Healing Materials by Modernizing the Industrial Edge

Machine learning at the edge is advancing steel production, port efficiency, and farming. Real-time control with Robotics Transformers. Self healing concrete and graphene.

Dec 18, 2022

Assembly Line

Capturing this week's trending industry 4.0 and emerging industrial technology media

U.S. Steel Looks to Forge High-Tech Future at Mills Both New and Old

Date: December 1, 2022
Author: Isabelle Bousquette
Organizations: US Steel

When U.S. Steel took full ownership of Big River last year, it also gained the plant’s artificial intelligence know-how and was a signal of the 120-year-old manufacturing giant’s commitment to advancing technology in its mills. But implementing the type of technology in use at Big River in the steelmaker’s other mills, some of which are over 100 years old, has proven a difficult task, according to the company’s chief information officer.

Big River uses advanced technology to make basic steel mill functions, such as the cooling of hot steel coils, more efficient. If the coils are too close to one another, they take longer to cool, which is why Big River’s machine-learning automated crane is so important. The temperature in parts of the mill can reach 150 degrees during the summer, so keeping things cool can be a challenge. Big River Steel recently installed a slushie machine to help employees cool off.

Big River also uses cameras to feed inputs into machine-learning algorithms that can detect defects in coil slabs or determine whether someone creates a safety hazard by getting too close to certain machines.

Read more at Wall Street Journal (Paid)

Maersk embraces edge computing to revolutionize supply chain

Date: December 12, 2022
Author: Paula Rooney
Organizations: Maersk, Microsoft, Databricks

Gavin Laybourne, global CIO of Maersk’s APM Terminals business, is embracing cutting-edge technologies to accelerate and fortify the global supply chain, working with technology giants to implement edge computing, private 5G networks, and thousands of IoT devices at its terminals to elevate the efficiency, quality, and visibility of the container ships Maersk uses to transport cargo across the oceans.

“Two to three years ago, we put everything on the cloud, but what we’re doing now is different,” Laybourne says. “The cloud, for me, is not the North Star. We must have the edge. We need real-time instruction sets for machines [container handling equipment at container terminals in ports] and then we’ll use cloud technologies where the data is not time-sensitive.”

Laybourne’s IT team is working with Microsoft to move cloud data to the edge, where containers are removed from ships by automated cranes and transferred to predefined locations in the port. To date, Laybourne and his team have migrated about 40% of APM Terminals’ cloud data to the edge, with a target to hit 80% by the end of 2023 at all operated terminals. Maersk has also been working with AI pioneer Databricks to develop algorithms to make its IoT devices and automated processes smarter. The company’s data scientists have built machine learning models in-house to improve safety and identify cargo. Data scientists will some day up the ante with advanced models to make all processes autonomous.

AI farming tool from BASF finds fertile ground in Japan's rice country

Date: December 11, 2022
Author: Taito Kurose
Topics: Machine Learning
Vertical: Agriculture

Yamazaki Rice, based near Tokyo in Saitama prefecture, began using BASF’s Xarvio Field Manager system this year with five workers on about 100 hectares of land.

Xarvio provides real-time analysis informed by satellite and weather data. Automated maps customize the amount of fertilizer recommended for each section of the farm. The data is fed to GPS-equipped farm equipment. The AI gives daily suggestions that Yamazaki Rice’s president said helped improve yields by up to 25% in some fields. Xarvio’s machine learning covers more than 10 years of crop data as well as scientific papers worldwide.

Bringing Next-Generation eBeam Technology Out of the Lab and into the Fab

⭐ RT-1: Robotics Transformer for Real-World Control at Scale

Date: December 13, 2022
Authors: Keerthana Gopalakrishnan, Kanishka Rao
Topics: Industrial Robot, Open Source
Organizations: Google

Tweet | Pocket | Instapaper

Major recent advances in multiple subfields of machine learning (ML) research, such as computer vision and natural language processing, have been enabled by a shared common approach that leverages large, diverse datasets and expressive models that can absorb all of the data effectively. Although there have been various attempts to apply this approach to robotics, robots have not yet leveraged highly-capable models as well as other subfields.

Several factors contribute to this challenge. First, there’s the lack of large-scale and diverse robotic data, which limits a model’s ability to absorb a broad set of robotic experiences. Data collection is particularly expensive and challenging for robotics because dataset curation requires engineering-heavy autonomous operation, or demonstrations collected using human teleoperations. A second factor is the lack of expressive, scalable, and fast-enough-for-real-time-inference models that can learn from such datasets and generalize effectively.

To address these challenges, we propose the Robotics Transformer 1 (RT-1), a multi-task model that tokenizes robot inputs and outputs actions (e.g., camera images, task instructions, and motor commands) to enable efficient inference at runtime, which makes real-time control feasible. This model is trained on a large-scale, real-world robotics dataset of 130k episodes that cover 700+ tasks, collected using a fleet of 13 robots from Everyday Robots (EDR) over 17 months. We demonstrate that RT-1 can exhibit significantly improved zero-shot generalization to new tasks, environments and objects compared to prior techniques. Moreover, we carefully evaluate and ablate many of the design choices in the model and training set, analyzing the effects of tokenization, action representation, and dataset composition. Finally, we’re open-sourcing the RT-1 code, and hope it will provide a valuable resource for future research on scaling up robot learning.

Material Evolution: Novel Geopolymers and Machine Learning

World-First Project to 'Self Heal' Cracked Concrete Using Sloppy Sludge Could Save $1.4 Billion Annual Repair Bill to Australia’s Sewer Pipes

Date: December 12, 2022
Author: Megan Craig
Topics: Materials Science, Corrosion

A world-first project led by University of South Australia sustainable engineering expert Professor Yan Zhuge is trialling a novel solution to halt unprecedented levels of corrosion in the country’s ageing concrete pipelines. Self-healing concrete, in the form of microcapsules filled with water treatment sludge, could be the answer.

Corrosive acid from sulphur-oxidising bacteria in wastewater, along with excessive loads, internal pressure and temperature fluctuations are cracking pipes and reducing their life span, costing hundreds of millions of dollars to repair every year across Australia.

“Sludge waste shows promise to mitigate microbial corrosion in concrete sewer pipes because it works as a healing agent to resist acid corrosion and heal the cracks,” Prof Zhuge says.

The role of temperature on defect diffusion and nanoscale patterning in graphene

Date: October 7, 2022
Authors: Ondrej Dyck, Sinchul Yeom, Sarah Dillender, Andrew R. Lupini, Mina Yoon, Stephen Jesse
Topics: Materials Science

Jesse said, “It heals locally, like the (fictitious) liquid-metal T-1000 in Terminator 2: Judgment Day.”

Graphene is of great scientific interest due to a variety of unique properties such as ballistic transport, spin selectivity, the quantum hall effect, and other quantum properties. Nanopatterning and atomic scale modifications of graphene are expected to enable further control over its intrinsic properties, providing ways to tune the electronic properties through geometric and strain effects, introduce edge states and other local or extended topological defects, and sculpt circuit paths. The focused beam of a scanning transmission electron microscope (STEM) can be used to remove atoms, enabling milling, doping, and deposition. Utilization of a STEM as an atomic scale fabrication platform is increasing; however, a detailed understanding of beam-induced processes and the subsequent cascade of aftereffects is lacking. Here, we examine the electron beam effects on atomically clean graphene at a variety of temperatures ranging from 400 to 1000 °C. We find that temperature plays a significant role in the milling rate and moderates competing processes of carbon adatom coalescence, graphene healing, and the diffusion (and recombination) of defects. The results of this work can be applied to a wider range of 2D materials and introduce better understanding of defect evolution in graphite and other bulk layered materials.

Capital Expenditure

Tracking this week's major mergers, partnerships, and funding events in manufacturing and supply chain

Chevron invests in carbon capture and removal technology company, Svante

Date: December 15, 2022
Topics: Funding Event
Organizations: Chevron, Svante, 3M

Chevron New Energies (CNE), a division of Chevron U.S.A. Inc., and Svante announced that Chevron is the lead investor in Svante’s Series E fundraising round, which raised $318 million that will be used to accelerate the manufacturing of Svante’s carbon capture technology.

Since its founding in 2007, Svante has developed carbon capture and removal technology using structured adsorbent beds, known as filters. This funding will support Svante’s commercial-scale filter manufacturing facility in Vancouver, which is anticipated to produce enough filter modules to capture millions of tonnes of carbon dioxide (CO2) per year across hundreds of large-scale carbon capture and storage facilities.

Cirba Solutions Secures Over $200 Million Investment to Grow its Closed Loop EV Battery Recycling & Materials Production in North America

Date: December 15, 2022
Topics: Funding Event

Cirba Solutions, the largest and most comprehensive battery management and materials processor for end-of-life batteries and gigafactory manufacturing scrap, has received a $245 million minority investment from the EQT Infrastructure V fund (“EQT Infrastructure”). The Heritage Group, a fourth-generation family-owned business managing a diverse portfolio of companies focused on a safer, more enriching, and sustainable world, will remain the majority owner of Cirba Solutions.

The new investment from EQT Infrastructure will support the expansion of an existing lithium-ion facility in Lancaster, Ohio that should generate enough recycled battery material to power 200,000 EVs annually when complete. Additionally, Cirba Solutions has announced a new lithium-ion processing site in Eloy, AZ and expects to announce future processing locations that will further enhance the broad battery collection and processing network that Cirba Solutions has already established. Cirba Solutions expects to invest over $1 billion into the market over the next 10 years to continue its leadership and be the largest and most comprehensive battery recycling company in North America.

Germany’s fruitcore robotics bags €23M to helps SMEs adopt digital robots, automation

Constance, Germany-based fruitcore robotics, a provider of industrial robotics and automation solutions, announced on Wednesday, December 14, that it has secured €23M in a Series B round of funding.

The company says it will use the funds to accelerate product innovation, sales, marketing, and international expansion. The robotic company is active in Germany, Austria, Switzerland, and Italy. “With the closing of our Series B financing, we are again a big step closer to our goal of making robotics and automation solutions accessible to the masses. We see the demand for high-quality and easy-to-use robots in the market is steadily increasing. Therefore, we will use the new capital primarily to serve the demand for our robotics and automation solutions as well as digital products in the European market,” says Jens Riegger, Managing Director (CEO) and co-founder of fruitcore robotics.

Robco links up with $14M led by Sequoia to bring modular robotics to industrial SMBs

Date: December 12, 2022
Author: Ingrid Lunden

Robco, a Munich-based startup that has built a platform for designing low-cost modular robots for small and medium industrial businesses, has picked up €13 million (about $13.8 million). The round — a Series A — is led by Sequoia, with Kindred Capital, Promus Ventures and Torsten Reil, Christian Reber and Daniel Dines all also investing.

Even with the hundreds of millions of dollars that have been poured into a variety of industrial automation and robotics companies over recent years (and some of the very notable failures that have inevitably come out of that) Robco believes that it has found a niche in the market by focusing on tricky tasks and building cost-effective solutions to address the needs of smaller manufacturers. In short, SMBs might need to scale up productivity at times but — either due to the economics of the need, or labor shortages, or both — are unable to hire people to fill those jobs on a permanent basis. This is an area that those making larger machines for bigger industrial clients had yet to address, he said.

Fast Radius assets acquired by SyBridge Technologies in $15.9m deal

Date: December 12, 2022
Author: Oliver Johnson
Topics: Acquisition
Organizations: Fast Radius, SyBridge Technologies

The sale comes a month on from Fast Radius filing for chapter 11 bankruptcy and nine months after its public listing. The sale to SyBridge is subject to bankruptcy court approval and is expected to close before the end of 2022. SyBridge is acquiring most of the operating assets of Fast Radius and says that it intends to make offers of employment to a majority of current Fast Radius employees.

According to SyBridge, Fast Radius will continue to operate and build its digital manufacturing and software business under its new owner and will go to market under the Fast Radius brand name. This acquisition is SyBridge’s fourteenth since the company’s inception in 2019.

ZEDEDA and Emerson Expand Relationship to Modernize the Industrial Edge

Date: December 14, 2022
Topics: Funding Event, Partnership
Organizations: ZEDEDA, Emerson

ZEDEDA, the leader in edge orchestration, today announced a strategic investment from Emerson Ventures, the corporate venture capital arm of Emerson (NYSE: EMR), a global technology, software and automation leader. The move extends the relationship between the two companies, as Emerson is also a ZEDEDA customer.

Emerson’s solution integrates ZEDEDA’s edge management and orchestration offer into its DeltaV™ automation system, enabling Emerson customers to extend DeltaV to the distributed edge. This expansion will deliver enhanced OT solutions while also continuing into the IT environment, providing software-defined automation and revolutionizing how customers can deploy and connect workloads within their distributed environments.

Exponential Industry

Discussion about this post