Manuel Kudruss

Heidelberg University

Papers

8

Total Citations

219

H-Index

5

About

Manuel Kudruss is a leading researcher in humanoid robotics and optimal control, whose work bridges the gap between theoretical optimization and real-world robotic motion. His primary research areas include nonlinear model predictive control (NMPC), whole-body motion generation, and learning-based control for complex robotic systems. Kudruss’s most impactful contribution is his pioneering demonstration that real-time NMPC can be implemented on position-controlled humanoid robots, as shown in his highly cited 2016 paper (97 citations), which introduced a reactive walking pattern generator that enables “walking without thinking.” He further advanced the field by developing a complete solution for fully-dynamic multi-contact motion generation (42 citations), expanding humanoid robots’ functional range beyond simple bipedal locomotion. Kudruss has also made significant strides in combining model-based and data-driven approaches, notably proposing a reinforcement learning framework to compensate for model-plant mismatch (37 citations) without requiring extensive real-world trials. His work on efficient derivative evaluation for rigid-body dynamics subject to kinematic constraints (2019) provides essential tools for gradient-based optimization in robotics. Through his research on the iCub and Leo humanoid platforms, Kudruss has demonstrated practical closed-loop control of walking and squatting motions, establishing himself as a key figure in making complex optimal control methods viable for real-time robotic applications.

Research Focus

Key Achievements

5
H-Index
8
Papers
219
Total Citations
27
Avg Citations/Paper
🏆 Most Cited Paper
A Reactive Walking Pattern Generator Based on Nonlinear Model Predictive Control
97 citations · 2016
📈 Most Prolific Year: 2017 (3 Papers)
🤝 Key Collaborators: 18
🏛 Institutions: Heidelberg University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago