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 Experiment Videos

A modular artificial neural net for controlling a six-legged walking system

H Cruse1, C Bartling, G Cymbalyuk

  • 1Department of Biological Cybernetics, Faculty of Biology, University of Bielefeld, Germany.

Biological Cybernetics
|January 1, 1995
PubMed
Summary

This study introduces a novel neural network model for controlling six-legged robots, inspired by stick insect leg movements. The adaptive gait control system ensures stable walking on irregular terrain.

Related Concept Videos

You might also read

Related Articles

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

Sort by
Same author

A large outbreak of the Kappa mutation of COVID-19 in Cork, Ireland, April-May 2021.

Irish journal of medical science·2022
Same author

Pre-stroke socioeconomic status predicts upper limb motor recovery after inpatient neurorehabilitation.

Annals of medicine·2022
Same author

Upgrade of the gated laser entrance hole imager G-LEH-2 on the National Ignition Facility.

The Review of scientific instruments·2021
Same author

Indirect impact of rotavirus vaccination on viral causes of acute gastroenteritis in the elderly.

Journal of clinical virology : the official publication of the Pan American Society for Clinical Virology·2021
Same author

On-site Multiplex PCR for CSF diagnostics in an Acute Hospital versus Referral to Reference Laboratories: Assessing Economic Factors, Length of Stay and Antimicrobial Stewardship.

The Journal of infection·2020
Same author

Massive clonal expansion of medulloblastoma-specific T cells during adoptive cellular therapy.

Science advances·2019

Area of Science:

  • Robotics
  • Biomimetics
  • Neuroscience

Background:

  • Legged locomotion systems require adaptive control for stable movement over uneven terrain.
  • Previous models often relied on algorithmic approaches, lacking biological realism.
  • Understanding insect locomotion offers insights into robust robotic control.

Purpose of the Study:

  • To develop a biologically inspired neural control model for a six-legged walking system.
  • To integrate realistic kinematics with adaptive gait generation.
  • To replace algorithmic control with artificial neural networks for leg and inter-leg coordination.

Main Methods:

  • A novel model based on biological data from stick insects was developed.
  • Artificial neural networks replaced algorithms for leg state and inter-leg coordination.

Related Experiment Videos

  • Each leg's control comprised three neural subnets: a superior net and two subordinate modules (swing and stance).

    • Main Results:

    • The 'swing module' demonstrated adaptability to various positions, mechanical disturbances, and obstacles.
    • The complete model achieved stable walking at different speeds on irregular surfaces.
    • The control system rapidly recovered a stable gait after disturbances.

    Conclusions:

    • Neural network-based control offers a robust and adaptive solution for legged locomotion.
    • Biologically inspired designs can significantly enhance robotic system performance.
    • The model provides a framework for understanding and replicating insect-like gait control.