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

Map matching algorithm based on deep learning

A map matching and deep learning technology, applied in the field of map matching to improve performance, enhance semantic expression ability, and improve effect

Active Publication Date: 2022-04-12
CHONGQING UNIV
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004]The technical problem of the present invention is that existing map matching algorithms cannot achieve satisfactory matching performance

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
  • Map matching algorithm based on deep learning
  • Map matching algorithm based on deep learning
  • Map matching algorithm based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0025] Embodiment one is basically as attached figure 1 As shown, a map matching algorithm based on deep learning (L2MM for short) is mainly composed of four components, namely, trajectory representation learning and enhancement component, pattern recognition and mining component, trajectory matching and generation component, and joint optimization component; L2MM includes Two working steps, offline training step and online inference step; in the offline training step, these four components work together to train a deep model for map matching; in the online inference step, the trained deep model is used , the trajectory representation learning and enhancement component, trajectory matching and generation component participate in the work and return the mapping result of a given trajectory based on test points.

[0026] In this embodiment, the offline training step includes the following steps:

[0027] A1. For a trajectory based on location points in the training data set, fi...

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 relates to the technical field of map matching, in particular to a map matching algorithm based on deep learning, which comprises four components and two working steps, and the four components comprise a track representation learning and enhancing component, a pattern recognition and mining component, a track matching and generating component and a joint optimization component. The two working steps comprise off-line training and on-line reasoning, in the off-line training step, the four assemblies work cooperatively to train a depth model used for map matching, and in the on-line reasoning step, point-based track inference is input to generate a real driving route. According to the model, the problem that the low-frequency trajectory data quality is poor is solved through trajectory representation learning, the driving route is inferred in a more cost-effective mode through pattern recognition and mining, and the map matching performance is improved.

Description

technical field [0001] The invention relates to the technical field of map matching, in particular to a map matching algorithm based on deep learning. Background technique [0002] In recent years, there have been increasing availability and applications of GPS trajectory big data, such as trajectory similarity computation, trajectory clustering, point of interest (POI) recommendation, and human mobility understanding. [0003] A ubiquitous but not yet well-solved problem is how to derive real driving paths from low-frequency GPS trajectories, also known as map matching. However, affected by many factors, there are usually three key problems in low-frequency GPS trajectories, namely noise, low frequency and non-uniformity. For example, high-rise buildings in an urban environment block GPS satellite signals, resulting in positioning errors and making GPS tracks noisy; due to the limitations of energy consumption and transmission bandwidth, existing GPS positioning equipment ...

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): G06F16/29G06F16/2458G06K9/62G06N20/00
Inventor 孟菲陈超江林丽陈超雄罗均蒲华燕李瑞远古富强郭松涛
Owner CHONGQING UNIV
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