Algorithms and Software for UAV Flight Planning for Monitoring the Stress Conditions of Plantations

• Published in: 2020 IEEE 6th International Conference on Methods and Systems of Navigation and Motion Control (MSNMC) pp. 146 - 149
• Main Authors: Komarchuk, Dmytro, Pasichnyk, Natalia, Lysenko, Vitaliy, Opryshko, Oleksiy, Shvorov, Serhii, Reshetiuk, Volodymyr, Udovenko, Oleg, Knizhka, Tetiana, Kharinova, Maryna
• Format: Conference Proceeding
• Language: English
• Published: IEEE 20-10-2020
• Subjects: Agriculture; Compaction; flight program; Indexes; Satellites; Software; Stress; technological stresses; unmanned aerial vehicle; Vegetation mapping
• DOI: 10.1109/MSNMC50359.2020.9255605

Remote monitoring technology is a mandatory component of the crop management concept. The available solutions allow determining the presence of plant stress but not identifying its causes. A particular danger is presented by stresses of a technological nature, and chemical poisoning of plants due to the aftereffect of herbicides, compaction of the subsoil, and the like. Plant stresses of a technological nature lead to a decrease in plant immunity and, accordingly, special measures are needed to restore their productivity. Laboratory methods for analyzing stress, in particular, chemical poisoning of plants, are technologically complex and expensive, which prevents their widespread use. Remote sensing technologies are capable of identifying areas with manifestations of technological stresses since such stresses have characteristic features. As our studies have shown, a promising method for identifying plant areas with signs of technological stress is the method of leaf diagnostics. For such areas, it is necessary to carry out monitoring with the highest image resolution, it is assumed in the UAV flight program. Taking into account the above, the aim of the work was to develop an algorithm and software for its implementation of UAV flight planning for the identification of plant stresses of a technological nature. The software was developed in the cross-platform programming language Python, and it allowed processing maps of the distribution of vegetation indices (for experimental studies, maps were used that were created using the Slantrange spectral sensor system). The use of the algorithm, implemented in the cross-platform programming language Python, made it possible to identify the paths of movement of technological equipment, the contours of areas with close values of the vegetation index, and the main features of areas with plant stress of a technological nature. The accuracy of identifying areas with technological stresses has been confirmed by ground surveys in production fields.