- Artificial Intelligence coordinates multi-agent systems
- Implementing European projects on the demonstrator in Kaiserslautern
The Chief Technology Officer of a major automation supplier once told me that an important technology I should keep an eye on was intelligent agents. Indeed, the poor little software object rarely gets star billing on the program. The technology does exist. This information came to me last month about multi-agent systems. It encompasses a European smart factory initiative. This initiative bears watching.
A consortium of seventeen European partners is developing multi-agent systems for autonomous modular production in the research project called MAS4AI (Multi-Agent Systems for pervasive Artificial Intelligence to assist humans in modular production environments). The European Union (EU) has funded the project with almost 6 million euros.
MAS4AI is a project focused on selected sectors of industry that plans for their smart digital transformation over the next three years using the tools of Artificial Intelligence (AI). The aim is to achieve resilient production that can react flexibly to changing requirements or disruptions in the added value networks. The underlying basis is the large variety of products in lot size 1 in complex manufacturing operations.
Single agents acting in concert
Multi-agent systems are an area of distributed artificial intelligence research, in which several differently specialized “intelligent” and mostly autonomous software components (agents or bots) act in a coordinated manner to jointly solve a problem. The researchers are working towards the long-term goal of stable production, which among other things, relies on Shared Production and Production-as-a-Service. Communication, synchronization, and coordination of skills (production capabilities) are needed in a production network in order to implement our vision. This coordination will be performed by AI processes in the future. The European project partners envision a future production that can be distributed in European networks (like GAIA-X).
People make the decisions
Scientists and engineers from Greece, Germany, Italy, Lithuania, the Netherlands, Poland and Spain are initially working on a modular system architecture and a communication structure to create the foundation on which to integrate industrial AI services for smart production. In the process, human participants will always retain control over the AI technologies. The prerequisite for this is to have AI processes designed in a way that is always understandable to the operator. Only then can they be validated, optimized, or modified. Demonstrators oriented on a series of industrial use cases are being developed in MAS4AI. The use cases are in European industrial sectors of high added value, such as companies from the automotive industry, contract manufacturing, bicycle production, or wood processing.
Production Level 4 as the visionary basis
“MAS4AI fits perfectly into our concept of Production Level 4, which is based on production-bots and modular networks. Our concept envisions future production resources that offer their capabilities (skills) to the networks and autonomously (self-directed) call up the products,” said Prof. Martin Ruskowski, Chairman of the Executive Board of SmartFactory-KL, Head of DFKI’s Innovative Factory Systems research, and Chair of the department of Machine Tools and Controls at TU Kaiserslautern. “The products in our vision know their attributes and their current production progress. Such products search their own way among the skills to complete their own production. This may take place in a facility, but also in a Europe-wide network.”
Four scientific and technological goals
The consortium is developing the following four topics:
- Multi-agent systems for the distribution of AI components at various levels of a hierarchy. The key idea is to control interaction between agents on a task-specific basis with agents integrated to form an overall system.
- AI agents that use knowledge-based representations with semantic web technologies. Every agent can detect what skills it has to offer and those of other agents and, in this way, decide what action should be executed. This also makes it easier to integrate people into the production, because the data is also prepared in a way that is understandable to them.
- AI agents for the hierarchical planning of production processes. Processes are broken down into individual steps and optimally reassembled according to the current requirements. Disturbances in the flow can be compensated.
- Model-based AI agents for Machine Learning (ML). These hybrid models are designed to combine human knowledge about physical processes with data acquired for machines.
A fundamental concept in MAS4AI is the integration of all smart components (machines with attributes like self-direction, self-description, and self-learning abilities) in a holistic system architecture. This facilitates easy development and use of industrial AI technologies. Software developers, system integrators, and end users will all benefit because the hurdle for the use of AI is low. “We expect this to generate revolutionary ideas for business models as well as brand new market opportunities,” said Ruskowski.
- Deutsches Forschungszentrum für Künstliche Intelligenz GmbH, Deutschland
- Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek, Niederlande
- University of Patras – Laboratory for Manufacturing Systems & Automation, Griechenland
- Fundacion Tecnalia Research and Innovation, Spanien
- Asociacion De Investigacion Metalurgica del Noroeste, Spanien
- University of Silesia, Polen
- Volkswagen AG, Deutschland
- SCM Group Spa, Italien
- SC Baltik Vairas, Litauen
- VDL Industrial Modules, Niederlande
- Fersa Bearings S.A., Spanien
- Semaku B.V., Niederlande
- Symvouloi Kai Proionta Logismikou, Griechenland
- flexis AG, Deutschland
- Sisteplant, S. L., Spanien
- D.M.D. Computers SRL, Italien
- Smart Manufacturing Competences Centre INTECHCENTRAS, Litauen
The conferences I have attended over the past five or six years have changed since my early days as a customer or sales engineer of industrial technology. While the automation companies have been changing with the trends in technology and application, IT companies have generated much energy and thought with advances in compute platforms, storage, memory, virtual machines, and the like.
I’ve tried to stir up, maybe not controversy, but at least conversation regarding using these powerful compute platforms for both control and data. Industrial Internet of Things (IIoT) connectivity has progressed to a point where networked I/O is not a problem.
That leads me to what I learned from today’s sessions about implementing IIoT and digital transformation and Edge Control from Emerson Virtual Exchange. This was discrete and industrial automation day.
Emerson’ Global User Exchange went virtual this year, and thinking outside the proverbial box, stretched it out over about six months. There are sessions January 5 (which you can see on demand) and January 7, as well as next week on January 12 and 14. These are well done as the platform companies continue to improve and provide better service. Emerson has always worked hard to present professionally done sessions. Virtual did not change things.
The opening “keynote” was introduced by Hakan Erdamar, Group President for Discrete & Industrial. Zach Gustafson, VP of Business Development for Machine Automation and Derek Thomas, VP of Strategy & Marketing for Machine Automation, discussed IIoT and digital transformation, and then introduced Warren Pruitt, VP of Global Engineering Services, Colgate-Palmolive, who related the recent digital transformation journey his company has been traveling.
The key messages include using IIoT connectivity for data, starting small with scalable components, and using local (on-prem) compute. Use cloud for longer term analysis.
The blurb you’ll see online goes like this: Zach and Derek go beyond the obstacles of the “new normal” that manufacturers face are the ongoing challenges of production pressure, operational efficiency, and sustainability measures. In this unique moment in industry the Industrial Internet of Things (IIoT) and the digital transformation it enables is poised to help unleash a new era of manufacturing that is more reliable, more efficient, more connected and more sustainable than ever before. Join Zach Gustafson and Derek Thomas to gain insight into how manufacturers are getting started with unlocking trapped data in individual machines and processes to identify quantifiable savings and benefit that can then quickly scale on global levels.
Warren’s presentation promo goes, A global consumer products leader, Colgate Palmolive is leveraging the power of Digital Transformation to help meet its aggressive global sustainability and operational performance goals. Not waiting for a silver bullet solution, Colgate’s approach is to “get started” with a test an learn mindset and scale successful lessons learned across their global footprint. Warren Pruitt shares Colgate’s implementation approach and lessons learned from the Digital Transformation efforts the organization has made to achieve the company’s objectives.
Thomas took us through a data path in his presentation going through the start small and scale theme and then landing on the latest “PACSystems” edge controller. This is a compute device that runs Linux, is open, makes extended use of Node-RED, also runs PLC control in the same processor/box, has extended open connectivity, and utilizes most standard industrial protocols.
Edge control is redefining the traditional control model by bringing together OT and IT technology in a single solution capable of logic, data management and analytics. This incredible evolution and transformation of PLCs and Industrial PCs is creating new possibilities for tackling today’s IIoT and digital transformation challenges by enabling customers to start small at a machine, scale to plant-wide analytics, and make manageable investments with a single platform. Learn how Emerson’s PACSystems RX3i CPL410 Edge Controller can operate as either an IIoT-enabled PLC or an advanced supervisory controller for your manufacturing operations. This industry leading controller combines the powerful PACSystems runtime, PACEdge IoT platform and Movicon WebHMI to deliver unmatched capability and possibilities for customers. Derek Thomas, Vice President, Marketing & Strategy, Emerson.
Continuing to catch up on 2020 news, here are some products meant to display at SPS Drives in Nuremberg, if there had been an SPS Drives trade fair. I am not a pneumatic geek, but I have visited Festo and seen some awesome displays of engineering talent there. Here are a couple of new modules.
Festo has expanded its intelligent pneumatic energy savings platform beyond the MSE6-E2M(E2M) with two new modules – the MSE6-C2M (C2M) and the MSE6-D2M (D2M). These modules can pay for themselves in less than a year with the energy savings accrued. The platform provides actionable information that supports quality production.
The modules in this platform automatically shut off the air supply to a machine when in standby mode, thus reducing energy consumption. They monitor system pressure and flow information in real time and enable faster response to compressed air leaks. These units flow up to 5,000 liters of compressed air per minute; program easily; connect to Festo MS series air preparation units, including the MS6-SV safety valve; and are suitable for new as well as existing machines.
The new C2M is an intelligent combination of a proportional pressure regulator, on/off valve, sensors, and fieldbus communication. It monitors the flow rate and, when production is not taking place, it automatically shuts off after a defined idle time. At the same time, the module prevents the system pressure from falling below a defined standby pressure level. The lower pressure level saves energy without completely depressurizing the system, which is essential for soft start and safety functionality.
The proportional pressure regulator also allows the user to define normal operating pressure. This feature means there is always control over the operating pressure and an adjustment point is eliminated on the machine, helping with tamper-proofing and with automating changeovers.
The C2M can be fully integrated into the machine network via PROFINET and also through the Festo CPX platform, which is compatible, as is the E2M, with major fieldbus protocols, including Ethernet/IP and EtherCAT.
Similar to the C2M, the D2M intelligent module monitors the compressed air supply and automatically shuts off the compressed air during breaks in production. Unlike the C2M, the D2M completely depressurizes the system. The D2M also provides leakage detection and process data acquisition for condition monitoring, as do the other energy saving modules.
OEMs and end users that do not require a fieldbus node for the energy savings unit will find the D2M an economical solution. This module integrates into the control architecture via the fieldbus node of the C2M or the CPX. Plug-and-play combinations of the D2M with a C2M or CPX can monitor the energy efficiency of two separate compressed air networks simultaneously.
Information from the C2M, D2M, and E2M give operations personnel comparative data over an extended period. Information on flow rate, air consumption, and pressure are continuously available. Data can help personnel determine historical trends on consumption, the amount of air consumed per product batch, and pressure and flow at the time of a malfunction or bad batch of product.
Bringing together control, IT, and the IoT to create an open, scalable system
The idea of “softPLCs”, software-defined control, decoupling software and hardware for control, and the like revisited my thinking thanks to commenting on a paper on that subject by IoT-Analytics. In the discrete, machine automation world, this is an old topic dating back more than 22 years.
It still hasn’t happened. Yes, the marketing arms of some companies trumpet the idea. The Open Process Automation Forum is pushing the idea in the continuous and batch process world. I have yet to see any disruption occurring because of it. Some of the companies that analysts think are disruptive have been around for upwards of 30 years and the three main incumbents are still leading—Siemens in German/Italy, Rockwell Automation in North America, Mitsubishi in Asia.
That doesn’t mean the idea has been shelved completely. There must be some sort of cycle where Bosch Rexroth recurs on my radar with a new platform. I don’t understand all the parts of this new “ctrlX Automation”, but it looks interesting.
These are quotes from Bosch Rexroth.
ctrlX AUTOMATION is based on a completely new software and engineering approach and means a complete departure from proprietary structures and systems. The automation platform includes all necessary software and hardware components for complete system solutions: high-performance controllers, compact drives, industrial PCs, safety solutions, I/O modules and HMIs.
ctrlX AUTOMATION enhanced with an open I/O range
The ctrlX I/O range from Bosch Rexroth offers new possibilities for users in terms of connectivity and networking. The solution represents a functional extension of the ctrlX CORE control platform and also enables horizontal and vertical integration. In the future, ctrlX I/O will offer comprehensive communication and performance enhancements as well as I/O modules geared towards future technologies such as 5G, TSN and AI.
Embedded controller ctrlX CORE now available
In addition, the embedded controller ctrlX CORE is ready to ship with the start of the fair. The modular, compact control is suitable for any automation application and, with its openness, offers users completely new degrees of freedom in setting up the functions. ctrlX CORE’s full scope of functions has already been tested by selected customers.
ctrlX SAFETY: redefining safe automation
With the safety solution ctrlX SAFETY, Bosch Rexroth is setting new standards for safe automation. The product is regarded as the fastest and most compact safety solution on the automation market. Its reaction time allows a more compact design and maximum safety during production.
ctrlX SERVICES for the automation of the future
The automation platform ctrlX AUTOMATION has recently been expanded to include ctrlX SERVICES. The ctrlX App Store now enables downloading of individually required software, while the ctrlX Device Portal facilitates easy and centralized administration of controls. The ctrlX AUTOMATION Community offers numerous functions such as support, know-how transfer and further training. Bosch Rexroth also guarantees 25 years of service availability for classic services.
ctrlX World for third party providers
Users can use apps from Bosch Rexroth, applications from third parties or apps they have produced themselves. ctrlX AUTOMATION users can access a broad portfolio of applications and download them easily via the App Store. In addition, more and more third-party providers are currently joining the ctrlX World, as they can provide their own apps on the platform and thus tap into new business potential.
The Open Process Automation Forum strives for a software-defined industrial control system where the hardware and software are dissociated. The specific reason is that upgrades become less expensive. Software must be upgraded more often than hardware in a control system. If the two are tied together as in all proprietary control systems, then upgrades run on a continuum from painful to impossibly expensive.
I’ve been puzzling out this press release from Schneider Electric about a new control software dubbed EcoStruxure Automation Expert. The company says, “it is the world’s first software-centric industrial automation system.” I’m not sure that claim would stand up exactly, but it seems to me that this is a step on that journey toward dissociating software and hardware in the control system. Executives have told me in the past few years that achieving this is an essential long-term strategy.
Any comments you all have about this are welcome (as long as they’re civil and enlightening).
The press release is written in the tone of a challenge to the rest of the industry to write “apps” that will run on this standards-based (IEC-61499) system.
Schneider Electric promises to unleash a new wave of innovation by championing the widespread adoption of open automation standards unveiling its vision for universal automation with EcoStruxure Automation Expert, “a new category of software-centric industrial automation system.”
Claims closed and proprietary automation platforms restrict the adoption of best-of-breed technologies, present challenges to integrate third-party components, and are expensive to upgrade and maintain. Industry has suffered from a lack of adaptability, modularization and interoperability, which is stunting innovation.
Universal automation is the world of plug and produce automation software components based on the IEC61499 standard that solve specific customer problems in a proven way. Adoption of an IEC61499-based standardized automation layer, common across vendors, will provide limitless opportunities for growth and modernization across industry.
By greatly extending the capabilities of existing IEC61131-based systems and enabling an app-store-like model for automation software components, Schneider Electric believes that the advancements possible in the Fourth Industrial Revolution will be fully realized. As its benefits become visible, Schneider Electric believes other vendors will adopt the universal automation approach, and end users will soon begin to demand it from their automation suppliers and ecosystem.
“The IT world has realized the benefits of open operating platforms; now it’s industry’s turn,” said Peter Herweck, executive vice president industrial automation, Schneider Electric. “Industrial automation architectures have done a good job of advancing industry to where we are today, but they are not capable of providing the agility and resilience that are paramount for modern industrial operations. To fully realize the promise of the Fourth Industrial Revolution, we need to reimagine our technology model by opening our platforms, decoupling software from hardware, and radically improving system agility and scalability.”
EcoStruxure Automation Expert is a new category of industrial automation system with IEC61499 at its core. EcoStruxure Automation Expert:
- Enables automation applications to be built using asset-centric, portable, proven-in-use software components, independent of the underlying hardware infrastructure.
- Allows the user to distribute applications to any system hardware architecture of choice —highly distributed, centralized, or both — with minimal to no additional programming effort.
- Supports established software best practices to simplify the creation of automation applications that interoperate with IT systems.
The first release of EcoStruxure Automation Expert supports traditional automation platforms, such as Modicon PLCs, and Altivar Variable Speed Drives and PCs. Completing the line-up, a virtualized software controller running in Docker-powered Linux containers supports distributed information and control systems across edge computing architectures.
Leveraging the object-oriented nature of IEC61499, software components known as Composite Automation Types (CATs) are used to model assets by combining real-time control functions with other facets, such as the human machine interface. This asset-centric approach delivers unprecedented cost and performance gains and frees engineers to innovate by automating low-value work and eliminating task duplication across tools. Benchmarking of EcoStruxure Automation Expert against today’s automation systems has demonstrated a 2 to 7X reduction in the time it takes to perform traditional automation tasks.
EcoStruxure Automation Expert’s support for mainstream IT best practices enables step-change improvements in asset and workforce efficiency using advanced technologies like predictive maintenance and digital twin. The system also reduces total cost of ownership by incorporating legacy systems with a wrap-and-reuse approach.
“EcoStruxure Automation Expert is the first step in the journey toward universal automation” said Fabrice Jadot, senior vice president, next generation automation, Schneider Electric. “To fully realize the potential of next-generation industries, we must embrace a new way of thinking. Working to common, open standards is vital to ensuring multivendor interoperability and seamless interfaces from supply chain through manufacturing and production to the end customer. Now is the time for all vendors to fully embrace open implementations with code and function portability to become more connected. Today is the first step in a new direction. We invite industrial developers everywhere to create their own software components and solutions based on the IEC61499 standard, which can easily interoperate with EcoStruxure Automation Expert.”
When there is a message for me on LinkedIn, it’s almost always a recruiter or SEO marketer trying to sell me something. So, there was a pleasant surprise the other day when it was a marketing person for a software company with a new take on the Edge, datacenters, and software. That company is NodeWeaver.
Here is a statement of the problem. Most of the software in the world runs outside of clouds or datacenters—it runs at the edge. But the edge is made of small systems deployed in tens of thousands of locations, in stores, inside industrial systems, on top of telecom towers. Places that may have limited connectivity, or be difficult to reach, all sharing the fact that they run critical systems, and if something stops, your users are not getting services, or production lines grind to a halt. What happens if something fails?
Existing solutions require manual interventions by skilled technicians to resolve problems. They are complex and difficult to manage. They are difficult to scale to thousands of locations. What is needed has the flexibility of the cloud, but the ability to run everywhere, even on the smallest devices, and run without requiring user intervention.
That’s the idea behind NodeWeaver—a platform that runs any application and manages the distribution, control, and operation thanks to its intelligent autonomous system. Each system learns from what happens on all the others. It becomes smarter the more it expands and able to do more on its own.
NodeWeaver is a software defined operating platform that installs on the bare metal of nearly any x86 hardware and enables the deployment of highly resilient, agile and scalable compute clusters capable of running multiple virtual machines and container-based workloads, optimized for running workloads at the edge fully autonomously, integrating self-management, self-optimizing, self-healing features that dramatically reduces cost of ownership.
NodeWeaver nano clouds consist of 1 to 25 x86 compatible servers of any manufacturer/configuration, from very small to quite large. Connecting a new server to the nano cloud layer 2 switch automatically adds the server components to the virtual resource pool and relevels all applications across the updated server pool.
NodeWeaver delivers full datacenter infrastructure/functionality, optimized for running workloads at the edge, taking less than 1.3GB of RAM to provide all services, leaving the maximum amount of system resources available for actual workloads. NodeWeaver integrates orchestration, software-defined storage, software-defined networking, multiple hypervisors all managed by the intelligent autonomous system.
Customers who need to manage a large fleet of deployments already have their own monitoring framework in place. NodeWeaver has a full API that allows them to monitor (and manage) their edge systems using their existing monitoring framework. Tools like Ansible, Puppet, Chef, Terraform, and OneFlow Services are for operating system and application automation and management. NodeWeaver fully supports those as well, via a combination of pre-built Marketplace VMs (in the case of Terraform) or built-in services (OneFlow), or simply via API and network connection.
The NodeWeaver marketplace enables users to quickly download complete, pre-configured application stacks [including operating system] and service templates, using any of the software products in the catalog, and deploy them with minimal effort; automatically load balanced across nodes in a highly resilient, agile and scalable compute cluster capable of running multiple virtual machines and container-based workloads.
Industrial control systems used to drive production equipment in factories and plants were installed more than 20 years ago and are now becoming outdated, presenting major business challenges. While this infrastructure has provided a stable platform for control systems for many years, it lacks flexibility, requires costly manual maintenance, and does not easily allow process information to be exported and analyzed. Virtualization overcomes the limitations of legacy control systems infrastructure and provides the foundation for the Industrial Internet of Things (IIoT).
Control functions that were previously deployed across the network as dedicated hardware appliances can be virtualized and consolidated onto commercial off-the-shelf (COTS) servers, which not only leverages the most advanced silicon technology but also reduces capital expenditure, lowers operating costs, reduces risk, and improves ability to manage change and implement continuous improvement.
One of the leading providers of solutions for large-scale industrial laundry systems has been a NodeWeaver customer for over 2 years. They control and monitor all processes, provide predictive analytics, as well as automated deployment and management of all systems.
With no IT staff at these locations, system resiliency and the ability to autonomically address failures and maintain uptime is crucial. Additionally, the environments in these locations are characterized by high temperatures and humidity, thus requiring fanless, ruggedized hardware that can withstand these conditions.
NodeWeaver’s software-only approach provided the flexibility to choose the hardware necessary for the application, and its lightweight codebase enables it to run on smaller devices that competing solutions simply can’t support, equating to an unmatched combination of reliability, flexibility, and time to value.