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

A method for accurate alignment of parts based on pd-like algorithm combined with icp algorithm

An algorithmic and precise technology, applied in the computer and mechanical fields, can solve problems such as being unfavorable for algorithmic solution, reducing computational efficiency, and difficult to meet accurate alignment of parts

Active Publication Date: 2022-04-19
BEIJING HANFLY AERO ENGINE CO LTD
View PDF7 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The Iterativa Clostest Point (ICP) algorithm requires a paired 3D point set. In practical applications, the part point cloud data measured by the probe will generate certain noise, which cannot guarantee a one-to-one correspondence with the CAD model point cloud data. The error between cloud data and CAD model point cloud data is relatively large, and the algorithm is easily trapped in a local optimal solution, so it is difficult to meet the requirements of accurate alignment of parts
[0005] When solving the position and attitude of a part, using the rotation matrix to represent the rotation of the part will be redundant and reduce the calculation efficiency; the rotation represented by the Euler angle and the helical axis has singularity, which is not conducive to the algorithm solution

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
  • A method for accurate alignment of parts based on pd-like algorithm combined with icp algorithm
  • A method for accurate alignment of parts based on pd-like algorithm combined with icp algorithm
  • A method for accurate alignment of parts based on pd-like algorithm combined with icp algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] Embodiment 1: as figure 1 and figure 2 As shown, a PD-like algorithm combined with an ICP algorithm for accurate part alignment method, including the steps of PD-like algorithm for rough registration and the ICP algorithm for fine registration;

[0060] The step of performing rough registration with the PD-like algorithm includes using the PD-like algorithm for rough matching of point cloud data, thereby providing an accurate initial position and attitude for the ICP algorithm;

[0061] The step of the ICP algorithm for fine registration includes the use of the ICP algorithm to fine-match the point cloud data to achieve the purpose of precise alignment.

[0062] The step of the rough registration of the PD-like algorithm is to use the machine learning algorithm of the PD-like adaptive step size adjustment to perform rough matching on the point cloud data.

[0063] The step of fine registration by ICP algorithm uses the ICP algorithm to fine match the matched point cl...

Embodiment 2

[0086] Embodiment 2: refer to figure 2 , a PD-like algorithm combined with an ICP algorithm for accurate parts alignment method, including the following steps;

[0087] The steps of the PD-like algorithm for rough registration also include the following:

[0088] Step 1), initialize the rotation matrix R 1 with the position vector t 1 .

[0089] Step 2), read the point cloud data obtained from 3D reconstruction A Point cloud data of P and computer CAD 3D model C P, calculate the registration objective function E(R 1 ,t 1 ).

[0090] Step 3), constructing the registration objective function Calculate the function output and expected error and get a new set of intermediate point cloud data sets B p.

[0091] Step 4), judging whether the objective function E is less than ε 1 , ε 2 , if less than ε 1 , ε 2 When the algorithm ends, the calculation result is output; otherwise, iterative learning is performed.

[0092] where ε 1 Indicates the expected error value, ε...

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 method for accurately aligning parts by combining a PD-like algorithm with an ICP algorithm, and belongs to the technical field of computers and machinery. It includes the steps of rough registration with PD-like algorithm and the step of fine registration with ICP algorithm. The PD-like algorithm is a machine learning algorithm for PD-like adaptive step adjustment, and the ICP algorithm includes fine matching of the matched point cloud data. The step of rough registration with PD-like algorithm includes using PD-like algorithm for rough matching of point cloud data, so as to provide accurate initial position and attitude for ICP algorithm; the step of fine registration with ICP algorithm includes using ICP algorithm to fine-tune point cloud data. Matching for precise alignment. The step of rough registration with PD-like algorithm is to use the machine learning algorithm of PD-like adaptive step adjustment to perform rough matching on point cloud data; the step of fine registration with ICP algorithm is to fine-tune the matched point cloud data with ICP algorithm. match.

Description

technical field [0001] The invention belongs to the technical field of computers and machinery, and relates to a method for accurately aligning parts using a PD-like algorithm combined with an ICP algorithm. Background technique [0002] At present, in the field of mechanical processing, the traditional six-point matching method is often used for automatic alignment of parts with complex spatial structures and surface shapes, such as aeroengine turbine blades. This alignment method is inefficient and difficult to realize automatic part processing. [0003] The PD-like machine learning algorithm is similar to the PD algorithm of the PID control law in the control system. It can adaptively adjust the step size of the machine learning algorithm, and obtain the point cloud data set with the smallest error with the CAD model through iterative calculation. [0004] The Iterativa Clostest Point (ICP) algorithm requires a paired 3D point set. In practical applications, the part poin...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/00G06T7/30G06N20/00
CPCG06T7/0004G06T7/30G06N20/00G06T2207/10028G06T2207/20081G06T2207/30164
Inventor 孙跃飞
Owner BEIJING HANFLY AERO ENGINE 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