Kevin Sebastian Luck
Aalto University, Technische Universität Darmstadt, Arizona State University, Decision Systems (United States)
Papers
11
Total Citations
130
H-Index
6
About
Kevin Sebastian Luck is a robotics researcher whose work spans robot learning, motion primitives, and the co-adaptation of morphology and behavior. He is perhaps best known for his contributions to contact-rich manipulation, particularly through the development of Residual Learning from Demonstration (rLfD), a framework that combines Dynamic Movement Primitives with reinforcement learning to enable robots to master challenging insertion tasks involving friction and contact forces — work that has accumulated nearly 50 citations since 2022. Earlier in his career, Luck made significant strides in policy search methods for high-dimensional robotic systems, introducing latent space approaches that reduce the curse of dimensionality in reinforcement learning, with foundational papers from 2014 and 2016 garnering 20 and 10 citations respectively. His research has also explored the principled co-adaptation of robot bodies and controllers using deep reinforcement learning, bimanual motor synergies, and bio-inspired design inspired by sea turtles. More recently, his Co-imitation framework advances simultaneous learning of robot design and behavior through imitation. Across these contributions, Luck's work consistently bridges theoretical machine learning with real-world robotics deployment, making him a notable figure in embodied robot intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2Latent space policy search for robotics20 citations · 2014
- 3
- 4Sparse Latent Space Policy Search10 citations · 2016
- 5
- 6
- 7Extracting bimanual synergies with reinforcement learning5 citations · 2017
- 8
- 9Residual Learning from Demonstration.5 citations · 2020
- 10Co-imitation: Learning Design and Behaviour by Imitation4 citations · 2023