On the application of CMA-ES to biped walk

Vista senzilla d'ítem
dc.contributor.authorSuárez Hernández, Alejandro
dc.contributor.authorMiralles, Carles
dc.contributor.authorMedina, Salvador
dc.date.accessioned2017-10-02T07:47:25Z
dc.date.available2017-10-02T07:47:25Z
dc.date.issued2017
dc.descriptionTechnical Report
dc.description.abstractThis project studies the applicability of Covariance Matrix Adaptation Evolution Strategies to enable a biped robot to walk with no previous knowledge. While most work on the subject is based on Genetic Algorithms, we believe that CMA-ES shows indeed great potential. Our work focuses on the Darwin robot, although it can be easily extrapolated to other humanoids. The evaluation of the solutions' goodness is performed via simulation to keep the real robot away from potential harm.
dc.description.versionPostprint (published version)
dc.format.extent19 p.
dc.identifier.citationSuarez, A., Miralles, C., Medina, S. "On the application of CMA-ES to biped walk". 2017.
dc.identifier.urihttps://hdl.handle.net/2117/108241
dc.language.isoeng
dc.relation.ispartofseriesIRI-TR-17-03
dc.relation.publisherversionhttps://intranet.iri.upc.edu/download/scidoc/1917
dc.rights.accessOpen Access
dc.rights.licensenameAttribution-NonCommercial-NoDerivs 3.0 Spain
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/3.0/es/
dc.subjectÀrees temàtiques de la UPC::Informàtica::Robòtica
dc.subject.lcshRobotics
dc.subject.lemacRobòtica -- Algorismes
dc.subject.otherartificial life
dc.subject.otherhumanoid robots
dc.subject.otherlearning (artificial intelligence). biped walking
dc.subject.otherevolutionary algorithms
dc.subject.othergenetic algorithms
dc.subject.otherevolution strategies
dc.titleOn the application of CMA-ES to biped walk
dc.typeExternal research report
dspace.entity.typePublication
local.citation.authorSuarez, A.; Miralles, C.; Medina, S.
local.identifier.drac21561109

Fitxers

Paquet original

Mostrant 1 - 1 de 1
Miniatura
Nom:
1917-On-the-application-of-CMA-ES-to-biped-walk.pdf
Mida:
10.85 MB
Format:
Adobe Portable Document Format

Col·leccions

Enviament des de DRAC