The July 2022 issue of IEEE Spectrum is here!

Close bar

Walmart and Five Elements Robotics Working on Robotic Shopping Cart

We like the idea of an autonomous shopping cart, but getting it to work will be a challenge

2 min read
Five Elements Robotics robotic shopping cart
Photo: Five Elements Robotics

It’s been a few years since we first met Five Elements Robotics at RoboBusiness, where they introduced Budgee, a sort of robotic stuff-carrier that will follow you around with up to 22 kilograms of your junk by homing in on a small ultrasonic emitter. Last week at the Bloomberg Technology Conference, Five Elements CEO Wendy Roberts announced that Walmart is evaluating a prototype of a new Five Elements robotic shopping cart called Dash. Dash is much more than an upgraded version of Budgee; it’s a completely new platform, specifically designed for autonomous shopping assistance.

There are a lot of things I like about this idea. There’s a clear value proposition to a robot that can carry groceries, guide shoppers directly to the items they want, and then handle paying for those items, since people hate doing all of those things. Also, unlike Budgee, you don’t have to worry about the cost of the robot, since the store absorbs all of that, spreading it out over many customers per day. 

However, there are some pretty serious safety and navigation challenges that Dash is going to have to handle, and so far, it’s hard to tell if the Dash prototype is going to be able to do what it has to do with what looks to be a single medium-range LIDAR system, a 3D camera, and some ultrasonic sensors. Grocery stores are often chaotic, with lots of people hurriedly moving all over the place while pushing carts or carrying stuff. Aisles are frequently blocked. There are things on the floor that have to be avoided, and often those things are (say) an employee with cases of items, restocking shelves.

I don’t have to belabour this point: you’ve been to a grocery store before, so you know exactly what kind of environment it can be, and getting a robot to autonomously, safely, and reliably navigate through all that at a useful pace is going to be a very difficult software problem. And the “following the customer to their car” feature, where you’d be asking the robot to navigate outdoors through a parking lot which consists of a big open space full of a bunch of nearly identical objects that effectively appear and disappear at random? I’ll believe it when I see it.

Again, I think that this could be a really great idea, but based on what we’ve heard from companies like Fetch and Savioke (who are deploying autonomous robots in significantly more structured and predictable environments than grocery stores), we want to emphasize how difficult, and likely unpredictably difficult, getting a robot like this to work is going to be. And just getting it to work isn’t enough: it has to work so well and reliably that inexperienced customers won’t get frustrated with it, that it won’t take constant minding from store employees, and that it’ll maintain enough uptime to justify however much it costs.

Dash is scheduled to enter production in early 2017.

[ Five Elements Robotics ] via [ Bloomberg ]

The Conversation (0)

How the U.S. Army Is Turning Robots Into Team Players

Engineers battle the limits of deep learning for battlefield bots

11 min read
Robot with threads near a fallen branch

RoMan, the Army Research Laboratory's robotic manipulator, considers the best way to grasp and move a tree branch at the Adelphi Laboratory Center, in Maryland.

Evan Ackerman
LightGreen

“I should probably not be standing this close," I think to myself, as the robot slowly approaches a large tree branch on the floor in front of me. It's not the size of the branch that makes me nervous—it's that the robot is operating autonomously, and that while I know what it's supposed to do, I'm not entirely sure what it will do. If everything works the way the roboticists at the U.S. Army Research Laboratory (ARL) in Adelphi, Md., expect, the robot will identify the branch, grasp it, and drag it out of the way. These folks know what they're doing, but I've spent enough time around robots that I take a small step backwards anyway.

This article is part of our special report on AI, “The Great AI Reckoning.”

The robot, named RoMan, for Robotic Manipulator, is about the size of a large lawn mower, with a tracked base that helps it handle most kinds of terrain. At the front, it has a squat torso equipped with cameras and depth sensors, as well as a pair of arms that were harvested from a prototype disaster-response robot originally developed at NASA's Jet Propulsion Laboratory for a DARPA robotics competition. RoMan's job today is roadway clearing, a multistep task that ARL wants the robot to complete as autonomously as possible. Instead of instructing the robot to grasp specific objects in specific ways and move them to specific places, the operators tell RoMan to "go clear a path." It's then up to the robot to make all the decisions necessary to achieve that objective.

Keep Reading ↓Show less