Adaptive Robot Control based on Neural Microcircuits

Adaptive Robot Control based on Neural Microcircuits

Wolfgang Maass (ORCID: 0000-0002-1178-087X)

Disciplines

Computer Sciences (75%); Mathematics (25%)

Keywords

    Robot Control, Neural Microcircuits, Learning, Plasticity, Real-Time Computing, Humanoid Robots

Final report

This project has produced new methods for the control of humanoid robots, based on results from neurobiology about neural circuits and adaptivity in biological organisms. In particular it has produced a new method for the balance control of humanoid robots in the presence of unpredictable forces. Computer simulations show that this method would even allow them to go surfing, i.e., to stand on an unpredictably moving platform (such as a surfboard in the water) and to withstand unpredictable external forces (such as wind). The publication on this new method received the Best Paper Award at the main international conference on humanoid robots, the IEEE Conference on Humanoid Robots in the USA.
Research institution(s)
International project participants

Research Output

  • 1740 Citations
  • 9 Publications
Publications
  • 2011
    Title Towards a theoretical foundation for morphological computation with compliant bodies
    DOI 10.1007/s00422-012-0471-0
    Type Journal Article
    Author Hauser H
    Journal Biological Cybernetics
    Pages 355-370
    Link Publication
  • 2009
    Title Learning complex motions by sequencing simpler motion templates
    DOI 10.1145/1553374.1553471
    Type Conference Proceeding Abstract
    Author Neumann G
    Pages 753-760
    Link Publication
  • 2009
    Title Echo state networks with filter neurons and a delay&sum readout
    DOI 10.1016/j.neunet.2009.07.004
    Type Journal Article
    Author Holzmann G
    Journal Neural Networks
    Pages 244-256
  • 2009
    Title State-dependent computations: spatiotemporal processing in cortical networks
    DOI 10.1038/nrn2558
    Type Journal Article
    Author Buonomano D
    Journal Nature Reviews Neuroscience
    Pages 113-125
  • 2008
    Title A Learning Theory for Reward-Modulated Spike-Timing-Dependent Plasticity with Application to Biofeedback
    DOI 10.1371/journal.pcbi.1000180
    Type Journal Article
    Author Legenstein R
    Journal PLoS Computational Biology
    Link Publication
  • 2007
    Title Biologically Inspired Kinematic Synergies Provide a New Paradigm for Balance Control of Humanoid Robots
    DOI 10.1109/ichr.2007.4813851
    Type Conference Proceeding Abstract
    Author Hauser H
    Pages 73-80
  • 2007
    Title Computational Aspects of Feedback in Neural Circuits
    DOI 10.1371/journal.pcbi.0020165
    Type Journal Article
    Author Maass W
    Journal PLoS Computational Biology
    Link Publication
  • 2007
    Title From memory-based decisions to decision-based movements: A model of interval discrimination followed by action selection
    DOI 10.1016/j.neunet.2007.04.015
    Type Journal Article
    Author Joshi P
    Journal Neural Networks
    Pages 298-311
  • 2011
    Title Biologically inspired kinematic synergies enable linear balance control of a humanoid robot
    DOI 10.1007/s00422-011-0430-1
    Type Journal Article
    Author Hauser H
    Journal Biological Cybernetics
    Pages 235-249
    Link Publication