JoyAI-RA 0.1: A Foundation Model for Robotic Autonomy

TL;DR

JoyAI-RA is a foundation model designed to improve robotic manipulation and generalization across different robot embodiments using diverse multi-source data and a unified training framework.

Contribution

It introduces a multi-source pretraining framework that bridges embodiment gaps, enhancing cross-embodiment generalization in robotic manipulation tasks.

Findings

Outperforms state-of-the-art methods in simulation and real-world benchmarks.

Effectively bridges embodiment gaps between human manipulation and robotic control.

Enhances cross-embodiment behavior learning through heterogeneous data integration.

Abstract

Robotic autonomy in open-world environments is fundamentally limited by insufficient data diversity and poor cross-embodiment generalization. Existing robotic datasets are often limited in scale and task coverage, while relatively large differences across robot embodiments impede effective behavior knowledge transfer. To address these challenges, we propose JoyAI-RA, a vision-language-action (VLA) embodied foundation model tailored for generalizable robotic manipulation. JoyAI-RA presents a multi-source multi-level pretraining framework that integrates web data, large-scale egocentric human manipulation videos, simulation-generated trajectories, and real-robot data. Through training on heterogeneous multi-source data with explicit action-space unification, JoyAI-RA effectively bridges embodiment gaps, particularly between human manipulation and robotic control, thereby enhancing cross-embodiment behavior learning. JoyAI-RA outperforms state-of-the-art methods in both simulation and real-world benchmarks, especially on diverse tasks with generalization demands.