About

Daniel H. Grollman is a pioneering researcher in robot learning from demonstration (LfD), with a particular focus on enabling non-expert users to teach robots new skills through natural interaction. His key contributions center on making robot learning more accessible and robust, especially through his work on learning from both successful and failed demonstrations. Grollman’s most influential paper, "Dogged Learning for Robots" (2007, 132 citations), introduced the concept of lifelong adaptation for ubiquitous robots, allowing them to modify their behavior in response to changing environments. He further advanced the field with "Robot learning by demonstration" (2013, 97 citations) and developed methods for incrementally learning subtasks from unsegmented demonstrations (2010, 87 citations). Perhaps most notably, Grollman challenged conventional LfD paradigms with his work on learning from failed demonstrations (2011, 85 citations; 2012, 46 citations), showing that mistakes can be valuable teaching moments. His research has been instrumental in developing intuitive interfaces, including Wiimote-based control systems for lifelong robot learning (2008, 24 citations), and has been applied to domains ranging from robot soccer to remote robotic laboratories. With over 600 total citations, Grollman’s work continues to shape how robots can learn from everyday human interaction.

Research Focus

Key Achievements

13
H-Index
23
Papers
725
Total Citations
32
Avg Citations/Paper
🏆 Most Cited Paper
Dogged Learning for Robots
132 citations · 2007
📈 Most Prolific Year: 2012 (5 Papers)
🤝 Key Collaborators: 30
🏛 Institutions: Brown University, John Brown University, École Polytechnique Fédérale de Lausanne, Butler Hospital, Vecna Technologies (United States)

Top Papers

  1. 1
    Dogged Learning for Robots
    132 citations · 2007
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago