De-Risking Robotic Welding for In-Space Metal Joining Through Physics-Grounded Digital Twins
The ISPARK project developed a physics-grounded digital twin for robotic welding to reduce risks in in-space metal joining. This framework combines a finite-volume thermal weld-pool model with Gaussian process and neural network surrogates. The system allows for screening welding conditions before physical tests, showing the surrogate model is accurate and fast enough for control loops.