Paul Koch

Aalborg University, Texas Instruments (United States), Fraunhofer Institute for Production Systems and Design Technology

Papers

Total Citations

109

H-Index

About

Paul Koch is a leading researcher in industrial robotics, with a career spanning from foundational servo control to cutting-edge human-robot collaboration. His work is defined by a focus on making robots more adaptable and effective partners for human workers, particularly in maintenance and manufacturing tasks. His most impactful contribution is the concept of a "skill-based robot co-worker," detailed in his 2017 paper (104 citations), which demonstrates a sensor-driven robot capable of working safely and flexibly alongside human operators in dynamic industrial environments. This work directly addresses the challenge of integrating robots into unstructured, human-centric workspaces. Earlier in his career, Koch laid critical groundwork in robot performance by designing joint servo development systems (1985), a fundamental contribution to optimizing robot motion control. Most recently, he is tackling the data bottleneck in modern robotics, proposing methods to autonomously generate training data for 6D pose estimation with minimal human input (2023). This forward-looking research aims to democratize deep learning for robot manipulation, reducing reliance on expert knowledge. Koch’s trajectory—from servo-level precision to autonomous skill acquisition—marks him as a pivotal figure in the evolution of collaborative industrial robotics.