Additive Snack · 14 Mar 2023 · 33m · Episode 4
Tim Simpson, a professor at Penn State and NASA consultant, discusses how additive manufacturing (AM) innovations in space, such as those on the Ingenuity helicopter and Perseverance rover, are transferring to terrestrial industries like aviation, medical, and EVtol. He highlights the multidisciplinary nature of his program, which integrates mechanical, industrial, and materials engineering to prepare students for complex challenges. Simpson also explores the intersection of AI and AM, noting how machine learning helps process vast amounts of data for real-time quality control, and introduces the concept of 'Manufacturing Fixation and Design' (MFD) to overcome traditional engineering biases.
Where this conversation describes AI being used to build or deliver a product. Part of the AI in practice tracker.
DesignAI-based algorithms (specifically in Autodesk Fusion 360)better design in a fraction of the time
AI algorithms optimized designs for lightweight structures, producing results that were better than those created by expert human designers in a fraction of the time.
“Let's use an AI-based algorithm to optimize the problem. Goodness, I'm sorry. Your 30 years experience. The AI gave me a better design in a fraction of the time.”
Speaker 2 · Hear it at 21:18
ManufacturingAI and machine learning468 terabytes of data for one build
AI and machine learning are used to analyze massive amounts of sensor data (468 terabytes per build) to understand process signatures, with the goal of enabling real-time closed-loop process control and qualifying parts based on monitoring rather than post-build inspection.
“we're able to gather. 468 terabytes of data for one bill. As a human. I'm not going to look through all of that. And so we have to rely on AI and machine learning to help make sense of the data, help understand the data, and then eventually, once once we built those models, then we can start doing sort of real-time closed loop process control…”
Speaker 2 · Hear it at 21:18
established companies
Other shows on the same ground, closest first.
Hosts Daniel and Forbode discuss three engineering stories: an AI model by NIST that predicts flashover in fires using sensor data and simulations; NASA’s use of ultrasonic additive manufacturing (UAM) via subcontractor Fibrosnic to create complex, dissimilar-metal heat exchangers; and Levitate, a company founded by an amputee in partnership with the Technical University of Denmark, which offers affordable, plug-and-play athletic prosthetics using proprietary composite materials.
Daniel Scott Mitchell and Farbod Moghaddam discuss three technological advancements, starting with a joint MIT, IBM, and Columbia University model that uses WordNet to enable AI abstract reasoning from video. They then examine how EOS and airlines like KLM and Air France are using additive manufacturing for complex, lightweight interior parts, citing Mitchell's experience at Form Labs during the pandemic. Finally, they analyze Penn State research on fruit fly gaze stabilization, discussing how active sensing could improve robotic flight and autonomous delivery systems like Starship Technologies.
Kristen Edwards, CTO and co-founder of Pensa Technologies and MIT PhD candidate, discusses integrating AI into hardware engineering workflows, emphasizing the need to track decision reasoning and manage data governance. She explains that while AI acts as a 'co-pilot' to augment human expertise, high-consequence tasks require human oversight due to liability and explainability concerns. Edwards highlights specific challenges in spatial reasoning, design-for-manufacturing translation, and IP security, suggesting techniques like federated learning and retrieval-augmented generation to protect sensitive engineering data. The conversation covers practical applications in CAD, automated testing, and quality assurance, noting that AI will likely replace repetitive tasks rather than creative engineering roles.
Blake Teipel, Cofounder and CEO of Essentium, explains how additive manufacturing has matured from hobbyist 3D printing to an industrial solution for agile, mid-range production, particularly in aerospace sustainment. He details Essentium's 'Flash Fuse' technology, which uses electromagnetic energy to create isotropic, strong parts from conductive plastics, solving a decades-old strength issue. Teipel discusses the company's pivot to government and aviation clients during the pandemic, the challenges of team composition and knowledge retention, and the critical need for better software tools to enable next-generation design.
Hosts Elliot Williams and Dan Maloney review a diverse range of hardware hacks, including the Mars Ingenuity helicopter's use of commercial Snapdragon processors, a Commodore 1702 monitor restoration using an Alibaba mainboard, and a deep dive into USB-C cable testing with KMOO3C analyzers. They also discuss 3D-printed LED signage using silicone diffusers, ancient Roman nanotech in the Lycurgus Cup, and the rapid evolution of high-speed 3D printing firmware like input shaping. The episode covers inflatable space habitats from Sierra Space and various community projects like flux making and portable antennas.
The summary, chapters and topics are generated from the episode’s audio by this directory, not written by Additive Snack. The episode itself belongs to its publisher.