This EArly-concept Grant for Exploratory Reserch (EAGER) award provides funding for the study of a computer network based manufacturing system that will allow a designer to specify, monitor and control the part fabrication process. Computer networks have evolved into a vital part of our everyday life, transforming commerce, communications, and social networking. However, the tremendous capabilities of computer networks have yet to have a substantial impact on the manufacturing sector. The current generation Internet is not suitable for manufacturing process monitoring and control. One of the major shortcomings is that real-time operation cannot be ensured. To overcome this shortcoming different protocols and architectures will be explored. Leveraging the experimental GENI platform, researchers can experiment with next generation networking protocols to determine how they can be most effectively applied to manufacturing applications.
If successful, the results of this research will lead to identifying the requirements of networks to be applied to manufacturing, and also to improve protocols and architectures to meet these requirements. The work will focus on small-lot manufacturing, which is of prime importance to the innovator. This research would enable the innovator to expand the entrepreneurial culture of the US at lower cost. A user would no longer need to contact a company and wait for them to fabricate and deliver a final product. They could just select a station, send the product information about the product, and control the process until completion. The goal is to have this occur automatically, without any part-specific human intervention or delay.
This project was centered on developing networked manufacturing by remote supervision. In light of that priority, work was focused on: designing application protocols for manufacturing enabled by high-speed networking, extending applications to edge networks (including wireless and sensor networks), defining service architecture for concurrent design and manufacture, and redesigning network layers for meeting guarantees and requirements for manufacturing applications. In order to develop networked manufacturing capabilities we designed and implemented a remotely supervised welding system. Application protocols for remotely accessing the local welding controller, to operate the welder, and remotely monitor the process, via high-resolution cameras, were developed. Guaranteeing bandwidth for both the control flow and monitoring video flow required configuring and implementing network flow and bandwidth control mechanisms. These network flow and bandwidth control mechanisms were implemented using an Openflow-based network. Experiments were conducted to demonstrate the effectiveness of the Openflow-based network for providing bandwidth and data flow guarantees. After extensive testing and calibration a proof-of-concept demonstration was performed. In the demonstration the automated welding system was started remotely. While the process was running, it was monitored remotely via a high-resolution camera and an infrared camera. The remote operator also had the ability to adjust some process parameters. The final proof-of-concept demonstration confirmed that remote manufacturing is fully realizable. The advent of this technology has the potential to open industrial and commercial manufacturing processes to small businesses and entrepreneurs.
