Smart Farming using Machine Learning
Predicting the amount of grass available for livestock
Organic, grass-fed livestock farming is a growth industry. However, moving away from artificial fertilizers makes it more difficult for farmers to predict when livestock should feed on a new pasture. Too much grass growth leads to an over usage, increases the likelihood for weed growth and ultimately reduces the grasslands by converting them back to bushes. Farmers therefore need to identify the best time to put livestock onto a new pasture.
Using weather data, coupled to historical data when a certain pasture was last used by livestock can be used to predict the optimal timing of re-opening a pasture. With AutoML, farmers can predict the amount of grass that should be on a pasture at a time. This allows the farmer to optimize the herding of livestock and make optimal usage of the available land resources.
Modulos AutoML can combine image data, tabular data, and time series data all in one. This allows the farmer to use all information available to train machine learning models at the farm without the need for elaborate computer infrastructure.