Flying inside greenhouses is challenging. Flying a drone indoors is in any case challenging given the absence of the GPS for X,Y positioning, which makes that a dedicated indoor positioning system is needed to provide the drone with the required positioning information for autonomous flight. Avular evaluated multiple indoor positioning systems for their respective feasible in greenhouses. Off-the-shelf systems like Ultra-Wide-Band did not provide the required accuracy given the challenging nature of the environment. Avular’s Starling system, an optical infrared beacon-based system, was also evaluated and performed very well in terms of accuracy but also seems to be the more viable option economically. Avular implemented the system in the greenhouse and proved the performance for indoor drone flight in greenhouses.

In turn, ADI has developed the artificial intelligence algorithms to analyze the visual data acquired by the drone. To do so, the AI was trained to recognize healthy and unhealthy plants. This was done by feeding the AI a large quantity of training pictures of healthy plants and plants that are infected with the top-5 plant diseases.

In the future, the functionalities of this autonomous monitoring system can be extended by gathering more types of data or by equipping the drone with end effectors that allow precision farming. Together with ADI, Avular will evaluate the next steps in bringing this innovation to the market.