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

Method for learning parameters of neural network, for generating trajectory of exoskeleton and for setting exoskeleton in motion

A neural network and exoskeleton technology, applied in the field of learning the parameters of the neural network, can solve problems such as inconvenient use

Pending Publication Date: 2022-05-10
WANDERCRAFT
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0038] - Currently, calculations are performed on remote servers and the results are then transmitted over the Internet, which forces the user of the exoskeleton to have a permanent and stable connection, which represents a great inconvenience for everyday use

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
  • Method for learning parameters of neural network, for generating trajectory of exoskeleton and for setting exoskeleton in motion
  • Method for learning parameters of neural network, for generating trajectory of exoskeleton and for setting exoskeleton in motion
  • Method for learning parameters of neural network, for generating trajectory of exoskeleton and for setting exoskeleton in motion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0067] architecture

[0068] According to three complementary aspects of the invention, the following are proposed:

[0069] - a method for learning the parameters of a neural network, in particular of the FNN (Feedforward Neural Network, feedforward neural network) type, for generating the trajectory of the exoskeleton 1;

[0070] - a method for generating the trajectory of the exoskeleton 1 (using a neural network, advantageously learned due to the aforementioned method); and

[0071] - A method for setting the exoskeleton 1 in motion (applying the trajectory generated due to the method according to the second aspect).

[0072] Due to the first and / or second server 10a, 10b, these three types of methods are figure 1 implemented within the architecture shown. The first server 10a is a learning server (implementing the first method), and the second server 10b is a server for generating trajectories (implementing the second method). It is entirely possible that the two serv...

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 relates to a method for learning parameters of a neural network for generating a trajectory of an exoskeleton (1), the method comprising implementing, by a data processing device (11a) of a first server (10a), the following steps: (a) a first learning database of periodic trajectories of a set of possible steps of the exoskeleton (1), learning parameters of a first neural network suitable for generating a periodic basic trajectory of the exoskeleton (1), each for a given step of the exoskeleton (1) defined by n-tuple step parameters; (b) using parameters from the first neural network from a second learned database of periodic basic trajectories and transitions of a set of possible steps of the exoskeleton (1), parameters of a second neural network suitable for generating a periodic base trajectory of the exoskeleton (1) and a transition from one periodic base trajectory of the exoskeleton (1) to another periodic base trajectory of the exoskeleton (1) are learned.

Description

technical field [0001] The invention relates to the field of robots of the exoskeleton type. [0002] More precisely, the invention relates to methods for learning the parameters of a neural network, methods for generating trajectories for an exoskeleton using a neural network, and methods for setting an exoskeleton in motion. Background technique [0003] More recently, for people with severe mobility problems (such as paraplegics), assisted walking devices called exoskeletons have emerged, which allow the operator (human user) to link the movements of the exoskeleton to their own The external robotic device is "put on" by a fastening system that rises. Exoskeletons for the lower limbs have several joints (usually at least at the knees and hips) to reproduce walking movements. Actuators make it possible to move these joints, which in turn move the operator. The interface system allows the operator to give commands to the exoskeleton, and the control system translates the...

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): G06N3/04G06N3/08A61H3/00B25J9/00B25J9/16B62D57/032
CPCG06N3/08A61H3/00B25J9/0006B25J9/161B25J9/163B62D57/032A61H2003/007A61H2201/165A61H2201/1642A61H2201/501A61H2205/10G06N3/048G06N3/044G06N3/045B25J9/1664
Inventor 亚历克西斯·杜伯克扬·舍瓦莱尔吉扬·博里斯
Owner WANDERCRAFT
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