mlsanityData to 7 Oct 2026

ExBody2: Advanced Expressive Humanoid Whole-Body Control

arXiv 2412.13196 ↗First posted 17 Dec 2024cs.RO cs.AI cs.LGcited by 75

Not on the map. Why?

Abstract (arXiv)

This paper tackles the challenge of enabling real-world humanoid robots to perform expressive and dynamic whole-body motions while maintaining overall stability and robustness. We propose Advanced Expressive Whole-Body Control (Exbody2), a method for producing whole-body tracking controllers that are trained on both human motion capture and simulated data and then transferred to the real world. We introduce a technique for decoupling the velocity tracking of the entire body from tracking body landmarks. We use a teacher policy to produce intermediate data that better conforms to the robot's kinematics and to automatically filter away infeasible whole-body motions. This two-step approach enabled us to produce a student policy that can be deployed on the robot that can walk, crouch, and dance. We also provide insight into the trade-off between versatility and the tracking performance on specific motions. We observed significant improvement of tracking performance after fine-tuning on a small amount of data, at the expense of the others.

Key facts. Cited by 75 papers in the map (data through 7 Oct 2026). Builds on 42 of its 64 references that are in the map. First posted on arXiv 17 Dec 2024.

Builds on

42 of 64 references.

References · 42

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
  3. 2019AMASS: Archive of Motion Capture as Surface Shapescited by 1,208
  4. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 882
  5. 2020Learning Quadrupedal Locomotion over Challenging Terraincited by 494
  6. 2021RMA: Rapid Motor Adaptation for Legged Robotscited by 430
  7. 2024Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperationcited by 377
  8. 2024Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robotscited by 373
  9. 2020HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimationcited by 304
  10. 2020Learning Agile Robotic Locomotion Skills by Imitating Animalscited by 229
  11. 2023MimicPlay: Long-Horizon Imitation Learning by Watching Human Playcited by 214
  12. 2021Character Controllers Using Motion VAEscited by 182
30 more
  1. 2024HumanPlus: Humanoid Shadowing and Imitation from Humanscited by 164
  2. 2024EgoMimic: Scaling Imitation Learning via Egocentric Videocited by 160
  3. 2024DexCap: Scalable and Portable Mocap Data Collection System for Dexterous Manipulationcited by 149
  4. 2023Real-World Humanoid Locomotion with Reinforcement Learningcited by 132
  5. 2024Expressive Whole-Body Control for Humanoid Robotscited by 117
  6. 2024Learning Human-to-Humanoid Real-Time Whole-Body Teleoperationcited by 112
  7. 2023Universal Humanoid Motion Representations for Physics-Based Controlcited by 109
  8. 2023Synthesizing Physical Character-Scene Interactionscited by 102
  9. 2024MaskedMimic: Unified Physics-Based Character Control Through Masked Motion Inpaintingcited by 101
  10. 2021Reinforcement Learning for Robust Parameterized Locomotion Control of Bipedal Robotscited by 95
  11. 2023CALM: Conditional Adversarial Latent Models for Directable Virtual Characterscited by 91
  12. 2022From One Hand to Multiple Hands: Imitation Learning for Dexterous Manipulation from Single-Camera Teleoperationcited by 79
  13. 2024HOVER: Versatile Neural Whole-Body Controller for Humanoid Robotscited by 76
  14. 2021Blind Bipedal Stair Traversal via Sim-to-Real Reinforcement Learningcited by 73
  15. 2022Adversarial Motion Priors Make Good Substitutes for Complex Reward Functionscited by 68
  16. 2024UMI on Legs: Making Manipulation Policies Mobile with Manipulation-Centric Whole-body Controllerscited by 67
  17. 2024Reinforcement Learning for Versatile, Dynamic, and Robust Bipedal Locomotion Controlcited by 63
  18. 2024Humanoid Locomotion as Next Token Predictioncited by 63
  19. 2024OKAMI: Teaching Humanoid Robots Manipulation Skills through Single Video Imitationcited by 62
  20. 2023Deep Imitation Learning for Humanoid Loco-manipulation through Human Teleoperationcited by 49
  21. 2024HRP: Human Affordances for Robotic Pre-Trainingcited by 40
  22. 2024ARCap: Collecting High-quality Human Demonstrations for Robot Learning with Augmented Reality Feedbackcited by 31
  23. 2023Learning Vision-Based Bipedal Locomotion for Challenging Terraincited by 30
  24. 2021The MIT Humanoid Robot: Design, Motion Planning, and Control For Acrobatic Behaviorscited by 28
  25. 2023HumanMimic: Learning Natural Locomotion and Transitions for Humanoid Robot via Wasserstein Adversarial Imitationcited by 23
  26. 2023Learning Whole-body Manipulation for Quadrupedal Robotcited by 20
  27. 2019Whole-Body Geometric Retargeting for Humanoid Robotscited by 19
  28. 2024Berkeley Humanoid: A Research Platform for Learning-based Controlcited by 19
  29. 2024Strategy and Skill Learning for Physics-based Table Tennis Animationcited by 17
  30. 2024Learning Multi-Modal Whole-Body Control for Real-World Humanoid Robotscited by 16

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 45; 2026: 30.

75 papers.

Sort:

By area:

  1. 2025ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skillscited by 104
  2. 2025Being-H0: Vision-Language-Action Pretraining from Large-Scale Human Videoscited by 94
  3. 2025TWIST: Teleoperated Whole-Body Imitation Systemcited by 86
  4. 2025InterMimic: Towards Universal Whole-Body Control for Physics-Based Human-Object Interactionscited by 67
  5. 2025HOMIE: Humanoid Loco-Manipulation with Isomorphic Exoskeleton Cockpitcited by 65
  6. 2025Visual Imitation Enables Contextual Humanoid Controlcited by 57
  7. 2025TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenizationcited by 46
  8. 2025Humanoid-VLA: Towards Universal Humanoid Control with Visual Integrationcited by 43
  9. 2025LeVERB: Humanoid Whole-Body Control with Latent Vision-Language Instructioncited by 40
  10. 2025KungfuBot: Physics-Based Humanoid Whole-Body Control for Learning Highly-Dynamic Skillscited by 39
  11. 2025WholeBodyVLA: Towards Unified Latent VLA for Whole-Body Loco-Manipulation Controlcited by 32
  12. 2025Humanoid Everyday: A Comprehensive Robotic Dataset for Open-World Humanoid Manipulationcited by 22
38 more
  1. 2025Learning Getting-Up Policies for Real-World Humanoid Robotscited by 21
  2. 2025Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skillscited by 21
  3. 2025HuB: Learning Extreme Humanoid Balancecited by 21
  4. 2025Towards Immersive Human-X Interaction: A Real-Time Framework for Physically Plausible Motion Synthesiscited by 20
  5. 2025PhysHSI: Towards a Real-World Generalizable and Natural Humanoid-Scene Interaction Systemcited by 17
  6. 2025From Language to Locomotion: Retargeting-free Humanoid Control via Motion Latent Guidancecited by 17
  7. 2025From Experts to a Generalist: Toward General Whole-Body Control for Humanoid Robotscited by 16
  8. 2025DreamControl: Human-Inspired Whole-Body Humanoid Control for Scene Interaction via Guided Diffusioncited by 13
  9. 2025Towards Embodiment Scaling Laws in Robot Locomotioncited by 12
  10. 2025KungfuBot2: Learning Versatile Motion Skills for Humanoid Whole-Body Controlcited by 11
  11. 2025A Unified and General Humanoid Whole-Body Controller for Versatile Locomotioncited by 10
  12. 2025Neural Brain: A Neuroscience-inspired Framework for Embodied Agentscited by 10
  13. 2025RL from Physical Feedback: Aligning Large Motion Models with Humanoid Controlcited by 10
  14. 2026FRoM-W1: Towards General Humanoid Whole-Body Control with Language Instructionscited by 10
  15. 2025SkillBlender: Towards Versatile Humanoid Whole-Body Loco-Manipulation via Skill Blendingcited by 9
  16. 2026ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulationcited by 9
  17. 2026MotionWAM: Towards Foundation World Action Models for Real-Time Humanoid Loco-Manipulationcited by 8
  18. 2025SENTINEL: A Fully End-to-End Language-Action Model for Humanoid Whole Body Controlcited by 7
  19. 2025Towards Adaptable Humanoid Control via Adaptive Motion Trackingcited by 6
  20. 2025RoboMirror: Understand Before You Imitate for Video to Humanoid Locomotioncited by 6
  21. 2025StyleLoco: Generative Adversarial Distillation for Natural Humanoid Robot Locomotioncited by 5
  22. 2025GentleHumanoid: Learning Upper-body Compliance for Contact-rich Human and Object Interactioncited by 5
  23. 2025UniAct: Unified Motion Generation and Action Streaming for Humanoid Robotscited by 5
  24. 2026MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstructioncited by 5
  25. 2026ZeroWBC: Learning Natural Whole-Body Humanoid Interaction from Human Egocentric Datacited by 5
  26. 2026ω-0: A Latent Predictive World Action Model for Concurrent Humanoid Loco-Manipulationcited by 5
  27. 2025Natural Humanoid Robot Locomotion with Generative Motion Priorcited by 4
  28. 2025Coordinated Humanoid Manipulation with Choice Policiescited by 4
  29. 2026Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Trackingcited by 4
  30. 2025Towards Synthesized and Editable Motion In-Betweening Through Part-Wise Phase Representationcited by 3
  31. 2025Task Tokens: A Flexible Approach to Adapting Behavior Foundation Modelscited by 3
  32. 2025JAEGER: Dual-Level Humanoid Whole-Body Controllercited by 3
  33. 2025Multi-Embodiment Robotic Retargeting via Guided Diffusion Modelcited by 3
  34. 2026InterReal: A Unified Physics-Based Imitation Framework for Learning Human-Object Interaction Skillscited by 3
  35. 2025FLAM: Foundation Model-Based Body Stabilization for Humanoid Locomotion and Manipulationcited by 2
  36. 2025MaskedManipulator: Versatile Whole-Body Manipulationcited by 2
  37. 2025RobotDancing: Residual-Action Reinforcement Learning Enables Robust Long-Horizon Humanoid Motion Trackingcited by 2
  38. 2025PolySim: Bridging the Sim-to-Real Gap for Humanoid Control via Multi-Simulator Dynamics Randomizationcited by 2

50 most cited of 75 All 75 (JSON)