Human-robot cooperative control method based on human body dynamic arm strength estimation model

An estimation model and collaborative control technology, applied in the field of human-computer interaction, can solve the problems of ignoring the nonlinear relationship between joint angles and electromyographic signals, the gap between the prediction accuracy of dynamic joint force estimation accuracy, and the low dynamic muscle force estimation accuracy. The effect of high estimation accuracy, improved real-time performance, and reduced labor intensity

Active Publication Date: 2021-07-02
SOUTH CHINA UNIV OF TECH
View PDF5 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In 2007, FaridMobasser used the fast orthogonal search (FOS) method using the EMG signal of the upper arm muscles related to the rotation of the elbow joint and the angle position and velocity of the elbow joint to study three operations in isometric, isotonic and light loads. Estimation of elbow joint force under conditions, but the estimation accuracy of dynamic muscle force is not high due to the limitation of alternative basis functions
The PCI model considers the dynamics and nonlinearity of the estimation model at the same time. It is an excellent muscle strength estimation model at present, but it adopts the interpolation fitting method for the influence of the joint angle, ignoring the nonlinear relationship between the joint angle and the EMG signal, so the dynamic The accuracy of joint force estimation is significantly diffe...

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
  • Human-robot cooperative control method based on human body dynamic arm strength estimation model
  • Human-robot cooperative control method based on human body dynamic arm strength estimation model
  • Human-robot cooperative control method based on human body dynamic arm strength estimation model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0045] In this embodiment, a human-robot collaborative control method based on the human body dynamic arm force estimation model, such as figure 1 shown, including the following steps:

[0046] 1) Signal acquisition and denoising stage

[0047] Such as figure 2 As shown, the surface electromyographic signals of the front, middle and rear of the deltoid muscle, biceps brachii and triceps brachii are separated by the electromyography sensor, and the shoulder joint rotation angle θ is determined by the goniometer. 1 and elbow joint angle θ 2 , using a force sensor to collect arm strength information, and using the collected data as training data for a human dynamic arm strength estimation model, the human dynamic arm strength estimation model includes a joint rotation compensation model, a long-short-term memory neural network, and this model is a follow-up human-robot The coordinated control of the arm provides arm strength information. The wavelet filtering algorithm is us...

Embodiment 2

[0118] This embodiment is a human-robot collaborative control method based on the human body dynamic arm strength estimation model. Compared with Embodiment 1, the difference is that the human body dynamic arm strength estimation model includes a joint rotation compensation model, data fusion and BP neural network. In the human body dynamic arm strength estimation stage, adopt BP neural network to estimate human arm strength; Described BP neural network structure is:

[0119]

[0120] is the input value of the fth input node of the BP neural network, y f is the value of hidden nodes, m is the number of hidden layer nodes, n is the number of input layer nodes, F h is the value of the output node (estimated arm strength), ω fi is the connection weight between an input node and a hidden node, ω kf is the connection weight between a hidden node and an output node, b in Input layer node threshold, and b hi Hidden layer node threshold.

Embodiment 3

[0122] This embodiment is a human-robot collaborative control method based on the human body dynamic arm strength estimation model. Compared with Embodiment 1, the difference is that the sliding window length N and the forgetting parameter η of the above-mentioned improved root mean square filter are carried out. Bayesian optimization, select the optimal parameters to preserve the delay and preserve the feature information to the greatest extent.

[0123] Using the joint rotation angle θ under no load and the extracted feature signal E LPF The correlation degree |ρ| is used as the evaluation standard for filter parameter selection. The closer |ρ| is to 1, the extracted feature signal E LPF The higher the correlation with the joint rotation angle θ, it indicates that the feature signal contains more useful information at this time, and the filtering delay is small.

[0124]

[0125] Cov(θ,E LPF ) is the feature extraction signal E LPF The covariance with the joint rotati...

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 human-robot cooperative control method based on a human body dynamic arm strength estimation model. The method comprises the following steps that denoising of collected upper limb muscle and corner data is conducted by employing a wavelet filtering algorithm, and Gaussian white noise in original data is removed; amplitude information of denoised electromyographic signals is extracted through an improved root-mean-square filter, and smoothing processing is conducted through a discrete low-pass filter; on the basis of a deep learning algorithm, the relation between the electromyographic signals and joint rotation is obtained, a joint rotation compensation model is established, and joint rotation equivalent electromyographic signals obtained on the basis of the joint rotation compensation model and the extracted electromyographic signal amplitude information are subjected to data fusion; the electromyographic signals obtained after data fusion serve as the input, arm strength information serves as the output, and the human body dynamic arm strength estimation model is trained through a long-short-term memory neural network; and finally, according to estimated arm strength, a PD control algorithm is combined, the displacement of the robot is adjusted, and cooperative operation of the the human body and the robot is finally achieved.

Description

technical field [0001] The invention belongs to the fields of human-computer interaction and artificial intelligence, and in particular relates to a human-robot cooperative control method based on a dynamic arm force estimation model of a human body. Background technique [0002] Since human-robot collaboration has higher flexibility than robots working independently, the relationship between humans and robots will be closer in the future, and humans and robots will share information and jointly complete complex tasks. At present, a widely used human-computer interaction interface is the force sensor, which collects the muscle strength information of the human tutor through the force sensor, and controls the robot to realize the intention of the human tutor (Agravante, D.J., Cherubini, A., Bussy, A., Gergondet, P ., & Kheddar, A. (2014). Collaborative Human-Humanoid Carrying Using Vision and HapticSensing. 2014 IEEE International Conference on Robotics and Automation (ICRA),...

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): B25J9/16B25J13/08
CPCB25J9/161B25J9/1661B25J9/1612B25J13/087
Inventor 张铁孙韩磊邹焱飚
Owner SOUTH CHINA 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