Safe collaborative framework for compliant industrial manipulators

doctoral thesis

• Title (croatian): Sigurno kolaborativno okruženje za podatne industrijske manipulatore
• Author: Bruno Marić
• Mentor: Matko Orsag (mentor)
• Committee member: Stjepan Bogdan (predsjednik povjerenstva)
• Committee member: Tamara Petrović (član povjerenstva)
• Committee member: Alan Mutka (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-18, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

This thesis is focused on developing a safe collaborative framework for compliant industrial manipulators. Industrial manipulators have been a cornerstone of modern industry for several decades. They are characterized by their precision, repeatability, and high payload capacity. However, they have very limited flexibility in terms of programming and deployment. For this reason, they are typically found in large-scale industries. On the other hand, as industries evolve and labor shortages become more prevalent, there is a growing need for robots to enter smaller-scale operations and handle delicate tasks that require adaptability and compliance in direct interaction with the environment. This is where collaborative robots have emerged as a promising solution. While collaborative robots offer features that enable soft contact and safe human-robot interaction, off-the-shelf models often lack the precision, repeatability, payload capacity, and reach required by industrial applications. To bridge this this gap, the goal of this thesis is to develop a collaborative framework that combines the strengths of both industrial and collaborative manipulators. By integrating compliant behaviour into industrial manipulators, we aim to enable safe interaction with humans and adaptability to dynamic environments, while retaining the precision, repeatability, and payload capacity essential for industrial tasks. Through this endeavor, we seek to enhance the versatility and applicability of manipulator systems across a wide range of industrial settings.

Abstract (croatian)

Glavni fokus ovog doktorskog rada je razvoj sigurnog kolaborativnog okruženja za podatne industrijske manipulatore. Industrijski manipulatori su temelj moderne industrije već nekoliko desetljeća. Karakteriziraju ih preciznost, izvrsna ponovljivost te visoka nosivosti i doseg. Međutim, njihova fleksibilnost u pogledu programiranja i implementacije je iznimno ograničena. Zbog navedenoga, obično su prisutni u industrijama velikog obujma. Međutim, razvojem industrije te sveprisutnim problemom nedostatka radne snage raste i potreba za robotima primjenjivim u industrijama manjeg obujma te primjenama koje uključuju delikatne zadatke i zahtijevaju prilagodljivost i podatnost u izravnoj interakciji s okolinom. S obzirom na navedeno, kolaborativni roboti postaju sve više popularni. Iako kolaborativni roboti nude značajke koje omogućuju mekani kontakt i sigurnu interakciju čovjeka i robota, modeli koji su trenutno dostupni na tržištu često nemaju preciznost, ponovljivost i nosivost potrebnu u industrijskim primjenama. Kako bi se premostila navedena praznina, cilj ovog rada je razvoj kolaborativnog okruženja koji spaja najbolje značajke industrijskih i kolaborativnih manipulatora. Dodavanjem podatnosti industrijskim manipulatorima, cilj je omogućiti sigurnu interakciju s ljudima kao i mogućnost prilagodbe dinamičkim okolinama zadržavajući preciznost, ponovljivost i nosivost kao bitne faktore inustrijskih zadaća. Ovi napori nastoje proširiti i poboljšati primjenu manipulacijskih sustava u širokom rasponu industrijskih primjena.

• Keywords: Compliance and Impedance Control; Force Control; Contact Modeling; Collaborative Robots in Manufacturing; Human-Centered Robotics; Learning from Demonstration
• Keywords (croatian): Podatno i impedantno upravljanje; Upravljanje silom; Modeliranje kontakta; Kolaborativni roboti u proizvodnji; Robotika usmjerena na čovjeka; Učenje putem demonstracije
• Language: english
• URN:NBN: urn:nbn:hr:168:199058
• Promotion: 2024
• Study programme: Title: Computer Science; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti (Doktor znanosti)
• Type of resource: Text
• Extent: 125 str.
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-08-30 07:27:30