Implementation of Software-Assisted Robot Programming for Wire-Arc Directed Energy Deposition

Additive manufacturing (AM) of metal parts has garnered significant attention for its ability to create near-net-shape components with complex geometries that are unachievable through traditional manufacturing methods. Among various AM techniques, wire-arc direct energy deposition (DED) stands out for its capacity to produce large parts at high deposition rates using the gas metal arc welding (GMAW) process with robotic systems to build a part layer-by-layer. For the robot to accurately execute programmed commands for AM, a digital twin—a virtual representation of the real robot—must be created. This allows engineers to pre-plan toolpaths, define parameters, and simulate and optimize the process completely within the virtual environment, reducing waste associated with experimental tests. The development of a digital twin is a complex process, therefore this work sought to develop a comprehensive framework for implementing a digital twin capable of generating and simulating robotic programs for printing complex geometries. This framework developed included the following steps: (1) 3D modeling of all robot components (tables, axis, fixturing) using computer-aided modeling (CAM); (2) Defining physical axis limits for all robot joints to program in XML and integrate into a software-assisted digital twin; (3) Creating an XML-based machine tool definition (.mtd) file that specifies the joint positions, movement vectors, movement priorities, user-defined variables, and process parameters; (4) Developing an XML-based post-processor to convert the virtual-created programs into the robot's operational language; (5) Conducting final experimental testing. This structured approach produced a reliable digital twin that accurately mimics the real robot and can be adapted for future implementations across various robotic configurations. Specific to the system tested with this work, the digital twin successfully simulated the printing of two complex geometries – a logo and a flange. Simulation in the virtual environment allowed for identification of potential defects and experimentation of printing strategy, thereby reducing physical trial-and-error.