Research On The Algorithm Of Drones Formation Control

With the rapid progress of the working methods of the UAV,the complexity of the UAV tasks has increased,and the workload and scope of the work have also increased.For large-scale tasks such as patrolling pastures,forest patrols and post-disaster search,A single UAV is not enough to complete the task.The UAV formation can effectively improve the work efficiency.The control content of the UAV formation which completes the task in cooperation mainly includes two aspects,one is how to save the total energy of the UAV formation,and the other is how to correctly complete the transformation of the specific formation of the UAV formation.The former saves energy for UAV formations to deal with unexpected events in the mission.The latter improves the efficiency of UAV formations to complete tasks and guarantees the safety of the formations.Therefore,the rapid and effective takeoff assembly of the UAV formations and the reconstruction of the UAV formations have become the focus of the UAV formation researchs.For the,this paper mainly studies the key technologies for the rapid take-off assembly of UAV formation and the solution speed of UAV reconstruction,gives a scheme for rapid take-off assembly of UAV formation based on circular trajectory tracking,a UAV formation reconstruction model based on distributed model prediction(DMPC)and two improved optimization algorithms for quickly solving the UAV formation reconstruction model.The main contributions of this paper are as follows:1.The autonomous take-off and assembly of UAV formation is the prerequisite for the UAV formation to perform the task.The traditional UAV takes off alone and chases to form an assembly cluster,which is undoubtedly a waste of energy for the entire UAV formation.Aiming at the problem of the long assembly time caused by the take-off of UAVS,this paper proposes a fast assembly take-off scheme for UAV formation based on circular trajectory tracking.This take-off and assembly solution reduces the time required for take-off and assembly of drones,the distance between drone formations,the drone formation communication consumption and provides a new way for fixed-wing drones to take off.2.In order to quickly and accurately solve the UAV formation control instructions required by UAV formation reconstruction,this paper proposes a UAV formation reconstruction method based on Distributed Model Predictive Control.The proposed Pre-DMPC-DE algorithm effectively implements the UAV formation reconfiguration function.At the same time,compared with other optimization algorithms,it has fast speed and high iteration efficiency and meets the requirements of UAV formation system real-time calculation.3.In order to further improve the speed and accuracy of the UAV formation reconstruction problem,this paper proposes a method for solving the UAV formation reconstruction based on the DMPC-QUATRE framework.This method uses a differential evolution algorithm(QUATRE)with reduced parameters to solve the UAV formation reconstruction model.At the same time,the initial population of the algorithm is optimized by the population operation to inject a high-quality control gene from the previous moment.The proposed Pre-DMPC-DE algorithm effectively realizes the reconstruction function of the drone formation.At the same time,compared with the improved Pre-DMPC-DE algorithm,it can achieve the faster iteration speed while ensuring the formation convergence of the drone formation fast.