Method and Algorithm of Autonomous Flight Trajectory Planning of an Unmanned Aerial Vehicle When Monitoring the Fire Situation in Order to Detect the Source of Ignition Early

Abstract

The purpose of the research is to develop a method and algorithm for autonomous planning of the flight path of an unmanned aerial vehicle when monitoring the fire situation, which are designed for early detection of the source of ignition. Since timely detection of a fire source at the stage of its development allows to reduce both material and human losses, the development of a method and algorithm for autonomous flight trajectory planning of an unmanned aerial vehicle when monitoring the fire situation in order to detect the source of ignition early is an urgent, important task.Methods. A method for detecting the source of ignition is proposed. The method is based on three flight plans. Plan A provides for a flyby of the monitored area by gas with determination of the concentration of a harmful substance in each pixel of the monitoring area. When a pixel is detected in which the concentration of a harmful substance exceeds the threshold level, the flight control of the UAV is carried out through plan B, which provides for local planning to achieve the pixel target, which is determined by calculating local differential operators in a nine–element mask. Plan B allows the UAV to fly directly to the source of the fire and determine its coordinates. Flight plan C provides for the return of the UAV to the point of departure from any pixel of the monitoring zone.Results. An algorithm for controlling the UAV flight path has been developed, which allows determining local target pixels and building a local flight plan on this basis. The basis for constructing a local flight plan is the rule "at least three pixels on the tack", which allows you to obtain a nine-element matrix with known concentrations of harmful substances in the target pixel and determine those pixels of the local flight plan in which it is not possible to obtain this matrix.Conclusion. Mathematical modeling of the UAV flight control algorithm according to the proposed test method was implemented in the Matlab 8.0 package and showed control stability and high speed of reaching the coordinates of the pixel of the ignition source, exceeding the speed of achieving the set goal by 1.5...2 times, depending on the location of the ignition source relative to the direction of the flyby of the monitored territory.