With over thousands of eggs to be collected daily from a poultry farm there is a pressure of resources and especially labour to complete the task effectively and efficiently. A welding robot in this test is of course just the fun element, but the fact is that we succesfully tested how a 60-gram, fragile mid-sized egg can be picked and placed by a 20-kg-payload Fanuc Arc Mate 120iC with an mGrip soft gripper from SoftRobotics.
We recieved many inquiries in regard for egg palletizing (pick & place) so we decided to to continue with our series of tests. A few months ago, we
tested a soft gripper application that handled chicken meat (wings and drumsticks). Since Easter is almost here, our testing team agreed immediately: Let's pack eggs!
In this quickly assembled test app, the eggs were delivered to the "robot cell" on a conveyor. Our goal was to pick up the eggs arriving on the tape one by one and place them into a paper egg holder without damage. The robotic arm we used was a 6-axis Fanuc Arc Mate 120iC with a load capacity of 20 kg and a repeatability of ± 0.03mm. Obviously we didn't run the test at maximum speed. The most common robotic application for this type of robot is welding, but our programming trainees tamed the structure in a matter of seconds: during the test, the robot treated the "workpieces" literally as Easter Eggs.
Just as last time, the test gripper was provided to us by one of the market's leading soft gripper manufacturers, SoftRobotics. This gripper, the
mGrip has variable number of fingers. Thus, gripper length and layout can be easily modified. During the test, we used it as a 2-finger parallel gripper. SoftRobotics grippers are also available in FDA-certified versions - regularly
tested by the manufacturer himself with a wide range of food and consumables industry products, starting from raw dough to banana or chips bags and toothpaste tubes. The gripper can be opened and closed to variable sizes with variable air pressures using a controller. When completed with a camera system, mGrip is absolutely excellent for sorting tasks.
We used an Omron FJ-SCG camera in the test. This type of camera, like most basic camera systems today, can be used for many functions: positioning, measuring, shape recognition, and more.
There is complete agreement among professionals today regarding the key to the development of "visual" sensing. The future of robotic pick and place apps is defined clearly by image processing based on artificial intelligence. Join us for a recent virtual
discussion of this topic with dr. Péter Galambos, Head of Antal Bejczy Robotics Center at the University of Óbuda, talking about self-learning algorythms and 3D bin picking at the
Robot-X Automation Webcast.