Research On Fault Diagnosis And Fault-tolerant Control Technology For UAVs Based On Observer

Since 1990 s,UAV technology is changing rapidly.In the national defense field and civil field,UAV is playing a more and more important role.Facing the more arduous task and the more complex information environment,single unmanned aerial vehicle(UAV)has not been able to meet the task demand.The United States military also divided the fully autonomous UAVs as the highest level in the division of the autonomous control level of the UAV.Research on UAVs is becoming a hot spot in the world.With the increasing number of UAVs in cluster,the possibility of fault occurrence and the difficulty of fault diagnosis is greater,and the impact of accidents is also more serious.Therefore,it is a challenging and meaningful task to design an effective fault diagnosis and fault tolerant control method for UAVs.In this paper,based on the observer method,the fault diagnosis and fault-tolerant control technology of UAV group under actuator failure is studied,and the engineering application value of the method is verified on the Quanser physical verification platform.First,the research background and significance of the research are introduced,and the research status of the fault diagnosis and fault tolerant control of UAVs is summarized.Then the inner-loop and the outer-loop model is established by analyzing the position relationship between the leader and the follower and the UAV's nonlinear model.Then the stability of UAVs is proved through a combination of the inner-loop and the outer-loop model.Secondly,according to the inner-loop UAV nonlinear model,the disturbance and actuator failure is considered.Under the condition of satisfying the reasonable assumption,the original system is separated into two subsystems by linear transformation.Then the reduced dimensional sliding mode observer is designed for fault diagnosis.Then based on the on-line obtained fault estimat,the compensation control method is used to carry out the fault tolerant control.Finally,the stability and accessibility of the reduced dimensional sliding mode observer are proved by the Lyapunov stability theorem,and the effectiveness of the method is verified by simulation.Thirdly,In order to fit the real situation,the relative output estimation error is introduced in UAVs.In this paper,the directed graph is used to describe the communication topology and the error conduction process,and a sliding mode observer is designed for a single UAV system.At the same time,in order to overcome the limitation of the reduced dimensional sliding mode observer to the actuator failure,the adaptive law is added to further improve the observer.Due to the existence of the relative output estimation error,considering the stability of UAVs through a single UAV is relativelyineffective.Therefore,by using Kronecker product,the stability and accessibility of the adaptive sliding mode observer are proved globally.The parameters of the observer are calculated by LMI.Finally,in order to verify the engineering applicability of the designed method,the experimental verification was carried out on the Quanser physical verification platform.Based on the existing experimental equipment conditions of the project group,the ground air joint formation of two ground robot Qbot2 and an air vehicle Qball-X4 is tested,and the single actuator fault and two actuator failure are considered.The experimental results show that the proposed adaptive sliding mode observer has good fault estimation performance,and the UAVs can keep the formation stability in the case of failure.