The Research On Fault Detection And Fault Tolerant Control In AUV Cooperative Localization

Under the improvement of effectiveness and reliability about Autonomous Underwater Vehicle(AUV),they have been widely used in scientific,military etc.Because of the limitation of volume,energy and the navigation environment,a long time,high precision positioning performance is the important technical challenge of AUV.Multiple AUV cooperative localization,which is the trending of the underwater positioning,not only can improve the positioning accuracy,but also has many advantages the single AUV do not have.In the complex underwater environment,safety and reliability are the challenges for underwater positioning system.To improve the reliability and security of the system,this paper mainly studied the cooperative localization algorithm,fault diagnosis and fault-tolerant control technology of AUV.Main work includes:1.The AUV kinematics equation is built based on the traditional two-dimensional Cartesian coordinates,the distance and azimuth observation information method for AUV for cooperative localization was adopted.Lie derivative was used to analyze the observability of system.According to the characteristics when AUV perform rotating motion,this paper proposes a new AUV cooperative localization algorithm under polar coordinates,and the system observability is analyzed by using lie derivative Considering the space characteristics of the AUV motion,kinematics equations of the AUV expanded from plane to space,and then the kinematics equation of AUV is set up under the spherical coordinates,and the cooperative localization results are analyzed.2?The drawback of traditional Multiple-Model Adaptive Estimation(MMAE)for fault detection and identification is derived,combined with six degrees of freedom nonlinear dynamic equations of the follower AUV,an Extended Multi-Model Adaptive Unscented Kalman Filter(EMMAUKF)was designed for the nonlinear fault detection and identification in the AUV cooperative localization system,where the failure of actuator or sensor information are extended in the state variables.In order to further improve the estimation precision,an Extended Multiple-Model Adaptive Cubature Kalmam Filter(EMMACKF)was designed for nonlinear fault detection and identification.3.Extended Multi-Model Adaptive Estimation algorithm cannot well deal with the problem when the system switching from one model to the other model,so a kind of the Extended Interacting Multiple Model based on CKF for nonlinear fault detection and identification algorithm is proposed.In order to improve the accuracy of fault diagnosis,this paper proposes an Extended Interacting Multiple-Model Strong Tracking Cubature Kalman Filter(EIMM-STCKF)to detect and identify the fault during AUV cooperative localization,in order to further improve the estimation precision,this paper proposes Extended Interacting Multiple-Model Multiple Fading Factors Strong Tracking Cubature Kalmam Filter(EIMM-MSTCKF)algorithm to detect the fault during AUV cooperative localization.4.The nonlinear fault tolerant control technology is studied during the AUV cooperative localization system.The principle to justifying the existence of fault-tolerant controller in the AUV system is proved,and then the error model of AUV system is deduced,according to the characteristics of the error model,we design a new nonlinear fault-tolerant controller design scheme for the system,where state equation can be divided into two parts,one with input item and the other without input item,and then according to different sub-block,its corresponding feedback gain can be designed,then feedback gain of the whole system can be obtained.The stability of closed-loop system is proved by using Lyapunov function.During the cooperative localization process,a special type of fault(route in case of mobile obstacles change issues)is studied;the geometry method is used to preplanning the path of follower AUV,it is also can be namely the fault-tolerant control design of follower AUV,the validity and feasibility of the algorithm is verified by simulation.5.In view of the limitation of test conditions,EIMM-MSTCKF algorithm is used to detect two kinds of failures(failure of contains DVL sensors and underwater acoustic modem)during AUV cooperative localization process.The redundant fault-tolerant control system is designed for the fault during AUV cooperative localization process.