VIGOR: Visual Goal-In-Context Inference for Unified Humanoid Fall Safety

TL;DR

This paper introduces VIGOR, a unified approach for humanoid fall safety that leverages a teacher-student training paradigm to enable robust, zero-shot fall recovery across diverse terrains using only egocentric vision and proprioception.

Contribution

It proposes a novel unified framework that integrates perception and action for fall safety, trained via distillation from a human demonstration-based teacher model.

Findings

Robust fall recovery demonstrated in simulation and real-world tests.

Zero-shot generalization to complex, non-flat terrains.

Effective use of a teacher-student training paradigm with goal-in-context representations.

Abstract

Reliable fall recovery is critical for humanoids operating in cluttered environments. Unlike quadrupeds or wheeled robots, humanoids experience high-energy impacts, complex whole-body contact, and large viewpoint changes during a fall, making recovery essential for continued operation. Existing methods fragment fall safety into separate problems such as fall avoidance, impact mitigation, and stand-up recovery, or rely on end-to-end policies trained without vision through reinforcement learning or imitation learning, often on flat terrain. At a deeper level, fall safety is treated as monolithic data complexity, coupling pose, dynamics, and terrain and requiring exhaustive coverage, limiting scalability and generalization. We present a unified fall safety approach that spans all phases of fall recovery. It builds on two insights: 1) Natural human fall and recovery poses are highly constrained and transferable from flat to complex terrain through alignment, and 2) Fast whole-body reactions require integrated perceptual-motor representations. We train a privileged teacher using sparse human demonstrations on flat terrain and simulated complex terrains, and distill it into a deployable student that relies only on egocentric depth and proprioception. The student learns how to react by matching the teacher's goal-in-context latent representation, which combines the next target pose with the local terrain, rather than separately encoding what it must perceive and how it must act. Results in simulation and on a real Unitree G1 humanoid demonstrate robust, zero-shot fall safety across diverse non-flat environments without real-world fine-tuning. The project page is available at https://vigor2026.github.io/