Aigro - smart horticulture system

Important results from the project

The goal of the project was to develop an AI-based cultivation platform (AIGRO) for real-time monitoring of Svegro´s six most cultivated products, which account for 80% of turnover. The purpose of the cultivation platform is to optimize energy use and reduce cultivation wastage by consistently using AI models to assess size and deviations in the cultivation and thereby take measures that reduce the risk of wastage without using too much energy in the greenhouses.

Expected long term effects

The project has resulted in a commissioned visualization tool, AIGRO, which is used in the daily work to monitor the size and quality of the products and, based on the forecasts, take measures to reduce energy use and ensure the right size at harvest. The measurement precision of the predictions from the AI models is currently +/-0.5 cm, which is more than accurate enough for Svegro´s products. During the part of 2022 (April-December) that AIGRO has been used, energy savings of 6% per grown product have been made which can be attributed to AIGRO.

Approach and implementation

Infrastructure with cameras was set up in the greenhouse and a system for labeling batches with barcodes was implemented to relate image data to cultivation data. A flexible data system architecture was developed that can handle different types of input data. The data collected is consumed by the AI models, which were trained on images of the different crops in different growth phases to give a good prediction of the size. Based on the predictions and other cultivation parameters, a forecast model was developed for size at harvest and deviations that are visualized in the web tool AIGRO.