Object grasp detection algorithm based on improved Keypoint RCNN model

被引：0
作者：
Xia H. [1 ]
Suo S. [1 ]
Wang Y. [1 ]
An Q. [1 ]
Zhang M. [1 ]
机构：
[1] Department of Mechanical Engineering, Tsinghua University, Beijing
来源：
Yi Qi Yi Biao Xue Bao/Chinese Journal of Scientific Instrument | 2021年 / 42卷 / 04期
关键词：
Attention module; Grasp detection; Grasp representation; Improved Keypoint RCNN model; Optimized grasp; Overlap rate; Weight of loss;
D O I：
10.19650/j.cnki.cjsi.J2107383
中图分类号：
学科分类号：
摘要：
There are two difficulties in the application of robot grasping in industry. How to detect the graspable object accurately and how to select the optimized grasp target among the detected multiple objects. In this paper the homoscedastic uncertainty is introduced into Keypoint RCNN to learn the weights of various losses, the attention modules are integrated into feature extractor, which composes the improved Keypoint RCNN model. A two-stage object grasp detection algorithm is proposed based on the improved Keypoint RCNN model. In the first stage, the improved model is used to predict the masks and keypoints. In the second stage, the masks and keypoints are used to compute the grasp representation and overlap rate of the object, the overlap rate represents the level of collision while grasping. According to the overlap rate, the optimized grasp target can be selected from multiple graspable objects. Comparison experiment indicates that the performances of the improved Keypoint RCNN model are improved in object detection, instance segmentation and keypoint detection compared with those of original model, and the average precisions (AP) on the self-built dataset reach 85.15%, 79.66% and 86.63%, respectively. Robot grasping experiment proves that the proposed grasp detection algorithm can accurately calculate the grasp representation, select the optimized grasp and guide the robot to grasp the target with collision-free grasp. © 2021, Science Press. All right reserved.
引用
收藏
页码：236 / 246
页数：10
相关论文
共 22 条
  • [1] DU G, WANG K, LIAN S, Et al., Vision-based robotic grasping from object localization, object pose estimation to grasp estimation for parallel grippers: A review, Artificial Intelligence Review, 54, pp. 1677-1734, (2021)
  • [2] HINTERSTOISSER S, LEPETIT V, ILIC S, Et al., Model based training, detection and pose estimation of texture-less 3D objects in heavily cluttered scenes, Asia Conference on Computer Vision, pp. 548-562, (2012)
  • [3] DROST B, ULRICH M, NAVAB N, Et al., Model globally, match locally: Efficient and robust 3D object recognition, Computer Vision and Pattern Recognition, pp. 998-1005, (2010)
  • [4] HINTERSTOISSER S, LEPETIT V, RAJKUMAR N, Et al., Going further with point pair features, European Conference on Computer Vision, pp. 834-848, (2016)
  • [5] MELLADO N, AIGER D, MITRA N., Super 4PCS fast global pointcloud registration via smart indexing, Computer Graphics Forum, 33, 5, pp. 205-215, (2014)
  • [6] HE K, GKIOXARI G, DOLLAR P, Et al., Mask R-CNN, International Conference on Computer Vision, pp. 2980-2988, (2017)
  • [7] REDMON J, DIVVALA S, GIRSHICK R, Et al., You only look once: Unified, real-time object detection, Computer Vision and Pattern Recognition, pp. 779-788, (2016)
  • [8] LIU W, ANGUELOV D, ERHAN D, Et al., SSD: Single shot multibox detector, European Conference on Computer Vision, pp. 21-37, (2016)
  • [9] HERNANDEZ C, BHARATHEESHA M, KO W, Et al., Team delft's robot winner of the amazon picking challenge 2016, RobotCup2016: Robot World Cup, XX, pp. 613-624, (2017)
  • [10] ZENG A, YU K, SONG S, Et al., Multi-view self-supervised deep learning for 6D pose estimation in the amazon picking challenge, International Conference on Robotics and Automation, pp. 1386-1393, (2017)
123