METHOD OF JOINT ROUTE PLANNING FOR A GROUP OF UNMANNED AERIAL VEHICLES

In the paper, an analysis is presented of existing methods for enhancing the efficiency of the joint route planning process for a group of UAVs using well-known scientific research techniques—namely systems analysis, object search theory methods, and mathematical modeling and optimization techniques. It is determined that the Differential Evolution Algorithm (DEA) operates stably, exhibits a high convergence rate and normal parallelism, and effectively solves global optimization problems with continuous variables. However, joint route planning for a group of UAVs involves certain specific tasks that are difficult to address solely using DEA methods, as investigated in the paper. In the article, it is determined that joint route planning for a group of unmanned aerial vehicles is a complex scientific task that involves taking into account a number of constraints: the variable dynamics of UAV movement and the target point, the maximum flight path length, the minimum/maximum UAV speed, the maximum flight time, the maximum UAV payload, the temporal sequence between key mission route points, and the waiting time. The main ideas and distinctions of the improved method presented in the article are: the implementation of a procedure for avoiding local optima, the formation of a three-dimensional space of the admissible region, and the method’s adaptability with respect to the mutation strategies of the genetic algorithm. This method eliminates the limitations associated with balancing exploratory and exploitative functions, increases the potential coverage of the search domain, and enhances the efficiency of locating key route points. A comparison was conducted between the improved joint route planning method for a group of UAVs and well-known methods such as Multi-Strategy Fusion Differential Evolution, Multi-Objective Genetic Algorithm, and the Full-Scale Adaptive Differential Evolution Algorithm. On average, the improved method reduces the route length by 7–9% while increasing computational speed fourfold through its adaptive mechanism, which enables its use onboard UAVs during coordinated group operations.