Robotic welding digital twins help plan metal joining in space
De-Risking Robotic Welding for In-Space Metal Joining Through Physics-Grounded Digital Twins
Robotics
Summary
Welding metal parts in space is harder than on Earth due to different gravity and pressure, which are not well studied yet. The authors created a computer model that predicts how welding works in space by combining physics and machine learning. Their model runs quickly and matches detailed simulations well, helping to test welding setups and controls before real experiments. This reduces the risks and costs of developing space repair and construction technology.
What this means in practice
- •For spacecraft manufacturing teams: Evaluate and optimize welding parameters and control strategies for in-orbit metal joining before physical trials.
- •For industrial robotic integrators: Simulate and adjust robotic welding paths and power settings near complex seam geometries using fast surrogate models.
Authors
Zhou Daniel Hao, Umer Masood Chaudry, Robert Shaw, Andrew Cheney, Nick Ludford, Mark Sims, Hongbiao Dong
Abstract
Growing interest in in-space servicing, manufacturing, and assembly has renewed attention on reliable metal joining for modular construction and on-orbit repair. Robotic welding is promising, but reduced-gravity and low-pressure operation remains poorly characterised and flight-like tests are limited. The UK Space Agency-supported ISPARK project is developing a physics-grounded digital twin for structured process de-risking. This paper evaluates a model-based prototype combining a finite-volume thermal weld-pool model, Gaussian-process (GP) and neural-network surrogates, and a simulated six-axis manipulator. The reduced model includes solid-liquid phase change, effective thermocapillary transport, and a local-pressure parameter. Gravity is assessed separately through a hydrostatic-to-capillary pressure ratio and does not enter the thermal energy equation. Across repeated five-fold cross-validation, the GP has lower normalised mean absolute error than the neural network for all eight outputs. In 10 held-out boundary cases, 74 of 80 solver responses fall within nominal 90% GP predictive intervals. A three-output GP query takes 249 microseconds, including scaling and predictive variance. In a simulated UR5 L-shaped-seam study, a proportional controller adjusts the synergic power-source setting as travel speed falls near the corner. Finite-volume re-evaluation finds all 12 closed-loop operating points and 9 of 12 open-loop points within an illustrative 1.0-3.0 mm root-side-width band. These simulation results support using the framework to screen welding conditions and control strategies before experimental tests. Thermal-vacuum, targeted reduced-gravity, and robotic hardware-in-the-loop studies are needed to assess the model and controller.