YaCuAi | Problem? We solve it.

Problem? We solve it.

Problem? We solve it.

To continue talking about the YaCuAi project, we gathered this article on the problems that our project is aimed to solve.

Let us first emphasize that cleaning itself, i.e. the action of removing dirt, i.e. its tool, is easily automated - it is not possible completely automate the movement which is necessary to move the cleaning tool. And here we find the first oddity, probably inherited from domestic vacuum cleaner robots, many teams involved in "depiloting" industrial floor cleaning machines get ahead of themselves, without having time to separate the tasks of controlling the cleaning in general and controlling its movement - something that is strictly and obviously separated in the well-established huge industry of automation. Remember, the process of cleaning a serviced area different from milling out workpiece material only by the level of openness of the world in which the tool moves (we use that term on purpose).

However, the automation of industrial cleaning is not considered by the community of specialists as a part of industrial automation and it is not aimed at the established concepts and technological paradigms of the industry, where the key is that the result (task) is controlled by man, and the process of getting it - the machine. We are dealing with a formula of a home vacuum cleaner robot: turn it on and forget about it, cleaning it well - fine, bad - well, okay, but look impressive.

The underlying architectural and operational software solutions, their forms and specific implementations, as well as levels of abstraction, also different. It is accepted that a cleaning robot should make all decisions by itself, "build" a map and routes instantly, navigate by this map, etc., a clear analogy with a human cleaning robot is made and no objections are allowed.

We agree, but in the whole field of industrial automation they are directly allowed and extremely successful, no CNC machine decides for itself how to handle the component, and this allows both a clear decomposed automation solution and controllability, as well as important predictability and repeatability of results.