TWIST2: Scalable, Portable, and Holistic Humanoid Data Collection System

TL;DR

TWIST2 is a scalable, portable humanoid data collection system that uses VR and a low-cost robot neck to enable efficient, holistic control and autonomous visuomotor policies for complex humanoid tasks.

Contribution

Introduces TWIST2, a novel mocap-free, portable humanoid data collection system with full-body control and a hierarchical visuomotor policy framework.

Findings

Collected 100 demonstrations in 15 minutes with high success rate.

Demonstrated long-horizon dexterous and mobile humanoid skills.

Successfully implemented autonomous whole-body visuomotor control.

Abstract

Large-scale data has driven breakthroughs in robotics, from language models to vision-language-action models in bimanual manipulation. However, humanoid robotics lacks equally effective data collection frameworks. Existing humanoid teleoperation systems either use decoupled control or depend on expensive motion capture setups. We introduce TWIST2, a portable, mocap-free humanoid teleoperation and data collection system that preserves full whole-body control while advancing scalability. Our system leverages PICO4U VR for obtaining real-time whole-body human motions, with a custom 2-DoF robot neck (cost around $250) for egocentric vision, enabling holistic human-to-humanoid control. We demonstrate long-horizon dexterous and mobile humanoid skills and we can collect 100 demonstrations in 15 minutes with an almost 100% success rate. Building on this pipeline, we propose a hierarchical visuomotor policy framework that autonomously controls the full humanoid body based on egocentric vision. Our visuomotor policy successfully demonstrates whole-body dexterous manipulation and dynamic kicking tasks. The entire system is fully reproducible and open-sourced at https://yanjieze.com/TWIST2 . Our collected dataset is also open-sourced at https://twist-data.github.io .