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

A data processing method and device for an end-to-end automatic driving system

A data processing device and automatic driving technology, which is applied in the computer field, can solve the problems of too many images, difficult to store, limit the development of deep learning, etc., and achieve the effect of improving learning efficiency

Active Publication Date: 2020-08-04
BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Due to the large number of images collected in real time in front of the automatic driving system and difficult to store in the limited storage space, the development of deep learning in the field of automatic driving is limited.

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
  • A data processing method and device for an end-to-end automatic driving system
  • A data processing method and device for an end-to-end automatic driving system
  • A data processing method and device for an end-to-end automatic driving system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] In the existing technology in this field, the image collected by the high-precision acquisition vehicle is stored in the HDF5 file for use by machine learning and control software. This method will cause the HDF5 file storing the image to take up more storage space, and will significantly increase the overhead of network I / O, so the traditional data processing method is not conducive to the deep learning of the automatic driving system.

[0044] Therefore, this embodiment proposes another data processing method for an end-to-end automatic driving system, combining figure 2 , including the following steps:

[0045] S210. Convert the multiple images collected in real time into predetermined resolutions and store them in HDF5 files.

[0046] The resolution of the original image data is reduced by 1 / 3 to ensure that the data model can be trained normally in limited storage space. The reduced image data can be stored in HDF5 files with the extension of .h5. And multiple ...

Embodiment 2

[0064] In the existing technology in this field, the image collected by the high-precision acquisition vehicle is stored in the HDF5 file for use by machine learning and control software. This method will cause HDF5 files storing images to occupy more storage space, and will significantly increase the overhead of network I / O. Image storage will also cause too many stored files, which is not conducive to editing and management. Therefore, the traditional data processing method Not conducive to deep learning for autonomous driving systems.

[0065] Although the storage space occupied can be reduced by compressing images, when these files need to be read, an additional decompression process is required, which makes it difficult to improve the efficiency of deep learning. Therefore, this embodiment proposes a data processing method for an end-to-end automatic driving system, combining image 3 , including the following steps:

[0066] S310. Adjust the original image collected in...

Embodiment 3

[0086] In the existing technology in this field, the image collected by the high-precision acquisition vehicle is stored in the HDF5 file for use by machine learning and control software. This method will cause the HDF5 file storing the image to take up more storage space, and will significantly increase the overhead of network I / O, so the traditional data processing method is not conducive to the deep learning of the automatic driving system.

[0087] Therefore, this embodiment proposes yet another data processing device for an end-to-end automatic driving system, combining Figure 5 As shown in , including the following devices:

[0088] A device (hereinafter referred to as "transformation storage device") 510 for converting multiple images collected in real time into a predetermined resolution and storing them in an HDF5 file;

[0089] A device for converting GPS standard time to Coordinated Universal Time (hereinafter referred to as "time conversion device") 520;

[0090...

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

Provided by the present invention are a data processing method and device for an end-to-end automatic driving system, wherein the method comprises: converting a plurality of images acquired in real time into a pre-determined resolution, and then storing the images in a Hierarchical Data Format version 5 (HDF5) file; storing in the HDF5 file the Coordinated Universal Time of a predetermined navigation system, a Gaussian projection corresponding to a speed value extracted from the predetermined navigation system and a curvature value corresponding to Global Positioning System (GPS) data extracted from the predetermined navigation system. By means of storing in an HDF5 file images having a reduced resolution, the Coordinated Universal Time of a navigation system, a Gaussian projection corresponding to a speed value and a curvature value corresponding to GPS data, the present invention may store a large amount of data by using a smaller storage space so as to establish a better automatic driving data model, and thereby increase the learning efficiency of deep learning in the field of automatic driving.

Description

technical field [0001] The invention relates to the field of computers, in particular to a data processing method and device for an end-to-end automatic driving system. Background technique [0002] With the rapid development of deep learning and the in-depth research of artificial intelligence, the automotive industry has undergone revolutionary changes. Realizing automatic driving through end-to-end deep learning is a major research direction in the field of automatic driving. In the prior art, an automatic driving system usually uses a model established by real-time images collected ahead, output steering angle, speed and other data for deep learning. The more data collected, the better the resulting model is for deep learning. Due to the large number of images collected in real time in front of the automatic driving system and it is difficult to store them in the limited storage space, the development of deep learning in the field of automatic driving is limited. Cont...

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 Patents(China)
IPC IPC(8): G06T3/00G06T1/60G06F16/50
CPCG06F16/51G06T1/60G06T3/0056
Inventor 闫泳杉郁浩郑超唐坤张云飞姜雨
Owner BAIDU ONLINE NETWORK TECH (BEIJIBG) CO LTD
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