Multi-sensor Based Autonomous Landing Guidance And Control System Of A Fixed-wing Unmanned Aerial Vehicle

In recent years,with the unmanned combat system(UCS)in Iraq,Afghanistan and several other local wars of widespread use,unmanned combat system is showing the great military value.As an important component of unmanned combat system,unmanned aerial vehicles(UAVs)are increasingly assigned with reconnaissance,surveillance,communication relay,battlefield evaluation and attack guidance.In order to expand the operating radius of UAV,improve combat effectiveness,the demand of autonomous landing on a destroyer or carrier is continuing to grow.Although UAVs take off and landing on the standard airport is quite common,navigation in the above maneuvers relies mainly on the Global Navigation Satellite System(GNSS).In the future A2 / AD battlefield environment,the satellite navigation system alone can not meet the UAVs' landing demand.This paper aims at the above research background,the main work and innovation are as follows:(1)A ground / ship-based multi-sensor UAV recovery and guidance system is proposed.The system consists of two independent guidance units,each with a two degrees of freedom PTU(Pan/Tilt Unit)and visible light camera,infrared camera or other sensors.The configuration of the two independent guidance units can be optimally setup according to the size of the UAV and the proposed detection distance.In this thesis,the feasibility of using the above two guidance systems in the process of UAV landing is proved by the theory of target position solution,error analysis and experimental verification,which is based on the characteristics of the system distributed on both sides of the runway independently.(2)Design and implement the real-time target tracking and location algorithm in UAV landing process.Aiming at the problem of tracking the UAV,size rapidly changing from tiny to large scale,the method of image preprocessing based on morphological filtering,TLD target tracking framework and target position updating based on active contour are improved.During the landing process,the Motion position and unmanned aerial vehicle motion estimation,can accurately calculate the position of UAV relative to the ship's position.(3)Design and implementation of the non-linear model predictive control(NMPC)and total energy control(TESC)of the UAV landing control system.Due to the computation constraints of UAV on-board equipment,the inner-loop controller and the outer-loop controller are designed to realize the autonomous landing.The inner loop controller is mainly composed of PI controller and PD controller.The outer loop controller mainly consists of nonlinear model predictive controller(NMPC)and total energy controller(TESC)in order to tracking the landing curve which generated by Dubins Path algorithms.(4)Design and implement the simulation system and experimental verification system for UAV shipboard landing problem.Based on the Robot Operation System(ROS)and Gazebo simulation environment,a software in the loop simulation system(SITL)and hardware in the loop simulation system(HIL)is constructed.The simulation environment can meet the verification requirements of the above algorithms.By selecting PTUs and related sensors,autonomous guidance and landing of the UAV can be achieved at ground and airfields.Open source code please refer https://github.com/weiweikong/rotors_simulator.