Gerhard Neumann

About

Gerhard Neumann is a prominent researcher in robot learning, motion generation, and human-robot interaction, whose work has fundamentally shaped how robots acquire and generalize complex behaviors. He is perhaps best known for his development of Probabilistic Movement Primitives (ProDMP), a framework enabling robots to learn modular, reusable motion patterns from data while naturally handling uncertainty — a contribution that has garnered over 400 citations and become a cornerstone methodology in the field. Extending this work, Neumann introduced Interaction Primitives, allowing robots to coordinate fluidly with human partners during collaborative tasks, accumulating nearly 400 citations across related publications. His influential survey on imitation learning, cited close to 750 times across versions, has become an essential reference for researchers navigating the rapidly evolving landscape of learning-from-demonstration. Beyond motion primitives, Neumann has made notable contributions to dexterous in-hand manipulation using tactile sensing, hierarchical skill learning for multi-phase tasks, and deep reinforcement learning for swarm robotics. With a body of work spanning foundational theory and real-world robotic applications, Neumann stands as a leading figure bridging machine learning and intelligent autonomous systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Probabilistic Movement Primitives

   413 citations · 2013
2. 2
   An Algorithmic Perspective on Imitation Learning

   379 citations · 2018
3. 3
   An Algorithmic Perspective on Imitation Learning

   370 citations · 2018
4. 4
   Interaction primitives for human-robot cooperation tasks

   200 citations · 2014
5. 5
   Using probabilistic movement primitives in robotics

   196 citations · 2017
6. 6
   Probabilistic movement primitives for coordination of multiple human–robot collaborative tasks

   192 citations · 2016
7. 7
   Learning robot in-hand manipulation with tactile features

   165 citations · 2015
8. 8
   Towards learning hierarchical skills for multi-phase manipulation tasks

   109 citations · 2015
9. 9
   Guided Deep Reinforcement Learning for Swarm Systems

   100 citations · 2017
10. 10
    Learning multiple collaborative tasks with a mixture of Interaction Primitives

    96 citations · 2015

Key Collaborators

Contact & Links