MotionDisco discovers long-horizon humanoid loco-manipulation motions from scratch via LLM-guided evolutionary search, trajectory optimization, and pruning, then transfers them to real robots with RL policies.
hub
Omniretarget: Interaction-preserving data gen- eration for humanoid whole-body loco-manipulation and scene interaction
22 Pith papers cite this work. Polarity classification is still indexing.
22
Pith papers citing it
abstract
A dominant paradigm for teaching humanoid robots complex skills is to retarget human motions as kinematic references to train reinforcement learning (RL) policies. However, existing retargeting pipelines often struggle with the significant embodiment gap between humans and robots, producing physically implausible artifacts like foot-skating and penetration. More importantly, common retargeting methods neglect the rich human-object and human-environment interactions essential for expressive locomotion and loco-manipulation. To address this, we introduce OmniRetarget, an interaction-preserving data generation engine based on an interaction mesh that explicitly models and preserves the crucial spatial and contact relationships between an agent, the terrain, and manipulated objects. By minimizing the Laplacian deformation between the human and robot meshes while enforcing kinematic constraints, OmniRetarget generates kinematically feasible trajectories. Moreover, preserving task-relevant interactions enables efficient data augmentation, from a single demonstration to different robot embodiments, terrains, and object configurations. We comprehensively evaluate OmniRetarget by retargeting motions from OMOMO, LAFAN1, and our in-house MoCap datasets, generating over 8-hour trajectories that achieve better kinematic constraint satisfaction and contact preservation than widely used baselines. Such high-quality data enables proprioceptive RL policies to successfully execute long-horizon (up to 30 seconds) parkour and loco-manipulation skills on a Unitree G1 humanoid, trained with only 5 reward terms and simple domain randomization shared by all tasks, without any learning curriculum.
hub tools
citation-role summary
background 2
method 1
citation-polarity summary
years
2026 22representative citing papers
ReActor jointly optimizes motion retargeting and RL policy training with an approximate gradient to generate physically consistent robot motions from human references using only sparse body correspondences.
A DRL policy learns racing controls from depth spectral distributions using a non-geometric physics-informed reward, achieving 12% better performance than humans on out-of-distribution tracks with under 1% of baseline computation.
Rhythm transfers interactive whole-body behaviors from simulation to real dual Unitree G1 humanoids via interaction-aware retargeting and graph-reward RL.
SceneBot conditions a humanoid tracking policy on motion references and contact labels, using reconstructed scene-interaction data to unify free-space locomotion with contact-rich manipulation and terrain tasks.
A data-centric approach shows that less than 3% of AMASS motion data, filtered by physics feasibility, diversity, and complexity, yields better humanoid tracking policies than the full dataset.
PHASOR factorizes motion into an FFT-based phase manifold and pose branch with semantic distillation to produce a cross-embodiment, human-anchored action embedding space for humanoid robots.
Imagine2Real enables zero-shot humanoid-object interaction by unifying motions as 4D point trajectories, tracking only base/hands/object keypoints inside a BFM latent space, and training with progressive simple rewards for mocap deployment.
CEER proposes a compliant end-effector and root control interface that unifies loco-manipulation for humanoids via a distilled low-level policy and hierarchical planners.
SynAgent enables generalizable cooperative humanoid manipulation by transferring skills from solo human-object interactions to multi-agent scenarios via interaction-preserving retargeting, single-agent pretraining with multi-agent PPO, and a conditional VAE generative policy.
Test-time steering of pre-trained whole-body policies via sample-based planning lets legged robots generalize dynamic loco-manipulation to varied heavy objects and tasks without additional training or tuning.
A modular system uses motion matching to compose long-horizon human skill chains, trains RL experts, and distills them into a depth-based policy that lets a Unitree G1 humanoid autonomously climb, vault, and roll over obstacles up to 1.25 m tall.
HAIC enables robust humanoid interactions with underactuated objects by predicting their dynamics from proprioceptive history and using a world model for adaptive control.
HUSKY combines humanoid-skateboard dynamics modeling with adversarial motion priors and physics-guided lean-to-steer strategies to achieve real-world stable skateboarding on a humanoid robot.
HANDOFF is a distilled mixture-of-experts humanoid whole-body controller that follows a compact task-space interface, matches SOTA velocity tracking, provides large manipulation workspace on Unitree G1, and supports VLM-driven agentic planning with no task-specific data.
A hybrid motion-tracking and imitation-reinforcement pipeline produces a depth-based visuomotor policy that lets humanoids climb varied ladders zero-shot on hardware and perform teleoperated manipulation while climbing.
Humanoid-GPT is a causal Transformer pre-trained on a unified billion-scale motion dataset that tracks dynamic behaviors with zero-shot generalization to unseen motions and tasks.
ConstrainedMimic integrates operational space control and control barrier functions into RL tracking policies to enforce arbitrary runtime constraints on humanoid kinematics and dynamics while preserving contact modes and tracking goals.
HumanoidMimicGen automatically generates large loco-manipulation datasets from few source demonstrations using whole-body planning, enabling visuomotor policies that outperform real-data-only training by 20% on a new nine-task benchmark.
A multi-agent LLM framework for humanoid loco-manipulation that separates active spatial perception and task planning from generalizable action generation without task-specific real-robot data.
HTD, a multimodal transformer policy trained with behavioral cloning and touch dreaming to predict future tactile latents, achieves a 90.9% relative success rate improvement over baselines on five real-world contact-rich humanoid loco-manipulation tasks.
SplitAdapter factorizes adaptation into load-aware and dynamics-aware encoders using split world-model objectives, GRL regularization, and hierarchical FiLM, reporting higher full-task success than baselines across 2-6 kg masses and 0-60 cm heights.
citing papers explorer
-
Sumo: Dynamic and Generalizable Whole-Body Loco-Manipulation
Test-time steering of pre-trained whole-body policies via sample-based planning lets legged robots generalize dynamic loco-manipulation to varied heavy objects and tasks without additional training or tuning.
-
Learning Versatile Humanoid Manipulation with Touch Dreaming
HTD, a multimodal transformer policy trained with behavioral cloning and touch dreaming to predict future tactile latents, achieves a 90.9% relative success rate improvement over baselines on five real-world contact-rich humanoid loco-manipulation tasks.