| Project/Area Number |
12680371
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Intelligent informatics
|
|---|
| Research Institution | The University of Electro-Communications |
|---|
Principal Investigator |
TAKEUCHI Ikuo The University of Electro-Communications, Faculty of Electro-Communications, Professor, 電気通信学部, 教授 (90293109)
|
|---|
| Project Period (FY) |
2000 – 2002
|
| Project Status |
Completed (Fiscal Year 2002)
|
| Budget Amount *help |
¥2,200,000 (Direct Cost: ¥2,200,000)
Fiscal Year 2002: ¥900,000 (Direct Cost: ¥900,000)
Fiscal Year 2001: ¥600,000 (Direct Cost: ¥600,000)
Fiscal Year 2000: ¥700,000 (Direct Cost: ¥700,000)
|
|---|
| Keywords | real time / distributed cooperative system / multi-agent system / RoboCup / RoboCupRescue / limited communication / concurrent programming environment / RoboCup / 自律分散協調 / エージェント / シミュレーション / RoboCupRescue / Robo Cup / Robo Cup Resaue / RoboCup-Rescue |
|---|
| Research Abstract |
We investigated on real-time multi-agent systems in which a number of autonomous agents that can share only incomplete information about their dynamically changing environment but have to achieve their objective collectively by cooperating with each other in real time. We chose as the research test-beds RoboCup soccer and RoboCupRescue because these are best suited for research on logical and software aspects of real-time distributed cooperative systems.
The most remarkable point of our research results is that developed agents manage to solve these complicated real-time distributedcooperation problems in humanlike ways as far as possible. For example, our soccer agents use only humanlike short shouts for their communication to make tactical team play. In spite of this limitation, our agent teams won all official competitions in Japan held in this term.
We also developed effective visual debugging environments for agent development. Such agent development is, from the programming viewpoint, nothing other than concurrent programming effort, which is most difficult to test and debug because there will be no same situations repeated with the existence of other offending opponents.
|
