详细
The problem of providing an emergency landing of an unmanned aerial vehicle (UAV) helicopter type flying in a certain area of the target application is considered. The two-stage algorithm for finding an unprepared landing pad taking into account a set of requirements is proposed. At the first stage, using a digital elevation map placed on an UAV helicopter type board a route for overflying unprepared landing pads in terms of surface topography is calculated. A route formation is achieved by sequentially solving static optimization problems in order to minimize the average losses that occur when an UAV helicopter type flying from one an unprepared landing pad to another. At the second stage, which is implemented directly during an UAV’s helicopter type movement along the calculated route, the final choose of an unprepared landing pad is made based on the processing of ground penetrating radar data. A neural network classifier based on a two-layer perceptron is used to assess the suitability of an unprepared landing pad to the soil density requirement. An example that illustrates the operation of the proposed algorithm both under the conditions of a computational experiment and during a series of flight experiments is considered.