Unified Perceptive Fall Safety for Humanoids

Osher Azulay; Zhengjie Xu; Andrew Scheffer; Stella X. Yu

cs.ROFeb 18, 2026

Unified Perceptive Fall Safety for Humanoids

Authors: Osher Azulay, Zhengjie Xu, Andrew Scheffer, Stella X. Yu

Organizations: University of Michigan, Ann Arbor

Abstract

Reliable fall recovery is critical for humanoids operating in cluttered environments. Unlike quadrupeds or wheeled robots, humanoids undergo high-energy impacts, complex whole-body contact, and large viewpoint changes during falls, making recovery essential. Prior work treats fall safety as separate problems (avoidance, impact mitigation, stand-up) or learns end-to-end policies without vision, typically on flat terrain. More fundamentally, it couples pose, dynamics, and terrain into a single data space, requiring exhaustive coverage and limiting generalization. We present a unified visually grounded fall-safety approach spanning all phases of recovery. It builds on two insights: (1) human fall-recovery poses are structured and transferable across terrains via alignment, and (2) effective recovery depends on interpreting the recovery goal in relation to the surrounding terrain. We train a privileged teacher using sparse human demonstrations and simulated terrains, and distill it into a deployable student that relies on egocentric depth and proprioception. The student predicts a goal-in-context latent representing the recovery target in relation to its surroundings, guiding coordinated whole-body reactions. Experiments in simulation and on a Unitree G1 demonstrate robust zero-shot fall safety across diverse terrains. Project page: https://vigor2026.github.io/.

Figures & tables

Panel of Figure 1 — Fig. 1: Vision-enabled unified fall safety for humanoids . Left: fall-mitigation recovery after a push towards stairs demands vision-guided anticipatory arm contact to arrest the fall. Right: recovery behaviors across diverse terrain and initial conditions highlight the generality of vision-guided contact selection and whole-body rebalancing.

Fig. 2: Factorized data generation yields sample-efficient imitation and scalable adaptation for humanoid fall safety learning. Rather than treating pose , time , and terrain as a single monolithic data space requiring exhaustive coverage, we factorize it into a small set of human pose trajectories from real-world demonstrations on flat terrain and independently varying terrain geometry in simulation.

Fig. 3: Overview of VIGOR . 1) Human fall–recovery demonstrations collected on flat ground are kinematically retargeted to the robot. 2) Reference poses are used directly on flat terrain and coarsely projected onto uneven terrain to provide sparse tracking targets. 3) A privileged RL teacher learns a goal-in-context latent representing the immediate recovery target in relation to local terrain . 4) A student learns through latent supervision and action imitation, using only egocentric depth and proprioceptive history at deployment.

Motion imitation
Reward	Definition	Weight
Pos. track.	$\exp(-\\|\Delta x\\|^{2}/\sigma)$	$1.25$
Rot. track.	$\exp(-\\|\Delta\theta\\|^{2}/\sigma)$	$0.5$
Lin. vel. track.	$\exp(-\\|\Delta v\\|^{2}/\sigma)$	$0.125$
Ang. vel. track.	$\exp(-\\|\Delta\omega\\|^{2}/\sigma)$	$0.125$
Joint pos. track.	$\exp(-\\|\Delta q\\|^{2}/\sigma)$	$0.5$

TABLE I: Reward terms used for fall-recovery learning .

Fig. 4: Training terrains, from left to right: flat, wave, rough, slope, stairs, and inverted stairs. One representative difficulty is shown.

Fig. 5: Recovery across terrain conditions. Solid bars show SSR ; bars show SR (%). Dividers mark unseen rubble, excluded from training.

Fig. 6: Demonstration scaling. Recovery success (SR) and stable success (SSR) for visual students supervised by teachers trained with one, four, or nine human demonstrations, under Stand-Up and Fall-Recovery.

Panel of Figure 8 — Fig. 7: Stair recovery illustration: w.o Vision (top) and VIGOR (bottom), progressing from left to right.

Panel of Figure 9 — Fig. 8: Real-world recovery sequences from the project teaser: (A) diagonal stairs fall, (B) fall toward stairs, (C) stand-up with raised legs, and stand-up across a gap from (D) face-down and (E) face-up configurations. Four selected frames per example show the starting pose or fall, contact transitions, and recovery from left to right; insets show synchronized depth.

Fig. 9: Real-world recovery performance across surfaces and initial configurations. Top: stand-up. Bottom: fall recovery. Bars show success over 5 trials; denotes successful recoveries with unsafe upper-body contact.

Humanoid Robot Fall Recovery

See all 13 papers

CG-MuTra: Continuously-Gated Multi-Terrain Adaptive Recovery for Unified Humanoid Fall Recovery and Locomotion
Jun 8, 2026Xiaoyu Xu, Zhiming Chen, Yuenan Zhao +2Humanoid Robot Locomotion Robot Failure Recovery
School of Control Science and Engineering, Shandong University, Jinan, 250061, China · Key Laboratory of Machine Intelligence and System Control, Ministry of Education, China
Test-Time Motion Steering for Perceptive Humanoid Fall Recovery
Oct 21, 2026Osher Azulay, Cheng-Lin Hsieh, Stella X. YuHumanoid Robot Fall Recovery Humanoid Robot Control
University of Michigan, Ann Arbor
StableMimic: Smooth Human-Like Recovery for Humanoid Motion Tracking - Learning Beyond the Tracking Distribution for Structured Post-Fall Behavior
Aug 3, 2026Weihao Wu, Ming Huang, Ruofei Liu +3Robot Failure Recovery Humanoid Robot Fall Recovery
Southern University of Science and Technology (SUSTech), Shenzhen, China.

Jun 8, 2026J/K move · Enter open · S save

CG-MuTra: Continuously-Gated Multi-Terrain Adaptive Recovery for Unified Humanoid Fall Recovery and Locomotion

Xiaoyu Xu, Zhiming Chen, Yuenan Zhao, Ran Song+1

School of Control Science and Engineering, Shandong University, Jinan, 250061, China · Key Laboratory of Machine Intelligence and System Control, Ministry of Education, China

Humanoid Robot Locomotion Robot Failure Recovery Terrain-Aware Robot Locomotion Humanoid Robot Fall Recovery Robot Locomotion Robust RL+1

Open paper AskPDF arXiv

Falling is an inherent risk for humanoid robots operating in unstructured environments. Existing reinforcement learning methods that leverage expert motion priors are predominantly trained on flat-ground fall-recovery tasks and typically rely on hard switching between separate recovery and locomotion controllers. As a result, such policies struggle to achieve smooth and robust recovery behaviors when deployed on complex terrains such as slopes and gravel. This paper presents CG-MuTra, a unified continuously-gated multi-scale discriminator framework for multi-terrain adaptive fall recovery. CG-MuTra introduces a proprioceptively-derived continuous gate

α= f(z_{\mathrm{root}}, s)

that softly blends three discriminators operating at different temporal horizons: frame-level stability (

Φ_{\mathrm{frame}}

H=1

), temporal smoothness (

Φ_{\mathrm{seq}}

H=5

), and gait periodicity (

Φ_{\mathrm{gait}}

H=10

). This design enables seamless recovery-to-locomotion transitions without explicit mode switching. Furthermore, we propose a Terrain-Pose Risk Coupling Sampler (TPRCS) that explicitly couples dangerous edge initial poses with terrain dynamics during training, forming a closed-loop synergy with the terrain-privileged shaping term

Ξ_κ

. We validate CG-MuTra on a Unitree G1 humanoid across grass, slopes (

10^\circ

15^\circ

), and gravel in both simulation and hardware. Experimental results demonstrate that CG-MuTra achieves smooth, highly robust fall recovery and locomotion transitions across multiple terrains while maintaining a single deployable policy.

Abstract

Humanoid Robot Fall Recovery

Explore similar work

CG-MuTra: Continuously-Gated Multi-Terrain Adaptive Recovery for Unified Humanoid Fall Recovery and Locomotion

Test-Time Motion Steering for Perceptive Humanoid Fall Recovery

StableMimic: Smooth Human-Like Recovery for Humanoid Motion Tracking - Learning Beyond the Tracking Distribution for Structured Post-Fall Behavior