Russian scientists have created an autonomous drone on a wheeled platform that automatically moves around the garden, finds fruits hanging on trees and identifies sick or damaged ones among them. He received the name botANNIC. Based on the collected data, a sanitary map of diseases of agricultural plantings is formed. Tests in an operating fruit farm have shown that monitoring accuracy exceeds 80%. As conceived by the authors, the use of the development should increase productivity and minimize fruit losses due to diseases and pests. According to agronomists, the device can only be used as a human assistant, but not as a replacement. Adapting the invention to work in greenhouses will be difficult, since there are no tracks for wheeled vehicles.
Four classes
NUST MISIS specialists, together with colleagues from the Tambov State Technical University (TSTU), China University of Mining Technology and the Higher School of Economics, have developed a wheeled robotic platform called botANNIC. It autonomously monitors the condition of fruits in orchards, identifies diseased or damaged fruits, and, based on the data obtained, forms a map of horticultural diseases. As tests have shown, the accuracy of the device exceeds 80%.
“With the help of a stereo camera that uses neural networks to reproduce human vision, botANNIC scans the deciduous and fruitful parts of trees, detects apples in the tree crown, revealing their degree of ripeness and damage,” Ivan Ushakov, head of the Department of Physics at NUST MISIS, told Izvestia.
Taken from the air: the autorobot will provide environmental control in enterprises
How a small car will replace the environmental inspector
botANNIC independently orients itself in the garden using a three-dimensional map of the area, but if desired, the operator can set it a trajectory and the end point of the route. Platform dimensions – 1.5×1.5 m, load capacity – 200 kg, electric motor power – 500 W. The design uses lithium-ion batteries. The drone is equipped with a stereo camera and a hyperspectral sensor that allows you to carefully analyze the appearance of fruits and evaluate their quality based on the observation results. In the process of moving around the garden, the drone collects information, which is then uploaded to the on-board computer of the platform and the operator’s tablet.
The large load capacity of the wheeled platform makes it possible to install additional equipment on it. For example, a manipulator for fruit sampling or harvesting.
The system has successfully proved itself during tests on the territory of an industrial apple orchard of the Federal Scientific Center. I.V. Michurin, located in the Tambov region. The research results are published in the prestigious scientific journal Drones.
- To detect apples against the background of foliage in fluctuating sunlight, an algorithm has been developed for automatically selecting the foreground area. The system then further classifies fruit images using multivariate analysis into the following groups: healthy apple (class 0), rotten (class 1), spotted (class 2), damaged by insects (class 3) or fungal scab (class 4). The classification accuracy is at least 80%,” said Alexander Divin, professor at the Mechatronics and Technological Measurements Department at TSTU.
The system is designed to work in the so-called intensive gardens, in which trees are planted in a compacted pattern. Scientists intend to equip botANNIC with additional sensors and a laser emitter, with which it can destroy harmful insects and weeds. The application of technology should lead to an increase in farm productivity and a decrease in fruit loss that occurs for various reasons.
harvesting applesPhoto: RIA Novosti / Vitaly Timkiv
Help agronomist
Pavel Tyurin, the chief technologist of the Plodovoye company, told Izvestia that at the moment the quality control of fruits in fruit plantations is carried out manually. For this, there is a hotel position – a forecaster or entomologist. He constantly inspects the garden. His duty is to look after the condition of the trees and fruits. However, it must work preventively and prevent the occurrence of diseases and pests. For example, a special trap for the codling moth is installed in the garden, if insects are found there, then the trees are treated with pesticides. Also, according to the indications of the weather station, they monitor the state of the environment.
