Maximilian Durner
Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR), Institute of Robotics
Papers
21
Total Citations
319
H-Index
9
About
Maximilian Durner is a robotics researcher whose work spans autonomous planetary exploration, robotic perception, and intelligent manufacturing systems. He is perhaps best known for his contributions to the ARCHES Space-Analogue Demonstration Mission (2020, 98 citations), a landmark project advancing heterogeneous teams of autonomous robots capable of collaborative scientific sampling in planetary environments — a significant step toward reducing human dependency in deep-space exploration. Durner has made substantial contributions to robotic perception, developing methods for 6D object pose estimation, CNN-based 3D object classification, and uncertainty-aware terrain segmentation tailored for planetary missions. His photorealistic terrain simulation pipeline (2021, 36 citations) addresses the persistent challenge of scarce real-world training data for outdoor robotic systems, while his work on Bayesian active learning helps bridge the Sim-to-Real gap that hampers practical deployment. In manufacturing robotics, Durner has advanced assembly sequence planning through graph representation learning and pattern recognition techniques, enabling robots to adapt efficiently to complex, customized products. His earlier research on perceptual episodic memory and experience-based perception optimization further reflects a consistent drive to build robots that learn and adapt intelligently across long-horizon tasks. With over 270 cumulative citations, Durner's research meaningfully advances robotics across both space exploration and industrial automation.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 46D Object Pose Estimation from Approximate 3D Models for Orbital Robotics20 citations · 2023
- 5
- 6
- 7Selecting CNN features for online learning of 3D objects10 citations · 2017
- 8Experience-based optimization of robotic perception10 citations · 2017
- 9Bayesian Active Learning for Sim-to-Real Robotic Perception9 citations · 2022
- 10Uncertainty Estimation for Planetary Robotic Terrain Segmentation9 citations · 2023