Aloha Research

ALOHA 2 builds on the original ALOHA system and addresses many of its limitations through redesigned hardware and upgraded software.

Key intentions behind the redesign:

• Extend and improve large-scale data collection

• Provide a robust and flexible platform for complex manipulation tasks

• Improve ergonomics and long-term usability for operators

The hardware design is published as open-source, allowing research groups worldwide to:

• Rebuild the platform

• Modify components

• Contribute improvements back to the community

This modular, customizable architecture supports a wide range of research in robotic learning, teleoperation and human–robot collaboration.

Key Features

• Enhanced grippers - Mounted on low-friction rails for smooth and accurate motion - Designed for both delicate handling and higher-force operations - Optimized for low latency and high responsiveness during teleoperation

• Advanced gravity compensation - Passive mechanical gravity compensation replaces earlier rubber-band solutions - Reduces physical effort and fatigue for the operator - Improves stability and comfort during long teleop sessions

• Reinforced frame and high-resolution cameras - Aluminum extrusion frame: lightweight, stiff and expandable - Supports larger props and human co-workers in the workspace - Intel RealSense D405 cameras provide depth sensing, wide field of view and accurate 3D perception for manipulation tasks

The open-source philosophy behind ALOHA 2 encourages shared development of both hardware and software, helping the community push forward teleoperation and manipulation research. For detailed technical information, see the official project website.