Research On Fast Formation And Anti-collision Of UAV Swarm Assisted By Wireless Ultraviolet

The cooperative work of Unmanned Aerial Vehicles(UAV)swarm is based on reliable communication and real-time information sharing among UAVs.Only a stable formation communication network can guarantee the ability of swarm to perform tasks.In the process of UAV swarm performing tasks,formation gathering with high speed and low energy consumption can provide effective guarantee for UAV to perform the follow-up tasks.Threat perception and collision avoidance are also the core issues running through the whole operation process and ensuring the safety of UAVs.In this paper,Ultraviolet Communications(UVC)has good local security,strong anti-interference ability and low background noise,which is used as an auxiliary communication mode within the UAV swarm.The problem of UAV assembly and anti-collision is studied when the radio communication of UAV swarm is limited.The main research contents are as follows:(1)Aiming at the demand that UAV swarm need to assemble into a specified formation in a specified area,a control strategy of partition assembly is proposed.The area of near assembly point is divided into virtual independent partitions which do not affect each other,and the assembly information inside the swarm is obtained by using UVC,including the initial distribution position,assembly area and predetermined formation pattern of UAVs.Based on this constraint,each UAV is assigned a target assembly point that minimizes the total cruding range,and the assembly sequence of UAVs is planned according to certain rules.The simulation results show that when the total number of UAVs is constant,with the increase of partition number,the assembly time of the whole swarm decreases in a ladder shape,and the predicted collision probability decreases gradually.The safety of assembly can be effectively improved without additional collision avoidance measures.In addition,according to the law of experimental results,a method of selecting the best partition number according to different requirements is proposed.(2)According to the characteristics of UVC,a spherical ultraviolet MIMO(Multi-Input Multi-Output)model is designed.The ultraviolet transceiver is used to form an ultraviolet communication ad hoc network inside the swarm,and the state information of UAV is obtained and shared to provide environmental information for the improved artificial potential field method.By improving repulsion potential field,adding memory resultant force and additional force in vertical direction,the problems of target point unreachable,path oscillation and falling into local minimum easily occurred in traditional artificial potential field method are solved.On this basis,the judgment of threat source direction is further improved based on ultraviolet MIMO structure,and the number of repulsion source directions is simplified.In addition,this paper designs an autonomous development knowledge base for anti-collision,which is used to store the mapping of "threat pattern-calculation result of repulsion force" so that it can be called directly when it reappears.Experimental results show that the improved artificial potential field method based on wireless ultraviolet MIMO not only improves the tolerance of direction error of threat source,but also reduces the predicted collision times by 22.6%.When encountering the same threat pattern again,it can reduce the calculation time of collision avoidance path by 4.992%,and improve the real-time reaction speed of collision avoidance.In this paper,UVC is applied to UAV swarm control,which provides a new method for information transmission and relative position determination between UAVs under the condition of limited use of radio communication.The proposed UAV partition assembly control strategy and the improved artificial potential field method based on ultraviolet MIMO also provide new ideas for UAV assembly and threat perception collision avoidance research.