Artificial Intelligence (AI) was one of the buzzwords on the SPS/IPC/Drives 2018 at Nuremberg, Germany. ARC experts have been keeping an eye on this topic for some time as the topic has become a real hype over the last 12 months. Read our ARC blogs “Artificial Baby!” and “UK – Serious about AI” to get up to speed on recent developments and expect to see more insights from our side in the near future.
Siemens puts it like this: Artificial intelligence will be able to contribute considerably to cost reduction for engineering, supporting agile and flexible control logic as well as structuring production processes for greater precision in industry.
AI for Simatic controller and I/O system
On the SPS show, Siemens announced a module using artificial intelligence (AI), featuring a chip with AI capabilities. The chip will be integrated into the company’s Simatic S7-1500 controller and their ET 200MP I/O system. The neural processing unit, S7-1500 TM NPU, is equipped with the Intel Movidius Myriad X Vision processing unit to enable the efficient processing of neural networks. The module works with a trained neural system on an SD card and is equipped with USB 3.1 interfaces and a Gigabit Ethernet port. This enables data from a connected sensor system or a CPU program to be processed on site within seconds. The application of machine-learning algorithms, e.g. a trained neural system, allows for the optimization of e.g. visual quality checks in production plants or image-guided robot systems.
Applications and Vision for AI
With this module, Siemens is taking another step towards integrating future technologies into industrial applications. Users can connect compatible sensors such as cameras or microphones to the integrated interfaces of the S7-1500 TM NPU module. Data from these connected sensor systems can then be analysed against the trained model imbedded in the AI chip and the optimal outcome can immediately be executed.
Resulting advantages comes to mind in pick-and-place applications, where mobile robots must recognize, pick out and place components that are lying randomly in a crate. Human expert knowledge regarding parameters such as consistency, colour or quality of a product or a process can be transferred direct to the module through the continuous training of a neural network with assigned (image) data.
Siemens’ Future of Automation aims to provide insights into future automation products and the role of artificial intelligence in their Totally Integrated Automation (TIA) portfolio. Siemens’ vision is to build scalable AI solutions from the field to the controller and edge level up to the Cloud, allowing them to be scaled in terms of the environment and the target application.
ARC continues to cover all topics regarding the use of Artificial Intelligence in Manufacturing. At the time when automation in manufacturing is shifting towards use of intelligent systems to optimize and control production processes, products like Siemens’ Simatic controller supports manufacturers who want to achieve Industry 4.0 aims without taking on additional heavy technology costs.