Speedy, Automated Robotics Yield High Efficiency in Food Packing

Soft flexible grippers and computer vision bring specialized automation capabilities to the food packaging market. Under the guidance of artificial intelligence, robotic arms are lending a hand to pack perishables for shipping.

Article Key

Image credit: Soft Robotics mGrip

The adoption of interconnected technologies is fueling the food automation market. Automated systems improve operational safety and extend the shelf life of edible products because they can be packaged quickly. At the same time, automated systems reduce the likelihood of foodborne illnesses and cut labor costs.

Robotics with integrated artificial intelligence, computer vision, and advanced analytics are changing the food packing industry for the better. With a CAGR of 9.5 percent, the food automation market is projected to reach $29.4 billion by 2027.

Automated palletizing machines bring food suppliers improved efficiencies, but automation is making its way up the food chain. Technologies such as Soft Robotics’ mGrip are changing the way products move from the conveyor belt to the shipping box.

Assatec Sweet Potato Automation Solution with Soft Robotics mGrip and 3D vision system

Image credit: Assatec 

Robotic Food Packing: No Small Potatoes

Based outside Boston, MA, Soft Robotics builds automated picking machines that integrate 3D machine vision and AI software with soft, pressure-adjustable fingers. The end-of-arm tool uses off-the-shelf components, so the gripper can be integrated into multiple robotic arms. Integrators can customize solutions for the task, building a gripper with two or 10 fingers.

The fingers are coated in a food safe blue material and the pressure can be adjusted depending on what item the grippers are lifting. They are gentle enough to pick up marshmallow Peeps and strong enough to lift vacuum-sealed salami logs without damaging either one. The system is adaptable so the same gripper can pick multiple products or SKUs.

Soft Robotics’ mGrip system uses the soft fingers to collect items from the conveyor belt. The mGripAI Perception module provides the computer vision component through 3D cameras, and the mGripAI Control Module is the guidance system, coordinating the picking and packing.

The mGripAI uses real-time object tracking to ensure consistent and accurate picking, even in challenging picking scenarios, like variable shaped potatoes or a jumble of cucumbers or sausages. It can even pick and place bulk items, like chicken legs or steaks, or pick produce and bakery items from a bin for repackaging.

Sweet Success with 3D Cameras

Israeli food exporter Carmy uses the mGrip system to box sweet potatoes, which can be hard to pack due to odd shapes and sizes. The mGrip solution is integrated with technology from Assatec Robotics, NSix Vision, and Eshet Eilon.

Carmy’s packing solution uses image analysis from three 3D cameras from NSix Vision. One camera sits above the conveyor belt and the other two are over the box. That provides the software with a 3D image of everything in the box at any given point during packing. The images assess the size and shape of the sweet potatoes on the conveyor, so the software can tell the arm which potato to pick up and where to place it in the box, ensuring the boxes are packed consistently and at the appropriate weight. The guided pick-and-place process takes only 1.2 seconds per potato.

Soft Robotics mGrip picking up sweet potatoes

Image credit: Soft Robotics

Potato Picking and Palletizing

Robotic automation has been a game changer at Tasteful Selections, one of the largest growers and shippers of mini potatoes in North America. As its business grew, Tasteful Selections worked with Schneider Packaging Equipment Company, a systems integrator that specializes in packing and palletizing equipment. The company rolled out 11 FANUC palletizing machines. To gain further efficiencies, Schneider helped Tasteful Selections deploy five FANUC M-3iA Delta robots, each equipped with Soft Robotics’ mGrip technology.

The mGrip fingers pick up mesh bags of baby potatoes that are splayed on the conveyor belt in different shapes. The software tells the robotic arm which bag to collect and where to place it in the shipping box. It also knows how many bags go in each box. The robotic mGrip system can pack more than 45 bags per minute, a 15 percent increase in overall equipment efficiency. Within nine months, Tasteful Selections achieved its return on investment.

Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries.