Algorithm for Spatial Control of Swarm of Unmanned Aerial Vehicles with Leader Based on Synergistic Approach

Abstract

Formulation of the problem in general. It is noted that it is advisable to use unmanned aerial vehicles as part of swarms (groups) to perform tasks in a complex environment with obstacles in an automatic or remote-controlled mode, which requires algorithmisation of the process of building routes. It is shown that the problematic aspects of UAV swarm control are their coordination to perform a single task, ensuring the reliability and stability of communication, and avoiding complex static and dynamic obstacles along the flight route.
    
Analysis of recent researches and publications. The paper analyses models and algorithms for building routes, in particular those that use a synergistic approach based on the known location of all other swarm elements. However, this approach is considered only theoretically and with a typical example for movement in a two-dimensional plane.
    
Presenting the main material. An improved control algorithm based on a synergistic approach for three-dimensional space in the presence of a leader is proposed, which reduces the computational complexity and calculation time. An expression is presented that formalises the mathematical model for determining the acceleration of an unmanned aerial vehicle from a group (swarm) consisting of N elements. It is shown that in order to implement an improved algorithm for route construction and group control, the location of the driven elements of the group should be determined relative to the route positions of its leader. The implementation of the algorithm has been tested by mathematical modelling in the Ardupilot SITL environment and by conducting a full-scale (flight) experiment using a group (swarm) of three UAVs of the copter type. 
    
Conclusion. The proposed algorithm for spatial control of a UAV swarm with a leader based on a synergistic approach allows keeping the group elements in space within the target optimal distance between them. The simulation and practical experiment confirmed the ability of the algorithm to provide a linear process of increasing the number of iterations to reach the critical distance, unlike the prototype, for which this indicator increased quadratically.
    
The perspectives of future researches. Further research should include the development of a methodology for constructing routes for collision-free movement of groups (swarms) of UAVs based on the proposed algorithm.