Lemmy Today

1,493 readers
75 users here now

Welcome to lemmy.today!

About us

๐Ÿค— Thanks for joining our little instance here, located in Oregon. The idea is to have a fast, stable instance and allow users to subscribe to whatever content they want from here.

๐Ÿ˜Ž We dont block any other instances. We will keep it that way unless it becomes a moderation problem.

๐Ÿค  We will be around for a very long time, so you dont have to worry about us shutting down the instance anytime soon. We like performance and stability in our servers, and will upgrade the instance when its needed.

๐Ÿฅน Make sure to join a lot of remote communities to get a good feed going. How to do that is explained here.

Lemmy mobile apps

You should start using one of these ASAP since the web browser user interface is quite ugly, even with themes.

Optional Lemmy web browser user interfaces

Rules

Contact the admin

founded 1 year ago
ADMINS
1
 
 

Imitation learning from human demonstrations has shown impressive performance in robotics. However, most results focus on table-top manipulation, lacking the mobility and dexterity necessary for generally useful tasks.

In this work, we develop a system for imitating mobile manipulation tasks that are bimanual and require whole-body control. We first present Mobile ALOHA, a low-cost and whole-body teleoperation system for data collection. It augments the ALOHA system with a mobile base, and a whole-body teleoperation interface.

Using data collected with Mobile ALOHA, we then perform supervised behavior cloning and find that co-training with existing static ALOHA datasets boosts performance on mobile manipulation tasks.

With 50 demonstrations for each task, co-training can increase success rates by up to 90%, allowing Mobile ALOHA to autonomously complete complex mobile manipulation tasks such as sauteing and serving a piece of shrimp, opening a two-door wall cabinet to store heavy cooking pots, calling and entering an elevator, and lightly rinsing a used pan using a kitchen faucet.

Paper [GitHub, PDF]: Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation

Paper [arXiv]: Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperation | PDF

Code:

2
3
 
 

There is a discussion on Hacker News, but feel free to comment here as well.

view more: next โ€บ