The first self-checkout system was installed in 1986 at a Kroger supermarket just outside Atlanta. It took decades, but the technology eventually spread across the U.S. Given the automated direction supermarkets are taking, it seems that robotic bagging can’t be far behind.
MIT's CSAIL department presents this week RoboGrocery. It combines computer vision with a soft robotic gripper to wrap a wide range of different items. To test the system, researchers placed ten objects unknown to the robot on a grocery conveyor belt.
The products ranged from delicate items such as grapes, bread, kale, muffins and crackers to much sturdier items such as soup cans, meal boxes and ice cream containers. The vision system comes into play first, detecting the objects before determining their size and orientation on the belt.
As the gripper touches the grapes, pressure sensors in the fingers determine that they are in fact fragile and therefore should not reach the bottom of the bag – something many of us have undoubtedly learned the hard way. Then it notices that the soup can has a stiffer structure and sticks it into the bottom of the bag.
“This is an important first step toward letting robots pack groceries and other items in the real world,” said Annan Zhang, one of the study's lead authors. “While not quite ready for commercial deployment, our research demonstrates the power of integrating multiple sensing modalities into soft robotic systems.”
The team notes that there is still a lot of room for improvement, including upgrades to the gripper and imaging system to better determine how and in what order things should be packed. As the system becomes more robust, it can be scaled beyond the supermarket to more industrial spaces such as recycling plants.