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

Underwater robot trajectory tracking method based on double-BP network reinforcement learning framework

An underwater robot, BP network technology, applied in the direction of instruments, non-electric variable control, height or depth control, etc., can solve the problem of time-consuming and labor-intensive online optimization of controller parameters, and achieve the effect of overcoming time-consuming and labor-intensive

Active Publication Date: 2020-06-05
HARBIN ENG UNIV
View PDF6 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem that the online optimization of controller parameters in the prior art needs to rely on a large amount of expert prior knowledge to establish fuzzy rules, which leads to the time-consuming and labor-consuming problem of online optimization of controller parameters. A Trajectory Tracking Method for Underwater Robots Based on Double BP Network Reinforcement Learning Framework

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 robot trajectory tracking method based on double-BP network reinforcement learning framework
  • Underwater robot trajectory tracking method based on double-BP network reinforcement learning framework
  • Underwater robot trajectory tracking method based on double-BP network reinforcement learning framework

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0026] Specific embodiment one: a kind of underwater robot trajectory tracking method based on double BP network reinforcement learning framework described in the present embodiment, this method comprises the following steps:

[0027] Step 1. Determine the control parameter k to be designed according to the speed of the underwater robot and the control law of the heading control system 1 、k 2 、k 3 and k 4 ;

[0028] Step 2, construct double BP neural network structure, the structure of described double BP neural network comprises current BP neural network and target BP neural network, and the structure of target BP neural network is identical with the structure of current BP neural network;

[0029] The input of the current BP neural network is the current state s t , the input of the target BP neural network is the current state s t Execute the optimal action a t The state s at the next moment obtained after t+1 ; Both the current BP neural network and the target BP ne...

specific Embodiment approach 2

[0040] Specific implementation mode two: the difference between this implementation mode and specific implementation mode one is: the specific process of the step one is:

[0041] The control law of the speed and heading control system of the underwater robot is shown in formulas (1) to (3):

[0042]

[0043]

[0044]

[0045] Among them, τ u is the longitudinal thrust of the thruster, m is the mass of the underwater robot, x u|u| , N r|r| , N r are dimensionless hydrodynamic parameters, v is the lateral velocity of the underwater robot, r is the yaw angular velocity of the underwater robot, |r| is the absolute value of the yaw angular velocity of the underwater robot, u is the longitudinal velocity of the underwater robot, |u| is the absolute value of the longitudinal velocity of the underwater robot, u d is the longitudinal expected velocity of the underwater robot, represents the desired longitudinal acceleration of the underwater robot, τ r is the turning...

specific Embodiment approach 3

[0060] Specific embodiment three: the difference between this embodiment and specific embodiment two is: in the step two, the input of the current BP neural network is the current state s t , the current state s t Expressed as where: x e 、y e as well as Respectively represent the longitudinal position error, lateral position error and heading angle error of the underwater robot to be controlled and the reference underwater robot in the current state;

[0061] Both the current BP neural network and the target BP neural network determine the output action according to the state of the input, and the expression of the output action is a={k' 1 ,k′ 2 ,k' 3 ,k' 4}, where a is the output action;

[0062] Among them, k′ 1 , k' 2 , k' 3 and k' 4 respectively for the action value k′ 10 , k' 20 , k' 30 and k' 40 The result of division, select k′ 10 ∈[-1,1], k′ 20 ∈[-1,1], put k′ 10 and k' 20 Every 0.5 is divided into 5 action values, select k′ 30 ∈[-0.4,0.4], k′ 4...

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 an underwater robot trajectory tracking method based on a double-BP network reinforcement learning framework, and belongs to the technical field of underwater robot trajectorytracking. According to the method, the problems that in the prior art, when online optimization of the controller parameters is carried out, fuzzy rules need to be established depending on a large amount of expert prior knowledge, and consequently time and labor are consumed for online optimization of the controller parameters is solved. According to the invention, continuous interaction with theenvironment can be realized by using a reinforcement learning method, after the strengthening value given by the environment is obtained, and the characteristic of the optimal strategy can be found through loop iteration, a reinforcement learning method is combined with a double BP network, by adjusting the speed of the underwater robot and the relevant parameters of the control law of the headingcontrol system on line, the designed speed and heading control system can select the optimal control parameters corresponding to the environment in different environments, and the problem that in theprior art, time and labor are consumed when controller parameters are optimized on line is solved. The method can be applied to trajectory tracking of the underwater robot.

Description

technical field [0001] The invention belongs to the technical field of track tracking of underwater robots, and in particular relates to a track tracking method of underwater robots based on a dual BP network reinforcement learning framework. Background technique [0002] Underwater robots play an important role in tasks such as marine environment detection, submarine surveying and mapping, and submarine pipeline inspection. sex and economy. Usually, when an underwater robot performs a specific operation, the load of the underwater robot is required to be variable, such as the laying of submarine pipelines, and the sea wind, waves, and currents will change with time and location, making the operating environment of the underwater robot highly unique. Nonlinearity and uncertainty, these factors make it difficult to design a controller that can have good control effects in different situations. Therefore, the control system of the underwater robot should have the ability to ...

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): G05D1/06
CPCG05D1/0692
Inventor 孙延超张佩王卓秦洪德李鹏景锐洁曹禹张宇昂
Owner HARBIN ENG 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