mlsanityData to 14 Oct 2026

KungfuBot: Physics-Based Humanoid Whole-Body Control for Learning Highly-Dynamic Skills

arXiv 2506.12851 ↗First posted 15 Jun 2025cs.RO cs.AIcited by 52

Not on the map. Why?

Cited by 20 in Deep reinforcement learning methods, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Humanoid robots are promising to acquire various skills by imitating human behaviors. However, existing algorithms are only capable of tracking smooth, low-speed human motions, even with delicate reward and curriculum design. This paper presents a physics-based humanoid control framework, aiming to master highly-dynamic human behaviors such as Kungfu and dancing through multi-steps motion processing and adaptive motion tracking. For motion processing, we design a pipeline to extract, filter out, correct, and retarget motions, while ensuring compliance with physical constraints to the maximum extent. For motion imitation, we formulate a bi-level optimization problem to dynamically adjust the tracking accuracy tolerance based on the current tracking error, creating an adaptive curriculum mechanism. We further construct an asymmetric actor-critic framework for policy training. In experiments, we train whole-body control policies to imitate a set of highly-dynamic motions. Our method achieves significantly lower tracking errors than existing approaches and is successfully deployed on the Unitree G1 robot, demonstrating stable and expressive behaviors. The project page is https://kungfubot.github.io.

Key facts. Cited by 52 papers in the map (data through 14 Oct 2026). Builds on 48 of its 60 references that are in the map. First posted on arXiv 15 Jun 2025.

Builds on

48 of 60 references.

References · 48

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2024OpenVLA: An Open-Source Vision-Language-Action Modelcited by 2,887
  3. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  4. 2019AMASS: Archive of Motion Capture as Surface Shapescited by 1,236
  5. 2025GR00T N1: An Open Foundation Model for Generalist Humanoid Robotscited by 1,225
  6. 2021Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learningcited by 998
  7. 2024RDT-1B: a Diffusion Foundation Model for Bimanual Manipulationcited by 691
  8. 2023MotionGPT: Human Motion as a Foreign Languagecited by 527
  9. 2021Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learningcited by 512
  10. 2024Video Prediction Policy: A Generalist Robot Policy with Predictive Visual Representationscited by 289
  11. 2020Learning Agile Robotic Locomotion Skills by Imitating Animalscited by 251
  12. 2021Robust Motion In-betweeningcited by 229
36 more
  1. 2023Perpetual Humanoid Control for Real-time Simulated Avatarscited by 218
  2. 2024OmniH2O: Universal and Dexterous Human-to-Humanoid Whole-Body Teleoperation and Learningcited by 189
  3. 2024HumanPlus: Humanoid Shadowing and Imitation from Humanscited by 185
  4. 2022Capturing and Inferring Dense Full-Body Human-Scene Contactcited by 180
  5. 2024Expressive Whole-Body Control for Humanoid Robotscited by 139
  6. 2024World-Grounded Human Motion Recovery via Gravity-View Coordinatescited by 137
  7. 2024Learning Human-to-Humanoid Real-Time Whole-Body Teleoperationcited by 130
  8. 2025ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skillscited by 130
  9. 2024MaskedMimic: Unified Physics-Based Character Control Through Masked Motion Inpaintingcited by 116
  10. 2023Diffusion Model is an Effective Planner and Data Synthesizer for Multi-Task Reinforcement Learningcited by 115
  11. 20233D Human Pose Estimation via Intuitive Physicscited by 103
  12. 2025TWIST: Teleoperated Whole-Body Imitation Systemcited by 101
  13. 2024Humanoid Parkour Learningcited by 96
  14. 2020Contact and Human Dynamics from Monocular Videocited by 95
  15. 2024ExBody2: Advanced Expressive Humanoid Whole-Body Controlcited by 88
  16. 2024Move as You Say, Interact as You Can: Language-guided Human Motion Generation with Scene Affordancecited by 80
  17. 2025HOMIE: Humanoid Loco-Manipulation with Isomorphic Exoskeleton Cockpitcited by 76
  18. 2025InterMimic: Towards Universal Whole-Body Control for Physics-Based Human-Object Interactionscited by 72
  19. 2025Humanoid Locomotion and Manipulation: Current Progress and Challenges in Control, Planning, and Learningcited by 61
  20. 2024Humanoid-Gym: Reinforcement Learning for Humanoid Robot with Zero-Shot Sim2Real Transfercited by 53
  21. 2024Mobile-TeleVision: Predictive Motion Priors for Humanoid Whole-Body Controlcited by 51
  22. 2024Learning Humanoid Locomotion with Perceptive Internal Modelcited by 50
  23. 2025Learning Humanoid Standing-up Control across Diverse Posturescited by 46
  24. 2025BeamDojo: Learning Agile Humanoid Locomotion on Sparse Footholdscited by 46
  25. 2023An Introduction to Bi-level Optimization: Foundations and Applications in Signal Processing and Machine Learningcited by 41
  26. 2023Robust and Versatile Bipedal Jumping Control through Reinforcement Learningcited by 36
  27. 2024SkillMimic: Learning Basketball Interaction Skills from Demonstrationscited by 30
  28. 2024Learning from Massive Human Videos for Universal Humanoid Pose Controlcited by 29
  29. 2024Learning an Actionable Discrete Diffusion Policy via Large-Scale Actionless Video Pre-Trainingcited by 26
  30. 2025Learning Getting-Up Policies for Real-World Humanoid Robotscited by 26
  31. 2024Multi-Transmotion: Pre-trained Model for Human Motion Predictioncited by 21
  32. 2022Learning and Deploying Robust Locomotion Policies with Minimal Dynamics Randomizationcited by 17
  33. 2024MMVP: A Multimodal MoCap Dataset with Vision and Pressure Sensorscited by 17
  34. 2025Adversarial Locomotion and Motion Imitation for Humanoid Policy Learningcited by 12
  35. 2025Humanoid Whole-Body Locomotion on Narrow Terrain via Dynamic Balance and Reinforcement Learningcited by 11
  36. 2023Bridging the Sim-to-Real Gap from the Information Bottleneck Perspectivecited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 18; 2026: 34.

52 papers.

Sort:

By area:

  1. 2025BeyondMimic: From Motion Tracking to Versatile Humanoid Control via Guided Diffusioncited by 165
  2. 2025WholeBodyVLA: Towards Unified Latent VLA for Whole-Body Loco-Manipulation Controlcited by 36
  3. 2026Perceptive Humanoid Parkour: Chaining Dynamic Human Skills via Motion Matchingcited by 19
  4. 2025PhysHSI: Towards a Real-World Generalizable and Natural Humanoid-Scene Interaction Systemcited by 18
  5. 2025From Language to Locomotion: Retargeting-free Humanoid Control via Motion Latent Guidancecited by 18
  6. 2025DreamControl: Human-Inspired Whole-Body Humanoid Control for Scene Interaction via Guided Diffusioncited by 15
  7. 2025SoftMimic: Learning Compliant Whole-body Control from Examplescited by 14
  8. 2026HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videoscited by 14
  9. 2026TextOp: Real-time Interactive Text-Driven Humanoid Robot Motion Generation and Controlcited by 14
  10. 2025KungfuBot2: Learning Versatile Motion Skills for Humanoid Whole-Body Controlcited by 13
  11. 2025DemoHLM: From One Demonstration to Generalizable Humanoid Loco-Manipulationcited by 11
  12. 2026ZEST: Zero-shot Embodied Skill Transfer for Athletic Robot Controlcited by 11
38 more
  1. 2026FRoM-W1: Towards General Humanoid Whole-Body Control with Language Instructionscited by 10
  2. 2026HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulationcited by 10
  3. 2025Humanoid Goalkeeper: Learning from Position Conditioned Task-Motion Constraintscited by 9
  4. 2025Towards Adaptable Humanoid Control via Adaptive Motion Trackingcited by 8
  5. 2026ZeroWBC: Learning Natural Whole-Body Humanoid Interaction from Human Egocentric Datacited by 8
  6. 2025RoboMirror: Understand Before You Imitate for Video to Humanoid Locomotioncited by 7
  7. 2026MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstructioncited by 6
  8. 2025CHIP: Adaptive Compliance for Humanoid Control through Hindsight Perturbationcited by 5
  9. 2026Humanoid-GPT: Scaling Data and Structure for Zero-Shot Motion Trackingcited by 5
  10. 2025Chasing Stability: Humanoid Running via Control Lyapunov Function Guided Reinforcement Learningcited by 4
  11. 2026CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperationcited by 4
  12. 2026Spatial Memory for Out-of-Vision Manipulation in Vision-Language-Actioncited by 4
  13. 2025RobotDancing: Residual-Action Reinforcement Learning Enables Robust Long-Horizon Humanoid Motion Trackingcited by 3
  14. 2026Learning Human-Like Badminton Skills for Humanoid Robotscited by 3
  15. 2026InterReal: A Unified Physics-Based Imitation Framework for Learning Human-Object Interaction Skillscited by 3
  16. 2026SafeFlow: Real-Time Text-Driven Humanoid Whole-Body Control via Physics-Guided Rectified Flow and Selective Safety Gatingcited by 3
  17. 2026SUGAR: A Scalable Human-Video-Driven Generalizable Humanoid Loco-Manipulation Learning Frameworkcited by 3
  18. 2026PhysDrift: Bridging the Embodiment Gap in Humanoid Co-Speech Motion Generationcited by 3
  19. 2025A Scalable Whole-body Motion Transfer via Implicit Kinodynamic Motion Retargetingcited by 2
  20. 2026Flip Stunts on Bicycle Robots using Iterative Motion Imitationcited by 2
  21. 2026CoorDex: Coordinating Body and Hand Priors for Continuous Dexterous Humanoid Loco-Manipulationcited by 2
  22. 2026Demystifying Reinforcement Learning Post-Training of Language Modelscited by 2
  23. 2025HAFO: A Force-Adaptive Control Framework for Humanoid Robots in Intense Interaction Environmentscited by 1
  24. 2026AdaptManip: Learning Adaptive Whole-Body Object Lifting and Delivery with Online Recurrent State Estimationcited by 1
  25. 2026Generalizing from References using a Multi-Task Reference and Goal-Driven RL Frameworkcited by 1
  26. 2026HALO:Closing Sim-to-Real Gap for Heavy-loaded Humanoid Agile Motion Skills via Differentiable Simulationcited by 1
  27. 2026Switch: Learning Agile Skills Switching for Humanoid Robotscited by 1
  28. 2026LineRides: Line-Guided Reinforcement Learning for Bicycle Robot Stuntscited by 1
  29. 2026PressMimic: Pressure-Guided Motion Capture and Control for Humanoid Robot Imitationcited by 1
  30. 2026HumanTracker: Towards Comprehensive and Human-Aligned Motion Tracking Benchmarkcited by 1
  31. 2025Switch-JustDance: Benchmarking Whole Body Motion Tracking Controllers Using a Commercial Console Gamecited by 0
  32. 2025Kinematics-Aware Multi-Policy Reinforcement Learning for Force-Capable Humanoid Loco-Manipulationcited by 0
  33. 2026PCHC: Enabling Preference Conditioned Humanoid Control via Multi-Objective Reinforcement Learningcited by 0
  34. 2026Tree Learning: A Multi-Skill Continual Learning Framework for Humanoid Robotscited by 0
  35. 2026Re2MoGen: Open-Vocabulary Motion Generation via LLM Reasoning and Physics-Aware Refinementcited by 0
  36. 2026Learn Weightlessness: Imitate Non-Self-Stabilizing Motions on Humanoid Robotcited by 0
  37. 2026Predictive Style Matching: Natural and Robust Humanoid Locomotioncited by 0
  38. 2026Stubborn: A Streamlined and Unified Reinforcement Learning Framework for Robust Motion Tracking and Fall Recovery for Humanoidscited by 0

50 most cited of 52 All 52 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026HULK: Learning Whole-Body Forceful Loco-Manipulation for Humanoids
  2. 2025HuB: Learning Extreme Humanoid Balance
  3. 2025SMAP: Self-supervised Motion Adaptation for Physically Plausible Humanoid Whole-body Control
  4. 2025Generating Physically Realistic and Directable Human Motions from Multi-Modal Inputs
  5. 2026ULTRA: Unified Multimodal Control for Autonomous Humanoid Whole-Body Loco-Manipulation
  6. 2024I-CTRL: Imitation to Control Humanoid Robots Through Constrained Reinforcement Learning
  7. 2025Humanoid Whole-Body Badminton via an Annealed Reinforcement Learning Curriculum
  8. 2026TRACE: Efficient Transfer Learning for Neural Whole-Body Control Across Humanoid Embodiments ↗