John Deere is using Intel’s artificial intelligence technology to help solve an old costly problem in the manufacturing and welding process.
Deere is trialling a solution that uses computer vision to automatically find common defects in the automated welding process in its manufacturing facilities.
Andy Benko, Quality Director of the John Deere Construction and Forestry Department, said: “Welding is a complex process. This artificial intelligence solution has the potential to help us produce high-quality machines more efficiently than before.
“Introducing new technologies into manufacturing is opening up new opportunities and changing our perception of processes that have not changed for many years.”
In 52 factories around the world, John Deere uses the gas metal arc welding (GMAW) process to weld low-carbon steel to high-strength steel to manufacture machines and products. In these factories, hundreds of robotic arms consume millions of pounds of welding wire each year.
With such a large amount of welding, Deere has experience in finding solutions to welding problems and is always looking for new ways to deal with potential problems.
One of the welding challenges commonly felt throughout the industry is porosity, where cavities in the weld metal are caused by air bubbles trapped as the weld cools. The cavity weakens the welding strength.
Traditionally, GMAW defect detection is a manual process that requires highly skilled technicians. In the past, attempts by the entire industry to deal with weld porosity during the welding process were not always successful.
If these defects are found in the later stages of the manufacturing process, the entire assembly needs to be reworked or even scrapped, which can be destructive and costly for the manufacturer.
The opportunity to work with Intel to use artificial intelligence to solve the problem of weld porosity is an opportunity to combine John Deere’s two core values-innovation and quality.
“We want to promote technology to make John Deere’s welding quality better than ever. This is our promise to our customers and their expectations of John Deere,” Benko said.
Intel and Deere combined their expertise to develop an integrated end-to-end hardware and software system that can generate real-time insights at the edge, which exceeds the level of human perception.
When using a neural network-based reasoning engine, the solution will record defects in real time and automatically stop the welding process. The automation system allows Deere to correct problems in real time and produce the quality products that Deere is known for.
Christine Boles, vice president of Intel’s Internet of Things Group and general manager of the Industrial Solutions Group, said: “Deere is using artificial intelligence and machine vision to solve common challenges in robotic welding.
“By leveraging Intel technology and smart infrastructure in the factory, Deere is well positioned to take advantage of not only this welding solution, but also other solutions that may emerge as part of its broader Industry 4.0 transformation .”
The edge artificial intelligence defect detection solution is supported by the Intel Core i7 processor, and uses the Intel Movidius VPU and the Intel OpenVINO toolkit distribution version, and is implemented through the industrial-grade ADLINK machine vision platform and MeltTools welding camera.
Submitted as follows: manufacturing, news tagged with: artificial intelligence, deere, intel, john, manufacturing, process, quality, solutions, technology, welding, welding
Robotics and Automation News was established in May 2015 and is now one of the most widely read websites in this category.
Please consider supporting us by becoming a paid subscriber, through advertising and sponsorship, or purchasing products and services through our store, or a combination of all of the above.
The website and its related magazines and weekly newsletters are produced by a small team of experienced journalists and media professionals.
If you have any suggestions or comments, please feel free to contact us via any email address on our contact page.
The cookie settings on this website are set to “Allow Cookies” in order to provide you with the best browsing experience. If you continue to use this website without changing the cookie settings, or click “Accept” below, you agree.
Post time: May-28-2021
