Improving UAVs control systems reliability for environmental monitoring by applying a common diversity metric to modify the agreed by the majority vote algorithm

Abstract

The article considers the problem of control systems for unmanned aerial vehicles reliability increasing, which are used for environmental monitoring and agricultural needs. Software systems reliability improving can be achieved through the use of N-version software, which often uses an algorithm for voting by an agreed majority. However, this algorithm can select the presumably correct answer of a entire module randomly. This happens when several groups of versions of the same amount give different answers. To solve this problem, the article proposes to modify the voting algorithm by an agreed majority by applying a common diversity metric. The paper describes the common diversity metric, as well as a couple of particular metrics included in it (the programming languages diversity metric and the algorithm diversity metric). The method of modifying the voting algorithm by an agreed majority is also given. In the article conclusion the results were summed up, and a conclusion was made about the expediency of using a common diversity metric to modify the vote algorithm by an agreed majority in order to increase the reliability of unmanned aerial vehicles for environmental monitoring and agricultural purposes.