Iterative Closed-Loop Motion Synthesis for Scaling the Capabilities of Humanoid Control

TL;DR

This paper introduces an iterative closed-loop framework for humanoid motion control that generates diverse, high-quality motion data and progressively enhances policy difficulty, significantly improving performance with less data.

Contribution

It presents a novel automated data generation and difficulty iteration framework that surpasses traditional fixed datasets, enabling scalable, rich motion synthesis for humanoid control.

Findings

Reduced failure rate by 45% using only 1/10 of the dataset

Generated diverse motion data including martial arts, dance, and sports

Demonstrated advantages through ablation and comparative experiments

Abstract

Physics-based humanoid control relies on training with motion datasets that have diverse data distributions. However, the fixed difficulty distribution of datasets limits the performance ceiling of the trained control policies. Additionally, the method of acquiring high-quality data through professional motion capture systems is constrained by costs, making it difficult to achieve large-scale scalability. To address these issues, we propose a closed-loop automated motion data generation and iterative framework. It can generate high-quality motion data with rich action semantics, including martial arts, dance, combat, sports, gymnastics, and more. Furthermore, our framework enables difficulty iteration of policies and data through physical metrics and objective evaluations, allowing the trained tracker to break through its original difficulty limits. On the PHC single-primitive tracker, using only approximately 1/10 of the AMASS dataset size, the average failure rate on the test set (2201 clips) is reduced by 45% compared to the baseline. Finally, we conduct comprehensive ablation and comparative experiments to highlight the rationality and advantages of our framework.