Research On Active Fault Detection Method Of Unmanned Surface Vessel

With the continuous development of science and technology,the performance requirements of surface unmanned vehicles are increasing day by day,and there are higher standards for the reliability and accuracy of various systems.Therefore,in order to ensure the reliable operation of the system,fault detection has become a vital link.However,small faults will occur in the operation of unmanned vehicles.because of the hidden nature of small faults,the traditional fault detection methods are difficult to detect quickly.Over time,minor failures can turn into significant failures,which have a significant impact on the system.In order to detect minor faults of unmanned vehicles in time,this paper focuses on the active fault detection method of unmanned vehicles which is based on set membership estimation theory,interval matrix theory and mixed integer quadratic programming theory.The main research contents are as follows:(1)An active fault detection method for surface unmanned vehicles system with minor faults.Firstly,the set membership form of the unmanned vehicles control response model is established based on the set membership estimation of zonotope.Secondly.Through the mixed integer quadratic programming problem,the auxiliary signal is designed,the goal of the planning is to separate the output set in different scenarios,and optimize the output set,and design simulation experiments to prove the effectiveness of the method.(2)A new active fault detection method based on constrained multi-cell set membership estimation is studied.First of all,the set membership form of the mathematical model of unmanned vehicles motion based on constrained zonotopic estimation is established.The state of the system is estimated by designing an observer and the observer gain is calculated by taking the size of the zonotope as the optimization index.State is represented by constrained multicellular bodies.Finally,the mixed integer quadratic programming is used to solve the auxiliary input signal,and simulation experiments are designed to prove the effectiveness of the method.(3)An active fault detection method combined with interval matrix is studied for unmanned vehicles systems with parameter uncertainties.Firstly,the concept of interval matrix is introduced,the set membership form of surface unmanned vehicles system model based on zonotopic set membership estimation is established,and the interval matrix is used to express the parameter uncertainty of the system.Secondly,the observer gain is calculated by combining the observer and taking the size of the zonotope as the optimization objective.Finally,based on the output of the system observer,the system uses mixed integer quadratic programming to solve the optimal auxiliary input signal and designs simulation experiments to prove the effectiveness of the method.