Observer-based Actuator Fault Estimation For 3-DOF Helicopter

Unmanned aerial vehicles(UAVs)have gained much attention due to their widespread use in military and civil areas.Compared with fixed-wing UAVs,rotary-wing UAVs have the characteristics of small size,long hover time,strong vertical take-off and landing capability,flexible flight and so on.Helicopter,as a typical type of rotary-wing UAV,is widely used in many fields,such as aerial surveying and mapping,transmission line inspection,security and surveillance and so on.Because the helicopter is a highly complex system,system faults are generally hard to be avoided.In addition,as the helicopter is an unstable nonlinear system,any actuator fault or sensor fault will result in disastrous consequences.Therefore,fault estimation of helicopter system is of great significance to improve the stability and safety of the system.In this thesis,Quanser’s two-rotor three-degree-of-freedom helicopter is taken as the research object to study the observer-based fault estimation method.The main research contents and results are as follows:(1)Aiming at the problem of actuator fault estimation for 3-DOF helicopter system,three fault estimation methods based on common observers are adopted.Firstly,a linear model of 3-DOF helicopter is established.Then,Luenberger observer,Kalman filter and unknown input observer are designed to estimate the actuator faults.Finally,simulation and experimental verification are carried out on 3-DOF helicopter simulation system and actual system respectively.(2)Aiming at the nonlinear problem of 3-DOF helicopter system,fault estimation methods based on adaptive observer are proposed considering both sudden and time-varying actuator faults.Firstly,the Lipschitz nonlinear model of 3-DOF helicopter is established,and the fault estimation method based on the proportional adaptive observer is studied.Furthermore,in order to improve the dynamic performance of fault estimation of 3-DOF helicopter system,a fault estimation method based on proportional-derivative adaptive observer is proposed.Finally,the proposed fault estimation methods are validated by a 3-DOF helicopter simulation system.The results show that,compared with the proportional adaptive observer,the proportional-derivative adaptive observer has better accuracy and speed of fault estimation.(3)Aiming at the interaxis nonlinear coupling problem of a 3-DOF helicopter system under aggressive maneuvers,and considering the influence of measurement noises in the system,an unknown input observer based on linear parameter varying(LPV)model is proposed to realize actuator fault estimation.Firstly,considering the measurement noises in the actual helicopter system,an LPV model about the pitch angle of 3-DOF helicopter is established under the condition of large pitch angle.Then,an unknown input observer based on LPV model is proposed to realize actuator fault estimation of helicopter system.Finally,the performance of the proposed fault estimation method is verified by simulation and experiment.The results show that the proposed fault estimation method can effectively suppress the influence of measurement noises and realize the fault estimation of the helicopter under the condition of large pitch angle.