Reinforcement learning and optimization based path planning for thin-walled structures in wire arc additive manufacturing

A well-designed deposition path is one of the basic prerequisites for successful fabrication a component by based additive manufacturing processes. Three main approaches are currently used to determine path. First, these general templates that applied entire geometry. Nevertheless, this approach suffers from poor adaptability Second, they algorithms where it necessary divide geometry into sub-parts, which then filled either or paths derived, e.g., signed distance function. These often require human intervention and may fail find suitable Third, there planning strategies deal only with particular topologies, not transferable other geometries. developed framework named RLPlanner, makes use reinforcement learning as well automatized prepossessing Sequential Least Squares Programming optimization method, addresses drawbacks. This solution enables fully automatic thin-walled structures in wire arc manufacturing. In addition, able vary welding speed feed rate thus influence size weld bead leading better