Learning Humanoid Standing-up Control across Diverse Postures

TL;DR

This work tackles the challenge of humanoid standing-up control across diverse postures in real environments. It presents HoST, an RL framework with a multi-critic architecture, force-based exploration, and motion constraints, trained in diverse terrains and transferred directly to the Unitree G1. The approach yields smooth, stable, and robust standing-up motions under disturbances and payloads, demonstrating strong sim-to-real transfer and outdoor generalization. Limitations include perception integration and potential interference when combining prone and supine postures, with future work aimed at broader perception and system integration. Overall, HoST advances posture-adaptive, real-world deployable standing-up control for humanoid robots and enables fall-recovery and loco-manipulation capabilities.

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/.

Figures (13)

• Figure 1: Framework overview. (a) We train policies in simulation from scratch with multiple critics and motion constraints operationalized by rewards, smoothness regularization, and action bound (rescaler). (b) The trained polices can be directly deployed in the real robot to produce standing-up motions.
• Figure 2: Simulation terrains and real-world scenes. We design four terrains in simulation: ground, platform, wall, and slope to create initial robot postures that are likely to be met in real-world environments. Examples of these real-world environments are shown at the bottom of the figure.
• Figure 3: Motion analysis in simulation. (Left) UMAP visualization of joint-space trajectories demonstrates similar but distinct motion patterns on the terrains except for the wall. Besides, the trajectories of each terrain are overall consistent, with variation to handle the difference of starting postures. (Right) 3D trajectory visualizations reveal stable, coordinated hand-foot motion and dynamic posture adaptability, demonstrating effective whole-body coordination and validating the proposed framework. Point color in the plot indicates motion progression, with lighter shades for earlier and darker for later times.
• Figure 4: Robustness analysis in simulation. Evaluation of control policies under four environmental disturbances demonstrates the robustness of our controllers. The poor performance of HoST-History1 indicates the importance of historical information for robustness, while HoST-Bound0.25's high energy consumption reveals limitations in motion quality under disturbance, demonstrating the effect of curriculum setup of action bound.
• Figure 5: Trade-off analysis in simulation. Trade-offs between motion speed, smoothness, and energy across terrains. Results show the inverse speed-smoothness relationship, indicating the importance of constrained motion speed achieved by our method for real-world deployment.