Well, this is pretty awesome:
You’re actually watching the extent of iCub’s learning process: it took the robot all of 8 trials to figure out how to hit the center of the bullseye. iCub is using a learning algorithm called ARCHER (Augmented Reward Chained Regression), which is optimized for tasks that have an easily definable goal and measurable progression towards that goal. Basically, hitting the center of the target equates to a maximum reward, and the algorithm builds off of past experience to estimate how to alter iCub’s hand positions to improve the aim of the arrow. In this case, the distance between iCub and the target is only 3.5 meters, but there’s no reason it couldn’t be scaled up to larger distances. Or bigger arrows. Or rocket launchers.
This robot experiment was conducted by Dr. Petar Kormushev, Dr. Sylvain Calinon, and Dr. Ryo Saegusa at the Italian Institute of Technology (the same guys who brought you robot pancake flipping). You can read a bit more about it at the link below.
Thanks Dr. Kormushev!