Fault-tolerant Control Algorithm And Actuator Fault Detection Research For A Quadrotor

A quadrotor is simple in structure and has a great mobility,it can vertically take off and land in different space environment.Since it has so many advantages,its application in military and civilian fields is of great research significance.In this article,we treat a quadrotor as the research object,we firstly introduce how to establish the attitude control system model and the fault model of a quadrotor.Based on the principle of model reference adaptive control,a fault-tolerant controller has been designed.Finally we discuss a quadrotor fault-tolerant control problem in a specific situation,and put forward a design of rotative velocity measurement.The main content of this article is as follows:Firstly,in order to deduce the attitude control system model for a quadrotor,we define the coordinate system and body coordinate system,and based on the rigid body dynamics equation,the attitude control system model is simplified expressed.Then we discuss different faults for a quadrotor and deduce its actuator fault model.Secondly,this article introduces a quadrotor fault-tolerant control method and the adaptive control theory,and put forward a design of model reference adaptive fault-tolerant controller.The analysis which is based on the lyapunov stability criterion shows that the controller can guarantee attitude parameters of the quadrotor tracking the attitude parameters of the reference system on the condition of unknown external disturbances and actuator fault,and we make a simulate experiment on MATLAB,the results show that the controller has a good asymptotic stability.Finally,in order to determine the type and occurrence time of a fault and to provide accurate fault information before the design of a fault-tolerant controller,we discuss how to implement the real-time detection for the rotative velocity of the motors.This article designs a kind of rotative velocity measurement system based on photoelectric sensor,and completes the software and hardware implementation work,the experimental results show that the system has a high detection precision and a good real-time performance.