Currently, the fill level of the beet bunker is recorded using ultrasonic sensors, which, however, can only determine the volume and not the mass and are rather inaccurate. The KibEZ project is testing a solution based on artificial intelligence (AI) that is intended to be usable in the long term on different machine types from various manufacturers. Machine parameters and measurements such as power consumption during cleaning are used and evaluated with other data, such as soil type, etc., using AI. To do this, the AI must learn to draw the right conclusions through extensive calibration trials and derive accurate yield maps from them.

To create such a system, the project partners must bring their expert knowledge together: The practical experience from agricultural practice is just as necessary as the cooperation of the harvesting machine manufacturers, the exact execution and documentation of the trials, and not least the selection and implementation of the algorithms and the training of the AI.