RecoverFormer: End-to-End Contact-Aware Recovery for Humanoid Robots

• URL: http://arxiv.org/abs/2604.22911v1
• Date: Fri, 24 Apr 2026 17:52:06 GMT
• Title: RecoverFormer: End-to-End Contact-Aware Recovery for Humanoid Robots
• Abstract summary: We present RECOVERFORMER, a fully end-to-end humanoid recovery policy.<n>It learns when and how to switch among recovery behaviors.<n>The architecture combines a causal transformer over a 50-step observation history with two novel heads.
• Score: 5.584060970507507
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Humanoid robots operating in unstructured environments must recover from unexpected disturbances-a capability that remains challenging for end-to-end control policies. We present RECOVERFORMER, a fully end-to-end humanoid recovery policy that learns when and how to switch among recovery behaviors-including compensatory stepping, hand-environment contact, and center-of-mass reshaping-while maintaining robust performance under model mismatch. The architecture combines a causal transformer over a 50-step observation history with two novel heads: a latent recovery mode that enables smooth transitions among distinct recovery strategies, and a contact affordance head that predicts which environmental surfaces (walls, railings, table edges) are beneficial for stabilization. We evaluate RECOVERFORMER on the Unitree G1 humanoid in MuJoCo. Trained only on open floor, RECOVERFORMER transfers zero shot to walled environments, achieving 100% recovery success across 100-300 N pushes and across wall distances from 0.25-1.4m. Under zero-shot dynamics mismatch, RECOVERFORMER reaches 75.5% at plus +25% mass, 89% under 30 ms latency, 91.5% at low friction, and 99% under compound friction, latency and mass perturbation. The learned latent modes specialize across force regimes without mode-level supervision, validated by t-SNE analysis of 300 episodes. Taken together, these results show that a single end-to-end policy can deliver multi-modal, contact aware humanoid recovery that generalizes across perturbation magnitude, contact geometry, and dynamics shift.

Related papers

• GigaBrain-WBC-0.5: A Behavior World Model for Robust Whole-Body Control with Environment Interaction [68.67772166423812]
  We present GigaBrain-WBC-0.5, the first Behavior World Model (BWM) for humanoid whole-body control.<n>Rather than a purely reactive tracker, we train a causal Transformer to jointly predict its next action, next state, and the distribution over its next latent behavior command.<n>An automatic terrain-annotation pipeline recovers full 3D contact geometry from retargeted motion, enabling terrain annotation at the scale of existing motion datasets.
  arXiv  Detail & Related papers  (2026-08-18T18:21:36Z)
• PhyLatent: Learning Dynamics-Relevant Representations for JEPA World Models [12.761667671058767]
  PhyLatent is a dynamics-relevant training objective for JointEmbedding Predictive Architecture (JEPA) world models.<n>We identify three failure modes in JEPA world models: physical invariance collapse, physical identifiability collapse, and counterfactual dynamics collapse.<n>On OGBench-Cube, PhyLatent reduces the three failure rates from 15.60%, 6.71%, and 8.41% to 7.53%, 0.95%, and 4.62%, respectively, and improves model predictive control (MPC) success from 70.0% to 78.1%.
  arXiv  Detail & Related papers  (2026-08-06T08:02:15Z)
• StableMimic: Smooth Human-Like Recovery for Humanoid Motion Tracking - Learning Beyond the Tracking Distribution for Structured Post-Fall Behavior [2.9625318946782397]
  We present StableMimic, a unified tracker trained beyond the nominal tracking distribution.<n>Perturbed resets around multiple human get-up references expose prone, supine, off-balance, and intermediate ground-contact states.<n>A hidden successor-state objective teaches human-shaped recovery without exposing reference identity or phase to the deployed Actor.
  arXiv  Detail & Related papers  (2026-08-03T15:27:53Z)
• EgoRecovery: Acquiring Failure Recovery Ability Through Human Recovery Demonstration [75.13534797264485]
  We show that egocentric human data capturing failure recovery processes provides a scalable alternative to robot teleoperation.<n>By efficiently arranging task-level failure configurations and recording short recovery segments, human operators can generate more than 10x as much valid recovery data per hour.
  arXiv  Detail & Related papers  (2026-07-22T04:44:26Z)
• AutoPersonas: A Multi-Timescale Loop Engine for Open-Ended Persona Evolution [1.607532703630949]
  Long-term persona agents must remain identifiable while adapting to new events, relationships, evidence, and social conditions.<n>We identify self-locking as a runtime failure mode in continuing persona-life loops.<n>We introduce AutoPersonas, a life-environment engine for bounded persona-level self-evolution.
  arXiv  Detail & Related papers  (2026-07-09T08:56:30Z)
• ReactiveBFM: Reactive Closed-Loop Motion Planning Towards Universal Humanoid Whole-Body Control [54.267632784131415]
  ReactiveBFM is a real-time-loop planning-control framework for humanoids.<n>It mitigates fatal cumulative exposure bias via a scheduled prefix sampling curriculum.<n>It demonstrates unprecedented physical agility across a vast repertoire of text-conditioned closed-loop motions.
  arXiv  Detail & Related papers  (2026-06-29T14:27:27Z)
• RepSAM: Bridging Foundation Models to Robotic Vision via Representation-Guided Adaptation [0.0]
  RepSAM is a representation-guided parameter-efficient fine-tuning framework for adapting foundation models to robotic vision.<n>RepSAM achieves 97.9% of full fine-tuning performance (89.0% vs. 90.9% mIoU) while reducing trainable parameters by 158x (from 632M to 4.0M).
  arXiv  Detail & Related papers  (2026-05-25T06:56:42Z)
• SteadyTray: Learning Object Balancing Tasks in Humanoid Tray Transport via Residual Reinforcement Learning [4.827821660801356]
  We introduce ReST-RL, a hierarchical reinforcement learning architecture that explicitly decouples locomotion from payload stabilization.<n>In simulation, the residual design significantly outperforms end-to-end baselines in gait smoothness and orientation accuracy.<n>This modular approach demonstrates highly reliable zero-shot sim-to-real generalization across various objects and external force disturbances.
  arXiv  Detail & Related papers  (2026-03-11T01:05:46Z)
• Less is More: Robust Zero-Communication 3D Pursuit-Evasion via Representational Parsimony [7.6658082440595186]
  Asymmetric 3D pursuit-evasion in cluttered voxel environments is difficult under communication latency, partial observability, and nonholonomic maneuver limits.<n>We build on an inherited path-guided decentralized pursuit scaffold to improve communication-free coordination.<n>We show that explicitly severing redundant cross-agent channels can suppress compounding error cascades and improve robustness in latency-prone deployment.
  arXiv  Detail & Related papers  (2026-03-09T11:44:04Z)
• When World Models Dream Wrong: Physical-Conditioned Adversarial Attacks against World Models [54.08784776767683]
  We present Physical-Conditioned World Model Attack (PhysCond-WMA), the first white-box world model attack that perturbs physical-condition channels.<n>PhysCond-WMA induces semantic, logic, or decision-level distortion while preserving perceptual fidelity.
  arXiv  Detail & Related papers  (2026-02-21T07:22:37Z)
• BadCLIP++: Stealthy and Persistent Backdoors in Multimodal Contrastive Learning [73.46118996284888]
  Research on backdoor attacks against multimodal contrastive learning models faces two key challenges: stealthiness and persistence.<n>We propose BadCLIP++, a unified framework that tackles both challenges.<n>For stealthiness, we introduce a semantic-fusion QR micro-trigger that embeds imperceptible patterns near task-relevant regions.<n>For persistence, we stabilize trigger embeddings via radius shrinkage and centroid alignment.
  arXiv  Detail & Related papers  (2026-02-19T08:31:16Z)
• Learning to Get Up Across Morphologies: Zero-Shot Recovery with a Unified Humanoid Policy [0.0]
  Fall recovery is a critical skill for humanoid robots in dynamic environments such as RoboCup.<n>Recent techniques using deep reinforcement learning have produced robust get-up behaviors.<n>This paper presents a single DRL policy capable of recovering from falls across seven humanoid robots.
  arXiv  Detail & Related papers  (2025-12-13T07:59:52Z)
• Opening the Sim-to-Real Door for Humanoid Pixel-to-Action Policy Transfer [59.02729900344616]
  GPU-accelerated, photorealistic simulation has opened a scalable data-generation path for robot learning.<n>We develop a teacher-student-bootstrap learning framework for vision-based humanoid loco-manipulation.<n>This represents the first humanoid sim-to-real policy capable of diverse articulated loco-manipulation using pure RGB perception.
  arXiv  Detail & Related papers  (2025-11-30T20:07:13Z)
• Squeeze the Soaked Sponge: Efficient Off-policy Reinforcement Finetuning for Large Language Model [56.92219181993453]
  We propose Reincarnating Mix-policy Proximal Policy Gradient (ReMix) to enable on-policyRFT methods like PPO and GRPO to leverage off-policy data.<n>ReMix consists of three major components: (1) Mix-policy proximal policy gradient with an increased Update-To-Data (UTD) ratio for efficient training; (2) KL-Convex policy constraint to balance the trade-off between stability and flexibility; (3) Policy reincarnation to achieve a seamless transition from efficient early-stage learning to steady improvement.
  arXiv  Detail & Related papers  (2025-07-09T14:29:45Z)
• Learning to Recover: Dynamic Reward Shaping with Wheel-Leg Coordination for Fallen Robots [10.712715061996194]
  This paper presents a learning-based framework integrating Episode-based Dynamic Reward Shaping and curriculum learning.<n>We demonstrate that synergistic wheel-leg coordination reduces joint torque consumption by 15.8% and 26.2%.
  arXiv  Detail & Related papers  (2025-06-05T18:58:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.