IndustReal Reinforcement Learning and the Trillion Dollar Grand Slam
Shop Talk
Capturing this week's zeitgeist
NVIDIA began by serving a few Grand Slam breakfasts:
The origin story of the world’s most valuable chip maker, NVIDIA, begins 30 years ago, when three engineers found themselves meeting in a Silicon Valley diner, brainstorming ideas about computing and dreaming up the company that would change their lives over Grand Slam breakfasts.
In fact, Huang says he wouldn’t be the leader he is today if not for one job: waiting tables at a Denny’s.
Assembly Line
This week's most influential Industry 4.0 media
🦾 Transferring Industrial Robot Assembly Tasks from Simulation to Reality
By lessening the complexity of the hardware architecture, we can significantly increase the capabilities and ways of using the equipment that makes it financially efficient even for low-volume tasks. Moreover, the further development of the solution can be mostly in the software part, which is easier, faster and cheaper than hardware R&D. Having chipset architecture allows us to start using AI algorithms - a huge prospective. To use RL for challenging assembly tasks and address the reality gap, we developed IndustReal. IndustReal is a set of algorithms, systems, and tools for robots to solve assembly tasks in simulation and transfer these capabilities to the real world.
We introduce the simulation-aware policy update (SAPU) that provides the simulated robot with knowledge of when simulation predictions are reliable or unreliable. Specifically, in SAPU, we implement a GPU-based module in NVIDIA Warp that checks for interpenetrations as the robot is learning how to assemble parts using RL.
We introduce a signed distance field (SDF) reward to measure how closely simulated parts are aligned during the assembly process. An SDF is a mathematical function that can take points on one object and compute the shortest distances to the surface of another object. It provides a natural and general way to describe alignment between parts, even when they are highly symmetric or asymmetric.
We also propose a policy-level action integrator (PLAI), a simple algorithm that reduces steady-state (that is, long-term) errors when deploying a learned skill on a real-world robot. We apply the incremental adjustments to the previous instantaneous target pose to produce the new instantaneous target pose. Mathematically (akin to the integral term of a classical PID controller), this strategy generates an instantaneous target pose that is the sum of the initial pose and the actions generated by the robot over time. This technique can minimize errors between the robot’s final pose and its final target pose, even in the presence of physical complexities.
🦾 How Robots & Artificial Intelligence are Transforming Unilever's Material Innovation Factory
Each machine at the MIF is designed to crunch colossal amounts of data and maintain consistency across samples and testing. Take Ariana, for example. This robot prepares multiple consistent hair fiber samples in seconds. These perfectly prepped strands are then used for research and testing as part of scientists’ work to create haircare products for Unilever brands. Dove’s Intensive Repair line, now on sale in the UK and the U.S., was developed with Ariana’s assistance, resulting in Unilever’s patented Fibre Repair Actives technology that helps to reconstruct inner hair fibers, reducing breakage and repairing from within.
Artificial intelligence is one advance that’s helping Unilever make progress at pace, allowing scientists to explore vast quantities of data in record time and translate discoveries into new formulas. The vibrant yet fully vegan Hourglass Confession Red Zero lipstick is one such example. Red lipstick is usually formulated using carmine – a pigment requiring over 1,000 crushed beetles per product. But using AI, Unilever’s experts were able to analyze color combinations and possibilities that would have taken millions of physical experiments to replicate. The lipstick launched in 2021.
More than 200 patents were filed between 2020 and 2022 based on data generated at MIF, and Unilever has invested more than €100 million ($123.4 million) in the innovation hub in the past three years.
Improving Image Resolution At The Edge
🖨️ How Will The Apple Reality Pro Headset Boost 3D Printing?
While most AR/VR companies certainly rely on 3D printing to some extent, at least at the level of product design, Apple’s latest product, specifically, may kickstart a niche segment of the industry known as “additively manufactured electronics (AMEs).” To those who have been following the 3D printing industry, the most obvious method for squeezing electronics into small spaces is to use AMEs. With 3D printing, it’s possible to spray conductive traces onto curved surfaces using a technology called Aerosol Jet, from Optomec, which allows electronic features to be incorporated into the structure of a product, rather than force entirely separate components into already tight spaces.
The Sandia National Labs spinout has sold Aerosol Jet printers to Google, Meta, Samsung and has all-but-confirmed that Apple is using the process, as well. By 2016, Taiwanese manufacturer Lite-On Mobile used these systems to spray antennas onto millions of mobile phones before its then-senior manager of Technology Development for Antennas, Henrik Johansson, left to work for Apple.
However, it isn’t Aerosol Jet alone that may be used by these companies to shrink devices. In December 2022, Meta acquired optics firm Luxexcel with a goal of using its lens printing process to create AR glasses. Luxexcel’s method produces optically clear polymers with the ability to integrate waveguides, necessary for transparent displays, into its lenses. It’s no coincidence then that the social media-turned-metaverse giant will be releasing the newest version of its Quest Pro headset late this year, a device said to rival Apple’s Reality Pro.
🛥️ Smooth Sailing: Data’s Role in the Maritime Supply Chain
If a maritime vessel requires unplanned maintenance while transporting cargo, this causes significant delays and disruptions to the on-time arrival of goods. With such an impact on our daily access to goods and the global economy as a whole, the digitalization principles that are revolutionizing critical phases of the supply chain should also be applied to how we maintain and monitor the structural health of the vessels that transport goods across the sea.
Industrial robots can be deployed to scan and gather unprecedented data without the need for scaffolding or man-lifts which require workers to operate at hazardous heights. The collected data points are used to create actionable visualizations and digital twins to understand the vessel’s condition comprehensively. Data-driven decisions provide speed and efficiency, allowing decision makers to spend their time on more valuable initiatives. Digitalization of maintenance and inspections across the maritime supply chain will boost reliability, crew safety, and on-time delivery for fleets to optimize utilization.
3Din30: How Its Made – the Evolution of Tooling
Capital Expenditure
This week's top funding events, acquisitions, and partnerships across industrial value chains
♻️ Carbios bags €54M grant to build the world’s first PET bio-recycling plant in France
France-based Carbios, a green chemistry company that focuses on discovering and developing enzymatic bioprocesses applied to plastic and textile polymers, announced on Thursday that it has received grants totalling €54M. Carbios’ project to finalise the industrialisation of their “unique” PET bio-recycling technology was chosen as part of the national call for projects on “Plastics Recycling” run by ADEME2.
Carbios says it will use the funds to continue its research into the optimisation and continuous improvement of Carbios’ enzymatic technologies.
Red 6 Raises $30M in Series A at a $130M Post Cap Valuation to Solve National Security Gap in Military Training
Red 6, a revolutionary Augmented Reality technology firm at the forefront of synthetic air combat training, is proud to announce it has closed a $30 Million Series A round at a $130M post cap valuation. The round was led with a $25M investment by Snowpoint ventures, an existing investor in Red 6. An additional $5M was raised from existing and new investors. “I am delighted to announce our Series A today. This funding will further support Red 6’s mission to deliver a new paradigm of military training across all the domains of air, sea and land. We are excited to have strengthened our existing relationship with Snowpoint Ventures, who share our commitment to solve training inefficiencies for our warfighters and enhance our military’s level of combat readiness and lethality,” said Daniel Robinson Founder and CEO of Red 6.
🖨️ Intrepid Automation Secures Series A Fundraise to Accelerate Deployment Support Initiatives
Intrepid Automation, provider of industrial additive manufacturing solutions, is thrilled to announce the successful completion of its Series A funding round. The funding will primarily be allocated towards scaling up the company’s planned deployment support capabilities and enhancing customer onboarding processes.
🧪 XtalPi Announces Collaboration with Lilly, Using AI + Robotics to Uncover First-in-class Therapeutics
XtalPi Inc., a pioneering pharmaceutical technology company powered by artificial intelligence (AI) and automation, announced an AI drug discovery collaboration of up to $250 million in upfront and milestone payments with Eli Lilly and Company. The collaboration will leverage XtalPi’s integrated AI capabilities and robotics platform to de novo design and deliver drug candidates for an undisclosed target.
🖨️ Fortify Secures $12.5 Million in Funding From Investors, Including Lockheed Martin Ventures & RTX Ventures, to Accelerate Growth in Advanced Materials & Additive Manufacturing
Fortify, a leading full-stack materials science and additive manufacturing company, announced today that it has raised $12.5 million in a funding round from investors, including Lockheed Martin Ventures and RTX Ventures, the venture capital arms of Lockheed Martin and Raytheon Technologies, respectively. This strategic investment will enable Fortify to expand its capabilities and accelerate the development of its groundbreaking Digital Composite Manufacturing (DCM) platform.
🧠 SparkCognition and Shell Announce a Technology Collaboration Aimed at Accelerating the Pace of Exploration Through the Use of Generative AI
SparkCognition, a global leader in artificial intelligence (AI) software solutions perfected for business, today announced a technology collaboration with Shell to accelerate the pace of imaging and exploration of subsurface structures using generative AI technology. By using advanced AI algorithms to process large amounts of data and automate the analysis, we aim to improve the efficiency and speed of exploration workflows, leading to potentially increased production and higher success rates.
🧠 GE, Waygate Partner On AI-Powered Engine Inspections
GE Aerospace and non-destructive testing specialist Waygate Technologies are partnering to advance engine inspection methods. The companies have signed a joint technology development agreement to develop software and hardware inspection solutions for commercial aircraft engines.
The agreement is aimed at improving the efficiency, accuracy and quality of video borescope inspections, while also minimizing costs for MRO providers. The companies will also explore the use of machine learning and artificial intelligence (AI) within borescope inspections, which could improve reliability, cycle time and consistency of inspections.
🧠 Terrot Adds AI-Upgrade to New Machines with Smartex
Terrot and Smartex.ai are pleased to announce a first-of-its-kind, strategic partnership in which German textile engineering meets proven Artificial Intelligence. As the industry moves towards smart machines that both produce a wide variety of fabrics quickly as well as automate previously manual tasks, such as quality control and data collection, Terrot has strategically adapted its machine production process to ensure all new Terrot and Pilotelli Open-Width machines will be “Smartex-ready”. Smartex-ready means that new Terrot and Pilotelli machine customers can now easily upgrade their machines to add Smartex CORE to take advantage of AI-enabled, automated quality control and gain real-time insights into their machine performance.
🖨️ Materialise and Vuzix Announce Collaboration to Bring Smart Eyewear to Consumers
Vuzix® Corporation (NASDAQ VUZI), a leading supplier of smart glasses and augmented reality (AR) technology and products, and Materialise NV (NASDAQ MTLS), a 3D printing pioneer, have announced a collaborative effort to accelerate the design and production of smart eyewear using 3D printing technology. By integrating 3D printing into the manufacturing process, Vuzix and Materialise aim to accelerate the development of new innovations for enterprise applications, from warehouses to operating rooms. Moreover, the collaboration will enable third-party technology, sports, and entertainment brands to rapidly introduce functional, fashionable, and affordable smart eyewear to consumers, reducing the average manufacturing cycle from 18 months to just 3.