One of the biggest obstacles still facing robotics is adaptability. A robot can be programmed to perform one repetitive task on an assembly line, but is easily confused when confronted with irregularity. This makes it especially difficult when it comes to using robots in agriculture, specifically picking and sorting fruit, since nature creates objects organically and not all fruits are created equal.
But now, product design firm Cambridge Consultants has come up with a robot that can identify different types of fruit and sort them accordingly. I called up Chris Roberts, the head of industrial robotics at the UK-based company, to talk more about this impressive development and ask him how exactly the robot works.
"Here, the robot needs to be able to deal with natural variation," Roberts told me. "The way it does that is we've got a vision system… the robot takes a photo of the scene so we've got a picture of the fruit in a bowl, [and] it's got another sensor which it uses to turn that picture into a sort of 3d map."
I then asked him how it differentiated between the types of fruit. "It's got some processing that looks at shape—[as with] the bananas—and some processing that looks at color," Roberts explained. "And if the fruit rolls around in between actions, it can cope with that, and if the apple's not quite the same shape as the last apple, it can cope with that."
And as for how this robot would affect the job market? "I would imagine this is the sort of thing where it can replace some of the really low-skill jobs…but there are still plenty of jobs that humans are just much better at."
Well, at least for now.