Planet Arduino

Wevolver’s previous article about the Arduino Pro ecosystem outlined how embedded sensors play a key role in transforming machines and automation devices to Cyber Physical Production Systems (CPPS). Using CPPS systems, manufacturers and automation solution providers capture data from the shop floor and use it for optimizations in areas like production schedules, process control, and quality management. These optimizations leverage advanced data Internet of Things (IoT) analytics over manufacturing datasets, which is the reason why data are the new oil.

Deployment Options for IoT Analytics: From Cloud Analytics to TinyML

IoT analytics entail statistical data processing and employ Machine Learning (ML) functions, including Deep Learning (DL) techniques i.e., ML based on deep neural networks. Many manufacturing enterprises deploy IoT analytics in the cloud. Cloud IoT analytics use the vast amounts of cloud data to train accurate DL models. Accuracy is important for many industrial use cases like Remaining Useful Life calculation in predictive maintenance. Nevertheless, it is also possible to execute analytics at the edge of the network. Edge analytics are deployed within embedded devices or edge computing clusters at the factory’s Local Area Network (LAN). They are appropriate for real-time use cases that demand low latency such as real-time detection of defects. Edge analytics are more power-efficient than cloud analytics. Moreover, they offer increased data protection as data stays within the LAN.

During the last couple of years, industrial organizations use TinyML to execute ML models within CPU and memory-constrained devices. TinyML is faster, real-time, more power-efficient, and more privacy-friendly than any other form of edge analytics. Therefore, it provides benefits for many Industry 4.0 use cases.

Building TinyML Applications

The process of developing and deploying TinyML applications entails:

1. Getting or Producing a Dataset, which is used for training the TinyML model. In this direction, data from sensors or production logs can be used.
2. Train an ML or DL Model, using standard tools and libraries like Jupyter Notebooks and Python packages like TensorFlow and NumPy. The work entails Exploratory Data Analysis steps towards understanding the data, identifying proper ML models, and preparing the data for training them.
3. Evaluate the Model’s Performance, using the trained model predictions and calculating various error metrics Depending on the achieved performance, the TinyML engineer may have to improve the model and avoid overfitting on the data. Different models must be tested to find the best one.
4. Make the Model Appropriate to Run on an Embedded Device, using tools like TensorFlow Lite which provides a “converter” library that turns a model into a space-efficient format. TensorFlow Lite provides also an “interpreter” library that runs the converted model using the most efficient operations for a given device. In this step, a C/C++ sketch is produced to enable on device deployment.
5. On-device Inference and Binary Development, which involves the C/C++ and embedded systems development part and produces a binary application for on-device inference.
6. Deploying the Binary to a Microcontroller, which makes the microcontroller able to analyse data and derive real-time insights.

Leveraging AutoML for Faster Development with Arduino Pro

Nowadays, Automatic Machine Learning (AutoML) tools are used to develop TinyML on various boards, including Arduino boards. Emerging platforms such as Edge Impulse, Qeexo and SensiML, among others, provide AutoML tools and developers’ resources for embedded ML development. Arduino is collaborating with such platforms as part of their strategy to make complex technologies open and simple to use by anyone.

Within these platforms, users collect real-world sensor data, train ML models on the cloud, and ultimately deploy the model back to an Arduino device. It is also possible to integrate ML models with Arduino sketches based on simple function calls. AutoML pipelines ease the tasks of (re)developing and (re)deploying models to meet complex requirements.

The collaboration between Arduino and ML platforms enables thousands of developers to build applications that embed intelligence in smart devices such as applications that recognize spoken keywords, gestures, and animals. Implementing applications that control IoT devices via natural language or gestures is relatively straightforward for developers who are familiar with Arduino boards.

Arduino has recently introduced its new Arduino Pro ecosystem of industrial-grade products and services, which support the full development, production and operation lifecycle from Hardware and Firmware to Low Code, Clouds, and Mobile Apps. The Pro ecosystem empowers thousands of developers to jump into Industry 4.0 development and to employ advanced edge analytics.

Big opportunity at every scale

The Arduino ecosystem provides excellent support for TinyML, including boards that ease TinyML development, as well as relevant tools and documentation. For instance, the Arduino Nano 33 BLE Sense board is one of the most popular boards for TinyML. It comes with a well-known form factor and various embedded sensors. The latter include a 9-axis inertial sensor that makes the board ideal for wearable devices, as well as for humidity and temperature sensors. As another example, Arduino’s Portenta H7 board includes two asymmetric cores, which enables simultaneously runs of high level code such as protocol stacks, machine learning or even interpreted languages (e.g., MicroPython or JavaScript). Furthermore, the Arduino IDE (Integrated Development Environment) provides the means for customizing embedded ML pipelines and deploying them in Arduino boards.

In a Nutshell

ML and AI models need not always to run over powerful clouds and related High Performance Computing services. It is also possible to execute neural networks over tiny memory-limited devices like microcontrollers, which opens unprecedented opportunities for pervasive intelligence. The Arduino ecosystem offers developers the resources they need to ride the wave of Industry 4.0 and TinyML. Arduino boards and the IDE lower the barriers for thousands of developers to engage with IoT analytics for industrial intelligence.

Read the full article on Wevolver.

The post From Embedded Sensors to Advanced Intelligence: Driving Industry 4.0 Innovation with TinyML appeared first on Arduino Blog.

This is an edited version of a longer piece first published on Wevolver.

In recent years, industrial enterprises are accelerating their digital transformation and preparing themselves for the fourth industrial revolution (Industry 4.0). This digitization of production processes enables industrial organizations to implement agile and responsive manufacturing workflows, which rely on flexible Information Technology (IT) systems rather than on conventional Operational Technology (OT). This flexibility facilitates a shift from conventional Made-to-Stock (MTS) manufacturing to novel customizable production models like Made-to-Order (MTO), Configure-to-Order (CTO) and Engineering to Order (ETO). 

The implementation of Industry 4.0 compliant production systems hinges on the deployment of Cyber-Physical Systems (CPS) in the manufacturing shop floor. In essence, CPS systems comprise one or several internet-connected devices integrated with other production systems in industrial environments. This is the main reason why Industry 4.0 is also referred to as Industrial Internet of Things (IIoT).  IIoT includes the subset of IoT (Internet of Things) systems and applications that are deployed in industrial environments such as the manufacturing, energy, agriculture, and automotive sectors. According to recent market studies, the lion’s share of IoT’s market value will stem from IIoT applications rather than from consumer segments.          

The typical structure of IIoT applications is specified in standards-based architectures for industrial systems such as the Reference Architecture of the Industrial Internet Consortium. It comprises a stack of components that includes sensors and IoT devices, IoT middleware platforms, IoT gateways, edge/cloud infrastructures, and analytics applications. 

The Power of Embedded Sensors in the Manufacturing Value Chain

IT systems, enterprise applications (e.g., ERP and Manufacturing Execution System (MES)), and industrial networks for production automation have been around for decades. The real game-changer in Industry 4.0 is the expanded use of embedded sensors in the value chain. Embedded sensors transform manufacturing assets into cyber-physical systems and enable many optimizations that were hardly possible a few years ago. Overall, embedded sensors and other IIoT technologies empower increased efficiencies by transforming raw digital data to factory floor insights and automation actions. 

Some of the perceived benefits of IIoT and embedded sensors deployments in production operation include:

• Flexible Production Lines

• Predictive Maintenance 

• Quality Management

• Supply Chain Management

• Zero Defects Manufacturing

• Digital Twins

Data analysis options: Edge, Cloud, or combination?

Most IIoT applications include data analytics functionalities such as sensor data analysis based on machine learning techniques. Therefore, they typically collect and process information within cloud computing infrastructures. The latter facilitates access to the required data storage and computing resources. Nevertheless, IIoT deployments in the cloud fall short when it comes to addressing low latency use cases, such as applications involving real-time actuation and control. In such cases, there is a need to execute operations close to the field (i.e., the shopfloor) that cannot tolerate delays for transferring and processing data in the cloud. 

To address real-time, low-latency applications, industrial organizations are deploying IIoT applications based on the edge computing paradigm. The latter involves data collection and processing close to the field, within infrastructures like edge clusters (i.e., local cloud infrastructures), IoT gateways, and edge devices. A recent report by Gartner predicts that by 2023 over 50% of enterprise data will be processed at the edge. 

Edge computing deployments are best suited for real-time control applications while helping to economize on bandwidth and storage resources. Specifically, data processing within edge devices facilitates the filtering of IoT data streams and enables enterprises to selectively transmit to the cloud “data points of interest” only. Furthermore, edge computing provides better data protection than cloud computing, as data remains within local edge devices rather than being transmitted to cloud data centers outside the manufacturing enterprise. Moreover, edge analytics functions like AI algorithms on edge devices are much more power-efficient than cloud-based analytics. 

In practice, industrial enterprises employ both cloud computing and edge computing for their IIoT use cases. Specifically, they tend to deploy real-time functions at the edge and data-savvy industrial automation functions on the cloud. There is always an interplay between cloud and edge functions towards achieving the best balance between analytics accuracy, computational efficiency, and optimal use of bandwidth and storage resources. Thus, IIoT applications are usually deployed in the scope of a cloud-edge environment.

Nowadays, there are many ways to implement edge computing and its interactions with cloud infrastructures. Likewise, there are also many options for employing machine learning at the edge of an industrial network, such as federated machine learning techniques or even deployment of machine learning functions in embedded devices. The latter involves a convergence of embedded programming with machine learning, characterized as embedded machine learning or TinyML. 

State of the art cloud/edge computing paradigms support varying requirements of IIoT use cases in terms of latency, security, power efficiency, and the number of data points needed for training ML algorithms. Future articles in this series will shed light on the technical architecture and the deployment configurations of some of the above-listed cloud/edge paradigms. 

The Scaling of IIoT and the Path towards industry 4.0

Industry 4.0 has been around for over five years, yet we are still quite far from realizing the full potential of embedded sensors and the Industrial IoT. Many enterprises have started their deployment journey by setting up data collection infrastructures and deploying CPS systems and IoT devices on their shop floor. There are also several deployments of operational use cases in areas like asset management, predictive maintenance, and quality control. Nevertheless, many use cases are still in their infancy or limited to pilot deployments in pilot production lines or lab environments. Therefore, there is a need for evolving and scaling up existing deployments to enable industrial enterprises to adopt and fully leverage the fourth industrial revolution.

The scaling up of Industry 4.0 use cases hinges on addressing the following challenges technical and organizational challenges:

• Legacy compliance for brownfield deployments. 
• Alleviating data fragmentation in industrial environments.
• Addressing the IoT, BigData, and AI skills gap.
• Ensuring access to pilot lines and experimentation infrastructures. 
• Easing IIoT integration end-to-end i.e., from the embedded device to the manufacturing application
• Realizing a cultural shift towards Industry 4.0. 

Arduino Pro and Industry 4.0

Driven by these challenges, Arduino has recently created its Arduino Pro solution for Professional Applications. It is an all-in-one IoT platform, which combines:     

• Hardware boards for industrial control, robots, and edge AI applications.
• End-to-End secure connectivity solutions for deploying cloud-based applications. 
• Advanced development environments that enable low code application development. 
• Ease of Implementation and significant community support. 

Conclusion

This article has introduced the Industrial Internet of Things, including its main use cases and business value potential for industrial enterprises. It has shed light on how embedded sensors, cloud/edge computing, and Artificial Intelligence provide a sound basis for optimizing production operations in directions that can improve production time, quality, and cost, while at the same time boosting employees’ safety and customers’ satisfaction. 

Read the full version of this article, including references used here, at Wevovler.com.

The post Engineer’s guide to Industrial IoT in Industry 4.0 appeared first on Arduino Blog.

Arduino Pro is introducing a powerful new member of the Portenta product family, the Portenta Machine Control. It’s a fully-centralized, low-power, industrial control unit able to drive equipment and machinery. Plus, you can program it using the Arduino framework or other embedded development platforms.

Thanks to its computing power, the Portenta Machine Control enables a wide range of predictive maintenance and AI use cases. It enables the collection of real-time data from the factory floor, while supporting remote control of equipment, including from the cloud.

Key benefits include:

• Shorter time-to-market
• Enhance existing products
• Add connectivity for monitoring, as well as control
• Each I/O pin can be configured, so you can tailor it to your needs
• Make equipment smarter, as well as AI-ready
• Provide security and robustness from the ground up
• Open new business model opportunities (such as servitization)
• Interact with your equipment with advanced human-machine interfaces (HMI)
• Modular design for adaptation, expansion and upgrades

Business as a Service

The Portenta Machine Control allows companies to enable new business-as-a-service models. You can monitor customer usage of equipment for predictive maintenance while gathering valuable production data.

The device enables industry standard soft-PLC control. Because of this, it’s able to connect to a range of external sensors and actuators. For example, the following options are all available.

• Isolated digital I/O, 4-20mA compatible analog I/O
• Three configurable temperature channels
• Dedicated I2C connector. 

Multiple choices are available for network connectivity, including USB, Ethernet and WiFi and BLE. Furthermore, it offers impressive compatibility through industry specific protocols such as RS485. All I/O are protected by resettable fuses, but on-board power management ensures maximum reliability in harsh environments.

The Portenta Machine Control core runs an Arduino Portenta H7 microcontroller board. This is a highly reliable design operating at industrial temperature ranges (-40 °C to +85 °C). Firstly, it boasts a dual-core architecture that doesn’t require any external cooling. Secondly, thanks to this versatile processor, you can also connect external human-machine interfaces. These include displays, touch panels, keyboards, joysticks and mice to enable on-site configuration of state machines and direct manipulation of processes.

The Portenta Machine Control’s design addresses a large variety of use cases. It’s possible to configure a selection of the I/O pins in software. Because of this, it stands out as a powerful computer to unify and optimize production where one single type of hardware can serve all your needs. 

Additional Portenta Machine Control Features

Furthermore, it offers these other outstanding features.

• Industrial performance leveraging the power of Arduino Portenta boards
• DIN rail compatible housing
• Push-in terminals for fast connection
• Compact size (170 x 90x 50 mm)
• Reliable design, operating at industrial temperature rates (-40 °C to +85 °C) with a dual-core architecture and no external cooling
• Embedded RTC (real time clock), for perfect synchronization of processes
• Leverage embedded connectivity without any external equipment
• CE, FCC and RoHS certified

The Portenta Machine Control can be used in multiple industries, across a wide range of machine types. For example, labelling machines, form and seal machines, cartoning machines, gluing machines, electric ovens, industrial washers and dryers, mixers and more.

As a result, adding the Portenta Machine Control to your existing processes mean you become the owner of your own solutions in the market of machines.

The Portenta Machine Control is now available for €279/$335.

Take a look here for more information and complete technical specs.

The post Portenta Machine Control: Add a powerful brain to your machines appeared first on Arduino Blog.

The Arduino Pro lineup continues to grow with the introduction of the new Arduino Edge Control. This is a remote monitoring and control solution optimized for outdoor environments. Easy deployment makes it suitable for smart agriculture, precision farming, and other intelligent control applications in remote locations.

Featuring built-in Bluetooth, Arduino MKR boards can expand connectivity with 2G/3G/CatM1/NB-IoT modems, LoRa®, Sigfox and WiFi. With solar power capabilities you can place it anywhere while leveraging AI on the edge. Once installed in the field, it can then be managed remotely using Arduino IoT Cloud (or other services). 

Real-Time Monitoring with Edge Control Sensors

You can also connect sensors, provide real-time monitoring, and drive actuators — commonly used in agriculture — thereby reducing production-related risks.

Particularly aimed at smart agriculture, the sensors can collect real-time data. Weather conditions, soil quality, crop growth and any other data you need. Once sent to Arduino IoT Cloud, the data value chain becomes valuable analytics that support business processes at various levels. For example, crop yield, equipment efficiency, staff performance and so forth. The Edge Control can improve crop quality, reduce effort and minimize error by automating processes like irrigation, fertilization or pest control.

Remote Access and Maintenance

With its robust design, the Edge Control is a fitting solution for applications in any outdoor setting. For example, using it on construction sites or in real estate to automate access control. Similarly, swimming pool maintenance and cleaning companies could monitor and control the condition of pool water from remote locations. As usual, we expect the Arduino community to come up with countless ingenious ways to implement this new technology.

To learn more about how you can use the Edge Control, check out how to get started.

The Edge Control is now available for €169/US$199 on the Arduino Store.

The post Sense the Future of Smart Agriculture with Arduino Edge Control appeared first on Arduino Blog.

How can you help advance Industry 4.0 using the Arduino ecosystem? From robots and predictive maintenance to remote control and data acquisition, we’ve teamed up Distrelec to launch a new Automation & Robotics Contest challenging our community to create innovative solutions that can make the industry faster, cheaper, more flexible, and efficient.

Participants are required to tap into our extensive range of IoT boards like the MKR1000 WiFi and MKR GSM 1400, libraries, and online platform to bring their ideas to life. Industrial automation projects could target energy management, remote monitoring, machine safety, or predictive maintenance, for example, using Arduino Create to set up, control, and connect your Arduino, Intel, and Arm-based devices. Robotics projects could include designs for surveillance drones, robotic arms, rovers, or autonomous transportation, leveraging feature-rich boards like the Mega and Due to prototype advanced systems. 

How to Enter

• Create a free account on Arduino.cc (or log in if already a member).
• Register for the contest by clicking “Register as a participant.”
• Send your concept to the Arduino/Distrelec: Automation & Robotics Contest by June 29, 2018. The top 150 makers will receive a coupon for Distrelec online store. Moreover, there will be a series of micro contests, with weekly prizes handed out from Distrelec.
• Design, build, and submit your project by September 16, 2018. Winning projects will be selected based on their originality, quality, creativity, and social impact. 

Prizes

• 1st Place: Keithley DMM6500 Bench Top Multimeter and a trip to Maker Faire Rome 
• 2nd Place: Weller WT1010 Set: WT1 Soldering Station, WTP90 Soldering Iron, Safety Rest 200, XNT A Soldering Tip
• 3rd-10th Place: Grove Starter Kit for Arduino

Ready to get started? You can find more information on the contest here and browse Distrelec’s entire Arduino lineup on their website. To submit your ideas, please visit the Arduino Project Hub. And remember, projects must use an Arduino board in order to be eligible to win!