SCDP: Learning Humanoid Locomotion from Partial Observations via Mixed-Observation Distillation

TL;DR

This paper introduces SCDP, a diffusion-based policy that enables humanoid robots to learn locomotion solely from onboard sensor data, eliminating the need for complex state estimation and achieving high success in simulation and real-world tests.

Contribution

SCDP is the first approach to learn humanoid locomotion from partial observations using mixed-observation diffusion training, removing reliance on privileged state information.

Findings

Achieves 99-100% success in velocity control in simulation.

Attains 93% success in motion tracking in simulation.

Demonstrates robust real-world humanoid locomotion without external sensors.

Abstract

Distilling humanoid locomotion control from offline datasets into deployable policies remains a challenge, as existing methods rely on privileged full-body states that require complex and often unreliable state estimation. We present Sensor-Conditioned Diffusion Policies (SCDP) that enables humanoid locomotion using only onboard sensors, eliminating the need for explicit state estimation. SCDP decouples sensing from supervision through mixed-observation training: diffusion model conditions on sensor histories while being supervised to predict privileged future state-action trajectories, enforcing the model to infer the motion dynamics under partial observability. We further develop restricted denoising, context distribution alignment, and context-aware attention masking to encourage implicit state estimation within the model and to prevent train-deploy mismatch. We validate SCDP on velocity-commanded locomotion and motion reference tracking tasks. In simulation, SCDP achieves near-perfect success on velocity control (99-100%) and 93% tracking success in AMASS test set, performing comparable to privileged baselines while using only onboard sensors. Finally, we deploy the trained policy on a real G1 humanoid at 50 Hz, demonstrating robust real robot locomotion without external sensing or state estimation.