Robots are lending a helping hand this holiday season

Last year, the holiday shopping season was kind of a disaster. A lot more people were shopping online because of the pandemic, and millions of packages were delayed or delivered late.

This year could be worse. There are shortages of materials, products, shipping containers and workers. And Deloitte estimates that online sales will jump by 11% to 15% this season.

Retailers are encouraging people to shop as early as possible. Also, they and shipping companies are trying out robots that can help them sort through items.

Ken Goldberg is a professor of industrial engineering at the University of California, Berkeley, and a co-founder of Ambi Robotics, which makes a robot called Ambisort. He said the robots are getting better at sorting. The following is an edited transcript of our conversation.

Ken Goldberg: We’ve come a long way in a year. We’ve gotten a lot better at figuring out where to grasp objects, how to scan them with faster scanners, all the arm motions have been sped up. The system is now ready to do the job. And I think what’s going on is that many vendors around the country, they’re all gearing up and trying out new things. This is going to be a big test year for new technologies. Then, those that work, they’re going to invest very heavily in starting in the spring, so that this will be able to solve these problems continuously in the future.

Marielle Segarra: Can you, can you just sort of describe what it looks like for me?

Goldberg: Just imagine one big bin on one side, and then a blue robot that has a gripper on the end, in fact, it [has] suction cups, and then cameras are all around it. On the other side is another robot arm that sort of moves along and drops the packages into this big array of 25 other bins. So it takes up a fair amount of space, the size of a small room, and the packages come in on one end, and they come out on the other end in the smaller bins assigned to each ZIP code.

Segarra: And how does that speed things up?

Goldberg: Well, it can be very slow to have humans do this, partly because the bins are actually quite deep, so you have to reach way in, which leads to a lot of back injuries. And then a human has to look at each one, figure out where the bar code is, right? That scanning process takes time, and then walk it over to the appropriate bin. So this system speeds up all three aspects. It never gets tired, it can reach deeply into the bin because of the robot. It’s scanned with lots of cameras simultaneously, so it can instantly scan, find the bar code, and then drop it in this scan tray that basically moves it. While the robot goes down to pick up the next object, the first object is being dropped into the appropriate bin.

Segarra: Is this intended to replace a person?

Goldberg: Well, it’s intended to supplement a person. In fact, you know, I think the analogy is, you’re a journalist, right? And there’s probably certain jobs that you don’t particularly like doing, and I would imagine something like transcribing an interview like this. But nowadays, there are tools that can help you do that, that use [artificial intelligence]. They don’t replace the transcription, you would still have to go in and fix it in places, but it can do a lot of the drudgery for you. It just means that you can concentrate on the parts of your job that you do best. And that’s exactly the analogy that we believe is going to be the case for workers and warehouses.