Berezhnyi A. Methods and information technology for automated route planning of unmanned aerial vehicles to increase the efficiency of object searches

Українська версія

Thesis for the degree of Candidate of Sciences (CSc)

State registration number

0420U101294

Applicant for

Specialization

  • 05.13.06 - Інформаційні технології

18-09-2020

Specialized Academic Board

Д 73.052.04

Cherkasy State Technological University

Essay

The rapid development of unmanned aircraft in recent years has changed the organization of monitoring stationary and dynamic objects. This is due to the following: unmanned aerial vehicles (UAV) are much cheaper than manned reconnaissance aircraft; UAVs are easier to maintain and manage; UAVs can operate at night, in conditions of poor visibility, in situations that threaten the life of the pilot. The tasks of finding dynamic objects on the ground are not new. However, there are features for the forest-steppe zone. It contains significant areas of forest. The search for objects of observation in such areas is complicated. The process of selecting the optimal flight routes is particular difficulty. In this case, it is not only about completing purely civilian missions, but also about solving military tasks. Recently, the forms and methods of armed struggle are being transformed (Syria, Iraq, Libya, East Ukraine). The combat use of highly mobile, small groups, that can carry out terrorist tasks, comes to the fore. Modern terrorism is a complex socio-political phenomenon as a form of political struggle, as a form of behavior of individuals and organized groups. They, disrupting public order, destabilize the process of public life, contribute to the creation and development of conflict situations. Reliable and timely information is needed to counter the terrorist attacks of such groups. It can be quickly obtained by conducting active reconnaissance (monitoring). Planning aerial reconnaissance route using UAV in the forest-steppe area requires the identification and consideration of special patterns in dynamic objects actions. The intelligence plan itself should provide for advancing the enemy in action. A forecast of possible dynamic objects actions will give a result if information on their initial location (time, coordinates, state) is known. In this case, the results of aerial reconnaissance using UAV will provide information on the direction of movement, a possible target, the time dynamic objects leaves the target. This information will allow for the timely preparation of measures to counter possible terrorist attacks. The problem with monitoring stationary and dynamic objects is that there is a conflict between the size of the search area and the size of the objects. The solution of this contradiction is possible through the autonomous operation of the UAV, changing the route of its flight (flight task) by the results of automatic object recognition. Modern UAVs include an autopilot and a powerful on-board complex that can perform these tasks provided it is equipped with the appropriate information technologies. Thus, the relevant task of developing information technology for the automated detection and identification of stationary objects by unmanned aerial vehicles arises. The main results obtained in the dissertation are: 1. An improved method for automated UAV flight route planning for searching for a dynamic object, which, unlike the existing ones, takes into account the uncertainty in the actions of the opposite side, generates motion prediction options, estimates the vulnerability of a dynamic object on each possible route of their movement, which reduces the uncertainty of information about these routes and form a rational monitoring route for monitoring objects. 2. An improved method for automated UAV flight route planning for searching for stationary objects, which, unlike the existing ones, takes into account the connected structure of stationary objects at different topological levels and allows you to assess the importance of this type of object in the general structure. 3. The information technology of automated UAV action planning, which, unlike the existing ones, takes into account the results of object recognition, has been further developed, which allows real-time change of the flight route when performing an UAV mission. The results can be used directly to plan UAV flight routes to search for dynamic and stationary objects in advanced decision support systems and automated aviation control systems. Keywords: decision support system, dynamic and stationary objects, information technology, routing, unmanned aerial vehicle.

Files

Similar theses