Unitree G-1

The 'Unitree G1' dataset/benchmark contains motion data used to evaluate the effectiveness of whole-body tracking models in transferring learned behaviors across different humanoid robot embodiments.

Papers using Unitree G-1 (16)

Benchmarking Model Predictive Control And Reinforcement Learning Based Control For Legged Robot Locomotion In Mujoco Simulation2025 · 8 citesAny2Any: Efficient Cross-Embodiment Transfer for Humanoid Whole-Body Tracking2026DiT4DiT: Jointly Modeling Video Dynamics and Actions for Generalizable Robot Control2026Morphology-Consistent Humanoid Interaction through Robot-Centric Video Synthesis2026ZEST: Zero-shot Embodied Skill Transfer for Athletic Robot Control2026FRoM-W1: Towards General Humanoid Whole-Body Control with Language Instructions2026World-Coordinate Human Motion Retargeting via SAM 3D Body2025TrajBooster: Boosting Humanoid Whole-Body Manipulation via Trajectory-Centric Learning2025CLF-RL: Control Lyapunov Function Guided Reinforcement Learning2025From Tabula Rasa to Emergent Abilities: Discovering Robot Skills via Real-World Unsupervised Quality-Diversity2025MULE: Multi-terrain And Unknown Load Adaptation For Effective Quadrupedal Locomotion2025Learning Sim-to-real Humanoid Locomotion In 15 Minutes2025Robotdancing: Residual-action Reinforcement Learning Enables Robust Long-horizon Humanoid Motion Tracking2025E-SDS: Environment-aware See It, Do It, Sorted - Automated Environment-aware Reinforcement Learning For Humanoid Locomotion2025McARL:Morphology-Control-Aware Reinforcement Learning for Generalizable Quadrupedal Locomotion2025MULE: Multi-terrain and Unknown Load Adaptation for Effective Quadrupedal Locomotion2025