📚 Humanoid Robot Learning Paper Notebooks
Paper reading notes on humanoid robot reinforcement learning
Total: 135 papers
⭐️ open-source code link in note
📍 推荐学习路线图
No matching papers found.
Foundational RL (15)
Foundational RL theory and classic algorithms — essential prerequisites for understanding subsequent research
- Proximal Policy Optimization Algorithms (PPO)
- Advantage Weighted Regression (AWR) ⭐️
- DeepMimic: Example-Guided Deep RL of Physics-Based Character Skills ⭐️
- AMP: Adversarial Motion Priors for Stylized Physics-Based Character Control ⭐️
- PHC: Perpetual Humanoid Control for Real-time Simulated Avatars ⭐️
- ADD: Adversarial Disentanglement and Distillation ⭐️
- ASE: Adversarial Skill Embeddings for Large-Scale Motion Control ⭐️
- CALM: Conditional Adversarial Latent Models for Directable Virtual Characters ⭐️
- PULSE: Universal Humanoid Motion Representations for Physics-Based Control ⭐️
- Diffusion Policy: Visuomotor Policy Learning via Action Diffusion ⭐️
- BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusion
- Understanding Domain Randomization for Sim-to-real Transfer
- Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World
- Learning Smooth Humanoid Locomotion through Lipschitz-Constrained Policies (LCP)
- MimicKit: A Reinforcement Learning Framework for Motion Imitation and Control ⭐️
Motion Retargeting (3)
Bridging human motion data and humanoid policies — geometric IK, learned retargeting, and the engineering choices that determine downstream policy quality
High Impact Selection (23)
Curated high-impact and milestone papers in humanoid robot learning
Whole-Body Control Core (6)
- Expressive Whole-Body Control for Humanoid Robots ⭐️
- HOVER: Versatile Neural Whole-Body Controller for Humanoid Robots ⭐️
- ExBody2: Advanced Expressive Humanoid Whole-Body Control
- HugWBC: A Unified and General Humanoid Whole-Body Controller for Versatile Locomotion ⭐️
- SONIC: Supersizing Motion Tracking for Natural Humanoid Whole-Body Control ⭐️
- UH-1: Learning from Massive Human Videos for Universal Humanoid Pose Control ⭐️
Teleoperation & Imitation Learning (3)
Locomotion Classics (6)
- Real-World Humanoid Locomotion with Reinforcement Learning
- Humanoid Locomotion as Next Token Prediction
- Humanoid Parkour Learning ⭐️
- Learning Sim-to-Real Humanoid Locomotion in 15 Minutes ⭐️
- ECO: Energy-Constrained Optimization with Reinforcement Learning for Humanoid Walking ⭐️
- Learning Quadrupedal Locomotion over Challenging Terrain
Sim-to-Real & Foundation Model (4)
Simulation Platform & Tools (4)
- Humanoid-Gym: Reinforcement Learning for Humanoid Robot with Zero-Shot Sim2Real Transfer ⭐️
- BEHAVIOR Robot Suite: Streamlining Real-World Whole-Body Manipulation for Everyday Household Activities ⭐️
- ProtoMotions3: An Open-source Framework for Humanoid Simulation and Control ⭐️
- Isaac Lab: Unified GPU Simulation Platform for Robot Learning ⭐️
Loco-Manipulation and WBC (32)
- LATENT: Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data ⭐️
- Ψ₀: An Open Foundation Model Towards Universal Humanoid Loco-Manipulation ⭐️
- SteadyTray: Learning Object Balancing Tasks in Humanoid Tray Transport via Residual Reinforcement Learning ⭐️
- ZeroWBC: Learning Natural Visuomotor Humanoid Control Directly from Human Egocentric Video
- Embedding Classical Balance Control Principles in Reinforcement Learning for Humanoid Recovery
- ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulation
- OmniXtreme: Breaking the Generality Barrier in High-Dynamic Humanoid Control
- LessMimic: Long-Horizon Humanoid Interaction with Unified Distance Field Representations
- VIGOR: Visual Goal-In-Context Inference for Unified Humanoid Fall Safety
- Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matching
- MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction
- General Humanoid Whole-Body Control via Pretraining and Fast Adaptation ⭐️
- HAIC: Humanoid Agile Object Interaction Control via Dynamics-Aware World Model
- EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration ⭐️
- MOSAIC: Bridging the Sim-to-Real Gap in Generalist Humanoid Motion Tracking and Teleoperation with Rapid Residual Adaptation ⭐️
- Learning Human-Like Badminton Skills for Humanoid Robots
- TextOp: Real-time Interactive Text-Driven Humanoid Robot Motion Generation and Control ⭐️
- Humanoid Manipulation Interface: Humanoid Whole-Body Manipulation from Robot-Free Demonstrations
- HiWET: Hierarchical World-Frame End-Effector Tracking for Long-Horizon Humanoid Loco-Manipulation
- Learning Soccer Skills for Humanoid Robots: A Progressive Perception-Action Framework
- PDF-HR: Pose Distance Fields for Humanoid Robots
- HUSKY: Humanoid Skateboarding System via Physics-Aware Whole-Body Control ⭐️
- Embodiment-Aware Generalist Specialist Distillation for Unified Humanoid Whole-Body Control
- HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videos ⭐️
- TTT-Parkour: Rapid Test-Time Training for Perceptive Robot Parkour
- ZEST: Zero-shot Embodied Skill Transfer for Athletic Robot Control
- Robust and Generalized Humanoid Motion Tracking
- RoboStriker: Hierarchical Decision-Making for Autonomous Humanoid Boxing
- PILOT: A Perceptive Integrated Low-level Controller for Loco-manipulation over Unstructured Scenes
- Semantic Co-Speech Gesture Synthesis and Real-Time Control for Humanoid Robots
- GentleHumanoid: Learning Upper-body Compliance for Contact-rich Human and Object Interaction ⭐️
- HERO: Learning Humanoid End-Effector Control for Open-Vocabulary Visual Loco-Manipulation ⭐️
Manipulation (6)
- HumDex: Humanoid Dexterous Manipulation Made Easy ⭐️
- cuRoboV2: Dynamics-Aware Motion Generation with Depth-Fused Distance Fields for High-DoF Robots ⭐️
- DreamDojo: A Generalist Robot World Model from Large-Scale Human Videos ⭐️
- HumanoidVLM: Vision-Language-Guided Impedance Control for Contact-Rich Humanoid Manipulation
- Generalizable Geometric Prior and Recurrent Spiking Feature Learning for Humanoid Robot Manipulation ⭐️
- EgoMimic: Scaling Imitation Learning via Egocentric Video
Teleoperation (6)
- CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperation ⭐️
- ExtremControl: Low-Latency Humanoid Teleoperation with Direct Extremity Control ⭐️
- TeleGate: Whole-Body Humanoid Teleoperation via Gated Expert Selection with Motion Prior
- SEW-Mimic: A Closed-Form Geometric Retargeting Solver for Upper Body Humanoid Robot Teleoperation
- Learning Adaptive Neural Teleoperation for Humanoid Robots: From Inverse Kinematics to End-to-End Control
- HumanPlus: Humanoid Shadowing and Imitation from Humans ⭐️
Locomotion (8)
- Biomechanical Comparisons Reveal Divergence of Human and Humanoid Gaits
- APEX: Learning Adaptive High-Platform Traversal for Humanoid Robots
- Now You See That: Learning End-to-End Humanoid Locomotion from Raw Pixels ⭐️
- A Hybrid Autoencoder for Robust Heightmap Generation from Fused Lidar and Depth Data for Humanoid Robot Locomotion
- Scalable and General Whole-Body Control for Cross-Humanoid Locomotion (XHugWBC)
- Extreme Parkour with Legged Robots ⭐️
- ANYmal Parkour: Learning Agile Navigation for Quadrupedal Robots
- Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning ⭐️
Navigation (6)
- EgoActor: Grounding Task Planning into Spatial-aware Egocentric Actions for Humanoid Robots via Visual-Language Models
- FocusNav: Spatial Selective Attention with Waypoint Guidance for Humanoid Local Navigation
- Thinking in 360°: Humanoid Visual Search in the Wild ⭐️
- Quantum deep reinforcement learning for humanoid robot navigation task
- STATE-NAV: Stability-Aware Traversability Estimation for Bipedal Navigation on Rough Terrain
- NaVILA: Legged Robot Vision-Language-Action Model for Navigation
State Estimation (6)
- AutoOdom: Learning Auto-regressive Proprioceptive Odometry for Legged Locomotion
- InEKFormer: A Hybrid State Estimator for Humanoid Robots
- Physics-Informed Neural Networks with Unscented Kalman Filter for Sensorless Joint Torque Estimation in Humanoid Robots ⭐️
- An Empirical Evaluation of Four Off-the-Shelf Proprietary Visual-Inertial Odometry Systems
- Contact-Aided Invariant Extended Kalman Filtering for Legged Robot State Estimation
- Legged Robot State-Estimation Through Combined Forward Kinematic and Preintegrated Contact Factors ⭐️
Sim-to-Real (6)
Core theory and algorithms are under 01 Foundational RL (Domain Randomization, LCP). This category will host Sim-to-Real-specific works (Real-to-Sim, ADR, Privileged Learning, etc.).
- RAPT: Model-Predictive Out-of-Distribution Detection and Failure Diagnosis for Sim-to-Real Humanoid Robots
- LIFT: Towards Bridging the Gap between Large-Scale Pretraining and Efficient Finetuning for Humanoid Control ⭐️
- PolySim: Bridging the Sim-to-Real Gap for Humanoid Control via Multi-Simulator Dynamics Randomization ⭐️
- Contrastive Representation Learning for Robust Sim-to-Real Transfer of Adaptive Humanoid Locomotion
- Towards Bridging the Gap: Systematic Sim-to-Real Transfer for Diverse Legged Robots（PACE） ⭐️
- RMA: Rapid Motor Adaptation for Legged Robots
Hardware Design (6)
- Characteristics, Management, and Utilization of Muscles in Musculoskeletal Humanoids
- Fauna Sprout: A lightweight, approachable, developer-ready humanoid robot
- Antagonistic Bowden-Cable Actuation of a Lightweight Robotic Hand
- Olaf: Bringing an Animated Character to Life in the Physical World
- OSMO: Open-Source Tactile Glove for Human-to-Robot Skill Transfer ⭐️
- Unitree H1 Humanoid Robot Whitepaper & Specifications
Simulation Benchmark (6)
- Towards Motion Turing Test: Evaluating Human-Likeness in Humanoid Robots
- MolmoSpaces: A Large-Scale Open Ecosystem for Robot Navigation and Manipulation ⭐️
- Benchmarking Humanoid Imitation Learning with Motion Difficulty
- Generative World Modelling for Humanoids: 1X World Model Challenge Technical Report ⭐️
- GRUtopia: Dream General Robots in a City at Scale ⭐️
- HumanoidBench: Simulated Humanoid Benchmark for Whole-Body Locomotion and Manipulation ⭐️
Physics-Based Animation (6)
- Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid Control
- CRISP: Contact-Guided Real2Sim from Monocular Video with Planar Scene Primitives ⭐️
- Mimic2DM: Generating and Mimicking 2D Motions for 3D Character Control
- PhysHMR: Learning Humanoid Control Policies from Vision for Physically Plausible Human Motion Reconstruction ⭐️
- Learning to Ball: Composing Policies for Long-Horizon Basketball Moves ⭐️
- Character Controllers using Motion VAEs
Human Motion (6)
- EmbodMocap: In-the-Wild 4D Human-Scene Reconstruction for Embodied Agents ⭐️
- WHOLE: World-Grounded Hand-Object Lifted from Egocentric Videos
- MAGNet: Diffusion Forcing for Multi-Agent Interaction Sequence Modeling ⭐️
- Learned Motion Matching ⭐️
- Generating Diverse and Natural 3D Human Motions from Textual Descriptions ⭐️
- Control Operators for Interactive Character Animation ⭐️