I am a PhD student in the Autonomous Intelligent Systems Group at University of Bonn advised by Sven Behnke. I graduated with a bachelor's degree from TU Dortmund University in 2018 and a master's degree from RWTH Aachen University in 2020, both in mechanical engineering.

My research interests revolve around robot learning. Specifically, I am interested in how reinforcement learning can enable complex, dexterous control in autonomous robots.

[]  [github]  [linkedin] 

Selected Publications

A complete, up-to-date list of publications can be found on [google scholar].

Grasp Anything: Combining Teacher-Augmented Policy Gradient Learning with Instance Segmentation to Grasp Arbitrary Objects
Malte Mosbach, Sven Behnke
ICRA 2024   [bibtex]  [paper] 

Learning Generalizable Tool Use with Non-rigid Grasp-pose Registration             Malte Mosbach, Sven Behnke
CASE 2023 (Best Conference Paper Award)   [bibtex]  [paper] 

Efficient Representations of Object Geometry for Reinforcement Learning of Interactive Grasping Policies
Malte Mosbach, Sven Behnke
IRC 2022   [bibtex]  [paper] 

Accelerating Interactive Human-like Manipulation Learning with GPU-based Simulation and High-quality Demonstrations
Malte Mosbach, Kara Moraw, and Sven Behnke
Humanoids 2022   [bibtex]  [paper]  [code] 

Transfer of Hierarchical Reinforcement Learning Structures for Robotic Manipulation Tasks
Christian Scheiderer, Malte Mosbach, Andrés Felipe Posada-Moreno, and Tobias Meisen
CSCI 2020   [bibtex]  [paper]