of updating control rules would be a very demanding
task that ends up in replacement of each rule with three new
rules for each sample. So in addition to the need for a garbage
collection algorithm to eliminate the redundant rules, the task
of correction would be a time consuming process. To overcome
these problems we introduce SOANFIC that is a model free
controller and solves these problems. In addition to this
SOANFIC is a very easily applicable controller that can be
applied to MIMO systems as well as SISO ones.
Index Terms—ANFIS, fuzzy controller, intelligent controller,
SOC.
H. R. Saghir and S. B. Shouraki are with the Mechanical Engineering
Department, Sharif University of Technology, Tehran, Iran (e-mail:
Saghir.hr@gmail.com; bagheri-s@sharif.edu).
Cite:H. R. Saghir and S. B. Shouraki, "Self-Organizing Adaptive Network-Based Fuzzy Intelligent Controller (SOANFIC)," International Journal of Machine Learning and Computing vol.2, no. 6, pp. 864-868, 2012.