Together with seven other companies, robominds has developed a new, extensive concept: The first fully automated warehouse lifter, equipped with AI robotics, a mobile incoming and outgoing goods department and a full safety concept. For this purpose, a mobile packing station is set up in front of a warehouse lifter from EFFIMAT. This is equipped at the two front corners with two Universal Robots UR10e cobots and a safety system from SICK. The cobots work with the robominds LED Gripper and Piab's MX suction cup as well as with the AI-based robobrain.vision system including smart skills. These enable the cobots to recognize different stored objects, grab them out of the boxes and finally pick entire orders - independently and flexibly, without having to pre-teach the individual products.
The ready-packed boxes are placed on the autonomously driving MiR mobile robots - equipped with a roller conveyor from ROEQ - and thus transported to the workbench. In the back is a conveyor belt that can be used both manually and via the two cobots to load the warehouse. Ecosphere Intralogistics as system integrator ensures a smooth setup and operation.
order-picking by cobots equipped with AI skills
Automated incoming and outgoing goods, secured manual packing station
output of up to 250 boxes per hour, 3x higher picking performance than conventional vertical storage lifter
linking of all components of the system at all levels (manufacturer-independent)
high flexibility in stored products through AI-based picking - no need to teach-in the parts
By using the vertical storage lifter, up to 75% space can be saved
output of individual boxes from different levels within one job, loading can take place around the clock
The used skills are part of the robobrain.vision system, which also consists of the robobrain® and the robobrain.eye camera. robobrain® has the task of linking all hardware components such as robot arms, gripper and vision camera and equipping the robots with certain capabilities through flexible AI skills. The vision camera gives the robot vision and the skills enable it to recognize different objects, regardless of shape or color. Unlike conventional, rigid systems, robobrain.vision eliminates the learning process of individual objects because the AI skills are already pre-trained with a patented process.
Detection and handling of small to medium sized parts with vacuum grippers, as well as analysis of optimal gripping points.
› Supported grippers: Vacuum, a.o. Piab
› Supported objects: including boxes, bags, cans, bottles, tubes, etc.