🔭 I'm the Director of Embodied AI research at the China National and Local co-built Innovation Center of Embodied AI Robotics. (I'm looking forward to collaborating on exciting projects in this field!) Previously, I participated in 'Tianhe 天河' supercomputing projects in China and also had a very pleasant working experience at DJI.

🌟 I’m currently working on Humanoid Robots:

• TianGong Humanoid Robot (China Universial Humanoid Robot Platform)
• Humanoid Robot Adam
• Humanoid Robot GR1 (2021~2023)

🗞️ News

• 2024-12-30: Invited by the National Health Commission, I participated in the review of the artificial intelligence medical project as an expert reviewer.
• 2024-11-25: Interviewed by the Global Times. Advancements in AI for football and robot marathons
• 2024-11-16: Invited to attend the AGIROS Open Source Community Conference at the Institute of Software, Chinese Academy of Sciences, and delivered a keynote speech on 'Embodied Intelligence of Humanoid Robots'.
• 2024-11-9: Join the Intel China Academic Talent Program and delivered a keynote speech on "Research on Embodied AI of Humanoid Robots".
• 2024-11-5: Invited to attend the BAAI "ZhiYuan Forum - Embodiment and World Model Summit".
• 2024-9-30: Give a talk at the School of Computer Science, Peking University, on Multimodal Perception and Large Model Decision-Making for Humanoid Robots.
• 2024-7-29: Interviewed by Mango (Hunan) TV on the topic of "Embodied AI and Humanoid Robots Tian Gong".
• 2024-7-5: Invited by the China Internet Research Institute to draft a white paper on embodied intelligence in the industrial internet, presenting a report on the integration and innovation of new industrial networks and humanoid robot embodied intelligence technology.
• 2024-7-1: Invited to give a talk in XMech at Zhejiang University.
• 2024-5-29: Interviewed by the Beijing Association for Science and Technology.
• 2024-5-9: The humanoid robot platform 'TianGong' is continuously being iterated and developed.

📝 I’m currently learning Embodied AI, RL, Vision Perception, LLM, Control&Plan in Robotics. Here are some papers I've published recently :

• Whole-body Humanoid Robot Locomotion with Human Reference (IROS2024)
• DEL: Discrete Element Learner for Learning 3D Dynamics from 2D Observations(NeurIPS2024)
• Spiking Neural Network as Adaptive Event Stream Slicer(NeurIPS2024)
• Prompting Multi-Modal Tokens to Enhance End-to-End Autonomous Driving Imitation Learning with LLMs (ICRA2024)
• Prompt, Plan, Perform: LLM-based Humanoid Control via Quantized Imitation Learning (ICRA2024)
• Reinforcement Learning with Generalizable Gaussian Splatting (IROS2024)
• TriHelper: Zero-Shot Object Navigation with Dynamic Assistance (IROS2024)
• Spiking Diffusion Models (IEEE Transactions on Artificial Intelligence)
• Fully spiking neural network for legged robots (ICASSP 2025)
• Event Masked Autoencoder: Point-wise Action Recognition with Event-Based Cameras (ICASSP 2025)
• Masked Spiking Transformer (ICCV 2023)
• Spiking Denoising Diffusion Probabilistic Models (WACV 2024)
• RoboMIND: Benchmark on Multi-embodiment Intelligence Normative Data for Robot Manipulation
• NeuGPT: Unified multi-modal Neural GPT
• Recursive Cleaning for Large-scale Protein Data via Multimodal Learning
• Uncovering Vision Modality Threats in Image-to-Image Tasks
• Mamba Policy: Towards Efficient 3D Diffusion Policy with Hybrid Selective State Models
• Query-based Semantic Gaussian Field for Scene Representation in Reinforcement Learning
• Mamba as Decision Maker : Exploring Multi-scale Sequence Modeling in Offline Reinforcement Learning
• Simple Policy Optimization
• MAD: Multi-Alignment MEG-to-Text Decoding
• NeuSpeech: Decode Neural signal as Speech
• Multi-Floor Zero-Shot Object Navigation Policy
• Manipulation Facing Threats: Evaluating Physical Vulnerabilities in End-to-End Vision Language Action Models
• Multi-Floor Zero-Shot Object Navigation Policy
• E2H: A Two-Stage Non-Invasive Neural Signal Driven Humanoid Robotic Whole-Body Control Framework
• Typography Leads Semantic Diversifying: Amplifying Adversarial Transferability across Multimodal Large Language Models
• RoboTube: Learning Household Manipulation from Human Videos with Simulated Twin Environments (CoRL2022)

👯 I have hidden some of my previous work, welcome to chat. I am currently preparing my personal website and more, and I’m looking to collaborate on Humanoid Robots, Embodied AI, RL, Vision Perception, LLM, Control & Planning in Robotics

📫 How to reach me [email protected] | [email protected]