Case Study: Produce Labelling Cell

For low volume produce, such as pomegranates or melons, the cost to upgrade from manual labelling to automated has, until now, been unjustifiable – owing to the sheer physical diversity of such products, and the associated complexities of dealing with them.

However, we have developed an adaptable vision-based robotic system that is capable of affixing labels to such produce, and almost anything with a 3D profile, all on a single processing line, with product and batch changes possible at the a touch of a button.

Other Case Studies

Description

First, boxes of produce are loaded onto a conveyor. Next, they pass through a vision enclosure where it is scanned by a 3D imaging camera. This is analysed by the system to identify the produce and to provide the robot with the precise location the labels should be placed.

The labels are printed as required to minimise wastage. The labels can be of different shapes and sizes.

Label application is performed by two robots which cooperate to optimise the picking and placing of labels. The robot controllers track the movement of the conveyor, enabling the labels to be applied without slowing down the process.

Finally, the fully-labelled produce then leaves the cell via a roller conveyor.

Equipment Overview

Two KR6 R900 Agilus Six Axis Industrial Robots.
Two multi vacuum padded sticker applicator end effectors.
Two Sato label printers.
Cell control system integrating safety, monitoring, control, software, and data acquisition.
3D imaging camera
Conveyor system

Key capabilities

DIVERSE RANGE OF PRODUCTS

Product size can vary from 50 x 50mm to 450 x 250mm. System handles textured or smooth surfaces. Labels of different shapes and sizes can be designed and printed on.

SIMPLE BATCH CHANGEOVER

Switch from avocados to watermelons at the touch of a button. Ideal for low produce and seasonal items.

MULTIPLE LABELLING MODES

The system supports print and apply, apply only, and mixed modes of operation whereby one robot can be performing print and apply and a second robot apply only for pre-printed or promotional labels.

FEATURE AVOIDANCE

This system can be trained to recognise certain features and then avoid them. For example, it will avoid stalks, corners and awkward product areas where label placement would be unwanted.

Training Routine

The complexities of the system are hidden from the user with a simple software interface that can be used to train the system to identify new products. Using a browser-based setup interface, a new product can be added to a list of products to be identified even while the machine is in operation. Training requires that a lone specimen be placed on the conveyor and imaged by camera.

The set-up interface also allows a user to:

Create library of label types and sizes
Import image of label for visualisation
Customise label positioning via drag and drop interface
Create labelling recipe which determines for each product: the printer to be used, specific labels to be printed, label position, number of labels

The approach to training and target identification makes no assumptions about product positioning or presentation. Consequently, the system is impervious to changes in packaging such as the size of boxes, the number of products in a box and damaged boxes.

Robotic Labelling Machine with Two Robots

Vision Software showing melon being identified on the HMI

Robotic Labelling of produce like avocados and melons

KUKA robot labelling melons with dual ended padded vacuum end effector

Contact

If you are interested in talking to us about a similar project, please contact us.

More Information

You can also find more information in our full article on the Vision Systems Design website:

Read article

More case studies