Slip Detection with Combined Tactile and Visual Information
© 2018 IEEE. Slip detection plays a vital role in robotic manipulation and it has long been a challenging problem in the robotic community. In this paper, we propose a new method based on deep neural network (DNN) to detect slip. The training data is acquired by a GelSight tactile sensor and a camer...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021
|
| Online Access: | https://hdl.handle.net/1721.1/137954 |
| _version_ | 1826203358010015744 |
|---|---|
| author | Li, Jianhua Dong, Siyuan Adelson, Edward |
| author_facet | Li, Jianhua Dong, Siyuan Adelson, Edward |
| author_sort | Li, Jianhua |
| collection | MIT |
| description | © 2018 IEEE. Slip detection plays a vital role in robotic manipulation and it has long been a challenging problem in the robotic community. In this paper, we propose a new method based on deep neural network (DNN) to detect slip. The training data is acquired by a GelSight tactile sensor and a camera mounted on a gripper when we use a robot arm to grasp and lift 94 daily objects with different grasping forces and grasping positions. The DNN is trained to classify whether a slip occurred or not. To evaluate the performance of the DNN, we test 10 unseen objects in 152 grasps. A detection accuracy as high as 88.03 % is achieved. It is anticipated that the accuracy can be further improved with a larger dataset. This method is beneficial for robots to make stable grasps, which can be widely applied to automatic force control, grasping strategy selection and fine manipulation. |
| first_indexed | 2024-09-23T12:35:21Z |
| format | Article |
| id | mit-1721.1/137954 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T12:35:21Z |
| publishDate | 2021 |
| publisher | IEEE |
| record_format | dspace |
| spelling | mit-1721.1/1379542021-11-10T03:26:10Z Slip Detection with Combined Tactile and Visual Information Li, Jianhua Dong, Siyuan Adelson, Edward © 2018 IEEE. Slip detection plays a vital role in robotic manipulation and it has long been a challenging problem in the robotic community. In this paper, we propose a new method based on deep neural network (DNN) to detect slip. The training data is acquired by a GelSight tactile sensor and a camera mounted on a gripper when we use a robot arm to grasp and lift 94 daily objects with different grasping forces and grasping positions. The DNN is trained to classify whether a slip occurred or not. To evaluate the performance of the DNN, we test 10 unseen objects in 152 grasps. A detection accuracy as high as 88.03 % is achieved. It is anticipated that the accuracy can be further improved with a larger dataset. This method is beneficial for robots to make stable grasps, which can be widely applied to automatic force control, grasping strategy selection and fine manipulation. 2021-11-09T16:18:57Z 2021-11-09T16:18:57Z 2018-05 2019-09-27T17:17:30Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137954 Li, Jianhua, Dong, Siyuan and Adelson, Edward. 2018. "Slip Detection with Combined Tactile and Visual Information." en 10.1109/icra.2018.8460495 Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/pdf IEEE arXiv |
| spellingShingle | Li, Jianhua Dong, Siyuan Adelson, Edward Slip Detection with Combined Tactile and Visual Information |
| title | Slip Detection with Combined Tactile and Visual Information |
| title_full | Slip Detection with Combined Tactile and Visual Information |
| title_fullStr | Slip Detection with Combined Tactile and Visual Information |
| title_full_unstemmed | Slip Detection with Combined Tactile and Visual Information |
| title_short | Slip Detection with Combined Tactile and Visual Information |
| title_sort | slip detection with combined tactile and visual information |
| url | https://hdl.handle.net/1721.1/137954 |
| work_keys_str_mv | AT lijianhua slipdetectionwithcombinedtactileandvisualinformation AT dongsiyuan slipdetectionwithcombinedtactileandvisualinformation AT adelsonedward slipdetectionwithcombinedtactileandvisualinformation |