Attitude Calculation Of Quad-rotor UAV Based On Multi-sensor

An accurate navigation system is the basis for the good performance of a quadrotor UAV.Small quadrotor UAVs cannot obtain accurate attitude information due to the low accuracy and high measurement noise of a single sensor inside the body.In view of the above problems,this paper fuses the information of multiple sensors to reduce the measurement error of a single sensor and obtain accurate attitude information.The main contents include:This paper analyzes in detail the research status of multi-sensor information fusion technology in the application of quadrotor UAV at home and abroad,and in-depth discusses the multi-sensor information fusion methods,including complementary filtering,Kalman filtering.The working characteristics of the instrument and the magnetometer are analyzed,and the cause of the error is analyzed,and the calibration is carried out.Because the body vibration of the quadrotor UAV will have a great impact on the data collected by the accelerometer and gyroscope.The fast Fourier transform is used to analyze the accelerometer and gyroscope sensor data in the frequency domain to determine the vibration noise range of the airframe.Based on the detailed analysis of the characteristics of the IIR and FIR digital filters,the simulation verifies that the IIR digital filter removes airframe vibrations.Effectiveness of noise.An adaptive explicit complementary filtering algorithm is proposed,it is based on the attitude quaternion method,the algorithm assigns different weights to the accelerometer attitude estimation value and the gyroscope attitude estimation value according to the motion state of the quadrotor UAV.Fusion adaptation is performed to complete the attitude calculation of the quadrotor UAV.The accelerometer estimation value adopts the gradient descent method,which increases the iterative speed of the algorithm and reduces the measurement error of the accelerometer.The gyroscope attitude estimation value adopts the second-order complementary filtering algorithm to fuse the measurement value of the gyroscope and the compensation amount of the accelerometer.The measurement error of the gyroscope is reduced.The weight is determined by the ratio of the estimated value of the accelerometer to the two-norm of the gravitational acceleration,which reflects the adaptability of the algorithm.Compared with the complementary filtering algorithm and the extended Kalman filtering algorithm,the experimental results show that the algorithm can effectively improve the attitude angle calculation accuracy of the quadrotor UAV in the indoor environment.Using stm32f407,icm20602,ak8975 and quad-rotor UAV power device to build a semi-physical simulation platform for quad-rotor UAV,the platform can realize sensor data acquisition and basic flight tasks of quad-rotor UAV.Aiming at the difficulties in the debugging process of the platform,a virtual simulation model of quadrotor UAV based on Gazebo is designed,which includes the required sensors and real environmental noise.The attitude of the model are analyzed by QGC software,and extract the flight log in the form of data,which facilitates the secondary analysis of the attitude of the quadrotor UAV.