Vizijska aplikacija za otkrivanje kolizije unutar radnog prostora robota

undergraduate thesis

• Title (english): A vision application for collision detection within the robot workspace
• Author: Lovro Sever
• Mentor: Tomislav Stipančić (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2023-03-01, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

Sve širom upotrebom robota u industriji sve češće se čovjek i robot mogu nalaziti u zajedničkom radnom prostoru. Tema završnog zadatka je izrada vizijskog sustava za sprječavanje kolizije robota i čovjeka. Vizijska aplikacija izrađena je u programskom jeziku Python uz korištenje biblioteka OpenCV i MediaPipe. Biblioteka MediaPipe omogućuje prepoznavanje i lociranje čovjeka ili određenih dijelova čovjeka (npr. šaka, lice, oči, ruka). Kao senzor korišten za prepoznavanje i lociranje čovjeka odabrana je web kamera. Lociranje robota provedeno je kontinuiranom komunikacijom robota i vizijske aplikacije putem TCP protokola. Za potrebe testiranja vizijske aplikacije korišteni je kolaborativni robot UR5. Uz vizijsku aplikaciju dizajnirani je i izrađeni nosač kamere te je provedena analiza opterećenja robota u normalnom radu i prilikom kočenja. U uvodnom djelu rada ukratko se opisuje tema zadatka dok se kroz ostala poglavlja detaljno opisuje postupak izrade vizijske aplikacije, dinamičke analize i izrade nosača kamere.

Abstract (english)

With increasingly common use of robot sin industry, more and more often, man and robot can be located in a common workspace. The topic of the final task is the development of a vision system to prevent the collision between robots and humans. The vision application was created in the Python programming language using the OpenCV and MediaPipe libraries. The MediaPipe library allows you to identify and locate a person or certain parts of a person (e.g., hand, face, eyes, hands). A webcam was chosen as a sensor used to identify and locate a human. Robot locating was carried out through continuous communication of the robot and vision application via the TCP protocol. For the purpose of testing the vision application, the collaborative robot UR5 was used. In addition to the vision application, a camera mount was designed and built and a robot load analysis was carried out in normal operation and during braking. The introductory part of the paper briefly describes the topic of the task, while through other chapters the process of creating a vision application, dynamic analysis and making a camera mount is described in detail.

• Keywords: vizijski sustav; računalni vid; prepoznavanje čovjeka; OpenCV; MediaPipe
• Keywords (english): vision system; computer vision; human recognition; OpenCV; MediaPipe
• Language: croatian
• URN:NBN: urn:nbn:hr:235:654285
• Study programme: Title: Mechanical Engineering; specializations in: Design, Process and Energy Engineering, Production Engineering, Engineering Modelling and Computer Simulation, Marine Engineering, Industrial Engineering and Management, Materials Engineering; Course: Design; 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 resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2023-02-26 11:25:59