Yes, there appears to already be work in this area. Here is a recent example I ran across on twitter showing videos of 2 relatively low cost robot arms learning various very fine manipulation tasks apparently after just 15 minutes or so of demonstrations:
Introducing ACT: Action Chunking with Transformers
Yes, there appears to already be work in this area. Here is a recent example I ran across on twitter showing videos of 2 relatively low cost robot arms learning various very fine manipulation tasks apparently after just 15 minutes or so of demonstrations:
Introducing ACT: Action Chunking with Transformers
https://twitter.com/tonyzzhao/status/1640395685597159425
related website:
Learning Fine-Grained Bimanual Manipulation
with Low-Cost Hardware
https://tonyzhaozh.github.io/aloha/