AI-Based Container Inspection

In pharmaceutical manufacturing, industrial containers must be thoroughly cleaned to meet strict hygiene and quality requirements.

However, verifying that no contamination or residual material remains inside large containers can be challenging. Their dimensions and limited accessibility make comprehensive visual inspection difficult.

The objective was to develop a system capable of inspecting containers ranging from 1.5 to 2.5 meters in height and identifying remaining contamination after the cleaning process.

RAWDEV developed a custom mechanical positioning system that allows an industrial camera to inspect the interior of large containers.

The mechanism provides three degrees of freedom: vertical movement along the Z-axis and rotational movement around the Z- and Y-axes.

These movements allow the camera to capture images from different positions and viewing angles, improving the accessibility of areas that would otherwise be difficult to inspect.

The captured images are analyzed using a custom-trained machine learning model designed to detect and localize residual contamination.

RAWDEV was responsible for the mechanical design of the positioning system and the development of the camera end effector.

A functional prototype combining custom mechanical hardware with AI-based image analysis was developed.

The project demonstrated an approach to automating visual post-cleaning inspections in pharmaceutical manufacturing.