Papers
2
Total Citations
21
H-Index
2
About
Manuel Kaspar’s research sits at the intersection of reinforcement learning, robotics, and industrial automation, with a focus on enabling robots to operate more intelligently and flexibly in real-world settings. In his most cited work, “Sim2Real Transfer for Reinforcement Learning without Dynamics Randomization” (2020, 12 citations), Kaspar introduced a method that leverages the Operational Space Control (OSC) framework under joint and Cartesian constraints to train policies in Cartesian space. This approach not only accelerates learning with adjustable degrees of freedom but also allows for direct policy transfer to physical robots—bypassing the need for costly dynamics randomization. His earlier work, “Full automatic path planning of cooperating robots in industrial applications” (2017, 9 citations), tackled a pressing challenge in aerospace manufacturing: programming multiple robots to handle oversized carbon fiber reinforced plastic (CFRP) components. By automating path planning for cooperating robots, Kaspar addressed the complexity of manual programming for varied part geometries. Together, these contributions highlight his ability to bridge simulation and reality while solving practical, large-scale industrial problems—making his work a valuable resource for researchers exploring scalable, transferable robotic systems.
Research Focus
Key Achievements
Top Papers
- 1Sim2Real Transfer for Reinforcement Learning without Dynamics Randomization12 citations · 2020
- 2