MODELING OF THE TERRITORY SURVEY BY MEANS OF THE UNMANNED AERIAL VEHICLES ON THE BASE OF THE ANT COLONY OPTIMIZATION ALGORITHM (Ukr)

Abstract

The paper considers the problem of the optimal route length determination, that allows to perform the survey of the territory in the shortest possible time, that is very important for the monitoring of the forests, rivers, transport, buildings, agricultural lands, calculation of the objects, etc. For the solution of the given problem it is suggested to use the unmanned aerial vehicles and various methods of the route optimization, among which we can distinguish probabilistic methods of the solution searching with minimal time (MTS). Namely, heuristics, cross-entropic optimization, Bayesian optimization algorithm and genetic algorithms, methods of the swarm intelligence optimization on the base of the observations over the wild life (ants colonies optimization (ACO), artificial colonies of bees, flocks of bats, etc.). It is suggested to use the algorithm of the ant colony optimization, as this enables to maintain the balance between various, namely computational parameters of the unmanned aerial vehicles and optimal length of its route. Experimental studies of the territory survey by means of the unmanned aerial vehicles at different amount of the iterations on the base of the ants colonies optimization algorithm applying modeling in WeBots and tsp-problem-ga-aco-comparisson environments have been carried out, these environments are the simulators of various devices, in particular, unmanned aerial vehicles, that compensates the impact of the external environment on the control of the flight of the unmanned aerial vehicle by means of the embedded programming tools for maintaining the current routing of the unmanned aerial vehicles. It has been established that the usage of the ants colonies optimization algorithm enables to perform the survey of the territory during less time than the genetic algorithm, which is the standard algorithm of numerous control systems by default, finding the balance between the optimality of the route and computational resources.