The rise of the Internet of Things (IoT) has been spurred by the availability of faster, smaller, and more affordable processors, improved battery technology, and expanded wireless communications infrastructure. The ubiquity of cheap, compact sensors connecting networks of embedded devices to the physical world means that any device can be smart, automated, and portable. It is widely accepted that the "Internet of Things" will have a tremendous impact on our consumer devices, but we have to recognize that it will determine the future of our industrial processes and infrastructures.
The “Internet of Things” customer has the potential to change the way we interact with the world around us, from reducing the energy we use to cool or heat our homes to smart climate control systems that allow us to control the number of calories expended based on measured activity levels. Test and measurement platforms play a key role in evaluating these devices before they reach the market.
a second area of focus, the “Industrial Internet of Things”, where control, test and surveillance platforms are the end product, as opposed to smart consumer devices. Using billions of sensors distributed to collect data, the “Industrial Internet of Things” will transform how businesses work. For example, advances in machine-to-machine communication will revolutionize factory production through better process control and automation. Aggregating data from all stages of the production process will enable faster, smarter decisions that can immediately impact the operation of an entire factory floor, from sourcing to manufacturing to shipping.
The rapid shift to intelligent systems presents an enormous challenge to embedded design engineers. The creation of these complex systems implies not only the design of the embedded devices that allow to collect data from the sensors, but also the development of a method for the creation of networks of these devices and a sophisticated programming logic to make decisions in real time. based on the data collected. These systems generate huge amounts of data that have to be managed and also analyzed to detect macro-trends.
The number of embedded systems engineers in the industry cannot grow fast enough to keep pace with the growing demand for connected devices. Meanwhile, companies have to make the most of limited budgets and minimize their time to market within an ultra-competitive environment. To reach 50 billion connected devices by 2020, embedded design and data management tools must evolve to enable engineers and domain experts to develop systems with smaller human teams and in less time.
The evolution of embedded system design tools
The tools for engineering the "Industrial Internet of Things" must enable the rapid design, prototyping and deployment of embedded systems. A platform-based approach, pioneered at the University of California, Berkeley, uses a proven methodology for creating complex embedded solutions. According to a recent Morgan Stanley article, “The Internet of Things is now connecting the real economy”, the industry is expected to shift to development platforms built for it or at least optimized for the “Internet of Things”. . These platforms will automate much of the development work, allowing developers to focus on what is most important to the business rather than on the infrastructure needed to integrate the Internet of Things.
We are going to consider the possibility of using a traditional method to design custom embedded solutions. Multiple experts are needed to implement the digital, analog, and mechanical design of the embedded hardware, as well as the design of the embedded software that brings the system to life. Software design alone requires specialized knowledge to develop the operating system that supports the board, device drivers, the Application Programming Interface (API), and the application itself.
In addition, domain experts are needed to specify the requirements for this system. For example, a manufacturing process control expert must play an integral role in specifying the data that is required to make sound decisions in the process, with the ultimate goal of increasing the efficiency of the manufacturing system. To fully realize the “Industrial Internet of Things”, engineers need better tools to create complex embedded systems with fewer people.
A platform-based approach enables smaller human teams to develop more efficiently by providing a cohesive set of tools that simplifies system design complexities. In this way, the tooling platform works more efficiently allowing domain experts to focus on application-level challenges without getting bogged down in low-level implementation details, such as creating a custom support package. of the plate.
With the right platform-based system, designers can separate design challenges by defining platform elements that interact through a clear API, resulting in highly modular designs.
This method makes it possible to replace or upgrade items with off-the-shelf hardware to lower development costs. Similarly, designers can reuse these platform elements for future iterations, verification, and documentation.
Using a platform for rapid prototyping reduces time to market
Platform-based design can be used at all stages of the embedded design cycle, from modeling to validation testing. Developing a system prototype is an important part of this process, whether to demonstrate the technical feasibility of an idea or to show potential investors the value it brings to the business. When designing the types of systems that will drive the Internet of Things, a platform-based approach is particularly effective in enabling rapid prototyping.
In particular, development platforms that provide access to FPGA technology to domain experts have been a game changer in rapid prototyping. Teams can use FPGAs to rapidly develop a custom embedded hardware solution, without having to repeat the lengthy process of manufacturing custom ASICs each time they modify a design. Because FPGAs are reconfigurable, human teams can quickly iterate on their design, whether to fix bugs or add functionality, and modify the FPGA circuit in a matter of hours instead of weeks.
An example of a company that is doing this is Airbus, a world-leading manufacturer of commercial aircraft. The company is embracing the "Internet of Things" to revolutionize its manufacturing processes through what they call the "Factory of the Future." Because aircraft production requires the assembly of large and heavy equipment, precise alignment, quality control, and traceability, many of these processes are largely done manually. The “Factory of the Future” is a research and technology project, focused on using new technologies to make it possible to meet these demanding requirements and increase quality and productivity.
Airbus believes that a platform-based approach is essential to make the “Factory of the Future” a reality. Initially, they tried to solve each problem in isolation, which made communication and code reuse very difficult.
To meet this challenge, a small team of engineers decided to create a hardware and software platform that could scale across different types of devices thanks to specific algorithms, such as computer vision, filter design, and schedule scheduling. movements.
NI is uniquely positioned to enable these designers to quickly solve the IoT challenge with a highly integrated hardware and software-based approach. This method focuses on LabVIEW, a powerful system design software tool for programming commercially available hardware with embedded processors, FPGAs, and a wide range of I/O modules. Using this platform, smaller human teams can develop complex embedded systems that would have traditionally required twice as many staff. This consistent platform-based approach to embedded system design provides engineers with the right tools to efficiently create the “Industrial Internet of Things”.
