New algorithms automate tool-path design in industrial 3D manufacturing
Product fabrication already took a large step forward with the invention of additive manufacturing, the large-scale industrial counterpart to 3D printing. Now, in a new project funded by Natural Sciences and Engineering Research Council of Canada (NSERC) Alliance and Ontario Centre of Innovation (OCE) Voucher for Innovation and Productivity (VIP) programs, two Ryerson professors are pushing the boundaries of the field even further.
Computer scientist Yeganah Bahoo and mathematician Konstantinos Georgiou are crossing departmental lines and collaborating with CAMufacturing Solutions (external link) , a Canadian firm that develops software used in additive manufacturing. Together, they’re developing new algorithms to automatically design the multi-axis tool-paths that guide industrial 3D printing machines.
“We are really excited about this project,” says Duncan Lam, General Manager at CAMufacturing Solutions. “We have high hopes that it will not only move our software forward, but also be a part of the movement that is bringing additive manufacturing to the mainstream of manufacturing. We want to do our part to put Ontario on the map.”
The results could lead to optimized manufacturing processes, with less cost, less human error, and more precise, high-quality end-products. The innovation may not only give CAMufacturing Solutions a competitive advantage in North America and Europe, but also play a role in developing and retaining Canadian intellectual property, and attract further investment into the field.
Automating tool-path design
Before an industrial 3D printer can fabricate a product, a tool-path that guides the machinery must be programmed. Manufacturing process planners develop the tool-paths for the printers’ control system software. Tools to automate toolpath development exist, but these still require a great deal of human intervention and tuning.
“In this project, we’re automating that entire process of thinking through and designing the tool-path that drives the machines. Our algorithms will create safe, high-quality, precise tool-paths than with minimum user input. This improved automation will alleviate current process bottlenecks, enable greater flexibility in design, and create cost savings by reducing human involvement.”
To solve the problem of automating tool-path design, Bahoo and Georgiou are using mathematical tools, including geometry, graph theory and optimization. Tessellation is a prime technique, in which triangles are tiled together in a mesh that graphically represents a 3D surface.
Initially, Bahoo and Georgiou will design tool-path algorithms for 3D surfaces. Once complete, they’ll move onto 3D volumes, using a technique that departs from the current state-of-the-art.
“Existing techniques currently achieve 3D printing by slicing the target 3D design into very thin 2D layers and then fabricating each layer on top of the previous one,” Bahoo explains. “What we’re now doing is creating algorithms to print arbitrary volumes, in which entire complex, three-dimensional geometries are not sliced and assembled, but printed all together.”
Optimism for cross-disciplinary, industry collaborations
Both Ryerson professors are experts in computational geometry, algorithm design and data structures — Bahoo as a theoretical computer scientist and Georgiou as a mathematician. The current project leverages diverse strengths across the Faculty of Science, and showcases the power and translatability of fundamental knowledge.
Prior to joining the faculty at Ryerson, Bahoo was a postdoctoral fellow at CAMufacturing Solutions working on the foundations of this research. The company’s founders were impressed with her knowledge and enthusiasm for applying mathematical theory for practical problems, and were enthused to keep a relationship with her after her appointment at Ryerson.
“People often assume that it’s not possible for mathematicians or theoretical computer scientists to work with industry — that there’s no practical or immediate application,” Georgiou reflects frankly. “But we’re showing how two theoretical scientists can come together and create something new, with concrete, real-world applications.”
Looking ahead, Georgiou is optimistic about the future of collaborative and partnership research opportunities: “We’re eager to identify common interests across departments, and to apply our expertise to solve complex, real-world problems for industry partners. There’s a lot of potential here at Ryerson, and I’m very proud to be part of this project.”