Most certainly because they've shown that imitation learning worked inspire of randomness inherent in a natural settings (colours, visuals, obstacles etc).
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.
4 comments
[ 0.22 ms ] story [ 21.2 ms ] threadSeems like a great starting point for open source human scale robotics.
Most certainly because they've shown that imitation learning worked inspire of randomness inherent in a natural settings (colours, visuals, obstacles etc).
for ~$32k, you can have a teachable manipulator that can go around a 1-story house and do virtually all household tasks.
seems like a game changer. many of the costs can trivially be brought down, as well.