Research On UAVs Cooperative Control Technology Based On Multi-target Recognition

As one of the main aircrafts in the future battlefield,Unmanned aerial vehicle(UAV)has the function of all-weather and multi-angle detection attack.With the development of science,technology and industry technology,the theory and technology of intelligent and autonomous UAV are constantly proposed and updated.Especially in the future battlefield,it is almost impossible for the UAV to carry out the combat task alone.The large-scale UAV group will be widely used in the military field,as well as in the civil field.So it is very important to study the cooperative control of UAV in the cluster.How to keep the UAV in a certain attitude and how to reach the target in the flight process has practical and theoretical significance.In order to achieve cluster cooperative control under multi-target recognition,a method based on multi-target detection and recognition is proposed in this paper.The target relative parameters are given to the cluster.Multi-target detection and recognition is based on how to detect and recognize multi-target by vision or other means.For example,in the military field,UAV group can identify ground armor,artillery and vehicle.The target position and distance result after identification are assigned to UAV group as tasks.Then how to plan the path for the UAV group to perform the task.Around the above problems,the research work of this paper is as follows:(1)Multi-target detection and recognition are studied.In this paper,a simplified multi-target recognition method is proposed.In the radar detection,the UAV in all aircraft groups can detect the distance of the target.In this way,the result is easy to get,and the result is reliable.At the same time,based on the visual image processing,using the spot detection and surf detection technology,the target recognition is completed,and the coordinates of the target are determined,which provides convenience for task allocation.(2)The task allocation of UAV in cluster is studied.According to the traditional task allocation,it can be divided into centralized task allocation and distributed task allocation.In addition,the constraint conditions for solving the assignment model are specified.At the same time,using the above model,task allocation,using the multi business travel method,Hungarian algorithm and contract task method to solve.The model of task assignment is obtained.(3)The paper studied path planning.From A* algorithm,PRM algorithm and RRT algorithm for path planning,and put forward based on embedded system using MCU to complete obstacle avoidance.In particular,it focuses on the flight path of multiple UAVs,and then compares the performance,summarizes how to use the algorithm in different situations.In this paper,the integration of multi-target detection and recognition,task allocation and path planning,so that the fleet can achieve independent planning,cooperative control to the task point after obtaining the information of multi-target.Thus,the two kinds of problems are integrated to solve the problem of cluster cooperative control under multi-target recognition.