Non-cooperative unmanned aerial vehicle track distribution prediction method based on flight state division

Abstract

The invention discloses a non-cooperative unmanned aerial vehicle track distribution prediction method based on flight state division, and the method comprises the following steps: constructing a rasterized airspace for the range of a selected monitoring airspace; dividing the flight state of the unmanned aerial vehicle; screening a similar trajectory data set, and constructing a trajectory prediction model based on data migration; generating a track reachable space; the index track covers the grid coordinates; a trajectory probability distribution is generated. According to the method, the flight state of the unmanned aerial vehicle is subdivided, a trajectory prediction method based on data migration is combined, the motion considering the uncertainty of the intention of an unmanned aerial vehicle operator is modeled as Brownian motion, and the position distribution of the non-cooperative unmanned aerial vehicle is modeled by adopting a truncated Brownian bridge method, so that more accurate trajectory distribution prediction is realized; and a non-zero probability track space range is reduced.

Description

technical field [0001] The invention belongs to the technical field of unmanned aerial vehicle traffic management, and in particular relates to a non-cooperative unmanned aerial vehicle trajectory distribution prediction method based on flight state division. Background technique [0002] With the rapid popularization of drones in various industries, the rapid development of low-altitude drones and the increasing demand for flying, air traffic activities are becoming more and more frequent. The research on UAV trajectory prediction technology can promote the safe, orderly and efficient operation of UAVs in limited low-altitude airspace. Calculating the trajectory distribution of UAVs at a certain moment in the airspace is the basis for building a dangerous behavior recognition model. . [0003] Most of the existing trajectory prediction methods obtain the estimated point of the future position of the UAV, not the position interval where the UAV is located. The error exists...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the above-mentioned prior art, the object of the present invention is to provide a non-cooperative unmanned aerial vehicle trajectory distribution prediction method based on flight state division, to solve the problem that the trajectory prediction method in the prior art is ineffective when facing non-cooperative targets. The intention of the man-machine operator is difficult to predict, and the prediction accuracy of future trajectory points is not high, which leads to the problems of missing or false alarms; the method of the present invention divides the initial flight state of the UAV, and combines the probability distribution prediction to replace the future trajectory point prediction, and generates Trajectory distribution space, reducing the occurrence of missed alarms and false alarms, and at the same time narrowing the range of trajectory space with non-zero probability to achieve more accurate trajectory distribution prediction

Images