Multi-robot grid map splicing method based on optimal graph matching

A raster map, multi-robot technology, applied in graphics and image conversion, instruments, image analysis and other directions, can solve the problems of large amount of calculation, large overlap of images required for splicing, and high computational cost, and achieves a small amount of calculation and improves robustness. flexibility and scope of application, and the effect of high splicing costs

Pending Publication Date: 2021-03-30
GUANGDONG UNIV OF TECH
View PDF5 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the technical defects of the existing grid map mosaic method, such as large amount of calculation, high cost of calculation, and large overlap between images, and to provide a multi-robot grid map mosaic based on optimal graph matching method

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
  • Multi-robot grid map splicing method based on optimal graph matching
  • Multi-robot grid map splicing method based on optimal graph matching
  • Multi-robot grid map splicing method based on optimal graph matching

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0059] like figure 1 As shown, a multi-robot grid map mosaic method based on optimal graph matching includes the following steps:

[0060] S1: Extract ORB feature points from the raster map to be stitched to obtain key points and feature descriptors;

[0061] S2: Use the multi-probe local sensitive hashing algorithm to retrieve the key point with the closest feature descriptor for each key point to be matched as a rough matching point pair;

[0062] S3: Combine the Hamming distance between feature descriptors to construct a feature dissimilarity matrix; construct a median K-nearest neighbor map for key points and calculate a residual matrix; combine and normalize the residual matrix and feature dissimilarity matrix to generate Transmission cost matrix;

[0063] S4: Introduce the transmission cost matrix to construct the optimal transmission objective function, construct an augmented node to remove outliers, perform negative entropy regularization on the optimal transmission ...

Embodiment 2

[0067] More specifically, on the basis of Example 1, each step is further described.

[0068] More specifically, the step S1 includes the following steps:

[0069] S11: Perform Gaussian blurring on the raster map to be spliced ​​to filter the edge noise of the raster map and make the edge of the binary image generate a continuous and smooth gradient;

[0070] S12: Extract multi-scale FAST key points and extract BRIEF feature descriptors.

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 provides a multi-robot grid map splicing method based on optimal graph matching. The multi-robot grid map splicing method comprises the following steps: extracting key points and featuredescriptors of a to-be-spliced grid map; retrieving a key point with the most similar feature descriptor for each to-be-matched key point as a rough matching point pair; constructing a characteristicdissimilarity matrix and calculating a residual matrix; combining and normalizing the residual matrix and the characteristic dissimilarity matrix to generate a transmission cost matrix; introducing atransmission cost matrix to construct an optimal transmission target function, performing negentropy regularization on the optimal transmission target function, and solving optimal matching through aSinkhorn-Knopp algorithm; solving rigid body change parameters between the optimal matching points through a least square method, transforming the whole grid map, and obtaining a fused map. Accordingto the multi-robot grid map splicing method, the splicing speed and precision of the grid map are effectively improved, the number of correct matching point pairs in the result is large, and the to-be-spliced grid map can be spliced without a large overlapping area.

Description

technical field [0001] The present invention relates to the technical field of multi-robot map construction, and more specifically, to a multi-robot grid map splicing method based on optimal graph matching. Background technique [0002] Constructing a map for an unknown environment is a basic challenge in mobile robotics, and the construction of a map usually requires an accurate estimation of the robot's pose. Therefore, at present, mobile robots mainly use Simultaneous Localization and Mapping (SLAM) technology. There are currently many mature single-robot SLAM algorithms for constructing environmental maps. However, in large-scale unknown environments, the accuracy, efficiency and robustness of single-robot map construction are limited. In the past decade, multi-robot Collaboration has become a hotspot in current research, and the introduction of multi-robots can help break through the limitations of the above-mentioned single-robot SLAM algorithm. In the robot SLAM algo...

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 Applications(China)
IPC IPC(8): G06T3/40G06T7/33G06K9/62
CPCG06T3/4038G06T7/33G06F18/22
Inventor 黄小杭曾碧刘建圻
Owner GUANGDONG UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap