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

Fast Robotic Imitation Learning for Force-Torque Tasks

A robot and torque sensor technology, applied in the field of fast robot imitation learning, can solve problems such as unsatisfactory static calculation workload

Active Publication Date: 2017-11-03
GM GLOBAL TECH OPERATIONS LLC
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, existing task-execution-based training methods are not ideal in terms of the number of required training examples and the overall static computational effort

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
  • Fast Robotic Imitation Learning for Force-Torque Tasks
  • Fast Robotic Imitation Learning for Force-Torque Tasks
  • Fast Robotic Imitation Learning for Force-Torque Tasks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] Referring to the drawings, in which like reference numerals designate like or similar parts throughout the several views, in figure 1 An exemplary robotic system 10 is shown in FIG. The robotic system 10 includes a robot 12 and a controller (C) 20 programmed to perform the method 100 and thereby train the robot 12 to perform a force-torque task via human-assisted task demonstration. Robot 12 may include a torso 15 , an arm 16 having an end effector 21 , and possibly, in an exemplary anthropomorphic embodiment, a head 19 . End effector 21 may be configured as any suitable device for performing the exemplary task, such as a grasper or an anthropomorphic hand with fingers 22 attached to wrist 24 or other arm segment 17, as shown. Fingers 22 in such embodiments may be motorized fingers, extensions, or other grippers.

[0025] A non-limiting exemplary force-torque work task (used below for simplicity and consistency) is the grasping of an object 18 in the form of a light ...

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

A method of training a robot to autonomously perform a robotic task comprising moving an end effector through a plurality of states of a predetermined robotic task to demonstrate the task to the robot in a set of n training demonstrations. The method includes measuring training data including measuring at least linear force and torque via a force-torque sensor as the end effector moves through a plurality of states. Key features are extracted from the training data, which is segmented as a time series of control primitives. Identify transitions between adjacent time series segments. During autonomous execution of the same task, the controller detects transitions and automatically switches between control modes. A robotic system includes a robot, a force-torque sensor, a controller programmed to execute a method.

Description

technical field [0001] The present invention relates to fast robotic imitation learning for force-torque tasks. Background technique [0002] Robots are electromechanical devices that can manipulate objects using a series of robotic links. Robotic linkages are interconnected by joints, each of which can be driven independently or interdependently by one or more actuators. Each robot joint represents an independent control variable or degree of freedom. An end effector is an end linkage that directly performs a task such as grasping a work tool or stacking parts. Typically, the robot is controlled to desired target values ​​via closed-loop force-, impedance- or position-based control laws. [0003] In manufacturing, there is a need for more flexible manufacturing and processing that can produce new or more varied products with a minimum amount of downtime. To accomplish this goal, the robot platform should be able to quickly adapt itself to new tasks without reprogramming...

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): B25J9/16B25J19/00
CPCB25J9/1664G05B19/423G05B2219/36442Y10S901/03B25J9/163
Inventor J.W.威尔斯D.W.佩顿R.M.尤伦布洛克L.库
Owner GM GLOBAL TECH OPERATIONS LLC
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