New Mobile Cleaning Robot Disinfects Equipment, Production Spaces

Undoubtedly, hygiene zones and production lines should be impeccably clean. And complete cleanliness is very important wherever medical devices are handled and food is processed.

A new mobile cleaning device recently developed by Fraunhofer scientists disinfects production spaces and equipment to a reproducible standard.

The robot is integrated with autonomous and self-learning motility systems, which allow it to automatically detect the extent of fouling and choose the right cleaning process.

When it comes to hygiene in industrial food production, compromise is not an option. Manufacturing rooms and equipment should be regularly cleaned. Fouling deposits, including biofilms, should not be allowed to gain entry under any situations. Hygiene and food safety depends on the effectiveness of cleaning efforts.

While the task of cleaning equipment and plants is quite difficult and is a determining factor of quality, most of this task is still performed physically. However, in spite of all the attentiveness of the industrious cleaning crews, their work is prone to errors, difficult to reproduce, and takes a considerable amount of time.

Smart Robot Cleaners for Indoors and Outdoors

To address this problem, researchers at the Fraunhofer Institute for Process Engineering and Packaging IVV (Fraunhofer IVV) based in Dresden have created a modular cleaning robot, two models of which are currently being operated.

While one robot moves down the production line on a conveyor belt and disinfects the device from the inside, the other cleans the walls, ceilings, and floor of rooms, as well as the external surfaces of machines.

Equipped with a rotary jet cleaner, an extendable robot arm can expand to access high spots on the production line. Dubbed Mobile Cleaning Device 4.0 (MCD), this modular and mobile machine is capable of independently scooting over the shop floor.

The researchers at Fraunhofer IVV have collaborated with the Fraunhofer IOSB-AST based at Ilmenau in a new research project to analyze a multi-sensor system for adverse settings.

The system needs to be incorporated into the MCD. It uses a fascinating technique known as fluorosensing to detect contamination. The integrated sensors check and estimate the extent of fouling so that the robot can adaptively modify the cleaning parameters such as pressure and the quantity of the foam cleaning agent, required to suit the scenario.

A detector uses UV light to identify fluorescent particles such as fats, oils and proteins, and doses the foam and water according to the determined parameters, such as the layer thickness and dryness of the residue.

Max Hesse, Team Leader, Dresden Branch Lab for Processing Technology, Fraunhofer Institute for Process Engineering and Packaging IVV

“This is to be accomplished by a self-learning AI system that selects the suitable cleaning parameters and specifies the process steps,” added Hesse, describing how this process works.

Moreover, a simulation allows a virtual twin to produce the data while this procedure is ongoing. This virtual twin can be used to map and send the perceived fouling to a three dimensional (3D) model of the plant. Following this, the water pressure can be altered and decreased based on the distance between the surface and the device—all these are done to use resources in an efficient way.

Sophisticated Sensors

Powered by batteries, the robot travels autonomously and carries only a hose for the cleaning agent linked to the docking station. The robot is managed through Wi-Fi. Sophisticated sensors coupled with artificial intelligence (AI) allow this adaptive cleaning; for example, a radar sensor takes readings even through steam, mist, and spray, and an ultra-wideband sensor estimates the position in the room.

An optical fluorosensor—a third sensor—identifies fouling and sends an impression of the geometric properties of the object. This is referred to as visual odometry by experts.

After extrapolating the process parameters from the perceived levels of fouling and the fused sensor data, the system tracks the process on the go to ensure that the cleaning process is being performed correctly.

In the subsequent step, the system sends its scan results to the virtual twin with the self-learning potential. In this manner, it enhances itself with every pass to obtain exceptional outcomes while sparing resources.

Our tests have shown that this can save up to 50 percent on cleaning agents because no more than the actually needed amount is applied to the surfaces. The system can be trained to clean as resource-efficiently as possible within a given period, for example, during the nightshift downtime in a factory operating in two shifts.

Max Hesse, Team Leader, Dresden Branch Lab for Processing Technology, Fraunhofer Institute for Process Engineering and Packaging IVV

Hesse continued, “What’s more, there is considerable efficiency potential to be tapped if the skilled workers that had been doing the cleaning can perform other tasks while MCDs handle the paralleled cleaning processes.”

The benefits do not stop there: an automated process captures the whole cleaning procedure, and all that automation assures a highly reproducible procedure.

Agile and Flexible—the Industry-Agnostic Mobile Cleaning Device

The application range of the MCD is not restricted to the food sector alone. This intelligent robot can also be used on other lines of business, for instance, the agricultural, cosmetics, medical engineering, pharmaceutical, and industries. Such autonomous robot cleaners can benefit many different sectors during times of crisis, like the corona pandemic.

“Our automated system really shines when staff is in short supply. Around ten percent of employees in food production are tasked exclusively with cleaning,” concluded Hesse, an engineer by trade. This calls for skilled personnel who are hard to find even in normal times.

With ongoing research, both versions of the robots are evolving, which means they will be able to perform even more complex cleaning jobs.