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

Multi-AUV formation distributed control method based on reinforcement learning algorithm and unknown disturbance observer

A technology of interference observer and distributed control, applied in the field of multi-AUV formation distributed control, can solve problems such as poor control accuracy, achieve the effects of precise control, improve convergence speed, and improve exploration ability

Active Publication Date: 2021-06-11
HARBIN ENG UNIV
View PDF16 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The present invention aims to solve the problem of poor control accuracy in the existing control method for controlling AUV formations

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
  • Multi-AUV formation distributed control method based on reinforcement learning algorithm and unknown disturbance observer
  • Multi-AUV formation distributed control method based on reinforcement learning algorithm and unknown disturbance observer
  • Multi-AUV formation distributed control method based on reinforcement learning algorithm and unknown disturbance observer

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0022] A multi-AUV formation distributed control method based on a reinforcement learning algorithm and an unknown disturbance observer described in this embodiment includes the following steps:

[0023] S1. Establish AUV kinematics model and dynamics model:

[0024] Considering the influence of ocean current disturbance, let the disturbance be d, the dynamic model of AUV:

[0025]

[0026] For the control of the agent on the horizontal plane, the AUV dynamics model is simplified to the horizontal plane motion model. The simplification process is based on several characteristics of a certain type of AUV developed by the Key Laboratory of Underwater Robots of Harbin Engineering University:

[0027] (1) The center of gravity of this type of AUV coincides with the origin of the satellite coordinate system, the center of gravity is located below the buoyancy center and on the same vertical line as the buoyancy center, and it is assumed that gravity and buoyancy are balanced; ...

specific Embodiment approach 2

[0096] A multi-AUV formation distributed control method based on a reinforcement learning algorithm and an unknown disturbance observer described in this embodiment, the establishment process of the AUV horizontal plane kinematics model and dynamics model includes the following steps: AUV kinematics equation: The AUV kinematics equation essentially reflects the transformation relationship between the earth coordinate system and the satellite coordinate system. When an external force acts on the AUV, it will generate linear acceleration and angular acceleration, so that the linear velocity and the angular acceleration of the AUV When the angular velocity changes, in order to understand the final pose change of the AUV in the earth coordinate system caused by the change of linear velocity and angular velocity, the coordinate transformation matrix will be involved.

[0097] When converting from the earth coordinate system (that is, the inertial coordinate system) to the satellite ...

specific Embodiment approach 3

[0127] A multi-AUV formation distributed control method based on a reinforcement learning algorithm and an unknown disturbance observer described in this embodiment, the process of designing an adaptive distributed controller includes the following steps:

[0128] Before designing the structure of the parameter adaptive distributed cooperative control system based on the improved Actor-Critic algorithm, it is necessary to deduce the longitudinal and heading distributed controllers of AUV based on the backstepping method. The backstepping method belongs to the nonlinear control method. Its basic idea is to design the intermediate virtual control quantity according to the Lyapunov theory, and design the feedback control law under the premise of ensuring stability, so as to ensure that the tracking error gradually approaches zero. The mathematical derivation of the longitudinal and heading distributed controllers in the present invention is based on a simplified AUV horizontal mat...

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 multi-AUV formation distributed control method based on a reinforcement learning algorithm and an unknown disturbance observer, and belongs to the technical field of robot control. In order to solve the problem of poor control precision of an existing control method for controlling the AUV formation, AUVs in the multi-AUV formation are controlled by using a longitudinal and heading compound control system; the longitudinal and heading composite control system comprises a longitudinal distributed controller determined based on a longitudinal disturbance observer, a heading distributed controller determined based on a heading disturbance observer, and an Actor-Critic algorithm used for determining the control gain of the controllers. The Actor-Critic algorithm is composed of four networks, namely an Actor current network, an Actor target network, a Critic current network and a Critic target network, and the four networks all use RBF (Radial Basis Function) neural networks. The method is mainly used for controlling the underwater robot.

Description

technical field [0001] The invention relates to a multi-AUV formation distributed control method and belongs to the technical field of robot control. Background technique [0002] Autonomous Underwater Vehicle (AUV), as an important technical means to explore the ocean, has revolutionary applications in marine environment exploration, resource exploration and other fields in recent years, and has attracted widespread attention from all walks of life. AUVs are inseparable from various sensors when performing tasks. Through the real-time feedback information from the sensors, the robot can obtain motion information such as its own position and speed, as well as important data closely related to the task such as the external environment. However, due to the limited detection range of a single AUV sensor, it is obvious that a single AUV cannot complete tasks quickly and efficiently in tasks involving a large range. In order to compensate for the physical limitations of a single...

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): G05D1/06
CPCG05D1/0692Y02P90/02
Inventor 王卓吴淼孙延超邓忠超秦洪德王海鹏杨赫
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