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

Automatic parking method and system based on generative confrontation imitation learning

An automatic parking and parking technology, applied in machine learning, computing models, instruments, etc., can solve problems such as insufficient flexibility of parking and complex calculations, and achieve the effects of speeding up training, improving performance, and improving utilization

Active Publication Date: 2021-08-24
POLIXIR TECH LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Therefore, the technical problem to be solved by the present invention is to overcome the existing problems of complex calculation and insufficient flexibility of parking, thereby providing an automatic parking method and system based on generative confrontation imitation learning with simple calculation and flexible parking

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
  • Automatic parking method and system based on generative confrontation imitation learning
  • Automatic parking method and system based on generative confrontation imitation learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0025] Such as figure 1As shown, this embodiment provides an automatic parking method based on generative adversarial imitation learning, including step S1: obtaining expert sample data for successful parking and first-round sample data for parking failure; step S2: using strategy The generator in the learning unit interacts with the environment, collects the first round of sample data in the learning process, samples the first round of sample data and processes and extracts the first round of sample feature data, and at the same time, the expert The sample data is sampled and processed to extract the feature data of the successful sample, and the first round of failed sample data is sampled and processed to extract the feature data of the first round of failed samples; Step S3: the first round of sample feature data, the successful sample feature The data and the feature data of the first round of failed samples are respectively input into the discriminator for judgment, and ...

Embodiment 2

[0053] Based on the same inventive concept, this embodiment provides an automatic parking system based on generative adversarial imitation learning, the problem-solving principle of which is similar to the above-mentioned automatic parking method based on generative adversarial imitation learning, and repeated descriptions will not be repeated.

[0054] This embodiment provides an automatic parking system based on generative confrontation imitation learning, including:

[0055] A storage unit, the storage unit includes an expert sample storage unit and a failure sample storage unit, wherein the expert sample storage unit is used to store expert sample data of successful parking, and the failure sample storage unit is used to store parking failure samples data;

[0056] The image processing unit is used to process high-dimensional image data and extract features from the parking trajectory image;

[0057] The strategy learning unit learns a parking strategy similar to an exper...

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 present invention relates to an automatic parking method and system based on generative confrontational imitation learning. The original image data of the parking track is used to generate a corresponding parking strategy based on generative confrontational imitation learning, and the generated parking strategy is used in the actual parking process. The resulting parking trajectory in should be similar to a successful parking trajectory. Since this application is an online learning algorithm, it will conduct many experiments while learning, and will experience many failures before successfully learning an excellent parking strategy, but these failed parking trajectory data can be stored and used For further learning, this can speed up learning and improve sample utilization. Because the intelligent parking strategy learned in the present invention is not based on rules, but a relatively intelligent strategy, it is capable of automatic parking in different scenarios.

Description

technical field [0001] The present invention relates to the technical field of intelligent parking in automatic driving, in particular to an automatic parking method and system based on generative confrontation imitation learning. Background technique [0002] Standardized parking of vehicles according to the parking space line is conducive to the rational use of urban resource space, the safety of vehicle start and stop, and the planning of city appearance. Due to the huge number of modern cars and the wide variety of actual environments, parking problems often consume people's time and energy, and even cause safety accidents such as friction and collision between vehicles. Therefore, automatic parking systems have room for their existence and development. [0003] At present, most of the parking methods in the field of automatic parking are implemented in a rule-based manner with a large number of hardware support. In the field of artificial intelligence, such as reinforc...

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): B60W30/06G06N20/00
CPCB60W30/06B60W2900/00G06N20/00
Inventor 朱佳成章宗长
Owner POLIXIR TECH 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