Methods of teaching control systems for unmanned aerial vehicles by immersing them in virtual reality

Abstract

The subject of study is the implementation of control systems for unmanned aerial vehicles. As their solution, a method of teaching and testing of these systems by immersing the entire system and its individual components into a virtual reality as close as possible to real conditions is proposed. The advantages and difficulties of implementation in relation to each of the system modules involved are considered. For each of the difficulties the authors propose solutions. The most successful scopes of application are revealed, and also possibility of application of the given method to land and surface vehicles is allocated. In the framework of this work, the existing aircraft control systems and the use of virtual reality within the framework of training their individual parts are investigated and the option of extending the use of such methods to the entire control system with an analysis of the advantages and disadvantages of this approach is proposed. The novelty of this article lies in the training of control systems for unmanned aerial vehicles by immersing it in virtual reality. The completeness and flexibility of such a training system is able, on the one hand, to adapt to any equipment configuration, on the other hand, to provide the highest quality training. The most important aspect is ensuring that you can use a larger proportion of the learning algorithms than is possible in other cases. In addition, this approach to the crane is useful in the framework of video navigation in connection with the possibility of better implementation of computer vision. The article shows the relevance of the research and the effectiveness of this method in the framework of aircraft control systems and proposed its application to other vehicles.