Razvoj sistema strojnega vida za implementacijo ročnega montažnega mesta v pametno tovarno

Križaj, Žan (author), Herakovič, Niko (mentor), Šimic, Marko (co-mentor)

Diplomska naloga obravnava tematiko razvoja sistema strojnega vida za analizo pobiranja sestavnih delov na ročnem montažnem mestu pametne tovarne. Ker je kljub vse večji avtomatizaciji, ročno montažno mesto še vedno pomemben segment v tovarnah, ga je treba opremiti s sistemi, ki omogočajo digitalizacijo. Eden izmed teh sistemov je modul za strojni vid. S tem bo ročno montažno mesto primerno za implementacijo v pametno tovarno. V prvem delu so predstavljene smernice Industrije 4.0 (strojni vid in sorodni sistemi). Prikazan je tudi pregled različnih programskih jezikov in knjižnic za strojni vid. V drugem delu je prikazan razvoj sistema strojnega vida. Sistem smo razvili za računalnik Raspberry Pi. Uporabili smo programski jezik Python ter knjižnico OpenCV. S strojnim vidom smo nadzirali, če je delavec iz zalogovnika vzel sestavni del. To smo dosegli s pomočjo štetja slikovnih točk, ki se spremenijo, ko v zalogovnik seže roka. Rešitev, do katere smo prišli, se je testirala na testnem ročnem montažnem mestu in je delovala zanesljivo in ponovljivo v kontroliranem okolju.

• Language: Slovenian
• Keywords: strojni vid, Industrija 4.0, pametna tovarna, avtomatizacija, ročno montažno mesto, Raspberry Pi
• Work type: Bachelor thesis/paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FS - Faculty of Mechanical Engineering
• Publisher: [Ž. Križaj]
• Year: 2018
• PID: 20.500.12556/RUL-102817
• UDC: 681.518:658.5(043.2)
• COBISS.SI-ID: 16406043
• Publication date in RUL: 08.09.2018

Secondary language

• Language: English
• Title: Development of computer vision module for manual assembly implementation in smart factory

Abstract

Final thesis deals with development of machine vision system for analysis of correct picking of parts on manual assembly place in smart factory. Despite the evergrowing automatization the manual assembly is still an important part of industry and needs to be equipped with systems that enable digitalization. One of those is machine vision system. Then manual assembly can be implemented in smart factory. In the first part guidelines for Industry 4.0 (computer vision and related systems) are described. Then the review of different programming languages and libraries for computer vision are described. In second part the development of machine vision system is presented. System was developed for Raspberry Pi Single Board Computer where we used programming language Python and OpenCV library . We used computer vision to monitor if the worker has picked up a part from the grab container. We have achieved this by counting the pixel change when hand reaches in the grab container. Machine vision module was tested on manual assembly place and worked reliable and repeatable in controlled environment.

• Keywords: computer vision, Industry 4.0, smart factory, automatization, manual assembly, Raspberry Pi