Please use another Browser

It looks like you are using a browser that is not fully supported. Please note that there might be constraints on site display and usability. For the best experience we suggest that you download the newest version of a supported browser:

Internet Explorer, Chrome Browser, Firefox Browser, Safari Browser

Continue with the current browser

Inventors of the Year 2017

Outstanding Innovation

Teaching artificial intelligence to do something with the help of mountains of data is a pretty simple job, relatively speaking. But things really get interesting when you turn those mountains of data into molehills, as the inventor team has demonstrated: Volkmar Sterzing and Steffen Udluft of Corporate Technology in Munich.

Small device, great effect

Small device, great effect

Artificial intelligence knows how to find the sweet spot to ensure that a gas turbine runs optimally.  Today, Power Generation Services uses a system called GT-ACO (Gas Turbine Autonomous Control Optimizer) in pilot operations to control large Siemens gas turbines in the United States and South Korea. “We were surprised at how much better the gas turbines could be operated as a result thereof,” Sterzing says in describing the results of the first test. The continuous fine-tuning of the combustion valves optimized gas turbine operations in terms of emissions and wear by constantly searching for the best solution in real time. With a MindConnect box, Siemens transfers data from industrial applications to Mindsphere.

Pioneer in the development of artificial neural networks

Pioneer in the development of artificial neural networks

Volkmar Sterzing has been fascinated by artificial neuronal networks ever since he studied information technology and computer science in Chemnitz, a city in eastern Germany. After earning his degree, he got an opportunity to apply his knowledge at Siemens Corporate Technology. “This was a really new research area at the time,” he says. “My work focused on speeding up artificial neuronal networks with the help of special hardware.” As part of this work, Sterzing spent nine months in Silicon Valley, which was the center of research into artificial intelligence (AI) at the time. In the initial phase of his research, Sterzing worked with his Siemens colleagues to develop software called SENN, which serves as a basis for the development of neuronal models. Thanks to his elaborate network in the company's divisions, the research group had an opportunity to turn its data-efficient methods in real industrial systems like gas turbines and wind power units into marketable products.

Algorithms for Reinforced Learning are his specialty

Algorithms for Reinforcement Learning are his specialty

Steffen Udluft studied physics at the Ludwig Maximilian University of Munich. While earning his doctorate at the Max Planck Institute for Physics, he explored AI themes. In 2001, he joined Sterzing’s research group Learning Systems. The physicist specializes in developing reinforcement learning algorithms that use small amounts of data to teach complex systems to learn from data.

No one will buy ‘dumb’ devices and systems anymore when smart ones are available at comparable prices.

Volkmar Sterzing, Steffen Udluft, Head of research team Learning Systems, Researcher Reinforcement Learning, Corporate Technology Munich
Inventors Volkmar Sterzing, Dr. Steffen Udluft