Generating a Set of Reference Images for Reliable Condition Monitoring of Critical Infrastructure using Mobile Robots

Abstract

The aim of this work is to reduce the amount of computational cost when monitoring the state of critical infrastructure objects using flying mobile robots equipped with correlation-extreme navigation system, based on minimizing the number of fragments of reference images. The goal is achieved by establishing a minimum permissible degree of correlation between the individual images, which form a set of reference images. The most essential result is substantiation of the approach to formation of a set of selective images based on scene correlation analysis and sufficiency of conservation of correlation connection of images in limits 0.6 ... 0.7. This reduces the amount of computation and extends the operating time of mobile robots while maintaining accuracy. The significance of the obtained results consists in the possibility of solving a complex task of forming a set of reference images, depending on the information content and stochastic conditions of sighting of critical infrastructure objects. The solution of this task will increase efficiency of critical infrastructure objects state control due to optimization of reference images number used in the monitoring process, increase operability, and provide high control reliability in stochastic sighting conditions. The novelty of the work lies in the fact that the method of process formalized description of forming a reference images set to ensure reliable monitoring of critical infrastructure facilities using flying mobile robots for various sectors of the economy, the practical application of which will ensure reliable control and their condition assessment.