The July 2022 issue of IEEE Spectrum is here!

Close bar

Flying Inventory Assistants Are a Good Use for Drones

It's not particularly exciting, but inventory drones have a real chance at doing something near-term useful

3 min read
Flying Inventory Assistants Are a Good Use for Drones
Image: Fraunhofer IML

It’s starting to seem like “throw a drone at it” is the solution that everyone wants to somehow solve every single problem everywhere, ever. And in most cases, it’s not going to work anytime soon, for reasons that we continue to belabor. This is not to say that drones aren’t valuable tools that can solve many problems: the key is to find a problem that needs a drone, as opposed to having a drone and then desperately looking for some problem for it to solve.

The Fraunhofer Institute for Material Flow and Logistics, in Dortmund, Germany, may have found one of these problems: taking inventory in a warehouse. To do this efficiently, you need a mobile antenna that can navigate in three dimensions, and autonomous flying robots certainly fit the bill.

Inventory is awful. I say this from experience, having made the mistake of accepting a department store inventory job for a few weeks in high school. Taking inventory in a store or warehouse involves wandering around and recording the location of every single item, using an RFID antenna or optical scanner. Did I mention that doing this by hand is awful? Because it’s awful.

One option to make inventory less painful is to deploy an infrastructure of (say) shelves with built-in RFID readers, such that the shelf can tell what’s being stored on it. (Another, even better option is doing what Kiva Systems, now owned by Amazon, does: its inventory system keeps track of both the location and contents of every bin in the warehouse, so when you need to retrieve or restock something, you just send robots do get those bins for you.) This can work very well, but it’s expensive and hard to scale. Fraunhofer’s idea is to forget about the fancy shelves and instead replace what is usually a small army of inventorying humans with a small fleet of autonomous, inventorying drones that use RFID antennas or cameras to identify the location of items.

Drones are a good idea for inventory management for several reasons. First, they’ll be operating in a semi-structured (or entirely structured) environment. If they’re in a retail store, the environment is probably considered semi-structured, since humans can be kept out of the area while the robots do their work and the environment is generally static and well-defined. A warehouse might be a structured environment, since it can be completely restricted and mapped in advance with very little risk of change.

Also, an inventory drone can have an immediate and significant benefit on the inventory task in a way that would be hard to do otherwise. The reason a drone is so potentially useful is that warehouses maximize space utilization by stacking inventory as high as possible. A human would need a ladder just to read any identifying information, but for a drone, the height above the ground doesn’t matter all that much, especially if it’s equipped with a long-range RFID reader.

Fraunhofer’s InventAIRy Project (nice, right?) is developing “autonomous flying robots that are capable of independently navigating and conducting inventory.” The drones won’t be relying on an external navigations systems: it’ll all be onboard, using ultrasound sensors, 3D cameras, and laser scanners to perform continuous  simultaneous localization and mapping (SLAM). By mid-2015, Fraunhofer’s prototype system should be operating with partial autonomy, navigation around shelving and avoiding other obstacles. The next step will be to add RFID antennas, database integration, and (most challenging) an effective path-planning algorithm that allows the robot to reliably and efficiently catalog the objects in an arbitrary space.

We’re more optimistic about the realistic near-term useful potential of InventAIRy than we are about most of the drone-related ideas that we come across, but as with anything related to robots, there’s a huge step between good idea and good execution. We’ll keep you updated.

Via [ Fraunhofer ]

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

“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