The dissertation is devoted to the development of methods and software for group
flight control of unmanned aerial vehicles (UAVs). It is important to provide a group of
autonomous vehicles with a certain geometric organization, where they move as a single
whole. This approach is used to perform a large number of practical tasks. For example,
aircraft-type UAVs have high speed and maneuverability, which is useful for tasks where
duration and range are important.
To ensure such control, it is necessary to introduce methods that allow the apparatus
to act independently of each other, avoiding centralized control. This approach is often
compared to group management, where each apparatus makes its own decisions. Group
control in the context of unmanned aircraft is considered as the organization of a group
of vehicles in order to perform complex tasks.
The method of vector flight fields along a given trajectory is one of the ways to
achieve such group control, where the vehicles form and maintain the specified geometric
structures for joint performance of tasks.
The method of controlling a group of autonomous aircraft is based on a decentralized
consensus architecture and the use of a heterogeneous vector field for passing a straight
route. This approach is aimed at creating control algorithms that allow aircraft to maintain
a given position in the group while moving along a straight horizontal route. It is based
on the principles of consensus and the use of vector fields for the passage of the route,
which provides flexibility in choosing the desired form of the group, taking into account
the complex dynamics of the UAV.
The decentralized consensus architecture allows aircraft to coordinate their position
through the exchange of information, facilitating accurate tracking of a given trajectory and maintaining the relative positions of aircraft in a group. This method is essential for
use in areas where the joint functioning of unmanned aircraft in a group is key: monitoring
the earth's surface, search and rescue, and performing military missions. These methods
guarantee an asymptotic approximation of the relative positions in the group to the
specified ones, as well as an approximation of the speed of each aircraft to the average
cruise speed.
On the basis of the proposed methods, the algorithms of group control for the system
of unmanned aircraft using a simulation mathematical model were investigated and
evaluated.
The problem of group control of autonomous objects in real conditions is quite
relevant today. This is due to the complexity of control, dynamic changes in situational
parameters, restrictions on input control signals in real "autopilot-UAV" systems. A lot
of research and publications of the following foreign and Ukrainian scientists are devoted
to the issues of group control of unmanned aerial vehicles: A. Piccard, C. Ryan, C.
Peebles, G. Collins, A. Erickson, N. Baldock, M. R. Mokhtarzadeh-Dehghan, L. N. Craig,
R. Olfati-Saber, R.W. Beard, W. Ren, T.W. McLain, H. Yamaguchi, as well as L.
Artyushin, O. Kononov, O. Mashkov, D. Kucherov, T. Sheveleva, P. Pavlenko, D.
Bondarev, V. Golembo, A. Bochkarev, O. Martynyuk, V. Gerasimenko, O. Barabash and
others.
According to the generally accepted definition, group control means obtaining a
predetermined geometric shape by a group of autonomous dynamic objects. In the process
of further completion of the task, the group must maintain this form by acting as a rigid
body. Groups of UAVs are used in a large number of practical tasks. Therefore, the
problems of group control of UAVs have recently received great attention from
researchers around the world. The results of the research testify to the success of the
proposed management methods. The algorithms developed on their basis have
demonstrated the ability to maintain a stable group and accurately follow the given
trajectories of aircraft in various scenarios.
On the basis of the analysis, the existence of a contradiction in practice has been
revealed: between the requirements for increasing the efficiency of group control of unmanned aircraft, which requires an increase in the cost of developing mathematical and
software, and the requirements for reducing the cost of performing various tasks and
missions by unmanned aerial vehicles; and contradictions in theory: the need to maintain
the configuration of a group of UAVs at the stage of group flight under the influence of
external and internal destabilizing factors and the limited capabilities of existing methods
to ensure the stability of a group of autonomous UAVs under decentralized control.
