Papers
9
Total Citations
156
H-Index
7
About
Zhiao Huang is a leading researcher at the intersection of robotics, simulation, and deformable object manipulation. His work centers on enabling robots to perceive, simulate, and physically interact with complex soft and elasto-plastic materials—from shaping dough to cutting multi-material objects like avocados. Huang’s major contributions include the development of **PlasticineLab**, a differentiable physics benchmark for soft-body manipulation (24 citations), and **RoboCraft**, a graph-network-based framework for learning to see, simulate, and shape elasto-plastic objects in 3D (over 60 combined citations). He also pioneered **RoboNinja**, an adaptive cutting policy for objects with heterogeneous material properties (16 citations), and **DiffSkill**, which leverages differentiable physics for tool-based deformable object manipulation (14 citations). Beyond soft-body physics, Huang has advanced sensor realism with physics-grounded active stereo simulation to close the optical sensing domain gap (22 citations). Most recently, he co-developed **ManiSkill3**, a GPU-parallelized simulation and rendering platform for generalizable embodied AI (16 citations), enabling scalable, high-fidelity robot learning. His work consistently bridges simulation and reality, providing the tools and algorithms that empower robots to handle the messy, deformable world.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5RoboNinja: Learning an Adaptive Cutting Policy for Multi-Material Objects16 citations · 2023
- 6
- 7
- 8
- 9