531,979 active members*
2,708 visitors online*
Register for free
Login Register

Robots collaborate with Deep Learning

FANUC is exhibiting at CeBIT 2017 on the group stand of partner country Japan in Hall 12, Stand D04. The thematic focus is the IoT platform FIELD, with which FANUC is setting a milestone in industrial networking. At a production cell with two robots, there will be a demonstration of how the topic of "Deep Learning" can be used in a concrete application. In addition, the exhibit will show how with the help of "Deep Learning", a data-based forecast of component durability can enhance preventive maintenance. FANUC considers "Deep Learning" one of the building blocks of the Internet of Things.

The future of industrial manufacturing is networked. It is necessary for these networks to be open, as the "participants" – machines, robots, application processes – are highly diverse. With the FIELD system, FANUC provides a comprehensive, intelligent platform where data is exchanged seamlessly and in real time. FIELD stands for FANUC Intelligent Edge Link and Drive and is an open platform for manufacturing. Among the defining elements of this platform are modules and tools that have already been (partly) tested in practice. One of these modules is the operation management software "Linki", which records, sorts, hosts and evaluates machine data within the FIELD system. The "vital signs" of a machine can thus be monitored and appropriate steps determined for preventive maintenance. This subtask is handled by "Zero Down Time" (ZDT), a module already being successfully used in the automobile industry. The FIELD system also encompasses Industry 4.0 functionalities such as "Deep Learning" and supports the development of new applications.

"Deep Learning" is being shown in Hannover by means of a bin-picking cell. In this cell, two robots with an area vision sensor are placed in a bin with parts that the robot has to retrieve from the bin without specifically being taught. Each robot saves the "experiences" gained, for example in the internal cloud referred to as "Fog". Once stored there, this information is also available to other robots. If four robots are working at this bin, for example, they benefit from the "experiences" of the other robots, emptying the bin more quickly as a result. The learning curve indicates that after 1,000 attempts, the robot has a success rate of 60%, after 5,000 attempts it can already pick up 90% of all the parts – without a single line of program code having to be written.

All FIELD functions work with or are based on very large volumes of data. A cloud is an appropriate means of storage ("Big Data"), to which all participants in a manufacturing environment, for example, have access. In order to structure this access more selectively, FANUC and Cisco introduced a factory Internet cloud, and for simplicity's sake they call it "Fog". In this internal fog, there is significantly less data stored, so that the data generated in edge computing by machines, robots or applications on the process level can also be quickly available.

In the future, systems such as FIELD will have to be open for all the machines involved in the production process. While FANUC itself continues to use proprietary interfaces for its own products such as Robodrill, Robocut, Laser or Robots (which up to now has enabled an extremely high level of functional reliability), it will also support third-party controllers via OPC servers. It will thus be possible to connect third-party products to a FIELD system.

Responsible for the content of this press release: FANUC Deutschland GmbH


FANUC Deutschland GmbH
Bernhäuser Straße 22
73765 Neuhausen a.d.F.
+49 (0) 7158/187 - 400
+49 (0)7158/187-455

Route planner

Route planner