Chair of the Research Committee & Chief Researcher at the China National Innovation Center of Embodied AI Robotics
Deeply collaborating with Prof. Shaoqing Ren at USTC
- π€ Currently leading research on Humanoid Robots β building the TianGong Humanoid Robot Platform
- ποΈ Previously contributed to "Tianhe 倩河" Supercomputing projects and worked at DJI
- π Welcome to visit and apply to our collaborative program at USTC
- π Check out my Google Scholar for recent publications
- π€ Open to collaborations in Humanoid Robots, Embodied AI, RL, Vision Perception, LLM, Control & Planning
|
China Universal Humanoid Robot Platform |
PNDbotics Adam |
Fourier GR1 |
π 2025-08: We won the 100m championship, 400m 2nd & 3rd, 1500m 2nd, 4Γ100m 2nd, material organization championship, and material handling 2nd at WHR 2025 β the first World Robotics Conference!
πββοΈ 2025-04-19: The Tiangong humanoid robot made history β it successfully completed a half-marathon!
π½ Click to expand full news timeline
| Date | Event |
|---|---|
| 2025-11-29 | Our work SPO adopted as baseline RL algorithm by PI0.6 |
| 2025-07-29 | Released Humanoid Occupancy β Generalized Multimodal Perception Module |
| 2025-07-09 | Released TienKung Marathon Control Framework |
| 2025-05-24 | Published an article in People's Daily |
| 2025-05-08 | Featured on the "Innovation China" TV program |
| 2025-04-24 | Co-hosted embodied intelligent robots seminar with Peking University |
| 2025-04-10 | Interviewed by "Innovation China" column of China Association for Science and Technology |
| 2025-03-29 | Invited talk at CEAI 2025 |
| 2025-03-06 | Joined the CNR Finance Jin Ding Think Tank |
| 2025-01-08 | Participated in Beijing City's long-term robot planning and technical roadmap |
| 2024-12-30 | Invited by National Health Commission as expert reviewer for AI medical projects |
| 2024-11-25 | Interviewed by Global Times on AI for football & robot marathons |
| 2024-11-16 | Keynote on "Embodied Intelligence of Humanoid Robots" at AGIROS Conference, Chinese Academy of Sciences |
| 2024-11-09 | Intel China Academic Talent Program keynote on "Research on Embodied AI of Humanoid Robots" |
| 2024-11-05 | Attended BAAI "ZhiYuan Forum β Embodiment and World Model Summit" |
| 2024-09-30 | Talk at Peking University on Multimodal Perception & Large Model Decision-Making for Humanoid Robots |
| 2024-07-29 | Interviewed by Mango (Hunan) TV on "Embodied AI and Humanoid Robots Tian Gong" |
| 2024-07-05 | Invited by China Internet Research Institute to draft embodied intelligence white paper |
| 2024-07-01 | Invited talk at XMech, Zhejiang University |
| 2024-05-29 | Interviewed by Beijing Association for Science and Technology |
| 2024-05-09 | Featured on CCTV β TianGong continuous iteration |
I'm currently exploring Embodied AI, RL, Vision Perception, LLM, Control & Planning in Robotics.
π½ Click to expand pre-prints
| Title | Link |
|---|---|
| MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction | π |
| HAIC: Humanoid Agile Object Interaction Control via Dynamics-Aware World Model | π |
| RoboStriker: Hierarchical Decision-Making for Autonomous Humanoid Boxing | π |
| MVISTA-4D: View-Consistent 4D World Model with Test-Time Action Inference | π |
| DPL: Depth-only Perceptive Humanoid Locomotion via Realistic Depth Synthesis | π |
| PoseDiff: A Unified Diffusion Model Bridging Robot Pose Estimation and Video-to-Action Control | π |
| EgoDemoGen: Novel Egocentric Demonstration Generation Enables Viewpoint-Robust Manipulation | π |
| HumanoidVerse: A Versatile Humanoid for Vision-Language Guided Multi-Object Rearrangement | π |
| LOVON: Legged Open-Vocabulary Object Navigator | π |
| Survival Games: Human-LLM Strategic Showdowns under Severe Resource Scarcity | π |
| Occupancy World Model for Robots | π |
| RoboOcc: Enhancing the Geometric and Semantic Scene Understanding for Robots | π |
| The Meta-Representation Hypothesis | π |
| EmbodiedVSR: Dynamic Scene Graph-Guided Chain-of-Thought Reasoning for Visual Spatial Tasks | π |
| HumanoidPano: Hybrid Spherical Panoramic-LiDAR Cross-Modal Perception for Humanoid Robots | π |
| NeuGPT: Unified multi-modal Neural GPT | π |
| Recursive Cleaning for Large-scale Protein Data via Multimodal Learning | π |
| Query-based Semantic Gaussian Field for Scene Representation in RL | π |
| Mamba as Decision Maker: Exploring Multi-scale Sequence Modeling in Offline RL | π |
| MAD: Multi-Alignment MEG-to-Text Decoding | π |
| Manipulation Facing Threats: Evaluating Physical Vulnerabilities in End-to-End VLA Models | π |
| E2H: A Two-Stage Non-Invasive Neural Signal Driven Humanoid Robotic Whole-Body Control | π |
| Typography Leads Semantic Diversifying: Amplifying Adversarial Transferability across MLLMs | π |
| A Dual-Agent Adversarial Framework for Robust Generalization in Deep RL | π |
π½ Click to expand publications
| Venue | Title | Link |
|---|---|---|
| IROS 2024 | Whole-body Humanoid Robot Locomotion with Human Reference | π |
| CVPR 2026 | SwiftVLA: Unlocking Spatiotemporal Dynamics for Lightweight VLA Models | π |
| ICLR 2026 | Compose Your Policies! Improving Diffusion/Flow Robot Policies via Test-time Composition | π |
| ICLR 2026 | ArtVIP: Articulated Digital Assets for Robot Learning | π |
| ICRA 2026 | Physics-informed Diffusion Mamba Transformer for Real-world Driving | π¬ |
| ICRA 2026 | TopoNav: Topological Graphs as a Key Enabler for Advanced Object Navigation | π |
| ICRA 2026 | Learning Structural Latent Points for Efficient Visual Representations in Robotic Manipulation | π¬ |
| AAAI 2026 | What You See is What You Reach: Spatial Navigation with High-Level Human Instructions | π |
| ICASSP 2026 | NeuSpeech: Decode Neural Signal as Speech | π |
| ICML 2025 | Simple Policy Optimization | π |
| RSS 2025 | RoboMIND: Benchmark on Multi-embodiment Intelligence Normative Data for Robot Manipulation | π |
| IEEE TVCG | DEGS: Deformable Event-based 3D Gaussian Splatting | π |
| CoRL 2025 | Omni-Perception: Omnidirectional Collision Avoidance for Legged Locomotion | π |
| ICCV 2025 | What Makes for Text to 360-degree Panorama Generation with Stable Diffusion? | π |
| ICCV 2025 | Learning Null Geodesics for Gravitational Lensing Rendering in General Relativity | π |
| ACMMM 2025 | Transfer Attack for Bad and Good: Adversarial Transferability across MLLMs | π |
| IROS 2025 | Mamba Policy: Towards Efficient 3D Diffusion Policy with Hybrid Selective State Models | π |
| IROS 2025 | Distillation-PPO: Two-Stage RL Framework for Humanoid Perceptive Locomotion | π |
| ACL 2025 | MapNav: Novel Memory Representation via Annotated Semantic Maps for VLN | π |
| CVPR 2025 | Uncovering Vision Modality Threats in Image-to-Image Tasks | π |
| ICRA 2025 | Multi-Floor Zero-Shot Object Navigation Policy | π |
| ICASSP 2025 | Fully Spiking Neural Network for Legged Robots | π |
| ICASSP 2025 | Event Masked Autoencoder: Point-wise Action Recognition | π |
| ICME 2025 | ES-Parkour: Advanced Robot Parkour with Bio-inspired Event Camera & SNN | π |
| IJCAI 2025 π | Exploring Typographic Visual Prompts Injection Threats in Cross-Modality Models | π |
| PM2CE@IROS 2025 | Humanoid Occupancy: Generalized Multimodal Occupancy Perception for Humanoid Robots | π |
| H2R@CoRL 2025 | UniTracker: Universal Whole-Body Motion Tracker for Humanoid Robots | π |
| Sim2Real@Humanoids 2025 | LiPS: Large-Scale Humanoid Robot RL with Parallel-Series Structures | π |
| Sim2Real@Humanoids 2025 | Trinity: A Modular Humanoid Robot AI System | π |
| GenModels@ICLR 2025 | Modality-Composable Diffusion Policy via Distribution-level Composition | π |
| NeurIPS 2024 | DEL: Discrete Element Learner for Learning 3D Dynamics from 2D Observations | π |
| NeurIPS 2024 | Spiking Neural Network as Adaptive Event Stream Slicer | π |
| IEEE TAI | Spiking Diffusion Models | π |
| IROS 2024 | Reinforcement Learning with Generalizable Gaussian Splatting | π |
| IROS 2024 | TriHelper: Zero-Shot Object Navigation with Dynamic Assistance | π |
| ICRA 2024 | Prompting Multi-Modal Tokens for End-to-End Autonomous Driving with LLMs | π |
| ICRA 2024 | Prompt, Plan, Perform: LLM-based Humanoid Control via Quantized Imitation Learning | π |
| WACV 2024 | Spiking Denoising Diffusion Probabilistic Models | π |
| ICCV 2023 | Masked Spiking Transformer | π |
| CoRL 2022 | RoboTube: Learning Household Manipulation from Human Videos | π |
π‘ I have hidden some previous work β feel free to chat! Currently preparing my personal website.
Python |
C++ |
C |
JavaScript |
R |
Bash |
MATLAB |
|
PyTorch |
TensorFlow |
OpenCV |
Docker |
Linux |
Ubuntu |
RedHat |
GitHub |
Git |
GitLab |
AWS |
CMake |
Jenkins |
RPi |
|
VSCode |
PyCharm |
Vim |
Sublime |
HTML5 |
PS |
AI |