Detekcija kolizije u stvarnom vremenu i planiranje gibanja robota u kompleksnom radnom okruženju

master's thesis

• Title (english): Real-time collision detection and robot trajectory planning in a complex environment
• Author: Fran Juričan
• Mentor: Zdenko Kovačić (mentor)
• Committee member: Zdenko Kovačić (predsjednik povjerenstva)
• Committee member: Matko Orsag (član povjerenstva)
• Committee member: Tamara Petrović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

U ovom diplomskom radu obrađuje se rekonstrukcija dinamičke okoline pomoću 3D kamera, izbjegavanje kolizije robota i okoline u stvarnom vremenu te planiranje gibanja robota za obavljanje kompleksnih zadataka. Tijekom prve faze izrade diplomskog rada istražene su biblioteke za rad s vokselima i razvijeni su algoritmi za detekciju kolizije u stvarnom vremenu. U idućoj fazi implementirani su algoritmi za izbjegavanje kolizije u dinamičkom okruženju, pri čemu algoritmi razlikuju gibanje koje potencijalno dovodi do kolizije, od potpuno sigurnog gibanja bez kolizije. Posebna pažnja posvećena je na detekciji prisutnosti čovjeka u radnom okruženju robota. U zadnjoj fazi diplomskog rada razvijen je algoritam za planiranje i izvođenje gibanja robota za zadatke skeniranja. Razvijeni algoritmi ispitatni su u simulacijskom okruženju te na stvarnom robotskom sustavu.

Abstract (english)

This thesis deals with the reconstruction of a dynamic environment using 3D cameras, avoiding the collision of robot and the environment in real time and planning the movement of robot for complex tasks. During the first phase of the thesis, libraries for working with voxels were researched and algorithms for real-time collision detection were developed. In the next stage collision avoidance algorithms were implemented, where the algorithms distinguish motion that potentially leads to collision from completely safe motion without collision. Special attention was paid to the detection of human presence in the robot work environment. In the last phase of this thesis, an algorithm for planning and performing robot motion for scanning tasks was developed. The developed algorithms were tested in a simulation environment and on a real robotic system.

• Keywords: Vokseli; Detekcija kolizije; Planiranje trajektorije; KUKA; Kolaborativni robot; PCA
• Keywords (english): Voxels; Collision detection; Trajectory planning; KUKA; Collaborative robot; PCA
• Language: croatian
• URN:NBN: urn:nbn:hr:168:804109
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 12:35:12