Autonomous Distributed Scheduling by Using Multi-agent Reinforcement Learning for 5-axis Machine Tools
| Project/Area Number |
25820024
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Multi-year Fund |
|---|
| Research Field |
Production engineering/Processing studies
|
|---|
| Research Institution | Osaka Prefecture University |
|---|
Principal Investigator |
Iwamura Koji 大阪府立大学, 工学(系)研究科(研究院), 准教授 (40332001)
|
|---|
| Project Period (FY) |
2013-04-01 – 2016-03-31
|
| Project Status |
Completed (Fiscal Year 2015)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,200,000、Indirect Cost: ¥960,000)
Fiscal Year 2014: ¥1,040,000 (Direct Cost: ¥800,000、Indirect Cost: ¥240,000)
Fiscal Year 2013: ¥3,120,000 (Direct Cost: ¥2,400,000、Indirect Cost: ¥720,000)
|
|---|
| Keywords | 5軸加工機 / マルチエージェント強化学習 / 自律分散型生産システム / スケジューリング / 自律分散型スケジューリング |
|---|
| Outline of Final Research Achievements |
Process integration achieved by 5-axis machine tools have received attention in recent years from the viewpoint of reduction of production lead time. New autonomous distributed scheduling systems by using multi-agent reinforcement learning are proposed in this research. The proposed scheduling systems consist of 5-axis machine tools, 3DCAD database, customized 3DCAM corresponding to types of 5-axis machine tools, and the jobs.
The proposed scheduling systems carry out following 3 steps. At first, the individual 5-axis machine tools refer the 3DCAD database to obtain the 3DCAD data of jobs which can be carried out next machining process. Next, NC data and processing time of all jobs are created by 3DCAM. Finally, the individual machine tools and jobs carry out the autonomous distributed scheduling process by using multi-agent reinforcement learning.
It is shown that objective functions of manufacturing systems are improved by using proposed system through the case studies.
|
Report
(4 results)
Research Products
(13 results)
