Research On Foreign Object Debris Detection In Airport Flight Area By UAV

Foreign object debris in the airport flight area have always been an important hidden danger in the safe operation of flights.This paper combines UAV and airport FOD detection,through intelligent image recognition technology,can quickly and effectively inspect the airport target area.The whole operation process mainly includes several aspects of UAV inspection path planning,intelligent vision system and UAV trajectory tracking and monitoring system.Finally,a software and hardware integrated system is designed to detect and arrange FOD removal procedures in time.In order to realize the full coverage path scanning for the inspection surface,UAV is used to inspect the airfield area.Firstly,the coordinates are transformed by Gauss projection.To deal with the target area in two-dimensional plane,firstly grid the target area of the airport,express the patrol area and obstacles with 0-1 matrix.Finally,the full coverage path planning algorithm is used to plan the UAV track in each area.In order to enable the UAV to complete the whole cruise inspection task safely,autonomously and accurately in the airport flight area,the UAV flight control system is adjusted for the motor,maintain the UAV’s flight attitude,and carry high-precision GPS system and ultrasonic height sensor.The positioning information transmitted by ADS-B transmitter loaded by UAV and GPS device carried by UAV is received.The multi-mode interactive algorithm is used to describe the motion model,and the asynchronous track is extrapolated.Finally,the convex combination algorithm is used to fuse to realize the accurate track tracking of UAV.In the process of autonomous inspection of UAV,the inspection area is photographed and transmitted back to the base station through the equipped PTZ camera.Firstly,the received video stream is extracted from the image frame to be detected,and the image is spliced with the front and back frames on the time axis.Through the digital image preprocessing technology,the image is de-noising and multi-scale Retinex algorithm is used to enhance the image.Through the fusion of pixel separation and line detection,the marker lines in the image are separated,and the image is cut to reduce the background complexity.Referring to the classical ITTI model,we extract the brightness and color of the sample set.Canny operator is used to extract the edge features.Monte Carlo method and fuzzy mathematics are used to calculate the basic probability of each feature to detect the foreign object,and finally,we use D-S evidence fusion theory to fuse the features according to the size of the support degree to detect the significance of the processed image.Finally,an application software is developed based on Android studio to control the UAV to complete the patrol task independently.The software has the functions of UAV positioning correction,path planning,speed and altitude control.At the same time,Java is used to develop a UAV monitoring platform based on World Wind,which can be processed and approved in the background after receiving data.