Research On Unmanned Aerial Vehicle Heading Resolution Technology

In this paper,the key technologies of multi-rotator UAV heading solution are studied.On the basis of ellipsoid fitting and least squares method,the traditional method of sensor error compensation in heading reference system is improved.Based on the Quaternion Cubature Kalman Filter algorithm,the algorithm is improved so that the system covariance matrix Q and the observed covariance matrix R are updated adaptively and iteratively with the algorithm.The experimental results show that the improved algorithm is more suitable for multi-rotator UAVs in different motion states.Designed a navigation position calculation platform based on the field programmable gate array,carried out the multi-sensor error formation mechanism and impact analysis,multi-sensor output noise characteristics analysis,navigation position calculation error analysis and high accuracy,high practicability of navigation position calculation technology research,mainly including.Based on the inertial navigation technology and the overall measurement structure of gyroscope,accelerometer and magnetometer based on the micro-system,the error formation mechanism and influence degree of the three micro-system sensors are analyzed.The error models of the three inertial sensors are established,and the three sensors are calibrated according to the error models,and the key error items are obtained.By using the error model backstepping compensation algorithm,the errors of the three sensors can be reduced and the navigation position calculation can be guaranteed.A data fusion algorithm based on volumetric Kalman filter and constrained by quaternion is presented to solve the three attitude angle information of a multi-rotator UAV through continuous iteration.The covariance matrix of the system state vector and the covariance matrix of the measured state vector are derived from the numerical relationship of the parameters in the algorithm,and the covariance matrix is updated continuously during the iterative process of the cubature Kalman filter to meet the needs of the multi-rotator UAV in different states.A computing platform is built for the algorithms and schemes described in this paper by selecting the platform based on SOC hard core.A complete GPS navigation position calculation system is constructed by designing an interface that satisfies the external data transmission protocol,FIFO for caching data,unit for compensating sensor data,and data fusion calculation based on SOC hard core.