Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Deep learning-based quick automatic capture and placement method

A deep learning and fast technology, applied in neural learning methods, image data processing, instruments, etc., can solve problems such as ineffective processing, and achieve the effect of fast and convenient deployment, simple equipment, and fast and convenient deployment

Active Publication Date: 2018-08-14
HANGZHOU LANXIN TECH CO LTD
View PDF5 Cites 45 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When dealing with some complex or irregular items, it cannot be processed efficiently

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep learning-based quick automatic capture and placement method
  • Deep learning-based quick automatic capture and placement method
  • Deep learning-based quick automatic capture and placement method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0025] The present invention provides a fast automatic grasping and placement method based on deep learning, which specifically includes:

[0026] 1. Installation scheme. First install the robotic arm, and the 3D camera is installed above the side of the robotic arm. The field of view of the camera should overlap with the operating range of the robotic arm as much as possible, so as to maximize the operating space. Such as figure 1 shown. The robotic arm adopts a multi-axis robotic arm, which can complete various postures and spatial ranges. The 3D camera uses a structured light camera, which can accurately obtain color information and depth information in the field of view, and the error and noise of the depth information are small. If there is an unsatisfactory light situation, the scheme of placing a supplementary light above.

[0027] 2. Calib...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a deep learning-based quick automatic capture and placement method. The method can quickly and accurately determine positions and attitudes of articles in combination with a GPU and a 3D camera by adopting a deep learning scheme. Through a quick calibration scheme, a transfer matrix from a coordinate system of the 3D camera to a coordinate system of a mechanical arm can beobtained. The transfer matrix can convert the positions and attitudes of the articles to the coordinate system of the mechanical arm, and then operate the mechanical arm to perform capture. For accurately placing the articles in a specific mode, the articles are subjected to secondary attitude estimation. A secondary attitude estimation process comprises the steps of firstly capturing the articlesto an estimation position by fixed attitude; at the moment, segmenting out the articles more completely by only using depth information; projecting the articles to a placement plane; and performing accurate attitude estimation for information on the plane, thereby enabling the mechanical arm to perform placement.

Description

technical field [0001] The invention relates to the technical fields of computer vision, robotics, deep learning, artificial intelligence, and industrial automation, and in particular to a fast automatic grabbing and placing method based on deep learning. Background technique [0002] Industrial automation is of great significance to the improvement of production efficiency of industrial production lines, assembly lines, packaging lines, etc. With the rapid development of e-commerce and the increase of labor costs, industrial automation equipment and robots are used more and more widely. [0003] The current scenario where industrial robots are used is generally a multi-axis robotic arm. The automation method of the robotic arm is relatively simple. Generally, it can only handle specific and regular items, and can only grab and place items (parts, products, packaging boxes, etc.) according to the predetermined position and posture. When dealing with some complex or irregul...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06T7/70G06T7/80G06N3/04G06N3/08G06T3/00B25J9/16B25J9/10
CPCG06N3/08G06T7/70G06T7/80B25J9/10B25J9/16B25J9/163G06N3/045G06T3/067
Inventor 时岭郑卫军高勇
Owner HANGZHOU LANXIN TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com