A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked Environments
Grasp detection takes on a critical significance for the robot. However, detecting object positions and corresponding grasp positions in a stacked environment can be quite difficult for a robot. Based on this practical problem, in order to achieve more accurate object position detection and grasp po...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-12-01
|
| Series: | Micromachines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-666X/14/1/117 |
| _version_ | 1797438577895800832 |
|---|---|
| author | Xuefeng Dong Yang Jiang Fengyu Zhao Jingtao Xia |
| author_facet | Xuefeng Dong Yang Jiang Fengyu Zhao Jingtao Xia |
| author_sort | Xuefeng Dong |
| collection | DOAJ |
| description | Grasp detection takes on a critical significance for the robot. However, detecting object positions and corresponding grasp positions in a stacked environment can be quite difficult for a robot. Based on this practical problem, in order to achieve more accurate object position detection and grasp position detection, a new method called MMD (Multi-stage network for multi-object grasp detection algorithm) is proposed in this paper. MMD covers two parts, including the feature extractor and the multi-stage object predictor. The feature extractor refers to a deep convolutional neural network that can generate shared feature layers as well as the initial ROIs (region of interest). A multi-stage refiner serves as the multi-stage object predictor, which continuously regresses the initial ROI to obtain more accurate object detection and grasping detection results. Ablation experiments show that the proposed MMD has better grasp detection performance. The specific performance is that the recognition precision achieves a state-of-the-art 76.71% mAPg on the VMRD dataset. Moreover, test experiments demonstrate the feasibility of our method on the Kinova robot. |
| first_indexed | 2024-03-09T11:40:05Z |
| format | Article |
| id | doaj.art-2ffc29e6491047158c5e4af91846dfec |
| institution | Directory Open Access Journal |
| issn | 2072-666X |
| language | English |
| last_indexed | 2024-03-09T11:40:05Z |
| publishDate | 2022-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Micromachines |
| spelling | doaj.art-2ffc29e6491047158c5e4af91846dfec2023-11-30T23:33:26ZengMDPI AGMicromachines2072-666X2022-12-0114111710.3390/mi14010117A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked EnvironmentsXuefeng Dong0Yang Jiang1Fengyu Zhao2Jingtao Xia3Faculty of Robot Science and Engineering, Northeastern University, Shenyang 110169, ChinaFaculty of Robot Science and Engineering, Northeastern University, Shenyang 110169, ChinaFaculty of Robot Science and Engineering, Northeastern University, Shenyang 110169, ChinaFaculty of Robot Science and Engineering, Northeastern University, Shenyang 110169, ChinaGrasp detection takes on a critical significance for the robot. However, detecting object positions and corresponding grasp positions in a stacked environment can be quite difficult for a robot. Based on this practical problem, in order to achieve more accurate object position detection and grasp position detection, a new method called MMD (Multi-stage network for multi-object grasp detection algorithm) is proposed in this paper. MMD covers two parts, including the feature extractor and the multi-stage object predictor. The feature extractor refers to a deep convolutional neural network that can generate shared feature layers as well as the initial ROIs (region of interest). A multi-stage refiner serves as the multi-stage object predictor, which continuously regresses the initial ROI to obtain more accurate object detection and grasping detection results. Ablation experiments show that the proposed MMD has better grasp detection performance. The specific performance is that the recognition precision achieves a state-of-the-art 76.71% mAPg on the VMRD dataset. Moreover, test experiments demonstrate the feasibility of our method on the Kinova robot.https://www.mdpi.com/2072-666X/14/1/117grasp detectionmulti-stage networkmulti-taskstack scenariosVMRDKinova robot |
| spellingShingle | Xuefeng Dong Yang Jiang Fengyu Zhao Jingtao Xia A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked Environments Micromachines grasp detection multi-stage network multi-task stack scenarios VMRD Kinova robot |
| title | A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked Environments |
| title_full | A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked Environments |
| title_fullStr | A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked Environments |
| title_full_unstemmed | A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked Environments |
| title_short | A Practical Multi-Stage Grasp Detection Method for Kinova Robot in Stacked Environments |
| title_sort | practical multi stage grasp detection method for kinova robot in stacked environments |
| topic | grasp detection multi-stage network multi-task stack scenarios VMRD Kinova robot |
| url | https://www.mdpi.com/2072-666X/14/1/117 |
| work_keys_str_mv | AT xuefengdong apracticalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments AT yangjiang apracticalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments AT fengyuzhao apracticalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments AT jingtaoxia apracticalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments AT xuefengdong practicalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments AT yangjiang practicalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments AT fengyuzhao practicalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments AT jingtaoxia practicalmultistagegraspdetectionmethodforkinovarobotinstackedenvironments |