Robotsveising av aluminiumskrog

The project has 4 Ph.D. students who work on the main goal of the project, which is to develop technology and solutions for robotic welding of large aluminium structures, with focus on shipbuilding with aluminium hulls. Ph.D. Student Jaime M. Rider studies the use of laser scanners to localize welding joints for aluminium structures with polarized laser light. This is done with a specialized camera which polarizes the laser light in 4 angles. This has been investigated in experiments, and the results have been presented at the IEEE Sensors Conference. Ph.D. Student Tuan Anh Tran has developed a system for seamless use of CAD data for robot programming of welding operations. This includes synchronization with the welding equipment and online adjustment of welding parameters. This system is currently being tested in the Manulab robot lab at the department. In addition to this Ph.D. Student Xintong Wang works on strength calculation for different solutions for welding of ship hulls. This includes strength calculations for welding involving extruded aluminium panels. The first publication on this is to be submitted for publication by the ed of 2021. A fourth Ph.D. Student, Ola Alstad works on convolutional neural networks (CNN) to eliminate reflections in laser line scanning of welding grooves. In this work the geometry of the parts and the reflection properties of the laser light is modelled in a graphical rendering system. This graphical rendering system is the used to train the CNN based on a large number of simulated workpieces. The first publication on this has been submitted. Experiments in the new Manulab robotic welding lab started in the summer of 2021. The research is done in close contact with the industry partners. There have been several visits to the companies, and the contact is now handled with monthly Teams meetings.

The proposed project will develop methods and technology for digital design and production of aluminium ships. The focus will be on digital design systems and robotic welding of aluminium ship panels, which will lead to significant costs reductions for aluminium shipbuilding. Robotic welding of aluminium using TIG and MIG technology is well developed in the research community and to some degree in industry. There are certain challenges with robotic welding and sensor technologies for aluminium, and it will be investigated how robot welding solutions for aluminium can be implemented in the ship-building industry. The project will also develop methods for modular ship design and production to achieve efficient robotic welding solutions. Moreover, this will include the development of specialized ship designs that are optimized for robotic welding, which will lead to further savings in production cost and time. It is proposed to use an integrated approach where the digital design system transfer design data to the production system, so that robotic welding operations can programmed and executed efficiently. In particular, it is interesting to investigate if welding paths and welding parameters can be included in the design data, so that time consuming programming of welding paths can be eliminated in the production phase.