The Design And Implementation Of Fault Diagnosis And Fault Tolerant Control System For Unmanned Helicopter

Recently,unmanned aerial vehicle has been widely used in multiple fields,as intelligent control technology develops.Large-scale unmanned helicopter has unique advantages in concrete civil,commercial and military applications,such as outstanding endurance capacity and load capacity.Complicated mission and environment demand more stable control systems and better independence ability.Fault Diagnosis and Fault Tolerant Control is an effective means to advance the stability and reliability of control systems.This research is raised from the practical need of some national defense pre-research foundation,which is the design of Fault Diagnosis and Fault Tolerant Control towards some actual unmanned helicopters mainly on sensors and actuators.The main contributions are as follows.Firstly,this research introduces the classic unmanned helicopter,and analyzes the whole control systems with practical hardware configured.This research also establishes the mathematical model of an unmanned helicopter for simulation verifying of the proposed method.This thesis puts forward a sensor Fault Diagnosis and Fault Tolerant Control method based on Wavelet Transform and multi-sensor comprehensive analysis with practical sensor data characteristic considered for some unmanned helicopters;multi-sensor comprehensive analysis utilizes analytical relationship between sensors for fault location and judgement.The proposed sensor Fault Diagnosis utilizes Wavelet Transform to analyze continuous sensor data.The multi-sensor comprehensive analysis utilizes analytical relationship between different sensors to realize fault detection.Considering sensor redundancy of large-scale unmanned helicopters,this thesis designs different sensor Fault Tolerant Control methods for difference redundancies.For multi redundancy configurations,this thesis proposes a distributed data fusion method to achieve optimal estimation of state whether some sensor fail or not.For single redundancy configurations,we redesign the control architecture taking sensor fault tolerance and flight safety in consideration.Simulation and experiment with actual flight data verify the effectiveness of the proposed method.Taking fault states of actuators as extended control state,a state estimation method based on Central Differential Kalman Filter to acquire real-time estimation of healthy status of actuators is designed,with all softy faults of actuators considered.We design a reconfiguration controller to reconfigure control output according to the actual healthy status of actuators,which eliminate the passive effectiveness of actuator faults.Simulation verifies that the presented method can detect soft faults of actuators quickly and accurately,and it also eliminate the negative impact of faults.