Research On Modeling Method Of Large UAV Collision Model

With the gradual development of UAV industry to information and large-scale,the air traffic flow increases significantly,which leads to the increasingly complex flight airspace environment of UAV.The structure of large-scale UAV is complex,the amount of information transmission is large,and the configuration of flight control system is demanding.How to ensure the flight safety of large-scale UAV in the limited airspace resources has become an urgent problem.Therefore,this paper establishes the collision model for the flight safety of large UAV,analyzes the collision probability and the minimum safe distance,and divides the safety level of the aircraft.Based on the classic event collision model,this paper improves the collision box and the Bay layer respectively.Considering the deflection trajectory of UAV during emergency avoidance in airspace,an improved collision box is proposed.The classic plane partition layer can not reflect the situation that two UAVs are approaching dangerously,but there is no flight collision.Based on this,the collision model of elliptical column collision template and surface spacer is established.Combined with the parameters of the model:longitudinal interval loss rate,vertical overlap probability and lateral overlap probability,the collision probability of the "two tailed scorpion" UAV is simulated in detail.Compared with the cuboid cylinder event collision model,the improved event collision model reduces the collision probability by about 33%,and effectively improves the utilization of airspace.In the free flight environment,the actual flight path and the planned flight path of UAV have flight yaw because of the CNS performance error.The yaw caused by the aircraft is equivalent to three-dimensional Brownian motion.The position,velocity and other parameters of UAVs in flight are analyzed.The collision probability between UAVs based on CNS performance is obtained by using the method of position error.It provides the necessary theoretical basis and data support for the next step of risk classification of UAV by CNS performance.According to the flight path,aircraft separation,CNS performance and flight intention,the risk classification of UAV Based on AHP is proposed.The relative weight of each influencing factor of risk classification of decision-making target UAV is obtained.The risk classification of each factor is divided and the risk score of each factor is superimposed According to the different scores,the risk level of UAV is obtained.