Marina Kollmitz

University of Freiburg, Leibniz University Hannover

Papers

7

Total Citations

178

H-Index

4

About

Marina Kollmitz is a leading researcher in human-aware robot navigation and social robotics, whose work bridges the gap between autonomous systems and human comfort. Her key research areas include socially compliant robot navigation, deep perception of people and mobility aids, and compliant mobile robot design. Kollmitz’s major contributions include pioneering a planning-based approach for human-aware navigation using layered social cost maps (89 citations), which enables robots to consider both safety and human comfort during movement. She has also advanced inverse reinforcement learning techniques to teach robots socially compliant behavior through physical interaction (33 citations), and developed deep 3D perception systems capable of recognizing people and their mobility aids in hospital environments (33 citations). Her whole-body sensory concept for compliant mobile robots (14 citations) challenges conventional collision-avoidance approaches by allowing safe physical contact with humans and environments. Kollmitz’s work has significant impact in healthcare robotics, where her detection algorithms help robots identify vulnerable individuals requiring cautious interaction. She was also part of the RoboCup@Work League winning team in 2012, demonstrating her practical expertise in competitive robotics. Her research continues to shape how robots navigate and interact in human-populated spaces.

Research Focus

Key Achievements

4
H-Index
7
Papers
178
Total Citations
25
Avg Citations/Paper
🏆 Most Cited Paper
Time dependent planning on a layered social cost map for human-aware robot navigation
89 citations · 2015
📈 Most Prolific Year: 2020 (2 Papers)
🤝 Key Collaborators: 16
🏛 Institutions: University of Freiburg, Leibniz University Hannover

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago