Papers

32

Total Citations

1,091

H-Index

17

About

Florian T. Pokorny is a researcher whose work spans robot learning, grasp planning, topological methods in robotics, and learning from demonstrations. He is perhaps best known for his foundational contributions to the Dexterity Network (Dex-Net) project, with his co-authorship of "Dex-Net 1.0" (2016) accumulating over 370 citations and establishing cloud-based grasp planning as a powerful paradigm for robust manipulation under uncertainty. His research consistently addresses the practical challenges of deploying robots in unstructured environments, from grasping objects in cluttered warehouses to surgical task segmentation in robot-assisted minimally invasive surgery. A distinctive thread running through Pokorny's career is the application of algebraic topology — particularly persistent homology and simplicial complexes — to robotics problems such as trajectory classification and gripper design for objects with holes, work that has attracted a dedicated following in the computational robotics community. He has also made significant contributions to imitation and reinforcement learning, developing algorithms like SWIRL and SHIV that reduce the burden on human supervisors while enabling robots to efficiently learn complex, multi-stage tasks. With a portfolio exceeding 800 total citations across these diverse areas, Pokorny represents a rare combination of mathematical rigor and practical engineering insight that continues to influence both theoretical and applied robotics research.

Research Focus

Key Achievements

17
H-Index
32
Papers
1,091
Total Citations
34
Avg Citations/Paper
🏆 Most Cited Paper
Dex-Net 1.0: A cloud-based network of 3D objects for robust grasp planning using a Multi-Armed Bandit model with correlated rewards
374 citations · 2016
📈 Most Prolific Year: 2016 (6 Papers)
🤝 Key Collaborators: 64
🏛 Institutions: University of California, Berkeley, KTH Royal Institute of Technology

Top Papers

  1. 1
  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 · 14 days ago