Non-contact control method of bionic manipulator based on human motion posture learning

A motion-stance, non-contact technology, applied in manipulators, program-controlled manipulators, manufacturing tools, etc., can solve problems such as inability to apply bionic dexterous hands

Active Publication Date: 2018-12-18
DALIAN UNIV OF TECH
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

It avoids the limitations of wearing devices such as data gloves, and solves the problem that the existing sensor-based control method can only obtain the pose of some joints of the hand, and cannot be applied to bionic dexterous hands with high degrees of freedom.

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
  • Non-contact control method of bionic manipulator based on human motion posture learning
  • Non-contact control method of bionic manipulator based on human motion posture learning
  • Non-contact control method of bionic manipulator based on human motion posture learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0103] Kinect2.0 is used as the acquisition device to acquire RGB-D images, and the acquired images are transmitted to the computer through the USB interface. The manipulator used is the SCHUNK SVH five-finger bionic manipulator.

[0104] Step 1, get the RGB-D image, where the color image is C, and the depth image is D.

[0105] Step 2, initialize parameters. Frame number frame=1, hand scale parameter (hand length L hand , width W hand , θ 1 ~θ 4 , L 1 ~ L 5 ), hand posture parameters (θ mcp_fe , θ pip , θ mcp，aa )k, where k={1,2,3,4,5} corresponds to the five fingers from the little finger to the thumb respectively.

[0106] Step 3, if frame=1, go to step 4, otherwise go to step 5.

[0107] Step 4, detect the hand area I and build a two-dimensional hand model C I and 3D model C H , including the following steps:

[0108] Step 4.1, obtain the binarized hand region image I according to formula (13);

[0109] Step 4.2, use the Sobel operator to extract the hand ...

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 non-contact control method for controlling a five-finger bionic manipulator by learning the movement posture of a human hand, and belongs to the field of intelligent control. This method proposes an adaptive three-dimensional hand modeling method, and according to the three-dimensional hand model, the movement posture of all the nodes of the controller's hand is tracked, and the relationship between the movement posture of the human hand and the action command of the manipulator is established through a mapping algorithm. The corresponding relationship between them enables the controller to control the five-finger manipulator in a natural way. In the case of using RGB-D images, the present invention establishes a three-dimensional hand model to describe the pose parameters of each joint of the human hand, and proposes an improved APSO algorithm to solve the pose parameters, effectively improving the solution efficiency of high-dimensional parameters Convergence speed avoids the limitations of wearable devices such as data gloves, and solves the problem that existing sensor-based control methods can only obtain the poses of some joints of the hand and cannot be applied to bionic dexterous hands with high degrees of freedom.

Description

technical field [0001] The invention belongs to the field of intelligent control, and relates to a non-contact control method for controlling a five-finger bionic manipulator by learning the motion posture of a human hand by using an RGB-D image as an input signal. Background technique [0002] With the continuous expansion of the application scope of robots, robots are playing an increasingly important role in the fields of industrial control and virtual assembly. At the same time, the scenarios and tasks of robot operations are becoming more and more complex. The robot's manipulator is the main device and tool for it to complete various tasks. Simple clamping devices and two-finger manipulators can no longer meet these application requirements. Manipulators are gradually developing into multi-finger, multi-joint and multi-degree-of-freedom mechanical dexterous hands. Although the current five-finger bionic mechanical dexterous hand is getting closer and closer to the human...

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): B25J9/22B25J13/08G06K9/00G06T7/20
CPCB25J9/163B25J9/1697B25J13/08G06V40/107
Inventor 孙怡屈雯魏诗白徐方杨奇峰
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap