From the number of different companiestrying their handat developingwarehouse picking robots, it’s fairly safe to assume that they’re going to be a Big Thing in Not Too Long. This latest design comes from Hitachi, and it’s notable because it’s from a big company, it’s got two arms, and it seems to actually work.
By “actually work,” what we mean is that this robot seems to be able to use vision to briskly grasp at least a couple different kinds of objects. So far, this is one of the fastest automated picking operations we’ve ever seen, especially since the robot is often grasping stuff in a gripper and not using suction for everything. Watch:
Those are Epson ProSix C4-A901S industrial robot arms mounted on scissor lifts, which are mounted on a mobile base. Each arm can lift about 1 kilogram. Somehow, “the robot can recognize where goods are and begin moving its lifting platforms and arms to grasp them even before it has arrived at the shelf location,” according to a Hitachi spokeswoman who spoke with IDG. “Because of this coordination, it takes about three seconds for the arm to pick up an item once it is in front of a shelf.”
Why two arms instead of one? There are a few things that a two-armed robot can do that a one-armed robot can’t, such as:
- Picking two objects at once
- Picking an object that’s too heavy or bulky for one arm
- Picking an object that necessitates manipulating another object
- Giving double high fives
In most warehouse environments, though, it’s unclear how valuable these capabilities would be. Obviously, double high fives are super important, we get that. But most picking tasks don’t really need two arms. For consumer goods, objects tend to be small and lightweight, and it’s rare that two different picks are situated such that the robot could reach both of them at once. Using one arm to pull out a box on a shelf to pick a bottle out of is a neat trick, but not having the box in the first place means you could just grab the bottle directly with one arm instead.
For some special applications, a two armed picking robot could certainly be valuable. Say, parts handling tasks, where you need to grasp, manipulate, and transport a variety of objects that are heavy or weirdly sized. The question is whether there are enough applications like these where replacing a human with a robot makes sense, and that’s going to depend on how much this robot costs. At this point, it looks like it’s probably super expensive, and Hitachi doesn’t have plans to sell it much sooner than five years from now.
Evan Ackerman is a senior editor at IEEE Spectrum. Since 2007, he has written over 6,000 articles on robotics and technology. He has a degree in Martian geology and is excellent at playing bagpipes.