A Fast Knowledge Acquisition for Multi-Input Systems Using Fuzzy Inference Network
| Autor: | Ichiro Hiraga, Takeshi Furuhashi, Shoichi Nakayama, Yoshiki Uchikawa |
|---|---|
| Rok vydání: | 1995 |
| Předmět: |
Adaptive neuro fuzzy inference system
Fuzzy classification Neuro-fuzzy business.industry Computer science Fuzzy control system computer.software_genre Defuzzification Fuzzy logic Theoretical Computer Science Computational Theory and Mathematics Artificial Intelligence Fuzzy set operations Fuzzy number Artificial intelligence Data mining business computer Software |
| Zdroj: | Intelligent Automation & Soft Computing. 1:379-388 |
| ISSN: | 2326-005X 1079-8587 |
| DOI: | 10.1080/10798587.1995.10750643 |
| Popis: | The authors have proposed a basic concept for the construction method of a fuzzy inference network for multi-input systems. This method is simple and quick in constructing complex nonlinear fuzzy models. This article studies an application of the fuzzy inference network to knowledge acquisition of complex control tasks: a collision avoidance problem with multiple moving obstacles. The avoidance task is gradually made complex by loading the acquired knowledge onto the moving obstacles. A small number of fuzzy rules is obtained out of more than 2.6 × 109 possible combinations. The fuzzy inference network is constructed from insufficient data within a very short computation time. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|