In an internationally renowned company in the chemical and pharmaceutical industry, the supply of chemical substances for processing in the downstream plant was organized manually for a long time: Employees had to lift bags weighing up to 25 kg from pallets and prepare the contents for the subsequent production steps. This process was not a good solution from both an ergonomic and an economic point of view. The fact that the contents of the sacks were also predominantly classified as hazardous substances led the company to decide to automate this part of the system as far as possible. The company duo ASA Automation and Vision On Line took on this task.

Gripping accuracy is crucial

ASA Automation and Vision On Line had to overcome a number of challenges when it came to handling the bags of chemicals. It quickly became clear to ASA Automation that the bags had to be picked up from the staging position on a pallet using a vacuum suction gripper on an appropriately designed robot. However, depending on the contents and properties of the bags, they sagged to varying degrees after being picked up. To be able to grip them reliably, an application-specific suction pad had to be developed and manufactured. To make matters worse, bags that are not picked up exactly in the center can tear open when they are lifted or during the subsequent movement. Special attention therefore had to be paid to high accuracy when determining the gripping positions. The precise detection of object positions is a typical task for industrial image processing systems. In this particular case, however, standard solutions were not sufficient: some of the sacks were lying twisted on the pallets or hanging over the edge. In addition, they are often arranged close together or even overlapping, especially in the lower layers, so that there is no clear gap between them as a boundary. Furthermore, the objects do not always have the same contour because they are sometimes dented or compressed. And last but not least, reflections and extraneous light problems can occur with the surrounding plastics, which further complicates the boundary conditions for image processing.

With AI to the solution

Two 3D scans of the pallets loaded with bags are combined by software and enable the exact gripping position of the bags to be calculated. © ASA Automation

ASA Automation implemented the mechanical part of the system using a 6-axis robot with an additional seventh axis for the travel path in the X direction. This setup ensured that the pallets, which were positioned in a defined location via corresponding conveyor elements, and the bagged goods on them were within reach of the specially developed vacuum suction gripper and could be fed to the downstream system in the exact position.

Vision On Line recommended the 3D picking solution 'EyeT+ Flex' from the Italian company IT+Robotics for the exact determination of the gripping point, as it covers the possibility of teaching objects with the help of artificial intelligence in addition to other tools. Due to the geometric variance of the sacks, their positions and the different materials, traditional image processing methods were ruled out as a possible solution.

The author: Peter Stiefenhöfer is the owner of PS Marcom Services in Olching. © PS Marcom Services

The eye of the system is a camera installed on the robot arm, which performs two 3D scans of the pallets filled with bags. Two 3D scans are necessary due to the small distance between the camera and the objects, which resulted from the low room height and the relatively high stack height: wide-angle images in just one scan were not feasible due to the distortion, and the use of a height sensor was also out of the question, as there was a risk of incorrectly measuring the object height at the lower edges of the sacks. This could have led to collisions between the gripper and the bags. Vision On Line was able to overcome this difficulty by taking two images with the camera and combining the image data using suitable software.