The focus topics of research at the Lab for Machine Learning lie in the following areas:
Robotics and machine learning in agriculture
Intelligent robots for agricultural applications bear great potential from economic and ecological standpoints. Automation can decrease costs and increase crop yield. Likewise, the use of crop protection products can be reduced. Possible applications for robots range from automated hoeing, mowing, and weed control to the detection of pest infestation or the degree of ripeness. Furthermore, pruning and harvest can be automated. For all of these steps, the robots require intelligent behaviour.
Robotics in intralogistics
The lab is currently working on solutions to use Reinforcement Learning for completing complex planning tasks. This enables robots to take on tasks formerly only fulfillable by humans. One of these planning tasks includes the packing of a stable dispatch pallet on which the space is used optimally by packaging items of different sizes.
Machine learning for electro-technical applications
Machine learning can be applied in various fields apart from robotics. The following topics have been addressed so far:
- Anomaly detection and predictions for energy networks
- Optimisation of e-machines by means of neural networks (it is possible to write a Master’s thesis on this topic!)