Abstract
A multiple fuzzy controller is proposed for self-organizing fuzzy control in which several “elemental fuzzy controllers” are processed in parallel and the degree of usage of each inferred consequent is determined by using a linear neural network. The inputs to the neural network are all the inferred consequents generated from the elemental fuzzy controllers, and the output of the neural network is the control input to the plant. The delta rule is used to update the connection weights for the network so that the square of plant output deviation is minimized. The present approach allows the elemental fuzzy controller to be used in situations in which the controller parameters are tuned incompletely; thus the number of trials and errors required for tuning the parameters can be decreased significantly. The effectiveness of the present fuzzy controller is illustrated by a simulation for the attitude control of a flexible satellite.
| Original language | English |
|---|---|
| Pages (from-to) | 2970-2976 |
| Number of pages | 7 |
| Journal | Nihon Kikai Gakkai Ronbunshu, C Hen/Transactions of the Japan Society of Mechanical Engineers, Part C |
| Volume | 58 |
| Issue number | 554 |
| DOIs | |
| Publication status | Published - 1992 |
| Externally published | Yes |
Keywords
- Automatic Control
- Computer Control
- Fuzzy Control
- Identification
- Iterative Learning Control
- Multiple Controllers
- Neural Networks
- Self-Organizing
ASJC Scopus subject areas
- Mechanics of Materials
- Mechanical Engineering
- Industrial and Manufacturing Engineering