Patrick van der Smagt
Deutsches Zentrum für Luft- und Raumfahrt e. V. (DLR), Technical University of Munich, University of Illinois Urbana-Champaign, Fortiss, Systems Dynamics (United States), Volkswagen Group (Germany), Munich University of Applied Sciences, Volkswagen Group (United States), Italian Institute of Technology, University of Amsterdam, Data:Lab Munich (Germany), Robotics Research (United States)
Papers
68
Total Citations
5,482
H-Index
27
About
Patrick van der Smagt is a pioneering researcher at the intersection of neuroscience, robotics, and machine learning, whose work has fundamentally advanced our understanding of how humans and machines can interface through neural and muscular signals. Best known for his landmark 2012 study on neural-controlled robotic arms enabling people with tetraplegia to reach and grasp — now cited over 2,700 times — van der Smagt has dedicated his career to restoring motor function and building biologically inspired robotic systems. His extensive work on electromyography (EMG) has produced robust, session-independent methods for decoding human muscle activity to control advanced prosthetic and robotic hands, with key contributions spanning from support vector machine-based finger detection to real-time 6D teleoperation. He has championed anthropomorphic design through antagonistic actuation and variable-impedance systems that mirror the mechanics of human musculature. More recently, his research has embraced deep learning and reinforcement learning for tactile and visual robotic control, as well as neurorobotics platforms that bridge spiking neural network brain models with embodied robotic environments. With a body of work accumulating thousands of citations across four decades, van der Smagt stands as a central figure in intelligent prosthetics and neurally-driven robotics.
Research Focus
Key Achievements
Top Papers
- 1Reach and grasp by people with tetraplegia using a neurally controlled robotic arm2,730 citations · 2012
- 2Surface EMG in advanced hand prosthetics422 citations · 2008
- 3Learning EMG control of a robotic hand: towards active prostheses246 citations · 2006
- 4
- 5Neural network control of a pneumatic robot arm156 citations · 1994
- 6Stable reinforcement learning with autoencoders for tactile and visual data142 citations · 2016
- 7
- 8
- 9EMG-based teleoperation and manipulation with the DLR LWR-III104 citations · 2011
- 10Antagonism for a Highly Anthropomorphic Hand–Arm System87 citations · 2008