aiqbench analysis
PTDL introduces a novel training framework that decouples recovery and locomotion phases while stratifying supervision across diverse terrains, enabling a single proprioceptive policy to handle slopes, gravel, and uneven ground without explicit terrain sensing at deployment. By validating this approach on a 29-DoF humanoid robot in both simulation and hardware, the work addresses critical gaps in prior methods that often compromise performance when trained on mixed environments. This advancement represents a significant step toward robust, real-world humanoid autonomy capable of self-recovery in unstructured settings.
PTDL:Multi-Terrain Fall Recovery via Phase-Terrain Decoupled Learning
arXiv:2606.08922v1 Announce Type: new Abstract: Humanoid robots can fall on slopes, gravel, and uneven ground in unstructured environments. We target integrated fall recovery and locomotion: rebuilding balance from a fallen state using proprioception alone and resuming velocity-commanded walking at the fall site. Prior methods often stop at quasi-static rise, neglect the post-fall ground-contact phase, or, when trained on mixed terrains without separating recovery and locomotion phases or per-surface constraints, collapse to a single compromise get-up across surfaces. We propose Phase--Terrain Decoupled Learning (PTDL), which decouples training supervision along phase and terrain axes while deploying one proprioceptive policy. On the phase axis, projected-gravity-gated dual motion-prior discriminators and a probe-to-walk transition link post-fall recovery to commanded walking. On the terrain axis, terrain-stratified recovery shaping assigns surface-specific training supervision on flat ground, gravel, and slopes; terrain labels are training-only and withheld from policy observations, enabling implicit post-fall strategy selection at deployment. We validate PTDL on a 29-DoF Unitree G1 across flat ground, gravel, and slopes up to 20 degrees in simulation and on hardware, achieving stable cross-terrain recovery, smooth recovery-to-locomotion transitions, and differentiated post-fall rise behaviors under one deployed policy.
Abstract:Humanoid robots can fall on slopes, gravel, and uneven ground in unstructured environments. We target integrated fall recovery and locomotion: rebuilding balance from a fallen state using proprioception alone and resuming velocity-commanded walking at the fall site. Prior methods often stop at quasi-static rise, neglect the post-fall ground-contact phase, or, when trained on mixed terrains without separating recovery and locomotion phases or per-surface constraints, collapse to a single compromise get-up across surfaces. We propose Phase--Terrain Decoupled Learning (PTDL), which decouples training supervision along phase and terrain axes while deploying one proprioceptive policy. On the phase axis, projected-gravity-gated dual motion-prior discriminators and a probe-to-walk transition link post-fall recovery to commanded walking. On the terrain axis, terrain-stratified recovery shaping assigns surface-specific training supervision on flat ground, gravel, and slopes; terrain labels are training-only and withheld from policy observations, enabling implicit post-fall strategy selection at deployment. We validate PTDL on a 29-DoF Unitree G1 across flat ground, gravel, and slopes up to 20 degrees in simulation and on hardware, achieving stable cross-terrain recovery, smooth recovery-to-locomotion transitions, and differentiated post-fall rise behaviors under one deployed policy.
Robotics (cs.RO)
arXiv:2606.08922 [cs.RO]
(or arXiv:2606.08922v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.08922 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 275 words
Related coverage
Stock Market Today Highlights, Sept 2: Sensex ends 373 pts lower at 76, 570, Nifty below 24, 000; Asian Paints, HDFC Bank fall most
thehindubusinessline.com · Sep 2, 2026
Unified Perceptive Fall Safety for Humanoids
arXiv · Oct 16, 2026
Restore LLM Inference Capacity in Seconds with Shadow Engine Recovery in NVIDIA Dynamo
NVIDIA · Aug 25, 2026
UniReLo: Learning a Unified Humanoid Policy from Fall Recovery to Locomotion across Diverse Terrains
arXiv · Jul 29, 2026