Research On Cooperative Target Allocation And Cooperative Trajectory Planning Algorithms For Multi-UAV In Complex Environments

The multi-UAV cooperative combat system is composed of multiple heterogeneous UAVs or UAV formations that are aware of each other and cooperate with each other.It is designed to execute multi-UAVs autonomously and cooperatively in complex combat environments.Complex task.In this process,in addition to considering the constraints of the UAV itself,it is also necessary to consider the coordination constraints between multi-UAV;At the same time,heterogeneous target allocation models and complex three-dimensional space structures lead to problems such as resource conflicts,difficulty in coordination,and repeated planning among multi-UAV.Therefore,in order to reduce the complexity of the system,this paper divides the multi-UAV cooperative combat system into two parts: multi-UAV cooperative target allocation and multi-UAV cooperative trajectory planning.The former defines a reasonable distribution relationship for the multi-UAV within the mission area,and the latter provides a practical flight path for the multi-UAV coordinated attack based on the former distribution relationship.The main contents are as follows:First,the distribution relationship between drones and target points in a complex environment is analyzed: N = M,N> M,and N <M,where N is the number of drones and M is the number of target points.Aiming at these three different distribution relationships,a unified voyage cost matrix is constructed.The objective functions of single-machine constraint conditions,cooperative constraint conditions,multi-UAV coordinated target allocation,and multi-UAV coordinated trajectory planning are established.Mathematical functions were established for the first time to replace the traditional flight-cutting method,which fully combined the planned trajectory with the complex mountainous environment.A unified allocation model was established,and the feasibility of the model in dealing with multi-UAV cooperative target allocation was preliminarily verified;This method overcomes the need for different allocation relationships using different allocation models,and effectively improves the system’s operating efficiency.Then,elaborate the difficulties of evolutionary algorithm in multi-UAV collaborative target assignment in high-dimensional complex environment.Construct an individual gene coding strategy,and combine the UAV,the target point and the corresponding voyage cost to form an individual’s gene position,so that the individual has more practical significance;Aiming at how to guide the evolution of the population according to the actual task cost,the method of dividing the population is analyzed,and the dynamic hybrid evolution strategy is used to balance the exploratory and developmental characteristics of the population,so that the algorithm can search a wider allocation space and find a reasonable allocation scheme faster;At the same time,archiving technology is used to replace individual random resets to avoid the optimal solution being overwritten as the number of iterations increases.This method solves the problem of balancing the individual’s exploratory and developmental roles in the evolution process by unifying the genetic coding method and selecting different mutation strategies according to different types of individuals,and improves the accuracy of multi-UAV cooperative target allocation.Finally,in order to solve the problems of large search space in three-dimensional environment,many turning points of trajectory and not meeting the maneuverability of UAV,a hybrid PSO key trajectory point algorithm was proposed.Map the target assignment relationship into a two-dimensional plane through coordinate transformation,set the feasible region of the trajectory points,and further reduce the search space;Introduce the coordination mechanism and information sharing mechanism between particles in the particle swarm optimization algorithm to optimize the search space and improve the search efficiency.In addition,the cubic B-spline curve and local correction method are used to smooth the trajectory,which can ensure that each trajectory conforms to the UAV’s maneuverability.This method cuts out complex three-dimensional space,improves the efficiency of selecting key trajectory points,and provides multiple safe,reliable and collision-free flight trajectories for multi-UAV.