|
ヤマザキ ヨシアキ
山崎 芳昭 所属 理工学部 総合理工学科 データサイエンス学環 職種 教授 |
|
| 言語種別 | 英語 |
| 発行・発表の年月 | 2022/01 |
| 形態種別 | 学術雑誌 |
| 標題 | Object recognition by RealSenseD435 using ROS and object holding with 6 DoF manipulator |
| 執筆形態 | 共著 |
| 掲載誌名 | Proceedings of the Joint Symposium of The Twenty-Seventh International Symposium on Artificial Life |
| 掲載区分 | 国外 |
| 出版社・発行元 | International Society of Artificial Life and Robotics |
| 巻・号・頁 | pp.968-972 |
| 総ページ数 | 5 |
| 担当区分 | 最終著者 |
| 著者・共著者 | Jaefun Seo and Yoshiaki Yamazaki |
| 概要 | In this study, we attached the depth camera Intel RealSense D435 to a 6DOF manipulator created in a previous study. Next, ROS (Robot Operating System) was used for control, and SSD (Single Shot multibox Ditector) and TensorFlow, which have relatively high processing speed and high accuracy, were used as tools for valve recognition.
Then, by detecting the valve from the learning data obtained by deep learning and transmitting the distance data obtained from RealSense D435 to the 6-DOF manipulator, the valve was automatically recognized and rotated by the gripper. Furthermore, by changing from a monocular webcam to a depth camera, we tried to simplify the object recognition method used in the previous research. |
| researchmap用URL | https://researchmap.jp/yamazakiyoshiaki/ |