As we continue on our thought study on what might be a successful approach to machine harvest of berries:
Great little note from The Economist on the limits of AI in the physical world. The brief article describes a test in Singapore using robots (actually robotic arms of the sort one sees in car factories) to put together a flat pack chair from IKEA. It took a while, and could only be accomplished with precise instructions like "pick up dowel" and "insert dowel into top-left hole". It just underlines that the physical dexterity we take for granted and execute from moment to moment unconsciously is computationally light years away from a machine that can play checkers or Go.
I am finding that this is far from being an untrammeled area of thought and study, and back in the 1980's Hans Moravec, Robert Brooks and Marvin Minsky came up with what ultimately came to be known as Moravec's paradox. It is the statement that higher level reasoning and pattern recognition of the sort that we currently are able to assign to robots doesn't take nearly the computational power than do lower level sensorimotor skills like moving around a cluttered space and picking up stuff.
Let's hear from Moravec himself what he means:
"Encoded in the large, highly evolved sensory and motor portions of the human brain is a billion years of experience about the nature of the world and how to survive in it. The deliberate process we call reasoning is, I believe, the thinnest veneer of human thought, effective only because it is supported by this much older and much more powerful, though usually unconscious, sensorimotor knowledge. We are all prodigious olympians in perceptual and motor areas, so good that we make the difficult look easy. Abstract thought, though, is a new trick, perhaps less than 100 thousand years old. We have not yet mastered it. It is not all that intrinsically difficult; it just seems so when we do it."
A bit of a sidebar, but the idea that the human reason we hold so dear as setting us apart from the rest of the Animal Kingdom is a "new trick" that we have "not yet mastered" is fascinating.
Fast conclusion here is that it's going to probably going to be a while before anyone builds a machine capable to being able to manipulate something like a soft fruit. Best to emphasize the strength of the machine, and not work to endow it with abilities which are so very difficult for it, but so very easy for us.