Pith. sign in

REVIEW 14 cited by

Learning Humanoid Locomotion over Challenging Terrain

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2410.03654 v1 pith:N4KNUKB4 submitted 2024-10-04 cs.RO cs.LG

Ilija Radosavovic , Sarthak Kamat , Trevor Darrell , Jitendra Malik This is my paper · ORCID
classification cs.ROcs.LG
keywords humanoidmodelterrainterrainscapablechallengingcontrollerslearning
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Humanoid robots can, in principle, use their legs to go almost anywhere. Developing controllers capable of traversing diverse terrains, however, remains a considerable challenge. Classical controllers are hard to generalize broadly while the learning-based methods have primarily focused on gentle terrains. Here, we present a learning-based approach for blind humanoid locomotion capable of traversing challenging natural and man-made terrain. Our method uses a transformer model to predict the next action based on the history of proprioceptive observations and actions. The model is first pre-trained on a dataset of flat-ground trajectories with sequence modeling, and then fine-tuned on uneven terrain using reinforcement learning. We evaluate our model on a real humanoid robot across a variety of terrains, including rough, deformable, and sloped surfaces. The model demonstrates robust performance, in-context adaptation, and emergent terrain representations. In real-world case studies, our humanoid robot successfully traversed over 4 miles of hiking trails in Berkeley and climbed some of the steepest streets in San Francisco.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 14 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. In vivo feasibility study of humanoid robots in surgery

    cs.RO 2026-07 accept novelty 7.0 of 10

    Teleoperated Unitree G1 humanoids using manual wristed instruments completed two in-vivo porcine cholecystectomies and dry-lab tasks with performance between manual laparoscopy and commercial surgical robots.

  2. Visual Imitation Enables Contextual Humanoid Control

    cs.RO 2025-05 conditional novelty 7.0 of 10

    A single policy trained from 123 monocular videos, fine-tuned in simulation, and distilled to heightmap plus root-direction inputs lets a Unitree G1 climb stairs and sit and stand on real furniture.

  3. MuJoCo Playground

    cs.RO 2025-02 conditional novelty 7.0 of 10

    An open-source, MJX-based robot learning framework with integrated batch rendering that provides fast training and demonstrates sim-to-real transfer on six robot platforms.

  4. Physics-Guided Biomechanical Gait Adaptation for Humanoid Locomotion on Extreme Sloped Terrains

    cs.RO 2026-07 conditional novelty 6.0 of 10

    A proprioceptive humanoid policy trained with slope-adaptive ZMP regularization plus biomechanical reward gating traverses outdoor grass slopes to 32.1° without online exteroception.

  5. PHUMA: Physically Reliable Humanoid Locomotion Dataset

    cs.RO 2025-10 conditional novelty 6.0 of 10

    PHUMA is a curated 73-hour humanoid locomotion corpus whose physical-reliability metrics are partly defined by the same losses used to optimize it, and whose imitation success claims are confounded by in-distribution ...

  6. KLEIYN : A Quadruped Robot with an Active Waist for Both Locomotion and Wall Climbing

    cs.RO 2025-07 conditional novelty 6.0 of 10

    KLEIYN, a quadruped with an active waist joint, climbs chimneys 800-1000 mm wide at roughly 150 mm/s using a reinforcement learning curriculum that eases the wall-floor transition from curved to vertical.

  7. GMT: General Motion Tracking for Humanoid Whole-Body Control

    cs.RO 2025-06 conditional novelty 6.0 of 10

    GMT trains a single unified humanoid policy using adaptive sampling and mixture-of-experts, achieving lower tracking errors than a re-implemented ExBody2 across diverse whole-body motions.

  8. TWIST: Teleoperated Whole-Body Imitation System

    cs.RO 2025-05 conditional novelty 6.0 of 10

    Human MoCap drives a Unitree G1 humanoid in real time through a single teacher-student RL+BC controller that transfers zero-shot from simulation.

  9. BeamDojo: Learning Agile Humanoid Locomotion on Sparse Footholds

    cs.RO 2025-02 conditional novelty 6.0 of 10

    A two-stage RL framework with a polygonal-foot foothold reward and double critic enables a Unitree G1 humanoid to traverse sparse footholds in simulation and the real world.

  10. Learning from Massive Human Videos for Universal Humanoid Pose Control

    cs.RO 2024-12 conditional novelty 6.0 of 10

    Humanoid-X contributes 163,800 text-annotated motion clips retargeted from human videos into humanoid robot poses, and UH-1 is an autoregressive transformer that maps text instructions to humanoid actions.

  11. Semantic Audio-driven Understanding for Dynamic Humanoid Whole Body Control

    cs.RO 2026-07 conditional novelty 5.0 of 10

    A multi-modal audio router maps streaming music and speech to imitation-learned whole-body policies for a Unitree G1 humanoid, achieving 84.8% chunk-level retrieval accuracy in simulation.

  12. Whole-Body Conditioned Egocentric Video Prediction

    cs.CV 2025-06 conditional novelty 5.0 of 10

    An autoregressive conditional diffusion transformer predicts future egocentric video from whole-body 3D pose sequences, trained on Nymeria, with atomic action and long-horizon evaluations.

  13. HiLo: Learning Whole-Body Human-like Locomotion with Motion Tracking Controller

    cs.RO 2025-02 conditional novelty 5.0 of 10

    A humanoid locomotion controller that combines open-loop reference tracking with a residual RL policy transfers to the real robot and changes gait style by reweighting the reference motion.

  14. Robust RL Control for Bipedal Locomotion with Closed Kinematic Chains

    cs.RO 2025-07 conditional novelty 4.0 of 10

    A reinforcement-learning gait controller that explicitly models closed kinematic chains outperforms one trained on a simplified serial model, both in simulation and on the physical TopA robot.

Pith tools