Creation of a biplan type uav model using a genetic algorithm in flight to help in solving the problem of fires in the Irkutsk region

Abstract

Since the end of July 2019, a sharp increase in the area of wildfires has been recorded in Siberia. The most difficult situation is in the Irkutsk region, where some areas suffer from floods, while others - from wildfires. In order to more effectively deal with natural disasters occurring on the territory of both the Angara region and Russia as a whole, the project is proposed for the development of unmanned aerial vehicles (UAVs) capable of operating autonomously and performing a number of tasks to collect information. The ecological situation in general in the catchment area of Lake Baikal continues to worsen every day. Not only natural disasters (forest fires and floods), but also unauthorized solid waste dumps, poaching, and tourist pollution of Lake Baikal pose a large-scale threat to the ecology of the drainage basin. Of course, constant and operational environmental monitoring of Lake Baikal drainage basin condition seems urgent, making quite useful the created model range of UAVs, including the model of a quadrocopter-ecologist, developed by our team earlier. This article is devoted to the design of a UAV model штеутвув to help monitor forests and water areas of the Irkutsk region. The design was carried out taking into account the fact that the device must operate in adverse climatic conditions and the on-board computer is capable of working in the interraction with a neural network. The choice of the UAV concept is substantiated - a biplane with swept wings. Computer three-dimensional modeling (CAD) is carried out in the universal program Autodesk Fusion 360. The work presents a drawing of the resulting computer model. The created CAD model will allow obtaining the drawings and G-code necessary for creating full-scale models for 3D printing and the possible use of a milling and laser machine tools. A description of the process of creating a test model and equipping it with electronics is given. The UAV software will be based on a genetic algorithm. The architecture of the neural network was chosen and the necessary mathematical calculations were carried out.