Grasp Pose Detection in Point Clouds

نویسندگان

  • Andreas ten Pas
  • Marcus Gualtieri
  • Kate Saenko
  • Robert Platt
چکیده

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 underlying idea is to treat grasp perception analogously to object detection in computer vision. These methods take as input a noisy and partially occluded RGBD image or point cloud and produce as output pose estimates of viable grasps, without assuming a known CAD model of the object. Although these methods generalize grasp knowledge to new objects well, they have not yet been demonstrated to be reliable enough for wide use. Many grasp detection methods achieve grasp success rates (grasp successes as a fraction of the total number of grasp attempts) between 75% and 95% for novel objects presented in isolation or in light clutter. Not only are these success rates too low for practical grasping applications, but the light clutter scenarios that are evaluated often do not reflect the realities of real world grasping. This paper proposes a number of innovations that together result in a significant improvement in grasp detection performance. The specific improvement in performance due to each of our contributions is quantitatively measured either in simulation or on robotic hardware. Ultimately, we report a series of robotic experiments that average a 93% end-to-end grasp success rate for novel objects presented in dense clutter.

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

3D Detection of Power-Transmission Lines in Point Clouds Using Random Forest Method

Inspection of power transmission lines using classic experts based methods suffers from disadvantages such as highel level of time and money consumption. Advent of UAVs and their application in aerial data gathering help to decrease the time and cost promenantly. The purpose of this research is to present an efficient automated method for inspection of power transmission lines based on point c...

متن کامل

Dex-Net 3.0: Computing Robust Robot Suction Grasp Targets in Point Clouds using a New Analytic Model and Deep Learning

Suction-based end effectors are widely used in industry and are often preferred over parallel-jaw and multifinger grippers due to their ability to lift objects with a single point of contact. This ability simplifies planning, and handcoded heuristics such as targeting planar surfaces are often used to select suction grasps based on point cloud data. In this paper, we propose a compliant suction...

متن کامل

Robotic Grasping of Unknown Objects

This work describes the development of a novel vision-based grasping system for unknown objects based on laser range and stereo data. The work presented here is based on 2.5D point clouds, where every object is scanned from the same view point of the laser range and camera position. We tested our grasping point detection algorithm separately on laser range and single stereo images with the goal...

متن کامل

Detection of some Tree Species from Terrestrial Laser Scanner Point Cloud Data Using Support-vector Machine and Nearest Neighborhood Algorithms

acquisition field reference data using conventional methods due to limited and time-consuming data from a single tree in recent years, to generate reference data for forest studies using terrestrial laser scanner data, aerial laser scanner data, radar and Optics has become commonplace, and complete, accurate 3D data from a single tree or reference trees can be recorded. The detection and identi...

متن کامل

Dex-Net 2.0: Deep Learning to Plan Robust Grasps with Synthetic Point Clouds and Analytic Grasp Metrics

To reduce data collection time for deep learning of robust robotic grasp plans, we explore training from a synthetic dataset of 6.7 million point clouds, grasps, and robust analytic grasp metrics generated from thousands of 3D models from DexNet 1.0 in randomized poses on a table. We use the resulting dataset, Dex-Net 2.0, to train a Grasp Quality Convolutional Neural Network (GQ-CNN) model tha...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

عنوان ژورنال:
  • I. J. Robotics Res.

دوره 36  شماره 

صفحات  -

تاریخ انتشار 2017