Papers
22
Total Citations
256
H-Index
11
About
Sven R. Schmidt-Rohr is a robotics researcher whose work sits at the intersection of robot learning, manipulation, and intelligent service robotics. His most significant contributions center on **Programming by Demonstration (PbD)**, where he has developed innovative frameworks enabling robots to learn, generalize, and execute complex manipulation strategies by observing human teachers. His partially symbolic representations of manipulation motions — a departure from purely subsymbolic approaches — have proven particularly influential, garnering over 40 citations and shaping how robots abstract and reuse learned behaviors across varied environments. Beyond manipulation learning, Schmidt-Rohr has made substantial contributions to human activity recognition for service robots, proposing optimized feature sets and classifiers that allow robots to proactively interpret user states — work cited nearly 40 times. His research also advances probabilistic decision-making, employing Partially Observable Markov Decision Processes (POMDPs) to help multi-modal service robots navigate uncertainty in real-world human-robot interaction scenarios. His flexible task knowledge representation frameworks further address the unpredictability of human-centered environments. With a cumulative citation count exceeding 200 across his top works, Schmidt-Rohr's research has meaningfully advanced the capability of autonomous robots to learn from, reason about, and adapt to human behavior.
Research Focus
Key Achievements
Top Papers
- 1
- 2Feature Set Selection and Optimal Classifier for Human Activity Recognition38 citations · 2007
- 3Advances in Robot Programming by Demonstration35 citations · 2010
- 4
- 5
- 6
- 7A Flexible Task Knowledge Representation for Service Robots.13 citations · 2006
- 8
- 9
- 10