© Siemens

Volkmar Sterzing (53) (left) and Dr. Steffen Udluft (48) from Siemens Corporate Technology in Munich-Perlach were the first in the world to develop a method for reinforcement learning that teaches machines to control complex processes themselves using just a small amount of data. The first area of application is the control of large Siemens gas turbines in the USA and South Korea. The continuous fine-tuning of the fuel valves optimizes the operation of the gas turbine in terms of emissions and wear by constantly searching for the best solution in real time. The two researchers were awarded Inventor of the Year 2017 in the "Outstanding Invention" category for their work.