Research On Attitude Estimation Of Multi-rotor UAV

In recent years,multi-rotor UAV have been gradually used in various industries due to their superior performance,resulting in huge economic benefits.Attitude estimation is the basis of multi-rotor UAV navigation and control.It plays an important role in the application of multi-rotor UAVs.However,with the wide application range of the multirotor UAV,the increasingly complicated working environment,and the high control performance requirements,there are higher requirements for attitude calculation technology of multi-rotor UAV.In this context,this paper further explores and studies the attitude estimation technology of multi-rotor UAV.This paper compares the three attitude updating methods of Euler angle method,direction cosine matrix method and quaternion method.In this paper,the three attitude updating methods of Euler angle method,direction cosine matrix method and quaternion method are compared.The theoretical analysis and simulation experiments show that the quaternion attitude updating method has better comprehensive performance and is more suitable for the multi-rotor UAV attitude estimation with low cost microprocessor.Using the quaternion as the attitude representation parameter,a multi-rotor UAV attitude estimation system model based on the combination of IMU sensor combination and MARG sensor is established.In order to solve the problem that the traditional attitude estimation algorithm is difficult to balance high precision and strong real-time,this paper studies the nonlinear Kalman filtering algorithm.Based on the derived unscented Kalman filter(DUKF)algorithm and based on the attitude solving model of IMU sensor group,a multi-rotor UAV attitude estimation algorithm based on DUKF is proposed.PIXHAWK sensor data and Matlab software is used to compare the algorithm with the traditional algorithm.The experimental results show that the algorithm has higher resolution,shorter solution time and excellent overall performance.The coupling problem of the attitude estimation system of MARG sensor group is studied.The mutual interference of attitude estimation system caused by the coupling problem is analyzed.A decoupling strategy based on quaternion state vector is proposed.Based on the decoupling strategy,an attitude solving algorithm based on derived unscented Kalman filter is applied,and a two-stage attitude solving algorithm with separation interference capability is proposed.The relevant simulation experiments are designed to verify the effectiveness of the algorithm.The robustness of multi-rotor UAV attitude estimation is studied,and an adaptive strategy based on measurement noise is proposed.According to this,an adaptive attitude calculation algorithm for multi-rotor UAV is designed.The simulation results show that the algorithm has good anti-interference ability.