Papers
30
Total Citations
1,297
H-Index
11
About
Nima Fazeli is a robotics researcher whose work sits at the intersection of robotic manipulation, physical simulation, and multimodal sensing. His research spans contact mechanics, grasping, and learning-based approaches to dexterous manipulation, with a particular focus on enabling robots to interact intelligently with objects in unstructured, real-world environments. Fazeli's most influential contributions include a robotic pick-and-place system capable of grasping and recognizing novel objects in cluttered settings without task-specific training data — work that has accumulated nearly 700 citations across multiple publication venues and represents a significant advance in generalizable robotic grasping. His 2016 dataset on planar pushing, "More Than a Million Ways to Be Pushed," has become a widely used benchmark resource (163 citations), offering high-fidelity experimental data that underpins research in contact modeling and manipulation planning. He has also explored hybrid simulation approaches, augmenting physics-based simulators with stochastic neural networks to better capture real-world uncertainty, and has advanced multisensory fusion techniques that combine tactile and visual signals for complex manipulation tasks. More recently, his VIRDO framework introduced implicit visio-tactile representations for deformable object manipulation. Across his career, Fazeli has demonstrated a consistent commitment to grounding robotic learning in rigorous physical understanding.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7A Summary of Team MIT's Approach to the Amazon Picking Challenge 201543 citations · 2016
- 8VIRDO: Visio-tactile Implicit Representations of Deformable Objects25 citations · 2022
- 9
- 10