Explicit Stair Geometry Conditioning for Robust Humanoid Locomotion

TL;DR

This paper introduces an explicit geometric parameter-based conditioning framework for humanoid stair climbing, improving generalization and robustness in real-world environments.

Contribution

It proposes a novel explicit stair geometry conditioning method that enhances the adaptability of locomotion policies to varying stair geometries.

Findings

Improved generalization to unseen stair heights in simulation.

Successful outdoor stair climbing with 33 consecutive steps without failure.

Robust indoor and outdoor traversal validated on the Unitree G1 robot.

Abstract

Robust humanoid stair climbing remains challenging due to geometric discontinuities, sensitivity to step height variations, and perception uncertainty in real-world environments. Existing learning-based locomotion policies often rely on implicit terrain representations or blind proprioceptive feedback, limiting their ability to generalize across varying stair geometries and to anticipate required gait adjustments. This paper proposes an explicit stair geometry conditioning framework for robust humanoid stair climbing. Instead of encoding terrain as high-dimensional latent features, we extract a compact set of interpretable geometric parameters, including step height, step depth, and current yaw angle relative to the robot heading. These explicit stair parameters directly condition a Proximal Policy Optimization (PPO)-based locomotion policy, enabling proactive modulation of swing-foot clearance and stride characteristics according to stair structure. Simulation experiments demonstrate improved generalization across unseen stair heights beyond the training distribution. Real-world experiments on the Unitree G1 humanoid validate reliable indoor and outdoor stair traversal. In challenging outdoor scenarios, the robot successfully ascends 33 consecutive steps without failure, demonstrating robustness and practical deployability.