A robot bagging system for autonomously bagging groceries has been developed by a team from Massachusetts Institute of Technology’s (MIT’s) Computer Science and Artificial Intelligence Laboratory (CSAIL) department.

The RoboGrocery reportedly combines computer vision with a soft robotic gripper capable of bagging a range of objects. To demonstrate RoboGrocery’s capabilities, the research team placed 10 objects previously unknown to the robot on a grocery-store like conveyer belt.

With products ranging from grapes, bread, kale, muffins and crackers to soup cans, meal boxes and ice cream containers, the RoboGrocery’s vision system activated first, enabling it to detect the objects before determining their size and orientation on the belt.

Once RoboGrocery’s grasper touched the grapes, its developers reported that pressure sensors in the robot's fingers determined that the product is delicate and should thus not go to the bottom of the bag, while the more rigid soup can was subsequently placed at the bottom of the bag.

An article detailing the system, “Real- Time Grocery Packing by Integrating Vision, Tactile Sensing, and Soft Fingers,” appears in the IEEE Conference Publication, IEEE Xplore.

For more information on the RoboGrocery bagging system, watch the accompanying video that appears courtesy of MIT CSAIL.