Sensor development and integration for robotized laser welding

D. Iakovou

    Research output: ThesisPhD Thesis - Research UT, graduation UT

    271 Downloads (Pure)


    Laser welding requires fast and accurate positioning of the laser beam over the seam trajectory. The task of accurate positioning of the laser tools is performed by robotic systems. It is therefore necessary to teach the robot the path it has to follow. Seam teaching is implemented in several ways: Offline Programming, Manual Point-to-point, Sensor Guided. The first two processes are time consuming; the second requires constant human interaction, whereas the last one is fast and fully automated. The most commonly used seam detection sensors are based on optical triangulation with a single structured light line. Such sensors impose restrictions on the laser tool orientation in relation to the seam trajectory. The shape of the seam trajectory in combination with the required speed, can force the robot into positioning errors due to robot dynamics. Furthermore, closed looped seam trajectories such as circles or rectangles are not possible to be taught with such sensors. A solution to these problems is given by the seam detection sensor of the integrated laser welding head. The developed perimetric sensor allows the detection and following of seam trajectories without restrictions on the relative position of the welding head. This reduces positioning errors due robot dynamics and allows looped seam trajectories. The developed integrated laser welding head fulfills two additional sensing functionalities: weld inspection and process monitor. The weld inspection sensor measures the weld surface properties and provides quality estimation according to ISO13919. The process monitor sensor allows the detection of the keyhole formation in full penetration welding. The integration of these sensors into one compact laser welding head required the fusion of their optical paths. Automated calibration procedures have been developed for the system’s integration with robots. Finally, a user friendly sensor software application has been developed that allows access and overview of all the sensors’ processes and handles all required data communications.
    Original languageEnglish
    QualificationDoctor of Philosophy
    Awarding Institution
    • University of Twente
    • Meijer, Johannes, Supervisor
    • Jonker, Jan B., Supervisor
    • Aarts, Ronald, Advisor
    Award date5 Feb 2009
    Place of PublicationEnschede, The Netherlands
    Print ISBNs978-90-365-2770-5
    Publication statusPublished - 5 Feb 2009


    • IR-60601
    • Metis-255990

    Cite this