2018 designer’s guide roundup

By Gina Roos, editor-in-chief

As part of Electronic Products’ role to help designers know what new electronic components, technologies, and tools are available, we are focused on providing guidance on the evaluation and selection of components to help with their design efforts.

Over the past year, we have covered a wide range of component categories from motor drivers and LTE-M modules to AI processors and accelerometers. These guides cover key specifications for selecting these components and discuss design trade-offs to help engineers make more informed design decisions. For those of you looking to advance your engineering skills, we also looked at engineering education opportunities.

Here is a rundown of our 2018 designer’s guides:

Designer’s guide: motor drivers evolving toward smart motion control
Getting motors to run is easy, but controlling their operation with any precision means a lot more work. With the rise in robotics, industrial automation, and the internet of things, the need for sophisticated motor control has dramatically increased. But sensorless operation and sophisticated algorithms such as field-oriented control have typically required complex programming and powerful computation, keeping them out of reach for many developers. Newer control motor driver chips are entering the market, however, bringing smooth, precise, and power-efficient motion control within the reach of the non-expert.

Designer’s guide: engineering education opportunities abound online, but choose carefully
The internet era brought with it unprecedented access to information from virtually anywhere in the world. The most efficient way of translating that information into knowledge and skill, however, is through training. Fortunately, there is also now an abundance of online training opportunities for electronic product developers. The key to maximizing results is to choose wisely from among the many options. The amount of online engineering training available is mind-boggling, ranging from short tutorials on basic skills to full master’s degree programs from world-renowned universities. Cost and availability also vary widely. At one extreme, there are individual courses that are free of cost and available on demand. At the other end, there are entire programs rigidly scheduled and costing tens of thousands of dollars to complete.

Designer’s guide: LTE-M modules expand range and mobility of low-power IoT
Many devices in the internet of things use Wi-Fi as their link to the cloud, but there is a host of applications needing both mobility and wide-area connectivity that Wi-Fi cannot offer. For such applications, a variety of low-power wide-area network (LPWAN) technologies are on offer, but most require creation of a proprietary service network to provide extensive coverage. Now, cellular phone networks are implementing a variation on the LTE services that they already have in place that will meet the IoT’s LPWAN needs, and pre-certified modules are the key for designers seeking to quickly enter such application spaces.

Designer’s Guide: Selecting AI chips for embedded designs
Developers and system designers have a number of options available to them for adding some form of neural-networking or deep-learning capability to their embedded designs. Early on — and even today — designers have successfully used graphics processing units (GPUs) and field-programmable gate arrays (FPGAs), both of which dovetailed nicely into the memory-intensive demands of deep learning. Even traditional x86 CPUs have found their way into AI applications. Many developers have discovered that these existing technologies aren’t always the best fit. As a result, over the past few years, a number of startups (as well as established chipmakers) are focused on building chips specifically for AI applications. These chips are built from the ground up to meet the computing power needs for AI algorithms and running applications. But like all technologies, there are advantages and trade-offs to every solution. Overall, designers need to choose the best technology based on their specific end application. AI chips are typically segmented into three key application areas: training on the cloud, inference on the cloud, and inference on the edge.

Designer’s Guide: Safety-critical processors
Many system designs, including industrial machinery, medical devices, and automobiles, are safety-critical and need to have an ability to detect their own operational failures in real time and react in a way to avoid harming the people using them. Creating a processor-based system to provide this functional safety thus requires using a combination of hardware error-checking, hardware self-test, and system redundancy to provide the software-independent fault detection and safe resolution that these systems need. Fortunately, there are processors available that handle much of the hardware heavy lifting needed for safety-critical systems.

Designer’s guide to industrial IoT sensor systems
Key components of modern industrial systems are the sensors that feed data to the controllers, monitors, and other operational technologies running the plant. Networking of sensors has been in use for years, but the advent of the internet has expanded both the opportunities and challenges of using sensor systems. The design opportunities and challenges have also expanded as sensors become part of the industrial internet of things (IIoT). Sensors play a variety of roles in the modern factory. In addition to providing data for process control, they assist in quality assessment, asset tracking, and even worker safety. The advent of powerful, cloud-based analytical software and artificial intelligence has also allowed the use of sensor data to lower production costs through process optimization and predictive maintenance. And once routed to the internet, sensor data can be put to a variety of uses from supply management to global coordination of production resources.

Engineer’s guide to embedded AI
If you are looking to take your first steps into the forest of deep learning, you are not alone and there are lots of resources. Deep neural networks are essentially a new way of computing. Instead of writing a program to run on a processor that spits out data, you stream data through an algorithmic model that filters out results. The approach started getting attention after the 2012 ImageNet contest, when some algorithms delivered better results identifying pictures than a human. Computer vision was the first field to feel a big boost. Since then, web giants such as Amazon, Google, and Facebook have started applying deep learning to video, speech, translation — anywhere they had big data sets that they could comb to find new insights.

Designer’s guide to accelerometers: choices abound
Also known as inertial sensors, accelerometers are critical elements in key applications such as automotive air-bag deployment, smartphone motion tracking, and industrial predictive maintenance. These varying needs have resulted in an even more varied array of accelerator products from which designers can choose. Fortunately, by focusing on a handful of key decisions, developers can quickly zero in on the right kinds of devices for their application.

Despite the multitude of options available, accelerometers are all based on the same basic principle: inertia. A proof mass within the accelerometer’s structure can readily move in at least one dimension. Because of inertia, that proof mass will tend to stay in place when the surrounding structure undergoes acceleration (i.e., changes its motion) along that same direction. Sensing systems within the accelerator detect the proof mass’s movement relative to the surrounding structure, and interface circuits deliver a corresponding signal to the outside. A spring of some kind provides a restoring force to return the proof mass to its initial position once the acceleration has ended.