electronica 2018 - Artificial Intelligence on the Edge

Connected driving, language assistance systems, Internet of Things: Artificial intelligence is no longer a vision of the future – it’s already moving into various technologies and affecting our everyday lives. Does it also represent an economic and social opportunity?

Dr. Reinhard Ploss, CEO, Infineon Technologies
"I think it’s a unique opportunity - for Germany, for Europe in the field of digital transformation and building up a digital industry alongside what we already have. It's not just an opportunity, it's a must. "

At electronica, experts are discussing the possibilities and potential applications of artificial intelligence. This would not be possible without electronics.
Producing things faster, cost-efficiently and error-free – that’s the claim of a fully automated factory. And that’s why Maxim Integrated is developing Go-IO, to bring intelligence closer to the “edge” of production. The plan is to completely control any production line, and make it flexible.

Jeffrey DeAngelis, Managing Director Industrial Communications, Maxim Integrated
"What we have done is, we integrated 12 of our industrial automation ICs, that control digital IO, analog IO and IO-linked technology, which talk to the different sensors throughout the factory. And what this enables us to do is to change the configuration of those sensors depending on the health and status information, that is being pushed to the cloud. So that we can adjust factory on the fly using this technology."
Safety and ideally customizable safety: that’s what the automotive industry is calling for, among others. Texas Instruments has responded with a small sensor chip that monitors the attentiveness of every possible driver via a camera, and can quickly detect various changes.

Brooke Williams, Business Manager, Texas Instruments
"It is a very small and low power chip that still is able to process the tremendous amount of image data. We are using artificial intelligence. Our third party partners are doing the algorithms are using deep learning and applying to the TDA 3 device. That is new as well. And in the application itself is new, that continue to improve the effectiveness of the algorithm and how is tracking the different features. That continues to evolve and improve to the state that requires the market."

Rutronik is also relying on so-called Deep Learning with its "Open Vino Toolkit." It enables video data for example from traffic monitoring to be analysed directly without any need for cloud access. Until now, all data had to be read in a cloud before an analysis, which takes up a lot of time.

Bernhard Estermann, Business Development Manager Intel, Rutronik
"Videos generate a huge amount of data, so it's important to analyze video data as soon as possible, and then bring only partial quantities to a server. Intelligence by the edge, that's the buzzword – that the intelligence might already be done inside the camera or somewhere near the camera to somehow recognize people or vehicles."

Because it's not only seconds but milliseconds that count when it comes to safety, for example.

electronica is showing the new possibilities being offered by Artificial Intelligence.