Learning to Get Up Across Morphologies: Zero-Shot Recovery with a Unified Humanoid Policy

TL;DR

This paper introduces a single deep reinforcement learning policy that enables humanoid robots of various sizes and shapes to recover from falls without additional training, demonstrating high transferability and generalization across different morphologies.

Contribution

A unified DRL-based fall recovery policy for multiple humanoid robots, enabling zero-shot transfer and outperforming specialized policies in some cases.

Findings

Achieves up to 86% zero-shot recovery success on unseen morphologies.

Unified policy generalizes across diverse robot sizes and dynamics.

Shared policy can outperform robot-specific baselines.

Abstract

Fall recovery is a critical skill for humanoid robots in dynamic environments such as RoboCup, where prolonged downtime often decides the match. Recent techniques using deep reinforcement learning (DRL) have produced robust get-up behaviors, yet existing methods require training of separate policies for each robot morphology. This paper presents a single DRL policy capable of recovering from falls across seven humanoid robots with diverse heights (0.48 - 0.81 m), weights (2.8 - 7.9 kg), and dynamics. Trained with CrossQ, the unified policy transfers zero-shot up to 86 +/- 7% (95% CI [81, 89]) on unseen morphologies, eliminating the need for robot-specific training. Comprehensive leave-one-out experiments, morph scaling analysis, and diversity ablations show that targeted morphological coverage improves zero-shot generalization. In some cases, the shared policy even surpasses the specialist baselines. These findings illustrate the practicality of morphology-agnostic control for fall recovery, laying the foundation for generalist humanoid control. The software is open-source and available at: https://github.com/utra-robosoccer/unified-humanoid-getup