Use Case


Potato diseases are a widespread problem that causes large economical losses for farmers and negative effects on the environment. Avular and Agrisim want to solve this problem by using robotics, artificial intelligence (AI) and big data for the prediction and early detection of potato diseases.

Our solution

Our solution is a ground robot that autonomously navigates through the potato fields to collect image data from the potato plants. Using AI, it is possible to detect potato diseases in an early stage or even to predict them, based on this image data from the plants and big data technologies. When the SoilMate has detected a problem, the farmer can act fast to prevent diseases from spreading through entire fields.

In this project, Avular will adapt its Ranger ground robot for driving through potato fields. These bumpy and muddy outdoor terrains have specific requirements for wheel suspension and other mechanical parts of the Ranger. When the ranger is able to drive through the fields, it should also know where to go. Therefore, Avular will develop software and hardware for autonomous navigation through the potato fields, in a row by row basis. Lastly, the Ranger will be equipped with a camera system to acquire high quality data that serves as input for the disease detection and prediction.

Agrisim will develop algorithms that predict which fields are likely to be affected by a certain disease. This algorithm uses environmental parameters like humidity, location, temperature (and more) that are merged though big data technologies. In parallel, Agrisim will establish AI for disease recognition. To do so, the algorithms will be trained with image data from healthy and unhealthy plants. To make as large an impact as possible in a short time, the focus of this project is on the recognition of the five most occurring potato diseases.

At the moment, Avular is searching for the best wheels, motors and batteries for use in potato fields, taking into account the slopes and bumps that are present in the fields. Agrisim is developing the AI to predict which fields are most likely to be affected by diseases using environmental information and big data technologies. When the Ranger is completely adapted for potato fields and when all AI for prediction and detection is developed, everything will be integrated to have a fully operational solution for potato farmers. In the future, we hope to expand to the detection of more rare diseases as well, to provide an even completer solution.

Subsidy program: MIT Gelderland R&D

Partner: Agrisim

Duration: 1 dec 2019 – 1 okt 2021

Photo: Kristina Paukshtite via Pexels