NRI-Small: Virtualized Welding: A New Paradigm for Intelligent Welding Robots in Unstructured Environment

Grants and Contracts Details


Welding is the final stage of manufacturing and US cannot afford to give up its leadership in this field. While industrial welding robots have been in use for several decades, they are pre-programmed actuators with limited, if any, intelligence. As a result welding robots are primarily used in well-controlled environments, such as assembly lines for mass production, in which the work pieces may be accurately prepared and positioned at reasonable costs. Given that manufacturing is moving towards more customized productions, the next generation of welding robots that can intelligently adjust to various welding tasks is urgently needed. Unfortunately, equipping robots with intelligence is challenging. We introduce a new human-machine welding paradigm, which we call “virtualized welding,” that can transfer human intelligence to welding robots. An existing “dumb” welding robot will be augmented with sensors to observe the work piece, as well as its surroundings. These sensors are able to record and reconstruct in 3D the welding process, which includes the working environment, the geometry of the weld pool, the pose of the welding torch, etc. The reconstructed data are transmitted to a control room and visualized with novel augmented reality techniques: A skilled welder can look at the welding process from different angles, as if he/she was right next to the actual welding. Welding parameters can be adjusted by the human (with intelligence) and executed by the robot (with precision). More importantly, the adjustment, together with the reconstructed welding process, will be recorded and analyzed. Novel system modeling techniques will be developed to correlate the human adjustment with the 3D reconstruction of the welding process. In this way, a welding robot can “learn by examples” the know-ledge and experiences of a human welder and make similar intelligent adjustments by itself in the future. Intellectual Merits In order to realize Virtualized Welding, we plan to (a) establish a novel sensing and display platform to record and visualize the welding process in real time. The 3D reconstruction task is particularly challenging because the weld pool is specular and cannot be directly observed due to the strong electric arc; (b) devise and evaluate a number of novel visualization schemes for remote welding and, more importantly, collaborative welding, in which a human welder is mainly monitoring the welding process and making minor adjustment as needed. Our distinctive projection-based techniques are expected to mitigate major problems associated with teleoperation; and (c) formulate novel algorithms to transfer skilled welders’ intelligence, embodied in the captured welding processes that lead to satisfactory welding products, to robotic system control algorithms so that eventually an intelligent welding robot can achieve or even exceed the capability of a skilled welder. Broader Impacts The primary use for this new technology is manufacturing. Successful completion of this proposed project paves the foundation for intelligent welding robots with closed-loop intelligent control. Such a robot can perform high-speed and high-precision welding while allowing more variations in the work pieces and environments. In addition, virtualized welding can be integrated with a mobile platform to allow welding in places that are hazardous or unsuitable for human welders. All these would significantly expand the use of welding robots, reducing manufacturing costs in spite of relatively high wages in the US. The vision outlined in our proposal can be achieved only through inter-disciplinary collaboration. We have assembled a strong team to carry out the research tasks, including experts in welding, 3D modeling and visualization, and human factor engineering.
Effective start/end date9/1/129/1/12


  • National Science Foundation


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.