Check patentability & draft patents in minutes with Patsnap Eureka AI!

RANSAC (Random Sample Consensus) improvement method suitable for simultaneous localization and mapping

A map construction and model technology, applied in the fields of computer vision, image processing and positioning and navigation, can solve the problems of high time cost, inability to achieve real-time SLAM, unavoidable cumulative error, etc., to achieve the effect of improving accuracy and robustness

Pending Publication Date: 2021-07-20
GUIZHOU POWER GRID CO LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Early visual SLAM uses SIFT features, and the positioning accuracy is relatively high, but the time cost is very high, and the real-time perfo

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
  • RANSAC (Random Sample Consensus) improvement method suitable for simultaneous localization and mapping
  • RANSAC (Random Sample Consensus) improvement method suitable for simultaneous localization and mapping
  • RANSAC (Random Sample Consensus) improvement method suitable for simultaneous localization and mapping

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0050] refer to figure 1 , is the first embodiment of the present invention, and this embodiment provides a kind of RANSAC improved method that is suitable for real-time positioning and map construction, comprises:

[0051] RANSAC (Random Sample Consensus) is an algorithm that calculates the mathematical model parameters of the data based on a set of sample data sets containing abnormal data, and obtains effective sample data.

[0052] S1: Randomly select n sample estimation models, use the sample estimation models to calculate all points, and obtain the number of internal points.

[0053] It should be noted that the sample estimation model can be a one-dimensional primary model, two-variable primary model or one-dimensional quadratic model, etc. The sample estimation model in this embodiment is:

[0054] y=ax+b

[0055] Substitute all the points into the sample estimation model, and the points conforming to the sample estimation model are considered as inliers, otherwise th...

Embodiment 2

[0078] In order to verify and illustrate the technical effect adopted in this method, this embodiment conducts experiments in various situations, utilizing the TUM of the indoor hand-held camera to collect the data set, the data set KITTI of the outdoor expressway and urban traffic, and the micro air vehicle (MAV) The collected data set EuRoC is used for comparative experiments.

[0079] The parameters are set as follows: the value of f is 0.99, the maximum number of iterations is 300, the number of randomly selected samples is 4, the probability d of selecting an inlier from the data set each time is 0.5, and the threshold for judging the maximum error is 5.991σ 2 , wherein, σ=0.3; the number of samples randomly selected during iteration is: min(14, I / 2), where I is the number of interior points in the previous round, and the multiple of threshold value change is 0.7.

[0080] (1) From a qualitative point of view, figure 2 and image 3 shows the matching of the two frames ...

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 an RANSAC improvement method suitable for simultaneous localization and mapping, which comprises the following steps of: randomly selecting n sample estimation models, and calculating all points by using the sample estimation models to obtain an interior point number; updating the optimal model according to the interior point number, and storing the interior points; estimating the updated optimal model by utilizing a least square method and based on a set threshold value, and storing a corresponding inner point; randomly selecting m samples from the corresponding inner points, iterating the samples to estimate the model, and if errors are reduced, updating the model until the best model appears or the highest number of iterations is reached. According to the method, the RANSAC is improved by combining the least square method, and the precision and robustness of the pose tracking thread of the SLAM are improved.

Description

technical field [0001] The invention relates to the technical fields of computer vision, image processing and positioning and navigation, in particular to an improved RANSAC method suitable for real-time positioning and map construction. Background technique [0002] The self-localization of mobile robots in unknown environments is closely related to the establishment of environmental models. The realization of positioning is inseparable from the environment model, and the accuracy of the environment model depends on the accuracy of positioning. In an unknown environment, the robot has no reference object and can only rely on its own inaccurate sensors to obtain external information. At this time, it is very difficult to achieve accurate positioning. It is easy to achieve positioning with an existing map and create a map with a known location, but it is impossible to achieve positioning without a map and create a map without positioning. In the existing research, the solut...

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): G01C21/32
CPCG01C21/32
Inventor 杨金铎曾惜王林波王元峰杨凤生王恩伟王宏远付滨
Owner GUIZHOU POWER GRID CO LTD
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com