Unwrap the Potential of Smart Manufacturing
Shop Talk
Capturing this week's zeitgeist
At $249, the new version of NVIDIA’s Jetson computer is the perfect gift for hobbyists and developers of robots and industrial automation to run sophisticated AI computations directly without connecting to a remote data center. Another option is the Qualcomm RB3 Gen 2 developer kits, designed to put advanced AI edge computing power into the hands of developers everywhere at $399. It’s never too early or late to invest in automation. Happy Holidays!
Assembly Line
This week's most influential Industry 4.0 media.
Vibration Compensation Improves Robot Performance
Any machine that moves deals with vibration, whether it’s a 3D printer, machine tool or robot. It’s typically managed to the point that many do not even realize that vibration is a problem, but with proper strategies to deal with vibration, machines can run faster and more efficiently. Ulendo is an Ann Arbor, Michigan-based startup that produces software solutions for manufacturing automation. The company launched with a product that’s now called Ulendo VC (for “vibration compensation”), which applies an algorithm to fused filament fabrication (FFF) 3D printers, counteracting the machine’s vibration patterns and enabling it to run up to five times faster while maintaining part quality.
Since the launch of Ulendo VC, the company has expanded into laser powder bed fusion (LPBF) additive manufacturing with Ulendo HC (heat compensation), which optimizes the path of an LPBF machine’s laser to reduce heat-induced deformation and stress. It’s also offering Ulendo Calibration-as-a-Service, which applies its vibration compensation to end-users’ extrusion 3D printers (as opposed to working with machine suppliers, as is the case with Ulendo VC). The company is also broadening its reach to include machines outside the 3D printing space with its fourth product: vibration compensation for robots.
Ulendo’s solution actually compensates for a machine’s vibrations instead of working around them. It does this by “tricking the machine,” as Okwudire describes it. Traditionally, Ulendo measures the machine’s vibration patterns and creates a “calibration map” that predicts how the machine will vibrate and calculates how to offset that motion. So, if a machine veers to the left instead of going straight, Ulendo VC-R (vibration compensation for robots) adjusts the path to the right, which cancels out the leftward motion and causes the machine to go straight.
Edmontonian developing 3D printer that could be 'manufacturing plant of the moon'
Madison Feehan, CEO and founder of Space Copy, said she realized that 3D printing could substantially reduce the significant cost and logistic hurdles of sending astronauts back to the moon during her five years as a contract worker for NASA.
Space Copy is developing what we call in situ logistics and additive manufacturing technology. Now, that is a bunch of fancy words for 3D printing, and essentially, it is a materials processing and 3D printing device all in one. Consider it the manufacturing plant of the moon. Essentially, what we’re doing is we’re taking materials like lunar soil and we are breaking it down into a fine powder. Then, we are using that powder to 3D print different kinds of infrastructure, ranging from astronaut habitats to repair parts and tools to repair rovers and anything that you could think of when it comes to living in space long term.
ROKAE XB7L Robots Streamline Apple Packaging with 1.9-Second Cycle Time
Unsupervised multimodal fusion of in-process sensor data for advanced manufacturing process monitoring
Effective monitoring of manufacturing processes is crucial for maintaining product quality and operational efficiency. Modern manufacturing environments often generate vast amounts of complementary multimodal data, including visual imagery from various perspectives and resolutions, hyperspectral data, and machine health monitoring information such as actuator positions, accelerometer readings, and temperature measurements. However, fusing and interpreting this complex, high-dimensional data presents significant challenges, particularly when labeled datasets are unavailable or impractical to obtain. This paper presents a novel approach to multimodal sensor data fusion in manufacturing processes, inspired by the Contrastive Language-Image Pre-training (CLIP) model. We leverage contrastive learning techniques to correlate different data modalities without the need for labeled data, overcoming limitations of traditional supervised machine learning methods in manufacturing contexts. Our proposed method demonstrates the ability to handle and learn encoders for five distinct modalities: visual imagery, audio signals, laser position (x and y coordinates), and laser power measurements. By compressing these high-dimensional datasets into low-dimensional representational spaces, our approach facilitates downstream tasks such as process control, anomaly detection, and quality assurance. The unsupervised nature of our method makes it broadly applicable across various manufacturing domains, where large volumes of unlabeled sensor data are common. We evaluate the effectiveness of our approach through a series of experiments, demonstrating its potential to enhance process monitoring capabilities in advanced manufacturing systems. This research contributes to the field of smart manufacturing by providing a flexible, scalable framework for multimodal data fusion that can adapt to diverse manufacturing environments and sensor configurations. The proposed method paves the way for more robust, data-driven decision-making in complex manufacturing processes.
New Product Introduction
Highlighting new and innovative facilities, processes, products, and services
Ideon Activates REVEAL™ for Caving Solution at Australia’s First Block Cave Mining Operation
Canadian subsurface intelligence leader Ideon Technologies has activated its supernova-powered solution for subsurface cave imaging at Evolution Mining Northparkes Operations in New South Wales, Australia. This is the world’s first commercial in-mine muon tomography system deployment, taking place just three months after the release of the REVEAL™ for Caving solution and successful field trial results.
The Ideon REVEAL™ for Caving solution allows mining companies to remotely monitor progression of cave propagation in 4D, significantly reducing geological uncertainty in understanding the rock volume and how it is evolving in and around the cave over time. It delivers a high-resolution, high-volume, high-velocity Earth model that delineates the location of the cave relative to both the background geology and the target ore body. In turn, this understanding gives mine operators the ability to mitigate emerging geotechnical risk, achieve greater production predictability, and proactively manage cave growth.
New physics sim trains robots 430,000 times faster than reality
On Thursday, a large group of university and private industry researchers unveiled Genesis, a new open source computer simulation system that lets robots practice tasks in simulated reality 430,000 times faster than in the real world. Researchers can also use an AI agent to generate 3D physics simulations from text prompts.
The accelerated simulation means a neural network for piloting robots can spend the virtual equivalent of decades learning to pick up objects, walk, or manipulate tools during just hours of real computer time.
The Genesis platform, developed by a group led by Zhou Xian of Carnegie Mellon University, processes physics calculations up to 80 times faster than existing robot simulators ( like Nvidia’s Isaac Gym). It uses graphics cards similar to those that power video games to run up to 100,000 copies of a simulation at once. That’s important when it comes to training the neural networks that will control future real-world robots.
Tesla begins operation at its South Texas lithium refinery, firing up its kiln for first time
Tesla Inc. has begun operations at its new lithium refinery near Corpus Christi, a plant CEO Elon Musk said eventually would produce enough of the material for batteries in 1 million electric vehicles per year — and could expand beyond that. Tesla North America said raw materials were successfully fed through the plant’s kiln for the first time last week.
At the May 2023 groundbreaking, Musk said the $375 million plant, the first large-scale refinery for battery grade lithium in the U.S., would begin production in 2025.
Business Transactions
This week's top funding events, acquisitions, and partnerships across industrial value chains.
Databricks is Raising $10B Series J Investment at $62B Valuation
Databricks, the Data and AI company, announced its Series J funding. The company is raising $10 billion of expected non-dilutive financing and has completed $8.6 billion to date. This funding values Databricks at $62 billion and is led by Thrive Capital. Along with Thrive, the round is co-led by Andreessen Horowitz, DST Global, GIC, Insight Partners and WCM Investment Management. Other significant participants include existing investor Ontario Teachers’ Pension Plan and new investors ICONIQ Growth, MGX, Sands Capital and Wellington Management.
The company has seen increased momentum and accelerated growth (over 60% year-over-year) in recent quarters largely due to the unprecedented interest in artificial intelligence. To satisfy customer demand, Databricks intends to invest this capital towards new AI products, acquisitions, and significant expansion of its international go-to-market operations. In addition to fueling its growth, this capital is expected to be used towards providing liquidity for current and former employees, as well as pay related taxes. Finally, this quarter marks the first time the company is expected to achieve positive free cash flow.
Fervo Energy Secures Additional $255 Million Funding to Meet Unprecedented Energy Demand
Fervo Energy, the leader in next-generation geothermal development, announced that it has secured $255 million in new funding and capital availability as it continues to build the world’s largest enhanced geothermal systems (“EGS”) power development. The announced funding comprises both opportunistic corporate equity and new debt financing.
Capricorn’s Technology Impact Fund II led the $135 million corporate equity round. Fervo is excited to welcome new investment and continued support from a suite of high-caliber investors, including Breakthrough Energy Ventures, CalSTRS, Congruent Ventures, CPP Investments, DCVC, Devon Energy, Galvanize Climate Solutions, Liberty Mutual Investments, Mercuria, and Sabanci Climate Ventures.
In addition to corporate equity, Fervo secured a $120 million letter of credit and term loan facility from Mercuria, one of the world’s largest independent energy and commodity groups. The facility will provide additional liquidity for Fervo to accelerate the deployment of EGS projects across the U.S. as it seeks to meet skyrocketing demand for clean, firm power.
Haber Raises $44M to Scale Smart Manufacturing Solutions in North America
Haber, a leading smart manufacturing startup, has successfully raised $44 million in its Series C funding round, which included $38 million in equity and $6 million in debt. The funding round was led by Creaegis, BEENEXT, and Accel, demonstrating strong investor confidence in the company’s innovative AI-driven solutions and its vision for global expansion.
This infusion of capital will fuel Haber’s expansion into North America — a region where the demand for cutting-edge manufacturing technology is soaring. Haber’s AI-powered automation solutions are designed to optimize manufacturing processes, enhance efficiency, and minimize environmental impact, transforming industries from the inside out.
Slip Robotics Raises $28M Series B Financing Led by DCVC
Slip Robotics, a next-generation provider of automated truck-loading robots-as-a-service, has raised a $28 million Series B financing led by DCVC, a leading Silicon Valley venture capital firm specializing in deep tech. The investment reflects Slip’s rapid growth with industry-leading customers – including John Deere, GE Appliances, Valeo, and Nissan – who are using SlipBots to load and unload any truck in just 5 minutes.
Slip Robotics is tackling one of the supply chain’s universal inefficiencies: truck loading and unloading. For over 100 years, truck trailers have been loaded and unloaded using the same basic methods. Because of this, truckers spend 23% of their workday idling at loading docks waiting for forklift operators to load and unload their trailers, while forklift operators rush to carry freight between the trailer and staging area despite 25% of all industrial accidents taking place at the loading dock.
Slip empowers people to auto-load and auto-unload any truck in just 5 minutes using SlipBots. SlipBots cut conventional 30 to 60-minute loading times to just 5 minutes, providing a faster, safer, and scalable solution for growing logistics demands. Unlike other automated solutions, SlipBots handle any type of freight, at any dock, with no infrastructure modification, and with zero Wi-Fi or complex IT integration required. This broad applicability and ease of deployment has fueled Slip’s rapid adoption and expansion with customers.
Valeo, a multinational automotive parts manufacturer, used SlipBots to reduce their trailer load/unload times from 30 minutes to just 5 minutes. SlipBots seamlessly integrated into their operations, enabling employees to increase dock throughput by 6x, reduce forklift traffic by 8x, and increase safety at the loading dock.
T-robotics Secures $5.4M Seed Funding and Applies AI to How Industrial Robots Understand, Learn, and Adapt to Complex Manufacturing Environments
T-robotics, a developer of programming for any robot using natural language and skill models with no code required, announced it has raised a $5.4 million Seed round, co-led by Emergent Ventures and Engine Ventures with Berkeley Skydeck and Raisewell participating. T-robotics will use the fresh capital to expand its U.S. and E.U. operations. Additionally, the company was selected as one of just two winners, from among a group of more than 100 global applicants, in the ABB AI Startup Challenge.
T-robotics’s software, ActGPT, is robot-agnostic and integrates with a variety of commercial robotic solutions, such as robotic arms or mobile robots, and leverages multi-modal data streams from vision, haptics, and language inputs to empower robots with enhanced problem-solving capabilities.
Mantis Robotics Secures $5M to Redefine Robotics through Physical AI
Mantis Robotics, a pioneer in Physical AI for robotic automation, announced it has secured $5M in new investments led by Emerald Technology Ventures with participation from the Amazon Industrial Innovation Fund. This cash infusion will enable Mantis Robotics to scale its operations, enhance its product portfolio, and expand its presence in the rapidly growing industrial robotics market.
Mantis has entered the next frontier in robotics by creating the first adaptive, human-centered robots. The company makes industrial robotic arms that are inherently safe for human interaction, eliminating the need for costly and complex external safeguards such as fences or laser barriers. Mantis’ technology learns and understands its environment based on sensor data from a proprietary built-in sensor-suite. This enables Mantis Robots to detect obstacles and humans in its direct environment and adjust the robot’s behavior accordingly in real-time, making robot arms safe and autonomous.
The robot’s intuitive deployment software with a 3D real-time Digital Twin, requires no technical or coding knowledge to use, making them accessible for diverse industries and users. By reducing operational complexity, Mantis’ robot systems can result in savings of as much as 75% in a typical robotic deployment. The company’s transformative technology has the potential to advance industries far beyond manufacturing and warehousing, paving the way for a new era of human-machine collaboration.
Apptronik Partners with Google DeepMind Robotics to Accelerate Advancement on AI-powered Humanoid Robots
Apptronik, the AI-powered humanoid robotics company, announced it had entered a strategic partnership agreement with the Google DeepMind robotics team. The partnership will bring together best-in-class artificial intelligence with cutting-edge hardware and embodied intelligence, advancing humanoid robots that can be more helpful to people in dynamic environments.
Google DeepMind’s robotics team has been at the forefront of AI and robotics. It brings together world-class expertise from machine learning, engineering, and physics simulation, around the hardest challenges facing the development of AI-powered robots. From pioneering advancements in foundation models to leveraging state-of-the-art AI models like Gemini, Google DeepMind’s robotics team is developing advanced AI systems that can support robotics applications, enabling reasoning and action in the real world.
NXP to Acquire Automotive Networking Pioneer Aviva Links
NXP Semiconductors N.V. (NASDAQ: NXPI), the worldwide leader in automotive processing and networking, announced that it has entered into a definitive agreement to acquire Aviva Links , a provider of Automotive SerDes Alliance (ASA) compliant in-vehicle connectivity solutions in an all-cash transaction valued at $242.5 million.
Aviva Links brings the industry’s most advanced ASA compliant portfolio, supporting SerDes point-to-point (ASA-ML) and Ethernet-based connectivity (ASA-MLE) with data rates up to 16 Gbps. The company has secured design wins at two major automotive OEMs and is sampling its devices to various OEMs and Tier-1 suppliers.