Existing data acquisition systems for additive manufacturing are typically proprietary, process specific, and limited to post-process data analysis. They often rely on isolated tools tied to specific hardware or sensors, which prevents synchronized, real-time insights and integration across platforms. InSitu-AM addresses these limitations by introducing a modular, Robot Operating System 2 (ROS 2) based communication and data infrastructure. Every connected device publishes standardized data streams within a shared communication architecture. These streams are time synchronized, interoperable, and recorded into a unified database schema, enabling correlation between machine states, sensor data, and process parameters in real time.
Example of projecting point cloud data from a stereo depth camera within the virtual engironment.
Infrared data taken from a fixed location during a single layer of a build.
Voltage point cloud data over a single layer of a buildin the virtual engironment.
InSitu-AM runs above the AM process, powered by ROS 2 allowing for flexible, data-driven solutions for AM data collection, real-time analysis, and feedback controls. This not only improves how data is collected and shared, but enables advanced analysis, modeling, and true digital twin development over time. InSitu-AM is currently deployed on wire-arc directed energy deposition processes utilizing ABB robots and Fronius weld sources. Development is ongoing to further the capabilities (for this process and hardware) and extend to additional AM processes, sensors/hardware, and motion systems.
Example of high order data streams in between layers. From left to right - Optical camera, Welding voltage point cloud, and infrared images.
Same as above, but during deposition, and without the optical camera feed.