On supporting science journalism
If you're enjoying this article, consider supporting our award-winning journalism by subscribing. By purchasing a subscription you are helping to ensure the future of impactful stories about the discoveries and ideas shaping our world today.
Anyone who has ever loaded a moving van knows how difficult it is to safely stack boxes of different sizes, weights and levels of fragility, all while minimizing the amount of space those cartons take up. Imagine an endless stream of such boxes and a business that lives and dies by the efficiency of their stacking, and you have an idea of the challenge facing warehouses that process large volumes of cargo onto pallets for shipping.
Automation can help, but the complexity of the task makes it difficult to compare the efficiencies of various robotic palletizing approaches. So a consortium of concerned parties is throwing the automated pallet-building challenge to university teams, who can explore the problem in a safe, low-cost virtual world. The Virtual Manufacturing Automation Competition, co-sponsored by the National Institute for Standards and Technology (NIST), the Georgia Institute of Technology and the electrical engineering association IEEE, aims to create standards for robotic stacking and packing approaches by which automated palletizers can be assessed.
Competing teams will design automated protocols for use on a simulated, computer-generated shop floor, which uses a video game engine as its backbone. The simulated environment comes complete with robot arms for lifting packages off a conveyor belt and placing them on pallets. With up to three robot arms, each working on up to three pallets at once, teams will have to fill an order of different-size boxes as quickly as possible.
"One of the things we're trying to determine is, what makes a good mixed pallet?" says Stephen Balakirsky, an engineer in NIST's Intelligent Systems Division in Gaithersburg, Md. As automated tasks such as stacking and delivering pallets proliferate throughout industry, Balakirsky says, it would be useful to have a metric to compare them. At present, companies that manufacture such robotic systems are "really not able to point to an international standard and say how good their performance is," Balakirsky says. "We feel that being able to do accurate measurements of how well you're doing fosters innovation."
NIST hopes to move the competition from software to hardware in May, at the IEEE International Conference on Robotics and Automation in Anchorage, Alaska. "We're talking with a robot vendor now about the possibility of setting up a work cell at the competition site in Anchorage," Balakirsky says. "Teams that show that they're able to solve this problem successfully will be able to run on that real robot cell." Balakirsky says that having five or six teams registered by the February 15 deadline would provide a solid, competitive field, and that he already has three entries in hand from universities in the U.S., Germany and Croatia.
