Research On Nonlinear Adaptive Fault-tolerant Control For A Quadrotor Unmanned Aerial Vehicle With Heterogeneous Faults

The quadrotor unmanned aerial vehicle(UAV)has the characteristics of taking off and landing vertically,hovering in a specified location and changing the flight direction easily.It can also adapt to various complex environments and perform a variety of tasks well.Thereby,quadrotor UAV has broad application prospects in both military and civilian.However,the dynamics of the quadrotor UAV is under-actuated,strong nonlinear and coupled,which brings great challenges for controller design.Moreover,as a complex system with multiple sensors and multiple actuators,it is inevitable that the quadrotor UAV will be subjected to different kinds of faults which usually occur in sensors and actuators in the actual flight.In order to ensure the stability and reliability of the control system,and complete established missions accurately and quickly,it is of great significance to study the fault tolerant control(FTC)of the quadrotor UAV.To this end,FTC will be developed for the under-actuated dynamics of quadrotor UAV.The following aspects are carried out respectively:For the attitude model of quadrotor UAV with four types of sensor faults,parameter uncertainty and external disturbance,a fuzzy adaptive FTC strategy based on dynamic surface control(DSC)is proposed.The sensor faults in Euler angle loop are transformed equivalently into a mismatched uncertainty vector.The unknown items involving fault,uncertain parameters and external disturbances in angular velocity loop are lumped into an unknown nonlinear function vector.Fuzzy logic systems are used to approximate the mismatched uncertainty and lumped nonlinearity online.Then,to control errors converge to a smaller region,DSC technique and sliding mode control(SMC)are combined to design the FTC controller.For the under-actuated model with six degrees of freedom of quadrotor UAV,sensor FTC controllers are designed.The sensor faults in displacement loop and Euler angle loop are transformed equivalently into mismatched uncertainty vectors.The unknown items involving fault,uncertain parameters and external disturbances are lumped into unknown nonlinear function vectors.Fuzzy logic systems are utilized to approximate the mismatched uncertainty and lumped unknown terms.Moreover,Nussbaum technique is used to deal with the unknown control gain.Then,FTC controllers are designed via command filtered-backstepping approach,in which sliding mode control is introduced in the last step of backstepping technique frame to establish asymptotical stability for the overall system.For the under-actuated model of quadrotor UAV with input saturation,considering uncertain parameters,external disturbance and actuator fault,a fuzzy adaptive FTC strategy based on Nussbaum technique is developed.Hyperbolic tangent function is used to approximate the saturation characteristic of actuator and transform it into affine formation on control input.Nussbaum technique is introduced to deal with the unknown control gain resulting from unknown fault,and singularity of controllers is avoided perfectly.In the system,unknown functions are approximated via fuzzy logic systems.Finally,in the design of backstepping approach,a modified sliding-mode filter and SMC are introduced to design FTC controllers for position and attitude loops respectively,and the asymptotical stability is established for the overall system.