Positioning method based on frequency domain analysis and convolutional neural network

A technology of convolutional neural network and frequency domain analysis, which is applied in the field of positioning based on frequency domain convolution analysis and product neural network, can solve the problems of difficulty in implementing multi-sensor fusion algorithms and the inability to simply and effectively eliminate the accumulated errors of inertial navigation systems, etc. Achieve the effect of reducing complexity, increasing feature mining, and improving robustness

Active Publication Date: 2020-09-25
UNIV OF ELECTRONIC SCI & TECH OF CHINA
View PDF7 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these methods have certain limitations, mainly in two aspects: 1) Not all types of carriers have obvious motion characteristics that can be used to eliminate the cumulative error of the inertial navigation system, such as unmanned aerial vehicle (UAV), vehicle There is no obvious motion law in the process of motion to achieve the same error correction effect as zero-speed update; 2)

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
  • Positioning method based on frequency domain analysis and convolutional neural network
  • Positioning method based on frequency domain analysis and convolutional neural network
  • Positioning method based on frequency domain analysis and convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] Such as Figure 1-2 As shown, a positioning method based on frequency domain analysis and convolutional neural network includes the following steps:

[0048] S1. Using the inertial measurement unit to collect the original motion data of the carrier;

[0049] S2. Preprocessing the original motion data to obtain corresponding motion data;

[0050] S3. Perform frequency domain transformation on the motion data in the time domain to generate a corresponding spectrum diagram;

[0051] S4. Extract the features in the spectrogram through the convolutional neural network, and output the pose transformation of the current carrier, thereby realizing positioning.

[0052] Step S1 of this embodiment is specifically: fix the inertial measurement unit on the carrier, set the sampling rate to 100 Hz, and sample the reading data of the accelerometer and gyroscope in the inertial measurement unit at the set sampling rate, and use it as the original motion data .

[0053] In the above ...

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 positioning method based on frequency domain analysis and a convolutional neural network. The method comprises the steps: based on an accelerometer and a gyroscope in a carrier inertia measurement unit, generating a spectrogram of discrete time signals through discrete time Fourier transform, designing a convolutional neural network, acquiring the pose transformation of acarrier by mining high latitude characteristics in the spectrogram, and then achieving the carrier positioning. According to the method, under the condition that the type and the position of a sensorcarrier are not limited, only original inertial measurement unit data is sent to the neural network in a sliding window mode, and the positioning function can be accurately and robustly achieved.

Description

technical field [0001] The invention belongs to the technical field of positioning methods, in particular to a positioning method based on frequency domain volume analysis and product neural network. Background technique [0002] The research on positioning technology has a wide range of application scenarios, and providing services with accurate and stable location information is extremely valuable for fields such as retail, autonomous driving, and robot control. [0003] The positioning system can be divided into indoor positioning system and outdoor positioning system according to the specific environment of the application. Among them, outdoor positioning can rely on Global Navigation Satellite System (GNSS), such as Global Positioning System (GPS), Beidou Satellite Navigation System (BDS), Galileo Satellite Positioning System (Galileo) and GLONASS (GLONASS). For indoor scenes. Due to the shielding and interference of building structures on satellite models, smart phon...

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): G01C21/18G06N3/04G06N3/08
CPCG01C21/18G06N3/08G06N3/045
Inventor 肖卓凌杨明堃朱然杜凯洋
Owner UNIV OF ELECTRONIC SCI & TECH OF CHINA
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