Learning Humanoid Standing-up Control across Diverse Postures

TL;DR

This paper introduces HoST, a reinforcement learning framework enabling humanoid robots to learn and perform stable standing-up motions across diverse postures and environments, with successful real-world deployment.

Contribution

The paper presents a novel RL-based approach with curriculum training and regularization techniques for sim-to-real transfer of standing-up control across postures.

Findings

Controllers achieve smooth, stable standing-up motions

Effective sim-to-real transfer demonstrated on Unitree G1 robot

Robust performance across laboratory and outdoor environments

Abstract

Standing-up control is crucial for humanoid robots, with the potential for integration into current locomotion and loco-manipulation systems, such as fall recovery. Existing approaches are either limited to simulations that overlook hardware constraints or rely on predefined ground-specific motion trajectories, failing to enable standing up across postures in real-world scenes. To bridge this gap, we present HoST (Humanoid Standing-up Control), a reinforcement learning framework that learns standing-up control from scratch, enabling robust sim-to-real transfer across diverse postures. HoST effectively learns posture-adaptive motions by leveraging a multi-critic architecture and curriculum-based training on diverse simulated terrains. To ensure successful real-world deployment, we constrain the motion with smoothness regularization and implicit motion speed bound to alleviate oscillatory and violent motions on physical hardware, respectively. After simulation-based training, the learned control policies are directly deployed on the Unitree G1 humanoid robot. Our experimental results demonstrate that the controllers achieve smooth, stable, and robust standing-up motions across a wide range of laboratory and outdoor environments. Videos and code are available at https://taohuang13.github.io/humanoid-standingup.github.io/.