#Unitree G1 (70)

BRACE: Adapting Whole-Body References for Force and Terrain Aware Humanoid Motion Tracking

Whole-body tracking has become the interface through which operators drive humanoid robots, yet the references it consumes are recorded on level ground and carrying nothing, so the tracker knows neither the forces the robot must exchange with objects nor the terrain it stands on. Existing controllers address one side of the gap: force-capable policies command an end-effector force but prescribe no whole-body pose, while terrain-adaptive trackers treat loads as disturbances to reject rather than wrenches to command. BRACE folds both into the reference the tracker follows: terrain conformance lifts footholds and root onto the local surface, and a wrench transformation resolves the hand displacement that produces a commanded force together with the center-of-mass and center-of-pressure shifts it induces, bounded by teacher-specific arm-effort limits. Separate exertion and compensation teachers are distilled by DAgger into one flow-matching student that runs on proprioception alone, without a height map or measured wrench, so a teleoperator can supply a flat-ground trajectory while the robot resolves slope and load onboard. On the Unitree G1, in simulation and on hardware, BRACE cuts global tracking error by 40% on average and improves success rate by 18.25 points.

Sudarshan Harithas, Chen Yu, Juan Borbon · 40 views · Oct 5, 2026

HumanoidWhole-Body ControlForce controlOct 5, 2026

2:43

PanTact: Teaching Whole-Body Dexterity with a Wearable Contact Interface

PanTact treats contact, not joint trajectories, as the transfer currency from people to humanoids. Wearable patches are assembled magnetically from 44x44 mm FSR modules; a custom 50x50 mm PCB streams 16 channels at 1 kHz over a CRC-checked RS-485 daisy chain, one Vive tracker per patch maps sensor centers into the world frame, and contact states are binarized against a per-channel adaptive baseline with threshold max(0.35 beta, r_min). The policy observes body-masked 3D egocentric point clouds, both wrist images and proprioception, and at 10 Hz predicts an action chunk of H=16 steps containing a look-at point, EEF poses with gripper widths, and per-sensor contact goals (binary state plus 3D position). A soft-minimum proximity d(p,q) with T = 0.02 m hands each active contact to whichever robot surface is currently nearest, and active goals are offset by delta along the pressing direction to form virtual goals that generate pressing force, so the G1 holds objects without any tactile sensor on its body. Three control formulations trade off differently: training-free upper-body IK, WBIK whose dense reference is executed by SONIC, and a sparse contact tracker trained with PPO over 4096 simulation environments for 30k iterations (50 Hz, 29 joints, S=8 upper-body / S=16 whole-body contact slots). On a 29-DoF Unitree G1: Grasp-and-Embrace reaches 85% with upper-body IK and 80% with the sparse tracker against 0% for Humanoid UMI; Sweep-and-Stow reaches 90.63% end to end while Humanoid UMI++ scores 18.75% on each stage and 0% end to end; Carry-and-Push is 90% versus 40%. In simulation replay, adding contact goals lowers contact distance for all three controllers at almost no cost in EEF accuracy.

Xu, Xiaomeng, Zhang, Han, Ruan, Zhian · 133 views · Oct 8, 2026

Whole-Body ManipulationTactile SensingContact-Rich ManipulationOct 8, 2026

HomeBody: A Humanoid That Explores, Remembers, and Acts on Its Own

Stanford TML's HomeBody drops the learned VLA middle layer: GPT Astra calls a library of composable skills on a Unitree G1, backed by a digital twin and persistent spatial memory the robot builds by exploring a home it has never seen, then tidies an unseen kitchen and fetches medicine with no task-specific training.

BLOG

Humanoid robotsEmbodied AIVLM AgentReal2SimMobile Manipulation

PredActor: Predictive Action Diffusion for Steerable Onboard Humanoid Control

Diffusion models provide a flexible framework for motion generation, but turning this flexibility into closed-loop humanoid control remains challenging. Hierarchical generator-tracker systems steer motion through reference trajectories, yet these references may exceed the capabilities of the downstream tracker, leaving physical feasibility and disturbance recovery largely to a separate control module. Action-only diffusion avoids this separation by directly generating executable actions, but provides no explicit future-state trajectory that can be steered toward test-time motion objectives. Joint state-action diffusion offers a natural alternative, but existing controllers often rely on privileged full-body states, while learned behavior selection and test-time motion steering remain only partially integrated. We present PredActor, a predictive action diffusion policy that unifies both steering modes in one directly executed policy using proprioception alone. Given proprioceptive history and optional task context, PredActor jointly predicts actions and an internal future-state trajectory that enables guidance: classifier-free guidance strengthens text-conditioned motion, while classifier guidance steers future states toward test-time objectives. Only actions are executed, requiring neither a motion-reference tracker nor privileged full-body states. In simulation, PredActor reaches 44 of 45 destination targets and achieves a text retrieval score of 0.539 versus 0.424 for conditional action diffusion, with similar disturbance survival. Rolling denoising and computation-preserving runtime optimizations reduce the complete callback to 16.790 ms median and 19.383 ms p95 on a Jetson Orin NX, within the 20 ms control period. Deployed on a Unitree G1, PredActor demonstrates text-conditioned motion, disturbance response, joystick control, and semantic interpolation in simulation and hardware.

Ye, Lei, Gao, Haibo, Li, Yitang · 137 views · Sep 21, 2026

Humanoid ControlDiffusion policiesWhole-Body ControlSep 21, 2026

HuMBLE: Human Motion-Driven Behavior Learning for Embodied Locomotion

Despite recent advances in humanoid locomotion, controllers optimized for command tracking and robustness tend to produce mechanical gaits, whereas controllers tied to human motion data often fail to generalize to commands outside the data distribution. This work introduces a learning framework that balances these competing objectives to synthesize real-time steerable, robust, and biomimetic locomotion policies from human data. Using an in-house curated locomotion dataset covering diverse speeds and directions, we first learn a natural locomotion prior policy through a teacher-student distillation process. Specifically, we train a full-body reference-conditioned policy with Reinforcement Learning (RL), then distill it into a lightweight prior policy conditioned solely on proprioception and a planar torso-velocity steering command. Next, we fine-tune the prior policy with multi-task RL to expand command coverage and robustness beyond the data distribution, pairing a goal-conditioned task that tracks arbitrary commands with a reference-guided task that tracks the human data as an explicit style regularizer. We validate our framework on three humanoid robots: the Boston Dynamics Atlas R1, Atlas D1, and Unitree G1. Experimental results demonstrate robust performance across real-world scenarios, including direct user-controlled locomotion in indoor and outdoor environments, and integration as the locomotion layer within hierarchical control stacks. Benchmarks against Tabula Rasa RL policies trained without human data and ablation studies confirm that our framework yields a lightweight, deployable policy that reconstructs coordinated whole-body behavior from a steering command, retaining the human gait characteristics while remaining robust and fully steerable.

Mike Zhang, Dongho Kang, Kevin Bergamin · 537 views · Oct 7, 2026

Humanoid LocomotionMotion ImitationReinforcement LearningOct 7, 2026

1:40

DexWeave: Learning Dexterous Humanoid Loco-Manipulation from Human Demonstrations

@ErenChenAI · 216 views · 2026-10-09

Mobile ManipulationHumanoidMotion retargeting

1:17

PredActor: One Diffusion Policy Unifies Humanoid Motion Generation and Control

@techniahqrobot · 202 views · 2026-10-08

PredActorDiffusion policiesHumanoid

2:50

OTRetarget: joint robot-and-object motion retargeting via optimal transport preserves contacts across morphologies

@CarnErwann · 551 views · 2026-09-30

HumanoidLoco-manipulationMotion retargeting

WB-WAM: Heterogeneous Body-Hand Pre-training for Humanoid Loco-Manipulation

WB-WAM from Tsinghua IIIS, Xiong'an Institute of AI and University of Melbourne (Hang Zhao group) injects explicit whole-body action supervision into generative video pre-training: a shared 72-D physical action space (body 29 + root 3 + two hands 20+20) unifies partial annotations from 1,880.2 hours of nine-source heterogeneous video and motion data via masked flow matching, followed by PICO egocentric mid-training (22 h / 73 tasks, GMR retargeting + constrained IK + MINT) and real-robot post-training with a forward-kinematics loss on Unitree G1 with Wuji hands. It wins all seven HumanoidArena tasks (81.9% mean), reaches 84.0% on five real tasks vs OpenWAM's 80.0% (ACT 20%, Fast-WAM 6%), and PICO mid-training lets 30 real demos hit 73.8% — beating 100-demo direct training at 65.0%, a 70% cut in robot data.

Chuan Qin, Shaoting Zhu, Siyuan Luo · 809 views · Sep 28, 2026

WAMWorld-action modelsHumanoidSep 28, 2026

1:07

Uni-VLaT: adding whole-body touch to pretrained VLAs lifts humanoid loco-manipulation from 32% to 75%

@stepjamUK · 523 views · 2026-09-29

Uni-VLaTTactile PerceptionTactile Sensing

0:45

HomeBody: a frontier VLM drives a Unitree G1 skill library directly, with no learned VLA in between

@CyberRobooo · 918 views · 2026-09-28

HumanoidUnitree G1VLM

Generate, Track, Improve: Perceptive Multi-Skill Humanoid Locomotion with RL-Fine-Tuned Motion Generators

General purpose humanoids require locomotion controllers that are multi-skill, perceptive, dynamic, and robust enough to go anywhere humans can. In this work, we present a two layer locomotion architecture: (1) a perceptive flow matching motion generator plans whole body trajectories from raw depth images while a (2) perceptive tracking policy trained with control-guided RL follows these motions. Both policies are trained on a library of terrain consistent motion clips created with dynamically optimized human data which yields both accurate velocity tracking and terrain consistent references. Our central contribution is a simple yet effective off-policy RL fine tuning loop that improves the motion generator. A structured search method is used with the generator to gather data for advantage weighted regression. This off-policy loop is much more sample efficient than on-policy residual fine tuning and improves terrain consistency on unseen geometries and skill compositions. We find that successful terrain traversals increased by up to 25 percentage points and skill selection improved by up to 80 percentage points. By using raw depth images to perceive the environment no odometry or height maps are needed, and outdoor deployment is easy. With two cameras, the policy can see terrain coming from further away and adjust its velocity regardless of the commanded speed so it can traverse the terrain. A single policy pair enables a Unitree G1 humanoid to walk, run, stand, jump on and off of boxes, and traverse stairs in outdoor environments. Project page: https://zolkin1.github.io/generate-track-improve/

Zachary Olkin, William D. Compton, Aaron D. Ames · 661 views · Sep 25, 2026

Humanoid LocomotionPerceptive LocomotionReinforcement LearningSep 25, 2026

0:28

100 Unitree G1s in one MuJoCo world: kung-fu tracking on a real 360-scanned plaza

@cuttini · 591 views · 2026-09-28

Unitree G1MuJoCoSimulation

4:15

Delta-0 Makes a Unitree G1 Play Music, Then Do Household Chores

@CyberRobooo · 632 views · 2026-09-28

HumanoidUnitreeUnitree G1

Rolling-WAM: World Action Models with Rolling Imagination

World Action Models (WAMs) couple action generation with future visual prediction for robotic manipulation. However, completing the joint video-action denoising process at each replanning cycle incurs substantial latency, delaying action updates and limiting closed-loop responsiveness. We present Rolling-WAM, a formulation that distributes joint denoising across successive replanning cycles. Our method maintains a sliding window of video-action chunks at staggered noise levels. At each step, a rolling noise schedule fully denoises the imminent action chunk for execution, while partially refining farther-future chunks. As the window advances with new camera observations, the retained future chunks continue their denoising process. This distributes the computational cost over time while carrying an evolving visual-action context across chunk boundaries. Evaluations on LIBERO, RoboTwin, and a real-world Unitree G1 humanoid show that Rolling-WAM achieves competitive manipulation performance. By removing the need to denoise the entire prediction horizon from scratch, it delivers a 4.5x steady-state replanning speedup over standard joint WAMs.

Yinghua Zhou, Junjie Ye, Yiqi Zhao · 731 views · Sep 24, 2026

Rolling-WAMWorld-action modelsWAMSep 24, 2026

ResSafe: Learning Safety Filtering with Residual Reinforcement Learning for Humanoids

ResSafe decouples performance and safety in humanoid control: a nominal policy optimizes task performance, then a residual RL policy trained on top learns only safety corrections, their sum acting as an implicit min-norm safety filter. On Unitree G1 extreme balance under random payloads, fall rate drops from 7.29% to 1.11% with near-nominal tracking error, generalizing across reference checkpoints and transferring to real hardware.

Qu, Gechen, Zhang, Tong, Zhang, Bike · 616 views · Sep 14, 2026

HumanoidSafety FilterResidual RLSep 14, 2026

BeyondRetarget: Learning Executable Humanoid Motions Directly from Monocular Video

BeyondRetarget (Nanjing University, arXiv:2609.29850) removes the reconstruct-then-retarget pipeline entirely: without any explicit SMPL intermediate representation, it learns a robot-oriented implicit shared motion representation directly from monocular RGB video and decodes full motions for a target humanoid via robot-specific heads. Three key designs: (1) decoders parameterize motion as residuals around learnable per-robot nominal states; a new embodiment adapts in ~1 hour by training only its DoF and root-orientation branches; (2) per-link non-uniform scaling analytically aligns SMPL annotations to robot proportions so human labels supervise robot motion directly, with 18 shared keypoints plus directed semantic links; (3) a contact predictor with a two-level contact/support gate triggers root height/horizontal corrections and lower-body IK, keeping support feet zero-slide. Results: on G1, MAMPJPE 26.36 mm with zero Fail65/Fail100 windows and 95.86% MuJoCo execution success, beating Oracle retargeting fed with SMPL ground truth (direct evidence of retargeting error); across eight humanoids MAMPJPE stays within 26.36-72.05 mm; end-to-end latency is 192.8 ms (vs ~1.3 s two-stage), enabling real-time monocular teleoperation. Code: bear-ty/BeyondRetarget.

Tianyu Xiong, Yi Lu, Jinrui Wang · 708 views · Sep 24, 2026

HumanoidMotion retargetingImitation LearningSep 24, 2026

0:32

TactileStep: Tactile Foot Control for Humanoid Locomotion

@YimingLi9702 · 673 views · 2026-09-26

TactileStepHumanoidTactile Sensing

6:12

Light-O1: Two Humanoids in One Continuous Household Run

@HumanoidRTech · 814 views · 2026-09-25

Light-O1Light OriginsWhole-Body Control

2:23

Light-O1: A Humanoid Foundation Model Pretrained on Human Video

@heetezition · 536 views · 2026-09-22

HumanoidFoundation ModelsAction Tokens