Robots in the fog: Fog-computing

Even autonomous robots have to be integrated into a wider control system – whether in production or in the service sector. That demands fast, reliable communications standards, and the resources of the cloud. The latest keyword being used in this context is “fog computing”, which brings resources closer to the robot.

For robots to move autonomously around an environment and perform tasks, they have to record and compute vast amounts of data. It doesn’t necessarily have to happen “onboard” though. For small robots, especially, the resources needed would be too costly. One alternative is to monitor the robots by external systems, and also to transmit control commands to them from an external source, such as by wireless radio communication.
The autonomous bionic eMotionButterflies developed by Festo, for example, are coordinated three-dimensionally by an external real-time communicating control and monitoring system. The communication and sensor technology creates an indoor GPS system which collectively controls the butterflies, guiding them so as to avoid collisions. Ten cameras installed inside the room monitor the spheres by way of their active infrared markers (LEDs) and relay the position data to a central master computer. The computed action commands are transmitted back to the objects, where they are executed remotely.

Communication in real time

In industrial applications especially, robots are not “island solutions”, but have to interact with the other production systems. That demands communication solutions which transmit the signals not just in real time, but synchronised too. Only in that way can a mobile robot and a fixed machine work together on a component for example. Previously, most real time-capable control applications were realised by non-standardised network infrastructures or through separate networks, which made it much more difficult – and sometimes impossible – to access machinery and data. So a task group of the Institute of Electrical and Electronics Engineers (IEEE) is currently working to standardise real-time functionality in Ethernet. The result is Ethernet TSN (Time Sensitive Networking), a technology which many companies see as the future of communications in the age of Industry 4.0 (Smart Manufacturing). TSN enables a conventional, open network infrastructure to be implemented with cross-manufacturer interoperability, and offers guaranteed feed and performance. The technology supports control and synchronisation in real time – such as between motor control applications and robots – over a single Ethernet network. TSN is also suitable for other commonly used production data transfer methods, however, meaning IT and Engineering are able to share a network. So the biggest advantage of TSN is that it converges and more efficiently interconnects technologies in order to provide the critical data required for Big Data analyses.

Fog instead of Cloud

This means all data relevant to the robot functionality can be combined in one digital model in real time. And it happens in the Cloud, implementing the new era of Industry 4.0. However, the increasing numbers of cloud computing services and the large numbers of machines with access to cloud resources are heightening concerns that the network will become overloaded, resulting in bottlenecks and delays in data processing. A technology solution known as fog computing is intended to provide a remedy. It creates an intermediate level of data processing. This means the data is not – as it was previously – uploaded to the Cloud or to a remote data centre without being processed at all, but rather undergoes processing in server systems, storage and networking components close to the IT infrastructure. These so-called edge devices provide services and perform tasks that previously came from the Cloud, and so reduce the volumes of data being transferred. So whereas the Cloud is a rather nebulous, remote place, the “fog” is down close to ground level, where the work is actually being done. “Fog computing brings analytical, processing and storage functions right to the edge of the network,” explains Kay Wintrich, Technical Director of Cisco Germany. “In the Internet of Everything – in a completely connected world – that is the only way of handling the huge volumes of data.”

Related Posts

  • Sight is a key factor in enabling robots to assist humans in their everyday lives. Equipped with various sensor systems, they not…

  • Robots assist personnel in the healthcare sector. Whether in hospital logistics, as household helpers, or in rehabilitation: service robots are being successfully…

  • It started around 50 years ago with just a narrow range of applications for the first robots. Since then, they have been…