Research On State Estimation Method Of Multi-rotor Aircraft Based On Multi-source Information Fusion

Multi-rotor aircraft is an ideal platform for indoor and outdoor multi-scene operations because of its small volume,good maneuverability and strong hover ability.With the rapid development of aircraft technology in recent years,multi-rotor aircraft has been widely used in various scenarios.For multi-rotor aircraft,the key technologies are state estimation and motion control in flight control system.In this thesis,the following two key technical issues are proposed for the safety requirements of state estimation of multi-rotor aircraft in harsh environment: 1)How to calibrate the error parameters of low-cost sensors to obtain accurate and reliable measurement values;2)How to integrate multi-source information to improve the accuracy and robustness of aircraft state estimation.Therefore,this thesis designs solutions including sensor calibration algorithm,state estimation algorithm and control algorithm for the above key technical problems with important engineering value.The specific contents and results of this study are as follows:（1）Aiming at the problem of how to calibrate low cost sensor error parameters,first of all,in view of the IMU error,sensor error model is given,and analyzes the advantages and disadvantages of various calibration algorithm,and through the experimental analysis,points out that calibration method based on the residual error and threshold segmentation under the circumstances of less sample will influence the accuracy of the calibration results.Therefore,a new optimal threshold calculation algorithm is proposed to ensure the optimality of calibration results and improve the precision of IMU calibration.（2）For the state estimation problem of multi-rotor aircraft,the method of single sensor can only play a role in a specific environment,and the current fusion framework does not support the fusion of multiple sensor information in a modular and scalable way.Therefore,a prediction-correction-delay fusion framework is proposed,which can deal with multifrequency and multi-delay sensor systems,and add or subtract sensors without changing the algorithm.（3）Aiming at the problem that multi-rotor aircraft is affected by motion acceleration and magnetic field interference during operation,magnetic field decoupling strategy and motion acceleration interpolation delay compensation algorithm are proposed,which can have good suppression effect on motion acceleration and magnetic field interference and improve the robustness of state estimation.（4）The control algorithm is designed for multi-rotor aircraft,and a complete flight control system is formed by combining the state estimation algorithm mentioned above.The effectiveness of the state estimation method for multi-rotor aircraft in harsh environment is proved through flight experiments.