MicroAI AtomML™ Enables Greater Greenhouse Productivity

The Industry Need

As industries move forward and become more sophisticated with increasing demands of output and efficiency, leaders must continue to find a way to differentiate themselves. When it comes to agricultural producers, specifically greenhouses, there are several ways forward to increase crop yield and catch problems early. In addition to ensuring the quality of the crops, fruits, and flowers grown within greenhouses, early detection of weeds and the detection of pests would also prove beneficial.

When all these issues are solved, greenhouse efficiency can be increased. To accomplish this, the solution would need to automatically detect the problems and then dispense immediate corrective behavior or raise alarms so that problems are identified, and corrective actions initiated.

The Solution

MicroAI’s MicroAI AtomML™ can accomplish all these goals and more. The solution can live directly on any currently present and available microcontrollers deployed in the greenhouse. If there are none, the solution functions on MCU’s and MPU’s so the client will have their choice of host.

MicroAI™ Atom learns the exact behavior of the greenhouse and can trigger corrective behavior and/or alerts and alarms when changes in the expected behavior occur. The only thing required for the solution to work is the selection of the proper sensors. Soil, temperature, and UV sensors would ensure that proper conditions for the plants can be monitored and maintained. Domain knowledge of the plant’s ideal conditions can be applied to engineer the sensor readings for increased accuracy. The soil sensors would also be useful in detecting the presence of weeds if hydration or nutrient levels are different than expected. From here, the developer can choose what responses the system will initiate.

Based off the results of MicroAI™ Atom, the system could automatically activate a pump to dispense more water or instruct the greenhouse’s climate control device to adjust temperatures based off the readings it is presented with. Finally, a motion sensor would allow for the detection of pests that would disturb the greenhouse’s yields.

The Impact

With MicroAI’s MicroAI AtomML™ monitoring and managing the greenhouse, the user can expect increases in product yield. In addition, human hours required to run the greenhouse can be minimized as they will only need to perform actions when alerted by the system. If the greenhouse is segregated, with each section requiring slightly different conditions, different MicroAI™ Atom engines can be applied to learn the different nuanced areas and create specific models to accommodate unique surroundings.