#Unitree (45)

X-WBC: A Cross-Embodiment Foundation Model for Humanoid Whole-Body Control

Scaling humanoid whole-body control toward general-purpose deployment requires large human motion corpora and training experience shared across robot bodies. Existing methods usually train one policy per robot, leaving motion experience isolated across embodiments. We introduce X-WBC, a cross-embodiment foundation framework that separates relatively shared human motion semantics from embodiment-specific physical execution. Human-centered command tokens align full human motion, robot reference motion, and sparse VR observations. A causal Transformer learns reusable temporal structure from mixed multi-robot rollouts, while lightweight robot-specific modules map the shared representation to each robot's proprioception and action space. Across nine simulated embodiments, external motions, and four real robots, experiments show that joint training improves tracking, the aligned representation supports consistent control across command sources, and the learned policy remains competitive beyond the training corpus. These results support heterogeneous humanoids as joint data sources and establish cross-embodiment joint training as a practical route toward whole-body control foundation models.

Juntong Zhang, Chun Gu, Li ZhangSep 14, 2026

HumanoidWhole-Body ControlCross-Embodiment LearningSep 14, 2026

0:30

X-WBC: A Cross-Embodiment Foundation Model for Humanoid Whole-Body Control

@vigram_void · 131 views · 2026-09-16

HumanoidWhole-Body ControlCross-Embodiment

World-Model-Augmented Visual Locomotion for Humanoids on Foothold-Constrained Terrain

Foothold-constrained terrain is characterized by sparse, discontinuous, or geometrically restricted feasible foot contacts, as encountered on stepping stones, across gaps, and on narrow stair treads. On such terrain, a single misstep often leaves little room to recover, so policies that base foot-placement decisions primarily on the immediately visible terrain are prone to failure. We ask whether a learned predictive summary of near-future observations and rewards can provide the anticipatory information required in such settings. We present World-Model-Augmented Visual Locomotion (WM-LOCO), which jointly trains a recurrent world model and a PPO policy. Conditioned on proprioception and a single onboard depth image, the world model produces a predictive recurrent feature that guides the policy, without explicit foothold labels. In simulation, WM-LOCO succeeds on gaps and stepping stones where a matched baseline fails completely, and matches the baseline's success rate on stairs while improving stride efficiency and reducing pelvis acceleration. We deploy the same policy onboard a physical Unitree G1 humanoid using onboard proprioception and a single depth stream; it traverses all three terrain classes with an average success rate of 93.3%.

Liu, Yuxi, Han, Lijun, Wang, ZimingSep 2, 2026

HumanoidLegged LocomotionWorld ModelsSep 2, 2026

LEGS: Fine-Tuning Teleop-Free VLAs for Humanoid Loco-manipulation in an Embodied Gaussian Splatting World

Training vision-language-action (VLA) policies for humanoid loco-manipulation is constrained by the high cost and complexity of collecting human teleoperation demonstrations. VLA policies fine-tuned in simulators have, until now, failed to transfer effectively in humanoid loco-manipulation tasks. We present LEGS (Loco-manipulation via Embodied Gaussian Splatting), a hybrid simulator that composites a mesh foreground (robot, objects, props) over a photorealistic 3D Gaussian Splatting (3DGS) background reconstructed from a handheld scene capture. LEGS uses a procedural motion-primitive generator to synthesize labeled demonstrations at scale without human teleoperation, and a deterministic two-stage color calibration to align the rendered 3DGS image to the robot's deployment camera. On a Unitree G1 humanoid robot, across three pick-and-place tasks of increasing whole-body difficulty and three VLA backbones (psi_0, pi_0.5, GR00T N1.6), a policy trained purely on LEGS data matches or exceeds one trained on human teleoperation demos on every experiment. It also outperforms a mesh-only simulation baseline that ablates the effect of the 3DGS background, showing that photorealistic rendering is a key enabler for synthetic data transfer. Humanoid motion is recorded independently of scene appearance in LEGS, allowing the same auto-generated demonstrations to be re-rendered under new backgrounds and object meshes–covering a new scene at more than 15x lower cost than teleoperation–to augment training data for robustness to scene variations. Under combined object-and-scene appearance shift, the policy trained on re-rendered LEGS-AUG data maintains task success while the baseline trained on teleoperation data fails entirely. Our project page is located at https://legsvla.github.io/.

Kim, Hojune, Chen, Timothy, Sun, JiankaiMay 31, 2026

HumanoidVLA3D Gaussian SplattingMay 31, 2026

FADA: Few-Shot Domain Adaptation via Dynamics Alignment for Humanoid Control

High-precision humanoid control is limited by target-domain dynamics mismatch, where the same control objective can induce different realized motions under changes in terrain, payload, or actuator response. Existing methods either pursue zero-shot transfer through domain randomization or in-context adaptation without target-domain specialization, or require heavy adaptation pipelines that leverage target-domain data, such as model calibration, residual learning, or policy retraining. In this paper, we present FADA (Few-Shot Domain Adaptation via Dynamics Alignment), a three-stage Planner-Inverse Dynamics Model (Planner-IDM) framework for few-shot adaptation in humanoid control. FADA first trains an oracle policy with privileged information and then distills the oracle behavior into a deployable Planner-IDM student through DAgger. At deployment, FADA freezes the planner and finetunes only the IDM using approximately 2 minutes of target-domain rollouts with standard supervised learning. Rather than requiring optimal demonstrations or rewards, FADA uses the paired actions and observations that are observed during these rollouts as supervision, aligning the IDM's action generation with target-domain dynamics. Experiments show that FADA outperforms both in-context and end-to-end adaptation baselines, improving task performance under dynamics shifts and enabling real humanoid robots to execute diverse high-precision whole-body tasks. Implementation details and qualitative hardware rollout videos are available at https://lecar-lab.github.io/FADA-humanoid/.

Angchen Xie, Nikhil Sobanbabu, Ishayu ShikhareJun 26, 2026

HumanoidFew-Shot LearningDomain AdaptationJun 26, 2026

1:36

FADA: 2-Minute Humanoid Dynamics Adaptation

@AngchenXie · 213 views · 2026-07-01

HumanoidDomain AdaptationInverse Dynamics

2:49

Unitree opens UnifoLM-WLA-1.0

@UnitreeRobotics · 231 views · 2026-09-10

HumanoidVLAUnitree

Video Friday: Digit Redecorates — 14 Real-World Clips, From a Couch Drag to a Grape Harvest Above 50 °C

IEEE Spectrum's Video Friday collected fourteen robotics clips this week. Nearly all of them show robots working in the real world rather than in a render. Agility's Digit drags a couch across a living room; SteadyTray's ReST-RL hierarchical reinforcement learning keeps unsecured payloads level on a Unitree G1, generalizing zero-shot sim-to-real across objects and external force disturbances; Figure argues that a robot in every home is a safety-and-cost problem rather than a data-and-compute one; CMU's APEX replaces choreographed motion with adaptive full-body maneuvers. NC State builds teardrop-shaped liquid-crystal elastomer soft robots that leap for as long as infrared light is on, with no battery, controller, or onboard sensing, and the University of Tokyo fits the musculoskeletal humanoid Musashi-W with a three-layer joint-covering skin carrying 44 pressure- and stretch-sensitive elements. DEEP Robotics' Lynx M20S hauls grape baskets in Turpan, where summer ground temperatures exceed 50 degrees Celsius, while ROBOTIS' first OH! GYM! student cohort goes from simulated motions to a physical AI Sapiens K1 in one month. The edition also carries the September-to-November event calendar: Humanoids Summit Seoul, IROS 2026, and CoRL 2026.

BLOG

Video FridayHumanoidLegged robotsSim-to-RealSoft robots

RoboGesture: Real-Time Semantic-aligned Co-Speech Gestures Generation for Humanoid Interaction

Enabling humanoid robots to respond to human speech with synchronized and semantically meaningful gestures is fundamental to natural human-robot interaction. However, this task faces three critical barriers: the scarcity of semantically rich datasets, the "modality eclipse" where models ignore audio cues in favor of kinematic inertia, and the sim-to-real gap regarding physical safety. We propose RoboGesture, a robot-centric framework that co-designs data, modeling, and control to power a complete interactive human-humanoid system in which the robot listens, responds, and gestures in real time. We first establish the RoboGesture dataset featuring over 300 gesture categories and develop an automated pipeline to synthesize large-scale collision-free, robot-specific audio-motion pairs. Our architecture features a Hierarchical Semantic-Acoustic Aligner that extracts multi-granular prosodic and semantic cues directly from raw audio tokens. These cues drive a Streaming Conditional Motion Generator based on a diffusion transformer with conditional flow matching. To ensure high responsiveness, we introduce Anti-Inertia CFG Masking, which prevents the model from collapsing into repetitive historical patterns by compelling it to proactively mine control signals from the audio modality. Finally, an MPC-based safety filter ensures real-time, collision-free execution on physical hardware. Experiments on a Unitree G1 humanoid demonstrate that RoboGesture generates safer, more rhythmic, and more semantically appropriate responses compared to state-of-the-art baselines.

Wang, Zifan, Ren, Ziang, Shi, PengyangAug 27, 2026

MPCSim-to-RealUnitreeAug 27, 2026