1993 Fiscal Year Final Research Report Summary
Study on Development of Intelligent Robot for Welding under Extreme Condition
| Project/Area Number |
04650670
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
溶接工学
|
|---|
| Research Institution | Keio University |
|---|
Principal Investigator |
SUGA Yasuo Keio University, Faculty of Science and Technology, Professor, 理工学部・機械工学科, 教授 (47266005)
|
|---|
| Project Period (FY) |
1992 – 1993
|
| Keywords | Extreme condition / Arc welding / Welding robot / Visual sensor / Fuzzy control / Intelligent robot / Seam tracking / Light projection method |
|---|
| Research Abstract |
In order to investigate the automatic welding under extreme condition, a welding robot with visual sensor and arc sensor system was constructed. Moreover, in order to detect weld line and make seam tracking stably a construction of intelligent robot was tried.
The main results obtained are summarized as follows :
1)In order to perform welding stably under high pressure atmosphere, fuzzy control of arc length was tried. And it is confirmed that, if optimum parameters are selected, the control system constructed can be applied to the arc length control of high pressure welding.
2)A three dimensional Cartesian coordinate welding robot was constructed and effects of surface condition of base metal and shape of joints on recognition of weld line are discussed. As the result, it is seen that, if optimum lighting system and reasonable image processing were applied, stable seam tracking can be performed.
3)A new method to detect weld line position using plural scanning lines (Continuous Multi-vector Connecting Method ; CMVC method) was proposed and the effectiveness of the method was confirmed.
4)A welding robot with laser lighting system which can move along a pipe autonomously was constructed. And it is confirmed that the system is effective on the automatic welding of butt and T-joint of pipes.
5)A new method to detect weld line and make seam tracking by the artificial neural network was proposed and a robot system was constructed. As the experimental investigation, the effectiveness of the system was confirmed.
|
