Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Handwashing III: During the Procedure and Post-Procedure Steps01:15

Handwashing III: During the Procedure and Post-Procedure Steps

1.5K
To wash hands properly, follow these steps:
1.5K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Do people feel safe in a robot's presence?

Science robotics·2026
Same author

Shape-morphing metamaterials with continuous relearning.

Science robotics·2026
Same author

Designing microrobots with embodied physical intelligence.

Science robotics·2026
Same author

Origami-inspired grasper for safe tissue manipulation.

Science robotics·2026
Same author

Grasshopper-inspired wing design improves gliding performance.

Science robotics·2026
Same author

Lightweight haptic ring delivers high force feedback.

Science robotics·2026

Related Experiment Video

Updated: Jul 11, 2025

Folding and Characterization of a Bio-responsive Robot from DNA Origami
07:59

Folding and Characterization of a Bio-responsive Robot from DNA Origami

Published on: December 3, 2015

14.6K

Teaching a single-arm robot to fold towels.

Melisa Yashinski1

  • 1Science Robotics, AAAS, Washington, DC 20005, USA.

Science Robotics
|November 15, 2023
PubMed
Summary

Reinforcement learning successfully trained a robot arm to perform complex towel manipulation tasks. This demonstrates the potential for AI in automating household chores and other physical activities.

Area of Science:

  • Robotics
  • Artificial Intelligence
  • Machine Learning

Background:

  • Robotic manipulation of deformable objects, such as textiles, presents significant challenges due to their complex dynamics.
  • Automating household tasks requires robust and adaptable robotic systems.

Purpose of the Study:

  • To investigate the efficacy of reinforcement learning in teaching a robotic arm to perform towel flattening and folding.
  • To develop a control strategy for complex manipulation tasks using a single robotic arm.

Main Methods:

  • A single-arm robotic manipulator was employed.
  • Reinforcement learning algorithms were utilized to train the robot.
  • The robot was tasked with learning to flatten and fold a towel through trial and error.

More Related Videos

Robotic Mirror Therapy System for Functional Recovery of Hemiplegic Arms
10:32

Robotic Mirror Therapy System for Functional Recovery of Hemiplegic Arms

Published on: August 15, 2016

15.5K
Development of a Novel Task-oriented Rehabilitation Program using a Bimanual Exoskeleton Robotic Hand
06:44

Development of a Novel Task-oriented Rehabilitation Program using a Bimanual Exoskeleton Robotic Hand

Published on: May 20, 2020

7.1K

Related Experiment Videos

Last Updated: Jul 11, 2025

Folding and Characterization of a Bio-responsive Robot from DNA Origami
07:59

Folding and Characterization of a Bio-responsive Robot from DNA Origami

Published on: December 3, 2015

14.6K
Robotic Mirror Therapy System for Functional Recovery of Hemiplegic Arms
10:32

Robotic Mirror Therapy System for Functional Recovery of Hemiplegic Arms

Published on: August 15, 2016

15.5K
Development of a Novel Task-oriented Rehabilitation Program using a Bimanual Exoskeleton Robotic Hand
06:44

Development of a Novel Task-oriented Rehabilitation Program using a Bimanual Exoskeleton Robotic Hand

Published on: May 20, 2020

7.1K

Main Results:

  • The robot successfully learned to flatten and fold a towel.
  • The reinforcement learning approach enabled the robot to acquire the necessary motor skills.

Conclusions:

  • Reinforcement learning is a viable method for training robots to perform intricate manipulation tasks on deformable objects.
  • This research paves the way for future advancements in domestic robotics and automation.