DC ElementWertSprache
dc.contributor.authorHensel, Marc-
dc.date.accessioned2023-08-30T11:35:21Z-
dc.date.available2023-08-30T11:35:21Z-
dc.date.issued2023-07-31-
dc.identifier.urihttp://hdl.handle.net/20.500.12738/14102-
dc.description.abstractDealing with artificial intelligence (AI), I was curious how to let computers learn to solve scrambled Rubik's cubes (3x3x3) and Pocket cubes (2x2x2). But wouldn't it be even more fun to demonstrate results by machines that mechanically solve physical cubes? Image to scramble a cube manually, and put it into the machine to unscramble it for you. In this context, I decided to develop a device for the Pocket cube, and I explicitly wanted to design a low-cost device, which is easy to rebuild and use for anyone interested. The design should be based mainly on 3D-printed parts and use two standard servos as motors, only. The device should be controlled by a Laptop, which connects via an Arduino board providing output pins for the servos. Programming applications should be simple by an easy to use Python software interface. The publication documents the device, consisting of the hardware, software for the Arduino board controlling the servo motors, and the Python software interface. A project repository and the 3D print files are available on GitHub and Thingiverse, respectively.en
dc.language.isoenen_US
dc.subjectReinforcement learningen_US
dc.subjectPocket cubeen_US
dc.subject.ddc004: Informatiken_US
dc.titlePocket cube solver : a motorized device for mini cubesen
dc.typeWorking Paperen_US
dc.description.versionNonPeerRevieweden_US
tuhh.oai.showtrueen_US
tuhh.publication.instituteDepartment Informations- und Elektrotechniken_US
tuhh.publication.instituteFakultät Technik und Informatiken_US
tuhh.publisher.urlhttps://github.com/MarcOnTheMoon/cubes/blob/main/docs/Hensel_2023_PocketCubeSolver.pdf-
tuhh.type.opusResearchPaper-
dc.rights.cchttps://creativecommons.org/licenses/by-nc-nd/3.0/en_US
dc.type.casraiWorking Paper-
dc.type.diniworkingPaper-
dc.type.driverworkingPaper-
dc.type.statusinfo:eu-repo/semantics/publishedVersionen_US
dcterms.DCMITypeText-
datacite.relation.IsSupplementedByhttps://github.com/MarcOnTheMoon/cubes/blob/main/docs/Lanz_2023_DeepReinforcementLearning.pdfen_US
item.creatorGNDHensel, Marc-
item.languageiso639-1en-
item.cerifentitytypePublications-
item.openairecristypehttp://purl.org/coar/resource_type/c_8042-
item.creatorOrcidHensel, Marc-
item.fulltextNo Fulltext-
item.grantfulltextnone-
item.openairetypeWorking Paper-
crisitem.author.deptDepartment Informations- und Elektrotechnik-
crisitem.author.orcid0009-0005-8888-3761-
crisitem.author.parentorgFakultät Technik und Informatik-
Enthalten in den Sammlungen:Publications without full text
Zur Kurzanzeige

Seitenansichten

646
checked on 28.11.2024

Google ScholarTM

Prüfe

HAW Katalog

Prüfe

Volltext ergänzen

Feedback zu diesem Datensatz


Diese Ressource wurde unter folgender Copyright-Bestimmung veröffentlicht: Lizenz von Creative Commons Creative Commons