Papers
2
Total Citations
121
H-Index
2
About
A. O'Neill is a leading researcher in robotics and embodied AI, whose work focuses on scaling robot learning through diverse, multi-embodiment datasets and modular system design. O'Neill's most impactful contribution is as a key collaborator on the Open X-Embodiment project, which produced the RT-X models and a massive, cross-robot dataset. This seminal 2024 paper, already garnering 119 citations, demonstrates that training large, high-capacity models on heterogeneous robotic data can yield generalist policies that transfer effectively across different hardware platforms—a paradigm shift akin to the rise of foundation models in NLP and computer vision. This work is rapidly becoming a cornerstone for the field, promising to consolidate pretrained robotic backbones. Complementing this large-scale approach, O'Neill also developed MANIP, a modular architecture that systematically integrates learned subpolicies with robust procedural primitives like Inverse Kinematics and Kalman Filters. By bridging data-driven learning with classical control, MANIP offers a principled framework for building reliable, interactive manipulation systems. Through these contributions, O'Neill is helping to define the next generation of general-purpose robotic intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2