High precision grasp pose detection in dense clutter

@article{Gualtieri2016HighPG,
  title={High precision grasp pose detection in dense clutter},
  author={M. Gualtieri and A. T. Pas and Kate Saenko and Robert W. Platt},
  journal={2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  year={2016},
  pages={598-605}
}
This paper considers the problem of grasp pose detection in point clouds. We follow a general algorithmic structure that first generates a large set of 6-DOF grasp candidates and then classifies each of them as a good or a bad grasp. Our focus in this paper is on improving the second step by using depth sensor scans from large online datasets to train a convolutional neural network. We propose two new representations of grasp candidates, and we quantify the effect of using prior knowledge of… Expand
Grasp Pose Detection in Dense Clutter Using Deep Learning
Recently, a number of grasp detection methods have been proposed that can be used to localize robotic grasp configurations directly from sensor data without estimating object pose. The underlyingExpand
Grasp Pose Detection in Dense Clutter
Recently, a number of grasp detection methods have been proposed that can be used to localize robotic grasp configurations directly from sensor data without estimating object pose. The underlyingExpand
Grasp Pose Detection in Point Clouds
TLDR
A series of robotic experiments are reported that average a 93% end-to-end grasp success rate for novel objects presented in dense clutter, an improvement in grasp detection performance. Expand
A Geometric Approach for Grasping Unknown Objects With Multifingered Hands
TLDR
This article proposes a method for grasping unknown objects from cluttered scenes using a noisy point cloud as an input using a shape complementarity metric and proposes an optimization-based refinement of the hand poses and finger configurations to achieve a power grasp of the target object. Expand
Combining Grasp Pose Detection with Object Detection
Recently, researchers have proposed various grasp detection methods that can be used to localize grasp configurations without estimating object pose [17, 7, 12, 16, 15, 2, 3, 1, 5, 10, 8, 20, 4].Expand
Robotic grasp detection based on image processing and random forest
TLDR
This paper proposes a very quick and accurate approach to detect robotic grasps by developing a morphological image processing method to generate candidate grasp rectangles set which avoids us to search grasp Rectangles globally. Expand
REGRAD: A Large-Scale Relational Grasp Dataset for Safe and Object-Specific Robotic Grasping in Clutter
TLDR
A new dataset named REGRAD is presented to sustain the modeling of relationships among objects and grasps, collected in both forms of 2D images and 3D point clouds to support comprehensive perception of grasping. Expand
Generating Robust Grasps for Unknown Objects in Clutter Using Point Cloud Data
For service robots in mundane environments, grasping unknown objects from clutter is an essential, yet complex, capability. In this work, we build on the recently popular Grasp Pose Detector systemExpand
Viewpoint selection for grasp detection
TLDR
The main findings are that the right viewpoint can dramatically increase the number of detected grasps and the classification accuracy of the top-n detections and the combination of the two methods can improve grasp success rates by as much as 12%. Expand
REGNet: REgion-based Grasp Network for Single-shot Grasp Detection in Point Clouds
TLDR
An end-to-end single-shot grasp detection network taking one single-view point cloud as input for parallel grippers, which significantly outperforms several successful point-cloud based grasp detection methods including GPD, PointnetGPD, as well as S$4$G. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 26 REFERENCES
Using Geometry to Detect Grasp Poses in 3D Point Clouds
TLDR
A set of necessary conditions on the geometry of a grasp that can be used to generate a set of grasp hypotheses are identified and shown, which helps focus grasp detection away from regions where no grasp can exist. Expand
Template-based learning of grasp selection
TLDR
A new grasp selection algorithm able to find object grasp poses based on previously demonstrated grasps able to improve over time using the information of previous grasp attempts to adapt the ranking of the templates. Expand
Deep learning for detecting robotic grasps
TLDR
This work presents a two-step cascaded system with two deep networks, where the top detections from the first are re-evaluated by the second, and shows that this method improves performance on an RGBD robotic grasping dataset, and can be used to successfully execute grasps on two different robotic platforms. Expand
One-shot learning and generation of dexterous grasps for novel objects
TLDR
A method for one-shot learning of dexterous grasps and grasp generation for novel objects using an incomplete point cloud from a depth camera and a product of experts, in which experts are of two types. Expand
Leveraging big data for grasp planning
TLDR
A deep learning method is applied and it is shown that it can better leverage the large-scale database for prediction of grasp success compared to logistic regression and suggest that labels based on the physics-metric are less noisy than those from the υ-metrics and therefore lead to a better classification performance. Expand
Real-time grasp detection using convolutional neural networks
TLDR
An accurate, real-time approach to robotic grasp detection based on convolutional neural networks that outperforms state-of-the-art approaches by 14 percentage points and runs at 13 frames per second on a GPU. Expand
Optimizing for what matters: The top grasp hypothesis
TLDR
This paper proposes a supervised learning method that is trained with a ranking loss that explicitly encourages that the top-ranked training grasp in a hypothesis set is also positively labeled and shows how this loss can be adapted to work with data that has binary labels. Expand
Empty the basket - a shape based learning approach for grasping piles of unknown objects
  • D. Fischinger, M. Vincze
  • Engineering, Computer Science
  • 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems
  • 2012
TLDR
This paper presents a novel approach to emptying a basket filled with a pile of objects that form, size, position, orientation and constellation of the objects are unknown, and introduces Height Accumulated Features (HAF) which provides an efficient way of calculating grasp related feature values. Expand
Supersizing self-supervision: Learning to grasp from 50K tries and 700 robot hours
TLDR
This paper takes the leap of increasing the available training data to 40 times more than prior work, leading to a dataset size of 50K data points collected over 700 hours of robot grasping attempts, which allows us to train a Convolutional Neural Network for the task of predicting grasp locations without severe overfitting. Expand
Learning a dictionary of prototypical grasp-predicting parts from grasping experience
TLDR
A real-world robotic agent is presented that is capable of transferring grasping strategies across objects that share similar parts, and the applicability of the prototypical parts for grasping novel objects is demonstrated. Expand
...
1
2
3
...