The first self-checkout system was introduced in 1986 at a Kroger grocery store just outside of Atlanta. Since then, this technology has slowly made its way across the U.S., paving the way for even more automated solutions in grocery stores. With the rise of automation in this industry, it’s only a matter of time before robotic bagging becomes a common sight.
Recently, MIT’s CSAIL department unveiled a groundbreaking new technology called RoboGrocery. This system uses computer vision in tandem with a soft robotic gripper to efficiently bag various items. To test its capabilities, researchers placed 10 different objects on a grocery conveyor belt for the robot to bag.
From delicate items like grapes, bread, kale, muffins, and crackers to more solid products like soup cans, meal boxes, and ice cream containers, the system had to adapt to each item’s unique characteristics. The computer vision component enabled the system to detect the objects, determine their size, and position them correctly on the belt.
As the robotic grasper interacted with each item, pressure sensors in the fingers gauged their fragility. For example, it recognized that grapes should not be placed at the bottom of the bag due to their delicate nature. On the other hand, the system correctly identified a soup can as a sturdier item and placed it securely at the base of the bag.
Lead author of the study, Annan Zhang, expressed optimism about the potential of RoboGrocery. He stated, “This is a significant first step towards having robots pack groceries and other items in real-world settings. Our research showcases the effectiveness of integrating multiple sensing modalities in soft robotic systems.”
While the system shows promise, the researchers acknowledge that there is still room for improvement. Enhancements to the grasper and imaging system could improve the accuracy and efficiency of the packing process. With further development, this technology could extend beyond grocery stores to industrial settings like recycling plants.