| Project/Area Number |
08650494
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
計測・制御工学
|
|---|
| Research Institution | SHIMANE UNIVERSITY (1997)
Okayama University (1996) |
|---|
Principal Investigator |
FUNABIKI Shigeyuki FACULTY OF INTERDISCIPLINARY OF SCIENCE AAND ENGINEEERING,Professor, 総合理工学部, 教授 (60108123)
|
|---|
| Project Period (FY) |
1996 – 1997
|
| Project Status |
Completed (Fiscal Year 1997)
|
| Budget Amount *help |
¥2,500,000 (Direct Cost: ¥2,500,000)
Fiscal Year 1997: ¥300,000 (Direct Cost: ¥300,000)
Fiscal Year 1996: ¥2,200,000 (Direct Cost: ¥2,200,000)
|
|---|
| Keywords | mobile robots / fuzzy / neural networks / folow-up control / learning / ファジィ推論 / 距離センサー |
|---|
| Research Abstract |
For mobile robots working under various conditions, the improvement of the system efficiency, the development of the tuning-less and maintenance-free system can be realized by obtaining the optimal traveling control scheme according to the traveling conditions. The goal of this study is to achieve the acquistion method of the traveling control of mobile robots for various traveling conditions by using the intelligent control methods, e.g.fuzzy theory and neural networks. The obtained results in this study is as follows ;
(1) Neural network based successive learning of steering control of mobile robots
The control results with neural networks will be unstable due to the over learning of neural networks. The real time tuning function is proposed for the control method of learning of neural networks, and then the stable steering control of mobile robots can be achieved in some trials.
(2) Automatic acquisition of follow-up control by using fuzzy theory and neural networks
The acquisition method of the follow-up control for mobile robots is proposed based on the fuzzy theory and neural networks. It is clarified by the computer simulation and the experiment that the developed control is very available for the follow-up control of mobile robots.
(3) Development of position sensor using a CCD camera
A monocular vision based position sensor using a CCD camera with neural networks is developed.
|
