Humanoid Robotics
Whole-body control, locomotion, manipulation, teleoperation, and robot platforms that make physical intelligence measurable.
HKUST(GZ) | Humanoid Robotics | Embodied AI
Humanoid Computing Lab
We study humanoid computing systems that connect robot bodies, world models, perception, control, and brain-inspired intelligence.
Director
Prof. Renjing Xu
Openings
PhD, postdoc, visitors
Mission
Building generalizable humanoid intelligence from hardware to cognition.
HCL explores how intelligent agents can sense, predict, move, and adapt in the physical world. The lab combines humanoid robotics, embodied AI, world models, learning-based control, and brain-inspired computing to pursue systems with human-level efficiency and adaptability.
Research
Four connected research lines
World Models
Predictive models that connect perception, physics, memory, planning, and long-horizon behavior in embodied agents.
Embodied Perception
Multimodal sensing, panoramic perception, tactile feedback, and geometry-aware representations for robots in open environments.
Brain-inspired Computing
Algorithms, devices, and efficient computing principles inspired by biological intelligence and adaptive behavior.
Inside HCL
Real-world spaces for building and testing embodied intelligence.
Lab Culture
Hardware-grounded, model-driven, and publication-ready.
We value complete research loops: build the robot or simulator, design the model, run the evaluation, and communicate the insight clearly.
Robot learning
Physical simulation
Embodied foundation models
Efficient intelligence
Recent Work
Latest 20 publications from Google Scholar
2026
RoboDojo: A Unified Sim-and-Real Benchmark for Comprehensive Evaluation of Generalist Robot Manipulation Policies
2026
PL-LIT: A LiDAR-Inertial-Thermal SLAM Using Point-Line Features and Thermographic Mapping
2026
Not All Actions Are Equal: Rethinking Conditioning for Dexterous World Model
2026
A Neuromorphic Reinforcement Learning Framework for Efficient Pathfinding in Robotic Mobile Fulfillment Systems
2026
HumanoidArena: Benchmarking Egocentric Hierarchical Whole-body Learning
2026
Action-Effect Memory Pretraining for Robot Manipulation
2026
VAIC: Vision-Guided Humanoid Agile Object Interaction Control via Decoupled Commands
2026
CoRe-MoE: Contrastive Reweighted Mixture of Experts for Multi-Terrain Humanoid Locomotion with Gait Adaptation
2026
GPU-Parallel Multi-Task Reinforcement Learning with Demonstration Guided Policy Optimization
2026
SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents
2026
Dynamic Token Masking in Spiking Neural Network: Y. Fang et al.
2026
Tabero: Learning Gentle Manipulation with Closed-Loop Force Feedback from Vision, Touch, and Language
2026
ParkourFormer: Integrating Predictive Supervision and Sequence Modeling into Parkour Locomotion
2026
MoSA: Motion-constrained Stress Adaptation for Mitigating Real-to-Sim Gap in Continuum Dynamics via Learning Residual Anisotropy
2026
OHP-RL: Online Human Preference as Guidance in Reinforcement Learning for Robot Manipulation
2026
What Limits Vision-and-Language Navigation?
2026
Sparse transmission in diffractive deep neural networks via optical spiking neurons
2026
RobotPan: A 360 Surround-View Robotic Vision System for Embodied Perception
2026
Beyond Viewpoint Generalization: What Multi-View Demonstrations Offer and How to Synthesize Them for Robot Manipulation?
2026
Morphology-Consistent Humanoid Interaction through Robot-Centric Video Synthesis
News
Latest news from HCL
RoboCup 2026 World Champions
Team Apollo3D won first place in the RoboCupSoccer Simulation League 3D competition at RoboCup 2026 in Incheon, Korea.
First Place at the ICRA 2026 WBCD Competition
Team MilkDragon won first place in the Deformable Track (Vienna) of What Bimanuals Can Do (WBCD) at ICRA 2026.
