Niklas Funk
Papers
9
Total Citations
203
H-Index
6
About
Niklas Funk is a robotics researcher whose work sits at the intersection of dexterous manipulation, tactile sensing, and robot assembly. His most impactful contribution is **SE(3)-DiffusionFields** (89 citations), a novel framework that learns smooth cost functions for joint grasp and motion optimization through diffusion models—addressing the fundamental challenge of multi-objective optimization in manipulation tasks. Funk has also pioneered tactile sensing with **Evetac** (45 citations), an event-based optical tactile sensor that overcomes the temporal resolution limitations of traditional optical sensors, enabling finer-grained feedback for robotic manipulation. His work on **benchmarking structured policies for dexterous object manipulation** (24 citations) using the TriFinger system has provided the community with reproducible, real-world benchmarks, while his research on **Learn2Assemble** and **graph-based reinforcement learning for 3D robot assembly** tackles the complex intersection of resource allocation and motion planning. Funk’s contributions extend to infrastructure, including the development of remotely accessible robot clusters for reproducible dexterous manipulation research. His work is characterized by a systematic approach to combining learning, optimization, and sensing—pushing the boundaries of what robots can achieve in unstructured environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2Evetac: An Event-Based Optical Tactile Sensor for Robotic Manipulation45 citations · 2024
- 3
- 4
- 5
- 6
- 7A Robot Cluster for Reproducible Research in Dexterous Manipulation3 citations · 2021
- 8
- 9Real Robot Challenge: A Robotics Competition in the Cloud2 citations · 2021