Learn About Machine Learning

April 18, 2019 Rich Nass

Artificial intelligence (AI) and machine learning (ML) are hot topics at the moment, and likely will remain in vogue for the next few years at least. In the embedded space, we tend to concern ourselves more with ML, as it’s the technology that lets the industrial systems morph themselves into data-generating platforms, able to adjust their characteristics on the fly based on intel received dynamically. That’s a mouthful, but basically, the machines learn about conditions and realign themselves to match those conditions.

To help developers in the process of building ML systems, the microcontroller vendors are making it easier than ever before by adding in the necessary hooks. For example, and as seen at the Embedded Technologies Conference, NXP’s Vice President of Software Engineering, Rob Oshana, will walk through the steps required to build an ML system for an embedded application. He’ll go through each step, the technologies available, how to choose the best technology, and how to integrate an embedded application to support machine learning. A case study is shown to demonstrate each step of the process, culminating in a real time embedded system machine learning application.

Using machine learning can add a level of intelligence to your embedded system that would otherwise be far too time consuming to implement with traditional methods. This could encompass neural networks to learn sensor and system behavior and identify when that behavior requires a human intervention.

To that end, Sai Yamanoor, an IoT applications engineer from Praxair, will discuss anomaly detection using machine learning at the conference. He will present the different toolsets that are available from the MCU vendors that can be used to implement machine learning algorithms on a microcontroller.

In many embedded/industrial applications, using ML in an MCU-based system can alter how the system works from a higher level. Nothing can be taken for granted. To address this issue, Intel’s Michael Clay will use vision as his design example and show just what capabilities are needed and what effects they have downstream.

While at the conference, you don’t want to miss Google’s take on machine learning, as explained by Hongyang Deng, a software engineer and technical lead manager of the machine learning team at Google. Hongyang will give an overview on the challenges and opportunities for running machine learning models on wearable and associated devices. The topics covered include use cases, constraints, model constructions and optimization, and tools to run the models with memory/MIPS restrictions.

About the Author

Rich Nass

Richard Nass is the Executive Vice-President of OpenSystems Media. His key responsibilities include setting the direction for all aspects of OpenSystems Media’s Embedded and IoT product portfolios, including web sites, e-newsletters, print and digital magazines, and various other digital and print activities. He was instrumental in developing the company's on-line educational portal, Embedded University. Previously, Nass was the Brand Director for UBM’s award-winning Design News property. Prior to that, he led the content team for UBM Canon’s Medical Devices Group, as well all custom properties and events in the U.S., Europe, and Asia. Nass has been in the engineering OEM industry for more than 25 years. In prior stints, he led the Content Team at EE Times, handling the Embedded and Custom groups and the TechOnline DesignLine network of design engineering web sites. Nass holds a BSEE degree from the New Jersey Institute of Technology.

Follow on Twitter Follow on Linkedin Visit Website More Content by Rich Nass
Previous Article
What can be Done to Improve IoT Security?

The scale and rate of growth are both huge; with significantly more connected nodes than there are people o...

Next White Paper
Debunking the Top Myths About Unsupported Linux for Embedded Development
Debunking the Top Myths About Unsupported Linux for Embedded Development

When it comes to Linux, you have options. Build and manage a system in-house, or partner with an open-sourc...