Florian Jaensch
Papers
4
Total Citations
29
H-Index
3
About
Florian Jaensch is a researcher at the forefront of applying reinforcement learning (RL) to industrial robotics and production automation. His work centers on bridging the gap between advanced machine learning algorithms and practical manufacturing systems, with a particular focus on automating complex control tasks that traditionally require human intuition. Jaensch’s major contribution lies in his pioneering use of software-in-the-loop (SITL) and virtual commissioning simulations as RL environments, enabling robots to learn control logic for tasks like cable handling without physical trial-and-error. His most-cited paper, "Reinforcement Learning of a Robot Cell Control Logic using a Software-in-the-Loop Simulation as Environment" (2019, 15 citations), demonstrates a hierarchical approach to automatic robot programming, where RL handles high-level decision-making while low-level skills are stored as domain knowledge. This work has laid a foundation for more applicable machine learning in production systems, as further explored in his "Test-Driven Reward Function for Reinforcement Learning" (2022, 3 citations), which addresses the critical challenge of designing effective reward functions. Jaensch also investigates the nuances of robotic soft tissue manipulation, as seen in his interdisciplinary case study on teleoperated drawing robots (2020, 2 citations). His research is steadily shaping how factories can leverage RL for flexible, autonomous manufacturing.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4