Mechatronics

“We combine tools from engineers and computer scientists to make the development and operation of mechatronic systems smarter with the help of virtual prototypes.” Prof. Dr.-Ing. Lars Mikelsons

Your partner for model-based methods in the development, production and operation of mechatronic systems

Our competencies:

 

  • AI-supported modeling and simulation of mechatronic systems
  • model calibration using Deep Learning
  • development of model-based methods for use in development, production and operation

Our team ...

 

... from the fields of mechanical engineering, electrical engineering and computer science conducts research at domain boundaries and generates innovations through industrial cooperation.

Learned probability distribution of expert states of an imitation learning task.
Parameter identification for end-of-line testing of MEMS sensors using BayesFlow.

Our offerings for you

  • Modeling of plants and systems based on physical relationships, with the help of machine learning and the combination of both
  • Analysis, evaluation & validation of simulation models
  • Acceleration of simulation calculations with the help of AI-based surrogate models
  • Calibration of simulation models
  • Design of model-based end-of-line tests
  • Model-based falsification of autonomous mechatronic systems
  • Hardware design of mechatronic systems
  • Model-based quantification of uncertainties

Projects

  • End-of-line-Testing of MEMS sensors via deep learning.
  • Combination of machine learning methods with physical modeling to simulate an electronic brake booster.
  • AI-based thermal simulation of a vehicle cabin.
  • Modeling of traffic agents using reinforcement learning.
  • Virtual validation of an autonomous micromobility vehicle.Design of a micromobility vehicle.

Contact

Lehrstuhlinhaber
Lehrstuhl für Mechatronik

Homepage:

Email:

To the website of the chair

Search