Application Research Of Tethered Six-rotor Aircraft Control System Based On Adaptive Neural Network

In recent years,with the large-scale rise of MEMS inertial sensors and the rapid development of aviation technology,multi-rotor aircraft with simple structure and strong mobility have been gradually applied to various fields.At present,most of the mainstream multi-rotor aircraft are powered by lithium batteries.However,lithium batteries have limited capacity and long charging time.These problems have been restricting the endurance and load of the multi-rotor aircraft,so the researchers increasingly incline to the mooring aerostats.Although it can work for a long time in the air,it has some shortcomings,such as large volume,poor maneuverability and so on.It can't fly according to the planned route.Therefore,this paper combines the characteristics of multi-rotor and moored the mooring aerostats,developing a set of tethered six-rotor flight control system with stable performance and prefect functions.Compared with conventional six-rotor aircraft,the tethered six-rotor can fly uninterruptedly through the power supply of the tethered cable.The attitude of multi-rotor aircraft is greatly influenced by cables in the air.The cable is in the towing aircraft state in the process of following the body movement.The inertia of the cable itself will interfere with the steady flight of the multi-rotor aircraft,which requires a higher attitude calculation and control algorithm for the moored multi-rotor aircraft.In this paper,a neural network adaptive control algorithm is designed and simulated.The main researches content are as follows:1.The design of the hardware system platform.The classical open source flight control system at domestic and international was analyzed.The hardware system is designed combine with the application requirements of the tethered six-rotor.According to the specific requirements of the functional modules,each module is designed and debugged to complete the design of the hardware system platform.2.The design of tethered six-rotor control system.The RBF neural network is integrated into the classical PID control to realize the adaptive adjustment of control parameters.The simulation analysis and comparison of PID,cascade PID and adaptive controller are carried out by using MATLAB.The experiments show that the neural network adaptive control has better effect.3.The design of Attitude Reference System.Firstly,all sensor data are corrected,and sliding filter is used to filter the data of the MEMS sensor.Then complementary filtering algorithm is designed to realize attitude calculation.The experiment shows that the algorithm has good performance.It guarantees the stable flight of tethered six-rotor aircraft.4.The design of the software system based on ARM-Linux.ARM-Linux development environment is build.Control system is divided and designed according to function and task.It includes transplanting Bootloader on the master control module,clipping and configuring Linux kernel,creating root file system and so on.5.The design of the power supply system for the six-rotor aircraft.The ground 220 V AC is used as input,and 400 V DC is output through DC voltage regulator module,transporting to lift-off platform by mooring cable.It provides uninterrupted power for mooring six rotors aircraft after depressurization.Finally,the aircraft conducts field test flight.