A trio of robotics engineers from Stanford University, working with colleagues at Google’s Deep Mind, leveraged Google’s ALOHA system to create a mobile robot capable of performing a wide variety of household tasks . They called it Mobile ALOHA.
Zipeng Fu, Tony Zhao and Chelsea Finn published an article describing the robot’s functionality on the arXiv preprint server, as well as several videos of the robot in action on YouTube.
The researchers note that the robot is trained using both a database and a supervised demonstration. The demonstrations are carried out using a device connected to the back of the robot that allows it to imitate the actions necessary to complete a given task, such as hanging a shirt on a hanger. Training sessions involve driving the robot and showing it how to do things, much like a puppet.
The robot, as its name suggests, is mobile. It is also untethered, with a large battery in its base that allows for long working periods and stability. It rolls around a work area to perform common household tasks. Videos of the robot in action show that it can be taught a wide range of tasks, from preparing a three-course meal to loading a dishwasher to performing all the tasks. laundry tasks.
The ALOHA mobile can also open and close the refrigerator, kitchen cabinets and doors. He also dusts, cleans and puts things away. In short, the robot can learn to perform almost any household task, including making coffee, feeding pets, replacing the lid on the toothpaste tube, and dispensing vitamins.
The research team points out that their system is also very inexpensive (around $32,000) compared to other robots created to help with household chores. Many of them, they note, are tabletop assistants, not mobile robots. The ALOHA mobile also demonstrates a high degree of freedom as well as extreme dexterity and finesse, as evidenced by its ability to crack eggs and open soda bottles.
The researchers acknowledge that they need to continue working on their robot’s design to both improve its accuracy in performing tasks and help it move more easily around a home.
More information:
Zipeng Fu et al, Mobile ALOHA: learning bimanual mobile manipulation with low-cost whole-body teleoperation, arXiv (2024). DOI: 10.48550/arxiv.2401.02117
Project website: mobile-aloha.github.io/
arXiv
© 2024 Science X Network
Quote: ALOHA mobile robot capable of helping with several household tasks (January 8, 2024) retrieved on January 8, 2024 from
This document is subject to copyright. Apart from fair use for private study or research purposes, no part may be reproduced without written permission. The content is provided for information only.