| Project/Area Number |
02556033
|
|---|
| Research Category |
Grant-in-Aid for Developmental Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
農業機械
|
|---|
| Research Institution | Okayama University |
|---|
Principal Investigator |
SHIBANO Yasunori Okayama Univ. Faculty of Agr. Prof., 農学部, 教授 (00033332)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
MONTA Mitsuji Okayama Univ. Faculty of Agr. Inst., 農学部, 助手 (80239714)
KONDO Naoshi Okayama Univ. Grad. school Inst., 大学院・自然科学研究科, 助手 (20183353)
MOHRI Kentaro Okayama Univ. Faculty of Agr. Prof., 農学部, 教授 (20081531)
伏見 吉正 岡山大学, 農学部, 助手 (90199274)
|
|---|
| Project Period (FY) |
1990 – 1992
|
| Project Status |
Completed (Fiscal Year 1992)
|
| Budget Amount *help |
¥5,200,000 (Direct Cost: ¥5,200,000)
Fiscal Year 1992: ¥600,000 (Direct Cost: ¥600,000)
Fiscal Year 1991: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 1990: ¥3,500,000 (Direct Cost: ¥3,500,000)
|
|---|
| Keywords | Agricultural robot / Fruit harvesting / Polar Coordinate Manipulator / Visual feedback / Berry thinning / Continuous path control / Grape / Most suitable wavelength band / 滴粒・整房作業 / 極座標マニピュレ-タ / ビジュアルフィ-ドバック |
|---|
| Research Abstract |
Grapevine is planted in trellis training in Japan. It is considered this training has much adaptability of robot working than other trainings. In this study, a manipulator with 5 degrees of freedom and a harvest hand were made as a trial based on the measurement of positions of grape bunches, frictional resistance and cutting resistance of rachis. The manipulator was a polar coordinate type its was controlled to move in continuous path. The hand had three functions that were grasping cutting rachis and pushing bunch.
As for visual sensor, spectral reflectances of parts grapevines were at first measures in order to select pairs of optical filters which were suitable for discrimination by the visual sensor. Experiments to discriminate among the parts were done by using R, G, B filters and interference filters. From the result, it was able to discriminate not only between different color parts but also between simular color parts of grapevines. Besides, experiments to detect positions of bunch and rachis and to harvest bunches were done by using the manipulator which CCD camera was attached to. It was obtained that the errors were reasonable as detecting error when the robot worked to harvest, since such errors were able to be absorbed in harvesting hand. Finally, the experiment to harvest individual bunches was done by using the robot in room and in the field. From the results, it was observed the robot was able to harvest satisfactorily.
In addition to that, berry thinning hand also manufactured in order to add ability to do other work to the robot. The hand could berries to make harvesting easy and could cut bunch to standardize its length.
|
