Supercharge Your Innovation With Domain-Expert AI Agents!

SLAM loopback detection method and system based on deep learning

A deep learning and detection method technology, applied in the field of computer vision, can solve the problems of relying on pre-trained dictionaries and low detection accuracy, and achieve the effect of improving accuracy and positioning accuracy

Pending Publication Date: 2020-07-10
的卢技术有限公司
View PDF2 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Loopback detection refers to the ability of the robot to recognize that it has reached a certain scene and make the map closed loop. Currently, it is realized through the word bag technology, but the word bag technology relies heavily on the pre-trained dictionary, and the detection accuracy is not high

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
  • SLAM loopback detection method and system based on deep learning
  • SLAM loopback detection method and system based on deep learning
  • SLAM loopback detection method and system based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0030] In visual SLAM, the error generated by calculating the current frame pose from the previous frame pose causes a cumulative error during the transmission of one frame, thus creating a loopback detection to reduce the cumulative error. Among them, the current frame and Establishing a pose constraint relationship in a previous frame is called loopback, and finding out the historical frame that established this pose constraint is loopback detection. When matching all frames with the current frame, the amount of calculation is too large, so the word bag technology is used to assist in screening information. However, the current word bag technology relies heavily on the pre-trained dictionary, which makes the loop detection ability not high and the accuracy is not enough. Therefore, , the present invention adopts the strategy of combining bag-of-words technology and deep learning detection technology to improve the accuracy of loop detection.

[0031] refer to figure 1 and ...

Embodiment 2

[0070] refer to image 3 , which is the second embodiment of the present invention, this embodiment is different from the first embodiment in that it provides a deep learning-based SLAM loop detection system, including a bag-of-words dictionary module 100, a deep learning detection module 200, a fusion Module 300, bag of words dictionary module 100 includes dictionary 101 and bag of words 102, dictionary 101 is constructed by descriptor clustering, contains all words, and is connected with bag of words 102, bag of words 102 is screened out and current frame by database. The key frame of common word, notice dictionary 101 statistics and current frame identical word quantity at the same time; Deep learning detection module 200 is connected with word bag dictionary module 100, and the loopback candidate frame and key frame that word bag dictionary module 100 detects are sent into deep learning detection When in the module 200, the deep learning detection module 200 starts the det...

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 SLAM loopback detection method and system based on deep learning, and the method comprises the steps: detecting a loopback candidate frame through a bag-of-words dictionarymodule, and transmitting the loopback candidate frame to a deep learning detection module; by the deep learning detection module, detecting and identifying whether the same object exists in a detection frame and the loopback candidate frame; separately acquiring detection probability data of the bag-of-words dictionary module and the deep learning detection module; fusing the detection probabilitydata of the bag-of-words dictionary module and the deep learning detection module by using a Gaussian probability model and a fusion module to obtain fused probability data; and determining whether the fused probability data meets the detection requirement or not. According to the method, the bag-of-words technology and the deep learning detection technology are combined, the loopback detection accuracy is improved, and therefore the positioning precision of the whole SLAM technology is improved.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a deep learning-based SLAM loop detection method and system. Background technique [0002] SLAM is simultaneous positioning and map creation. For example, if a robot is moving in an unknown environment, how to determine its own trajectory through observation of the environment and build a map of the environment at the same time. SLAM technology is the sum of many technologies involved in achieving this goal. Early SLAM was mostly realized by sensors such as sonar and single-line lidar. Since 2000, with the development of computer vision, visual SLAM using cameras has become a research hotspot, and has shown great application value in many fields. SLAM was proposed in 1986, and it has been developed for more than 30 years. From 1986 to 2004, the problem was converted into a state estimation problem, which was solved by means of extended Kalman filter, particle filter and...

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): G06K9/00G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V20/10G06V10/464G06N3/045G06F18/23G06F18/24
Inventor 马鑫军
Owner 的卢技术有限公司
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