Publication alert
New paper alert: Too hot to print, too slow to handle; finding optimal path characteristics for WAAM
In our latest work, "Too hot to print, too slow to handle; finding optimal path characteristics for WAAM" , we developed an automated path planning framework that takes Wire-Arc Additive Manufacturing (WAAM) to a new level. Unlike existing approaches that focus only on thermal control or productivity, we’ve cracked the code on optimizing both.
Our solution? A reinforcement learning agent that generates ultra-efficient paths to minimize deposition time, combined with a Monte Carlo tree search algorithm to ensure precise temperature control. The result: better productivity and improved thermal performance.
