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

Underwater vehicle bionic geomagnetic navigation method based on deep reinforcement learning

An underwater vehicle and reinforcement learning technology, applied in ground navigation, navigation through speed/acceleration measurement, navigation, etc., can solve problems such as difficulties in obtaining geomagnetic maps, inability to learn navigation, and lack of learning

Inactive Publication Date: 2020-11-27
NORTHWESTERN POLYTECHNICAL UNIV
View PDF8 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method relies on accurate prior geomagnetic maps, but in harsh environments such as underwater ocean currents and hidden reefs, it is extremely difficult to obtain accurate geomagnetic maps
Some researchers have turned their research direction to use magnetic trend for navigation. However, the current navigation method has certain randomness and lack of learning in heading decision-making. When AUV needs to go back and forth between two places (such as transportation, patrol ) cannot gain experience from the previous round, and cannot learn a better path for navigation

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
  • Underwater vehicle bionic geomagnetic navigation method based on deep reinforcement learning
  • Underwater vehicle bionic geomagnetic navigation method based on deep reinforcement learning
  • Underwater vehicle bionic geomagnetic navigation method based on deep reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] The purpose of the present invention is that the traditional evolutionary search algorithm has greater randomness in the selection of the heading angle, and when the AUV is performing tasks such as patrolling and transporting between two places, it cannot gain experience and obtain a better strategy. To solve this problem, a geomagnetic bionic navigation algorithm based on deep reinforcement learning is proposed, which uses deep reinforcement learning to obtain the magnetic trend law, predicts the heading angle of the AUV, and uses the constraints of the boundary function to make the AUV reach the target position faster. When the AUV needs to go back and forth between the two places multiple times, a better and shorter navigation path can be obtained by updating the strategy.

[0037] First, establish the AUV kinematics model.

[0038] Considering an autonomous underwater vehicle (AUV) as a particle in a two-dimensional plane, its equation of motion is:

[0039]

[...

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 provides a geomagnetic bionic navigation method based on deep reinforcement learning. The method comprises the steps: carrying out the course decision through employing a Deep-Q-Networkalgorithm under the condition of having no prior geomagnetic information, guiding an underwater vehicle to navigate to a target point, and improving the search efficiency of an AUV through employing aboundary function. Along with the increase of the number of times that the AUV arrives at the target point, the navigation path is gradually shortened and finally approaches the optimal path.

Description

technical field [0001] The invention relates to a bionic geomagnetic navigation method for underwater vehicles based on deep reinforcement learning, and belongs to the technical field of geomagnetic navigation for underwater vehicles. Background technique [0002] Underwater geomagnetic navigation and positioning has the characteristics of passive, non-radiation, all-time, all-region, etc., and can effectively overcome the disadvantages of traditional underwater navigation systems, such as the defect that the inertial navigation system has errors that accumulate over time, and the satellite navigation system has wireless The lack of rapid attenuation of signals in water is one of the ideal methods to realize real-time, continuous and accurate underwater autonomous navigation of underwater vehicles. [0003] At present, the main direction of geomagnetic navigation is geomagnetic matching navigation. The geomagnetic sensor installed on the carrier is used to measure the geomag...

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): G01C21/08G01C21/16G01C21/20
CPCG01C21/08G01C21/203G01C21/165
Inventor 刘明雍王聪牛云李嘉琦汪培新
Owner NORTHWESTERN POLYTECHNICAL 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