prikaz prve stranice dokumenta Planiranje hvatanja i robotska manipulacija
  Download
PDF 2.9 MB
undergraduate thesis
Planiranje hvatanja i robotska manipulacija
Zagreb: University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, 2022. urn:nbn:hr:235:063626
Vučković, Jurica
University of Zagreb
Faculty of Mechanical Engineering and Naval Architecture

Institutional repository: Repository of Faculty of Mechanical Engineering and Naval Architecture University of Zagreb

Cite this document

Vučković, J. (2022). Planiranje hvatanja i robotska manipulacija (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:235:063626

Vučković, Jurica. "Planiranje hvatanja i robotska manipulacija." Undergraduate thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:235:063626

Vučković, Jurica. "Planiranje hvatanja i robotska manipulacija." Undergraduate thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:235:063626

Vučković, J. (2022). 'Planiranje hvatanja i robotska manipulacija', Undergraduate thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, accessed 28 April 2024, https://urn.nsk.hr/urn:nbn:hr:235:063626

Vučković J. Planiranje hvatanja i robotska manipulacija [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture; 2022 [cited 2024 April 28] Available at: https://urn.nsk.hr/urn:nbn:hr:235:063626

J. Vučković, "Planiranje hvatanja i robotska manipulacija", Undergraduate thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, Zagreb, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:235:063626

Metadata
TitlePlaniranje hvatanja i robotska manipulacija
Title (english)Grasp planning and robotic manipulation
AuthorJurica Vučković
MentorMarko Švaco (mentor)
GranterUniversity of Zagreb
Faculty of Mechanical Engineering and Naval Architecture
Zagreb
Defense date and country2022-09-28, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Mechanical Engineering
Abstract
Problem robotskog univerzalnog hvatanja i manipulacije nepoznatih predmeta već se desetljećima pokušava riješiti, ali unatoč naporima brojnih istraživača još uvijek je neriješen. Zbog kompleksnosti problema i velikog broja poremećaja (npr. greška pozicije kontrole robota, šum u slici kamere) koji se javljaju u procesu hvatanja robotskom rukom, teško je analitički definirati i riješiti problem. Zbog toga se zajednica robotičara okrenula na metode strojnog učenja za rješenje problema hvatanja. U sklopu ovog završnog rada izabrana je metoda pronalaženja prostornog položaja hvata na nepoznatim objektima zasnovana na strojnom učenju zvana GPD (engl. Grasp Pose Detection). Izabrana metoda je implementirana na robotski sustav razvijen u simulacijskom okruženju. U radu je detaljno objašnjena konfiguracija i postavljanje svih programa i paketa korištenih za obavljanje zadatka hvatanja nepoznatih objekta koristeći robotsku ruku. Robotski sustav u sklopu ovog završnog rada je razvijen koristeći robotski operativni sustav (ROS) za komunikaciju i povezivanje njegovih funkcionalnih elemenata. Na kraju rada se objašnjava razvijeni upravljački program koji rezultate paketa GPD koristi za uspješno izuzimanje predmeta u simulacijskom okruženju, na temelju dubinske slike tih predmeta pomoću simulirane kamere. Upravljački kod je sinteza svega naučenog i implementiranog u ovome završnom radu.
Abstract (english)
Universal robotic grasping and manipulation is still an unsolved problem in robotics, despite the efforts of researchers in the field which span decades. Because of the complexity of the problem and a lot of unknown variables that can disrupt the grasping process (ex. positional errors of robot arm control, noise in the camera image) of a robotic arm, it is hard to analytically define and solve the problem. Because of that, the robotics community turned to grasping methods based on machine learning algorithms to try and solve the problem. In this undergraduate thesis an approach of finding grasp poses on unknown objects using machine learning techniques called GPD (Grasp Pose Detection) was chosen. The chosen method was implemented on a robotic system developed inside a simulation environment. This thesis explains the configuration and setup of all the used programs and packages for the task of grasping unknown objects using a robotic arm. The robotic system developed for the purposes of this thesis uses the Robot Operating System (ROS) for connections and communication between its functional elements. In the end of this thesis a developed program that uses the results from the GPD package for successful picking of objects inside a simulation environment is explained. This code is the synthesis of everything learned and implemented inside this undergraduate thesis.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:235:063626
Study programmeTitle: Mechanical Engineering
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence Attribution 4.0 International (CC BY 4.0) icon
Created on2022-09-23 09:34:03