Speed and process reliability thanks to AI-controlled gripper system

Core components of the fully automated solution for single-item picking are vision, robots and grippers, as well as the IT system and the conveyor connection to the entire intralogistics system. The smart process design and the seamless integration into the IT control system have a significant effect on the overall performance of robot picking and its economic efficiency.

The AI-supported vision system identifies suitable surfaces on the items in the source tote and calculates gripping points and collision-free paths for the robot within a few milliseconds. This enables the robot to pick up unknown items from bulk goods by vacuum suction and place them in the target tote or shipping carton. Prior teach-in of the individual items (e. g. by loading in CAD data) is not necessary. This means that new, unknown items can be added to the product range immediately without any additional effort and picked fully automatically by the robot.

The gripper can be individually designed for a wide variety of articles with different packaging units. The self-learning system recognises the gripping strategy that is best suited for each item. The range of articles that can be handled can also be expanded with an integrated gripper exchange station with different vacuum grippers.

The achievable picking performance for a process-reliable system depends largely on the properties of the objects to be gripped, yet it is well above 1 000 parts per hour. Together with AI support, this high picking speed ensures very high system resilience. Errors in the picking process are corrected independently, and individual repetitions during the gripping process have no significant impact on the overall performance.

A particular challenge in automated picking is the economic organisation of the overall process. Since robots cannot handle all items in every product range, challenges arise in terms of goods flow, inventory synchronisation, and consolidation. We provide our customers with comprehensive advice on the customised design and implementation of these processes.