碩士 / 國立海洋大學 / 機械與輪機工程學系 / 90 / In this thesis, we investigated the related problems of the application of the neural network to the positioning calibration of an underwater manipulator. Due to the influence of the ocean current and the dynamic effects of the underwater vehicle, high level of positioning accuracy of an underwater manipulator is generally unattainable. In this preliminary study, we did not consider the external loads such as the effects of the current, but focused on the inaccuracy induced by the process of fabrication, assembling, and the errors from the transmission systems of the underwater manipulator. First, we performed the positioning accuracy tests and repeatability tests to find out the factors of inaccuracy. Then, we regarded the manipulator’s inaccuracy as a system error, and performed the linear regression analysis to the system output. A database was resulted from the linear regression analysis. Finally, we proceeded the positioning calibration using the back-propagation algorithm of the neural network. The results from the calibration using neural network showed that, the positioning accuracy of the manipulator was greatly enhanced than that before calibration.
| Identifer | oai:union.ndltd.org:TW/090NTOU0491031 |
| Date | January 2002 |
| Creators | Chih-Chung Lu, 盧志忠 |
| Contributors | 林鎮洲 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 81 |
Page generated in 0.0131 seconds