Armin Lederer
Papers
7
Total Citations
91
H-Index
4
About
Armin Lederer is a leading researcher at the intersection of safe control, Gaussian process (GP) regression, and human-robot interaction. His primary contributions lie in developing rigorous, data-driven frameworks that enable autonomous systems to operate safely under uncertainty. Lederer’s seminal work on **uniform error bounds for Gaussian process regression** (2019–2020, cumulatively over 69 citations) provides the theoretical foundation for certifying the safety of learning-based controllers, a critical step for deploying robots in safety-critical domains. He further advanced online learning with his "Smart Forgetting" approach (11 citations), allowing models to adapt to changing dynamics while maintaining safety guarantees. In human-robot interaction, Lederer has pioneered methods for **vision-based uncertainty-aware motion planning** and **safe control of elastic joint robots** using control barrier functions. His recent work on data-driven force observers for series elastic actuators (2024) addresses the practical challenge of safe, compliant physical interaction. With over 90 total citations and publications spanning top venues, Lederer’s research is shaping the next generation of autonomous systems that can learn, adapt, and guarantee safety in real-world environments.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Smart Forgetting for Safe Online Learning with Gaussian Processes11 citations · 2020
- 4
- 5
- 6
- 7