Skip to main content
Strategy
31/10/2018

Edge computing in production?

A contribution from Prof. Dr.-Ing. Martin Ruskowski

The “Edge” concept has been swirling around production for several years now. What’s really be-hind it? The “Edge” terminus originates from mobile network technology. There it became clear some time ago that the available data rates not sufficient to relay data intensive computing tasks to centralised computer centres. This in turn means that IT technology must be provided at the edge of the network, in radio cell base stations.

The concept was transferred in the production sector without second thoughts. Here, however, we find a completely different architecture. Production control traditionally features a decentralised structure. Powerful controllers perform process control in real time, while on-site industrial PCs provide databases, interfaces, and evaluation functions.

Frankly speaking, we've always been performing edge computing in the production sector. However, there are two other new things besides just the terminus itself: the increasing use of open and standardised protocols, as well as today’s availability of small and inexpensive computers, the so called edge devices. There is yet another new aspect at hand: increasing vertical integration is leading to a convergence of central IT functions and decentralised plant control since modern production must adapt to requirements and requires timely production data in a flexible way. This necessitates new, and above all standardised, interface protocols like OPC UA.

Smart Factory KL
Photo: Smart Factory KL/Alexander Sell

In principle the PC-based controllers available today offer the option of running virtually any software on a computer. From high-level language programming in IEC61131 to virtual machines and hypervisor technology, the door is wide open here. However, in reality controllers in machines and plants do not usually have the internal software architecture needed to perform edge computing in the sense of decentralised preprocessing as per IT standards. A line controller of an automation application is not networked with the surrounding IT. In addition, for safety and availability reasons it should be as autonomous as possible, i.e. decoupled from IT processes.

Also, upgrading an existing system by adding even small functions already requires intervention in the controller. Often, just to enable new communication interfaces even the control computer would need to be exchanged.

At this point, dedicated edge devices for new I4.0 services can be easily deployed to upgrade existing equipment of all ages. An upgrade involving new functions is thereby possible without interference with a tested and certified system. Consequently, edge devices can, among other things, act as a kind of data gateway to bridge protocols and implement security functionality.

Prof. Dr.-Ing. Martin Ruskowski
Prof. Dr.-Ing. Martin Ruskowski

About the author:

Prof. Dr.-Ing. Martin Ruskowski holds the Chair of “Machine Tools and Controls” at the Technical University of Kaiserslautern and is the Head of the Innovative Factory Systems research department at the German Research Center for Artificial Intelligence. His research interests include industrial robots as machine tools, artificial intelligence in automation technology and novel control concepts for automation. Previously, he held several executive positions in industry, most recently as Head of Research & Development at the KUKA Industries Group.

Recommend article

Eingeschränktes HTML

  • Allowed HTML tags: <a href hreflang> <em> <strong> <cite> <blockquote cite> <code> <ul type> <ol start type> <li> <dl> <dt> <dd> <h2 id> <h3 id> <h4 id> <h5 id> <h6 id>
  • Lines and paragraphs break automatically.
  • Web page addresses and email addresses turn into links automatically.