| Project/Area Number |
14550453
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Control engineering
|
|---|
| Research Institution | Musashi Institute of Technology |
|---|
Principal Investigator |
NOHARA Tsutomu Musashi Institute of Technology, Faculty of Engineering, Professor, 工学部, 教授 (80339530)
|
|---|
| Project Period (FY) |
2002 – 2003
|
| Project Status |
Completed (Fiscal Year 2003)
|
| Budget Amount *help |
¥3,000,000 (Direct Cost: ¥3,000,000)
Fiscal Year 2003: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2002: ¥2,100,000 (Direct Cost: ¥2,100,000)
|
|---|
| Keywords | Immune System / Memory / Evolution / Study / Immune Response Network / Neural Network / Genetic Algorithm / Behaviorally Artificial Intelligence |
|---|
| Research Abstract |
In this study, the method, which acts as an arbitration mechanism, is established. The method of an arbitration mechanism is based on the artificial immune system, which mimics human bodies immunity. The following results are obtained.
(1) If we observe the immunity as the system of network, the problem is reduced to solve the simultaneous differential equations in the mathematical mean. Solving simultaneous differential equations in real time becomes to be a bottleneck in applications. We introduce the immune memory mechanism and combine with the immune network. The immune memory mechanism makes the computation load light.
(2) In the non-network immune system, we face on the difficulty of the design of antibodies as the antigens increase. We establish the automatic design method for antibodies using the genetic algorithm.
(3) We realize the evolutionary behavior arbitration system with x and f3 learning functions. The α learning function acquires appropriate antibodies and the f3 learning function obtains the highly arbitration task, which dissolves the conflict situations.
|
