Research On Fault-Tolerant Control Method Of Tilt-Rotor UAV

Multi-rotor UAVs have a wide range of applications in the fields of inspection,aerial photography,security,agriculture,forestry and plant protection due to their simple structure and control algorithm,easy modularization,and low operation and maintenance costs.With the rapid development of related industries,people have put forward higher requirements for the stability,safety and reliability of UAVs.When performing a flight mission,due to the complex and changeable external environment,the actuators and sensors of the UAV have the potential to encounter failures.Once the failure occurs,it will further affect the performance of the UAV system.Therefore,it is of great significance to study the related topics of multi-rotor UAV fault diagnosis and fault-tolerant control.Considering that the three attitude angles cannot be stable and controllable under the failure of the traditional six-rotor UAV,this paper takes the UAV under the tilt-rotor structure as the research object,and applies the interactive multi-model algorithm to the failure problem of a single actuator.The diagnosis module uses the strong tracking Kalman filter to perform parallel filtering processing on the UAV system,and obtains the diagnosis result by comparing the model probability and the fault detection threshold.After obtaining the fault information,the idea of the switching system is used to match the corresponding self-reconfiguration matrix and redistribute the motor speed.Due to the particularity of the tilt rotor structure,the remaining five rotors can compensate for the part of the torque lost by the faulty actuator rotor.The resultant moment is balanced,and finally the three attitude angles of the UAV are kept stable under the action of the integral sliding mode controller.The research content of the full text can be summarized in the following aspects:(1)Design the tilt-rotor structure,analyze the flight principle of the tilt-rotor UAV,introduce the definition and transformation of the geographic coordinate system and the body coordinate system,and establish the kinematic equations and dynamics of the tilt-rotor UAV based on the force analysis under the Newton-Euler method Learning equations,laying a theoretical foundation for the subsequent research on fault diagnosis and fault-tolerant control.(2)The fault diagnosis module is designed based on the interactive multi-model algorithm.The UAV system model is firstly linearized.After the fault model set is established,the strong tracking Kalman filter is used to monitor and track the UAV system in real time to improve the accuracy of system state estimation.The probability and fault detection threshold are used to judge the current state of the system and determine the specific location of the fault.(3)Considering that the tilt-rotor UAV system has the characteristics of nonlinearity,strong coupling and susceptibility to interference,the integral sliding mode algorithm is especially used to design the pose controller to improve the robustness of the system.After a brief analysis of the controllability of the tilt-rotor UAV under actuator failure,a fault-tolerant control scheme based on the idea of switching system is proposed,and the self-reconfiguration matrix under different modes is designed.When a fault occurs,the corresponding self-reconfiguration matrix is matched according to the set switching strategy,and the motor speed is solved and updated to rebalance the torque.