by Gary Mintchell | Apr 1, 2025 | Networking, Technology
Some final thoughts from the ODVA meetings in March in Florida. I’ve been thinking for some time about complexity. Sometimes we get into the weeds with our technologies, building routine atop routines, solving a multitude of problems through creating others.
The thinking began with considering my reporting from the March 2022 ODVA meetings.
Paul Maurath, Technical Director—Process Automation from Procter & Gamble’s Central Engineering, presented the user’s view of automation. I will dispense with suspense. His conclusion, ”Help us manage complexity.”
Maurath told the story of setting up a test process cell in the lab. They used it to test and demonstrate Ethernet APL devices and the network. They discovered that APL worked, the controller didn’t see any issues. The discouraging discovery was the amount of configuration required and the complexity of setup. He referred to an E&I technician working the shift on a Sunday morning at 3 am. Call comes in. Device is down. With a regular HART / 4-20 mA device, the tech has the tools. But with an Ethernet device configuration can be a problem.
Conclusion:
- There is a need for new technology to deliver functionality and simplicity
- Standards are great
- Please keep end users in mind when developing standards and technology
A paper presented this year by Paul Brooks, Rockwell Automation, Wolfgang Hoeferlin, Endress+Hauser, Sean Vincent, FieldComm Group, and Joakim Wiberg, ODVA discussed the complexities and difficulties following the acquisition of FDT technology by FieldComm Group (FCG). They noted the industrial automation community has the opportunity to enhance our technologies to allow a single device integration standard to be used through discrete, hybrid and process automation disciplines. Double work on business logic and user interface for a device across different technologies and for use in different applications can be eliminated.
In this paper we outline the use cases that FCG – together with ODVA, PNO and OPC Foundation – wish to address. We will look at some of the initial technical assumptions that allow this work to dovetail into device description improvements already underway within ODVA. We will discuss the framework that will allow ODVA members to contribute to and benefit from this work.
Disclaimer: I have not written a line of code in years. As they discussed the details of configuring and programming and data amongst all these technologies, I was impressed by the complexity and the difficulty of the work.
Another end user paper was presented by former GM engineer Gary Workman laying out reasons for some specification changes to define a control network in EtherNet/IP. Similar to the talk by Maurath above, he began with a discussion of the complexity of installing and implementing not individual EtherNet/IP devices instead looking at the network as a whole. He pointed to the problems of electricians and maintenance workers working with a network. His ask was whether ODVA could consider adding guidance for implementing an entire EtherNet/IP control network to help workers on the factory floor.
Most of these meetings discuss the basic technologies and extensions of the product. Whether agreeable or not, the point of view of the end user always serves as a call to step back and consider the problem from their point of view. (Maybe a third of my career was product development—considering the user’s need while developing a product. I sympathize.)
Back to Maurath—complexity is a friction point to the application of technology. It should be the task of the technology provider to remove as much friction as possible.
by Gary Mintchell | Mar 17, 2025 | Generative AI, Technology
Just in from The Association for Advancing Automation (A3) about a timely new conference. I’m not sure I can make it to Seattle for this conference, but it looks like a good place to explore timely topics.
The Association for Advancing Automation (A3), the leading voice in automation and robotics, today announced the launch of a new industry event, FOCUS: Intelligent Vision & Industrial AI Conference. Set to take place September 24-25, 2025, in Seattle, this conference will provide an in-depth look at the latest advancements in machine vision, imaging technologies, AI, and smart automation applications. Attendees will explore cutting-edge innovations in vision systems and imaging while also diving into real-world case studies on AI-driven automation across industries, including manufacturing, aerospace, agriculture, defense, energy, logistics and medical devices.
With AI-powered automation and vision systems rapidly improving quality control, predictive maintenance, and robotics capabilities, industrial leaders need actionable insights to stay ahead of the curve. Unlike broader industry conferences, the FOCUS: Intelligent Vision & Industrial AI conference will center specifically on the real-world applications of AI and vision technology, featuring expert-led sessions, in-depth case studies, and hands-on technology showcases.
Registration opens soon! Stay ahead of the curve—visit the FOCUS 2025 page and subscribe for updates to be among the first to know when registration goes live.
by Gary Mintchell | Mar 14, 2025 | Sensors, Technology
I’ve had a soft spot for visual systems ever since my introduction to the technology in the mid-1980s. Trends of more powerful video sensors plus AI have combined to form a number of interesting new products.
This news comes from a company called Cybord (that I previously wrote about here), who bills itself as “the leading provider of advanced AI-powered electronic component analytics.” They have announced the launch of its Real-Time Interception (RTI) solution, an advanced visual AI-powered software that prevents defective components from being assembled onto Printed Circuit Board Assemblies (PCBAs) in real time. By identifying and discarding faulty components within milliseconds before placement, Cybord’s RTI safeguards product quality, integrity, and compliance while significantly reducing manufacturing waste and costs. The solution, which is already integrated into Fuji’s NXT III placing machines, is currently expanding to manufacturing lines globally.
The solution provides:
- Instantaneous Detection and Rejection: The solution identifies every type of defect and discards defective or unauthorized components in real time before they are assembled onto electronic circuit boards, ensuring only top-quality and approved components are utilized.
- Seamless Manufacturing Integration: The flexible, drop-in software solution easily integrates into existing manufacturing lines to enhance quality control without disrupting production workflows.
- Data-Packed Insights: The platform provides manufacturers with crucial analytics and monitoring and ensures compliance with IPC standards.
- Value-Add for Machine Manufacturers: RTI allows machine manufacturers to empower their EMS customers with enhanced production efficiency and quality control by reducing rework and scrap. EMSs, in turn, prevent unnecessary revenue loss.
The RTI solution builds on Cybord’s successful Quality Component Inspection (QCI) and Traceability Component Inspection (TCI) offerings, addressing a critical need voiced by manufacturers: the ability to prevent faulty components from penetrating the assembly line rather than detecting them post-assembly. The RTI solution has already driven high demand from leading industry players and is currently integrated in Fuji America’s pick-and-place machines.
Powered by a database of nearly five billion components and counting, Cybord utilizes deep learning and AI algorithms to advance the next generation of AI in electronics manufacturing. During placement on the assembly line, the visual AI solution prevents defective, damaged, and counterfeit components from being assembled onto PCBA in real time with 99.9% accuracy. Cybord currently works with industry leaders including Fuji America, Siemens, and Flex.
by Gary Mintchell | Feb 28, 2025 | Technology
A PR person I’ve known for some time recently introduced me to Deepgram. This company’s application of AI is for speech-to-text. I use speech-to-text to dictate thoughts to Apple Notes on my iPhone while out walking in nature. I would certainly welcome all advancements in this area.
Deepgram announced the launch of Nova-3, its most advanced speech-to-text (STT) model to date. Nova-3 is said to be accurate in challenging audio environments. It can be customized for industry-specific needs. The company’s infrastructure includes text-to-speech (TTS) and full speech-to-speech (STS) capabilities.
Nova-3 is engineered for real-time use cases leveraging an advanced latent space architecture to encode complex speech patterns into a highly efficient representation.
Sample use cases:
- Adverse acoustic conditions – Accurately transcribes speech in distant, noisy, and multi-speaker scenarios, making it ideal for air traffic control, drive-thrus, and call centers.
- Real-time multilingual support – Enables real-time transcription across multiple languages—the first model of its kind to do so—making it ideal for emergency response, global customer service, and multilingual operations.
- Industry-specific accuracy – Recognizes domain-specific terminology for specialized fields like medical and legal transcription.
- Precision data handling – Ensures accurate numeric recognition for retail, banking, and finance while supporting real-time redaction of sensitive information for compliance and data privacy.
by Gary Mintchell | Feb 21, 2025 | Process Control, Technology
People often remind me of problems I tried to solve decades ago now made possible through technology advances and creativity.
Some 30 years ago, I tried to concept a system where we could monitor all the electrical power buses in a plant using the data to reveal potential problems with machines along the production line.
It didn’t fly.
One of the few companies that I would expect to be able to put that sort of system into operation would be Schneider Electric. I had the opportunity to interview Manishi Tiwari, Global Director of EcoStruxure Power and Process at Schneider Electric during the recent conference held in Orlando.
Her task is to lead the teams that will enable the sort of system I had envisioned with a customer way back in 1995. The technology now exists. Now the leadership exists to bring it all together.
She told me that about half of her 17-year career at Schneider Electric was in the process business with the other half working in the electrical power business.
Industries require electrical power to operate. But those systems are not built for information exchange between it and process control. Engineers must connect these systems considering latency, reliability, and continuity. Many systems are upgrades to brownfield sites. These most likely require upgrades to electrical equipment such as circuit breakers, relays and other such equipment. How these new components interact with the process must be carefully considered.
Conversely, process system upgrades, say a change in capacity, will affect the electrical power system. These must be studied and considered at design.
Cyber security also must be engineered into the system. Then steps must be taken to improve information to the operator to make their job better.
Schneider Electric not only has expertise in power systems and process systems, but it also has the AVEVA software portfolio to finish the loop.
I must add that in 27 years of interviewing I’ve seldom had a conversation where the other person jumped right in with background, problem statement, possible solutions, and, oh, with a minimum of marketing jargon. And when I sat down to write, I could construct a logical story.
by Gary Mintchell | Jan 30, 2025 | Generative AI, Technology
[I have two websites where I write blogs. This one started in December, 2003. My other one began in 2012. Across the two sites, this is my 7,000th post.]
New companies with new ideas continue to drift into my vision. This news relates to developers and managers who see the potentials for their products if a reliable Voice AI could be added. This technology whets my old product development appetites for new products or enhanced applications. The company is called Deepgram. It has some pretty impressive credentials. If you’re an innovative thinker looking for a way to jump start your application, take a look at this.
Deepgram marketing sent a quick summary of their 2024.
AI Company Ends 2024 Cash-flow Positive with 400+ Enterprise Customers, 3.3x Annual Usage Growth Across the Past Four Years, Over 50,000 Years of Audio Processed, and Over One Trillion Words Transcribed.
The company’s tools include speech-to-text (STT), text-to-speech (TTS), and full speech-to-speech (STS) offerings.
“2024 was a stellar year for Deepgram, as our traction is accelerating and our long-term vision of empowering developers to build voice AI with human-like accuracy, human-like expressivity, and human-like latency is materializing,” said Scott Stephenson, CEO of Deepgram. “Our product strategy from founding has been to focus on deep-tech first, and the work we have done in building 3-factor automated model adaptation, extreme compression on latent space models (LSMs), hosting models with efficient and unrestricted hot-swapping, and symmetrical delivery across public cloud, private cloud, or on-premises, uniquely positions us to succeed in the $50B market for voice AI agents in demanding environments requiring exceptional accuracy, lowest COGS, highest model adaptability, and lowest latency.”
Deepgram expects to end 2025 as the industry’s only end-to-end speech-to-speech solution built to solve the four critical challenges of enterprise-ready voice AI:
- Accuracy / audio perception: Enterprise use cases require high recognition, understanding, and generation of specialized vocabulary in often challenging audio conditions. Deepgram solves this through novel, non-lossy compressions of these spaces for rapid processing paired with generation, training, and evaluation on synthetic data that precisely matches Deepgram customers’ real-world conditions.
- COGS at scale: Deepgram customers need to profitably build and scale voice AI solutions. Deepgram delivers this through its unique latent audio model with extreme compression combined with deep expertise in high-performance computing.
- Latency: Real-time conversation requires near-instantaneous responses. Deepgram achieves this using streaming state space model architectures, optimized specifically for the underlying hardware to deliver minimal processing delays.
- Context: Effective conversations are deeply contextualized. Deepgram will pass the speech Turing test thanks to its ability to train on vast bodies of data that thoroughly represent its customers’ use cases and pass that context through the entire system and interaction.
