Primjena samoučećeg algoritma za crtanje robotskom rukom robota Baxter

undergraduate thesis

• Title (english): Learning algorithm for drawing using Baxter robotic arm
• Author: Marija Ančić
• Mentor: Marija Seder (mentor)
• Committee member: Marija Seder (predsjednik povjerenstva)
• Committee member: Filip Marić (član povjerenstva)
• Committee member: Luka Petrović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-09-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovome radu opisan je robot Baxter i njegove glavne karakteristike, posebno obraćajući pažnji na one karakteristike koje će biti korištene tijekom izrade projekta. Na samom početku rada opisuje se postupak analize slike pri kojem se koristi Pythonova biblioteka OpenCV. Korištene funkcije iz ove biblioteke su detaljno opisane i prikazani su dobiveni rezultati. Zatim se uvodi pojam dubokog učenja i zajedno s njim opisane su glavne karakteristike neuronskih mreža. Za treniranje neuronske mreže upotrjebljena je MINST baza podataka, a za razvoj algoritma koristi se Pythonova biblioteka Keras. Završni dio rada se bazira na teach-in tehnici učenja snimanja i reproduciranja trajektorija povučenih robotskom rukom. Na samom kraju su prikazani i opisan dobiveni rezultati.

Abstract (english)

In this paper, the robot Baxter and its main features are described, paying particular attention to those characteristic that will be used during the project. At the very beginning of the work, the process of image analysis using the Python OpenCV library is described. The functions used in this library are described in detail and the results are presented. Then the concept of deep learning is introduced and together with it are described the main features of the neural networks. For training neural networks was used MINST database, and for developing algorithm Python library Keras. The final part of the work is based on teach-in technique of recording and reproduction of trajectories withdrawn by robotic hands. At the very end, the results are shown and described.

• Keywords: Robot Baxter; duboko učenje; duboke neuronske mreže; računalni vid; OpenCV; MINST; Keras
• Keywords (english): Robot Baxter; deep learning; deep neural networks; computer vision; OpenCV; MINST; Keras
• Language: croatian
• URN:NBN: urn:nbn:hr:168:191952
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 18:36:24