Flex Depal

Note

The following requirements are needed for enabling this feature:

• On the hardware side, a standard or high performance Pickit processor, which have a dedicated GPU.

• On the software side, the Pickit Depal application license.

Flex Depal is a detection engine designed for depalletizing applications, where the parts to pick are mostly flat and stacked in layers. It can reliably detect boxes (mixed and unmixed), bags and other palletizable items.

Flex Depal uses a Vision Language Model (VLM) to detect objects based on simple text prompts, requiring minimal configuration. Because it works from prompts that identify entire categories of objects rather than individual SKU models, it is ideal for high-mix applications such as logistics and e-commerce, where new packaging arrives daily and cannot be modeled in advance.

For box depalletizing, the default configuration should be sufficient for most applications. In the Object type section, it specifies the box prompt, and good default values for the other parameters.

Changing the prompt allows you to detect other types of objects (like bag), and higher threshold values result in stricter matching for the selected object type. Optionally, the Expert settings give finer control over a rectangle that is fitted to the dominant object plane.

Also optional, the Filter objects section allows rejecting objects that deviate from specified dimension ranges.