When do our houseplants need fresh water, and how much water can they actually tolerate? Generally, plant care is based on estimates and the experiences that we collect over time. Modern precision agriculture replaces these experiences with measured values, and sensors are used to find out when the soil is too dry. However, one important factor is still being ignored. Plants are not always ready to absorb water, light, and nutrients. If a plant is in a photosynthesis phase, it needs care that is different from when it is sick. For this reason, startup Hortiya is training artificial intelligence (AI) to better understand the language of plants.
Small sensors
To listen to plants, AI needs sensors. For this purpose, Hortiya developed its own technology, which allows sensors to be installed near or directly on plants. These sensors provide AI with real-time data on soil moisture, as well as the amount of light that falls on the leaves.
Equipping plants with sensors is by no means a new development in agriculture. Most companies that are engaged in precision agriculture rely on sensor data to care for their crops. Computer programs combine this data with location-based data such as weather forecasts. Researchers from the USA already proved that this reduces herbicide usage by up to nine percent.
More efficient agriculture requires a more efficient usage of resources. Companies can use light sensors to intelligently control greenhouse lamps. Previously, automation was only possible through the usage of timers, nowadays the need for an artificial light source is determined through several factors. For instance, the lamps only switch on when there is not enough natural light available. In this way, the availability of natural light sources is optimally utilized, which saves energy. Also, these systems are solely based on environmental factors. Unlike these systems, Hortiy's plant AI adds the plants themselves into the equation.
When is a plant ready to get water?
"Our aim is to develop a universal data model that can be used to predict the condition of a plant," says Marc Weimer-Hablitzel, CEO of Hortiya. The model is already being used for greenhouse tomatoes. Hortiya is also focused on basil and cannabis.
It's crucial for farmers to know when tomatoes, basil, and cannabis are undergoing photosynthesis or when they suffer from diseases or water shortages. Automated greenhouses could use this information to more accurately assess the resources they add to plants. According to Weimer-Hablitzel, this efficient approach could replace the 'better too much than too little'- approach.
In the future, Hortiya would like to develop a generalist model for plants. In other words, one that can interpret the signals of all C3 plants rather than only those of specific plants. Like all conventional language models, plant AI requires a massive amount of data. Namely, the longer the model is used and the more users it has, the more precisely it works.
AI allows us to understand the language of plants
The availability of plant AI could enhance the efficiency of precision agriculture in the future, even if Hortiya does not provide the sensors itself. Systems that have already been implemented could be enriched, which would lead to more data and more efficient AI.
This is a great opportunity to reduce agricultural emissions and to prepare the sector for climate change. Plant AI could make the sector more resource-efficient, and thanks to better automation, it could counteract the impeding skilled labor shortage. Moreover, it can be used all over the world.
Over the past few months, the Chat-GPT language model has made the rapid development of AI tangible for many people. However, it's consistently ignored that AI requires a large amount of energy and that it involves CO2 emissions. Hortiya's plant AI aims to show that modern language models can have the opposite effect.
Source: en.reset.org