Automated Order Picking

About Automated Order Picking

Fizyr is specialized in segmenting, classifying, doing quality control and finding the best grasp poses for picking unknown items that vary in shape, size, color, stacking, etc. Thanks to our deep learning algorithms, robots are able to pick any object from bulk for e-commerce, warehouses, logistics and more.

Multiple system integrators and e-fulfilment providers have trusted Fizyr’s software in online retail, warehousing and parcel services. It is highly successful in various conditions and allows sensors to deal with: • Closely stacked or overlapping objects • Highly reflective items and apparel in poly-bags • White-on-white / Black-on-black flats • Transparent objects

Our deep learning algorithms are trained for: • Segmentation: Detecting doubles, overlaps and side-by-sides on a conveyor. • Classifications: Including hard and soft shells, non-conveyables and other relevant information. • 6 DoF Grasp Poses: Proposing the best grasp strategy based on all classifiers. You can configure your preferred multi-suction gripper. Our software will propose which (combination of) different suction cups to apply. The algorithm can provide the grasp poses of >100 baggage each second.

Keywords

• 3D Vision
• Ai
• Artificial Intelligence
• Automated Order Picking
• Classification
• Computer Vision
• Deep Learning
• Feeding Sorter
• Industrial Robotics
• Machine Vision
• Machine Vision Software
• Mixed-Bin Picking
• Order Picking
• Parcel Singulation
• Picking Apparel
• Robot Automation
• Universal Picking

Address

• Headquarters

  Julianalaan 67A2628BC DelftZuid-Holland, Netherlands