Rodriguez Brena, Ismael Valentin (2023) Knowledge Based Autonomous Planning of Industrial Assembly Tasks for Robotized Setups. Dissertation, Technische Universität München.
PDF
- Nur DLR-intern zugänglich
125MB |
Offizielle URL: https://mediatum.ub.tum.de/?id=1661435
Kurzfassung
In recent years, there has been an increasing need for autonomy and flexibility in robotic as- sembly systems. From industrial production to in-space satellite assembly, many applications benefit from a planner that is able to flexibly and automatically assemble different target as- semblies without human assistance and which can be transferred between different robotic systems. This thesis proposes a framework that enables robotic systems to plan and execute the assem- bly of a large variety of target assemblies. The implementation of this framework on different robotic systems shows that this framework can assemble a predefined set of mechanical com- ponents without any human intervention. Algorithms are developed to automate the assembly process starting from a user-friendly tar- get assembly specification, allowing non-experts to use the system. This automation results in a faster way of programming the assembling sequence of new product variants and avoids errors that human programmers may introduce when building the production line for a spe- cific assembly. The objective is to start from a database of building block objects, process descriptions, basic assembly skills of the robot, and the workcell layout. Given these inputs, the system automatically generates the programs for any specific product variant that is then downloaded to a robotic interface. The whole process consists of the following stages: - Generation and management of a knowledge database, including models of objects, processes, and the workcell layout. - Assembly planning on a symbolic level, based on the knowledge database information. - Mapping of the plan to specific robot skills and automatic generation of the robot pro- gram. - Analysis of the solutions on simulation to refine them through symbolic feedback. - Program execution with error recognition and handling. The central development platform is a system with two DLR/KUKA light-weight robot arms, which enable safe human interaction and permit highly sensitive compliant assembly. Also, this robot allows a wide variety of actions; therefore is a powerful tool to demonstrate the capability and flexibility of the system. Additionally, the developed planner system is also demonstrated in two space applications involving the assembly of satellites and telescopes.
elib-URL des Eintrags: | https://elib.dlr.de/195362/ | ||||||||
---|---|---|---|---|---|---|---|---|---|
Dokumentart: | Hochschulschrift (Dissertation) | ||||||||
Titel: | Knowledge Based Autonomous Planning of Industrial Assembly Tasks for Robotized Setups | ||||||||
Autoren: |
| ||||||||
Datum: | 2023 | ||||||||
Erschienen in: | Knowledge Based Autonomous Planning of Industrial Assembly Tasks for Robotized Setups | ||||||||
Referierte Publikation: | Ja | ||||||||
Open Access: | Nein | ||||||||
Seitenanzahl: | 125 | ||||||||
Status: | veröffentlicht | ||||||||
Stichwörter: | Assembly planning, robotics, hybrid planner, industry 4.0 | ||||||||
Institution: | Technische Universität München | ||||||||
HGF - Forschungsbereich: | Luftfahrt, Raumfahrt und Verkehr | ||||||||
HGF - Programm: | Raumfahrt | ||||||||
HGF - Programmthema: | Technik für Raumfahrtsysteme | ||||||||
DLR - Schwerpunkt: | Raumfahrt | ||||||||
DLR - Forschungsgebiet: | R SY - Technik für Raumfahrtsysteme | ||||||||
DLR - Teilgebiet (Projekt, Vorhaben): | R - Projekt Factory of the Future, R - Leichtbau-Robotik [RO] | ||||||||
Standort: | Oberpfaffenhofen | ||||||||
Institute & Einrichtungen: | Institut für Robotik und Mechatronik (ab 2013) > Kognitive Robotik Institut für Robotik und Mechatronik (ab 2013) | ||||||||
Hinterlegt von: | Rodriguez Brena, Ismael Valentin | ||||||||
Hinterlegt am: | 05 Jun 2023 07:27 | ||||||||
Letzte Änderung: | 05 Jun 2023 07:27 |
Nur für Mitarbeiter des Archivs: Kontrollseite des Eintrags