Bryden Wood are a multi-disciplinary company of creative technologists, designers, architects, engineers and analysts working in AEC. As part of their focus on Modern Methods of Construction (MMC) and Design for Manufacture and Assembly (DfMA) they are looking at the automated fabrication of a kit-of-parts for construction known as “Platforms”. As existing users of the HAL Robotics Framework for their initial R&D into manufacturing the critical bracketry for this system they called on HAL Robotics to support their welding trials and extract live data from the process.
To enable this we developed a lightweight application to communicate with an ABB IRC5 controller and monitor signal states, joint positions, TCP positions as well as welding power source current and voltage. This data was stored locally and streamed to a database for later analysis. To test the efficiency of the application, it was successfully run on a Raspberry Pi.
This kind of data can be critical to identify why any issues may have arisen during production. For example, should there be any spatter during the welding process then we can investigate whether there were any variations to process parameters at that moment. If so, then we can go several steps further and identify this kind of defect during runtime and adapt the process accordingly by changing either the tooling or robot parameters. This is a critical step in the creation of robust, self-correcting processes with increased productivity and reduced rework.