Distributed Optical Fiber Sensing Perimeter Security System Based On UAV Video Linkage System

The dual Mach-Zehnder interferometric fiber-optic disturbance warning and positioning system has been widely used in the field of perimeter security because of its high sensitivity,simple structure,and long-distance positioning in complex environments.Because a single perimeter security system will not be able to visualize the intrusion event like a camera,the security of long-distance oil and gas pipelines and complex terrain still has a big test.In order to solve the above problems,based on the dual Mach-Zehnder interference type optical fiber disturbance early warning and positioning system,this paper proposes a video linkage system based on the four-rotor drone dual Mach-Zehnder interference type optical fiber disturbance early warning and positioning system.This system uses distributed optical fiber disturbance sensing technology to locate the intrusion event and transmit the positioning information to the quadrotor drone.The drone surveys the intrusion location,records the intrusion video,and transmits the video to the ground station.At the same time,it is equipped with an ODROID-XU4 board to track the intruded target in real time and dynamically,and to return to land autonomously.This paper mainly carries out the following innovative work:1.According to the positioning principle of the dual Mach-Zehnder distributed optical fiber disturbance early warning and positioning system,based on the modular design idea,an instrumented platform is built to realize the monitoring and positioning function of the intrusion event.And according to the realization of the video linkage function with the disturbance warning and positioning system,the four-rotor UAV is introduced to perform the sensor module.The hardware platform of the four-rotor UAV was designed and built.2.According to the software framework and hardware interface definition of the pixhawk flight controller,the dynamic model of the quadrotor UAV is established.The position analysis of the quadrotor drone was analyzed in depth,and the pose and control algorithms were completed.A four-rotor unmanned aerial vehicle(UAV)hybrid controller model was established and converted into a motor speed to control the drone’s flight.At the same time,a ground station is set up,and the communication between the quadrotor UAV and the ground station is realized using the MAVLink protocol.3.A dynamic target tracking algorithm based on cascade detection classifier detection and Kalman filtering is proposed.The cascade detection classifier is used to detect intruders,and Kalman filters are used to track targets to improve the tracking effect of intruders.At the same time,the ODROID-XU4 board is equipped with a ROS operating framework to port Open CV and complete the dynamic tracking of intrusion targets.4.Video linkage experiment on four-rotor UAV and disturbance warning and positioning system to verify program feasibility.At the same time,data analysis is performed on the data of the four-rotor flight.Such as analysis of GPS search star positioning capabilities,control curve of horizontal direction speed control and roll angle and pitch angle under PID control to verify of flight stability and control accuracy of quadrotor UAV.The validation of the tracking effect of the intruder shows that the algorithm can track the intruder and has a certain robustness.This paper designs a visual landmark recognition scheme to achieve precise positioning of landing height and landing position.