| Aviation geophysical exploration plays an important role in the exploration of oil and gas resources,geological structure investigations,and military defense construction.Aeronautical geophysical exploration with UAV as a carrier is a new mode of operation.Because of its high efficiency and flexibility,it has become an important direction for the development of aviation geophysical exploration at home and abroad.Among them,the quality of geophysical data is closely related to the attitude measurement technology of drones.Although the nine-axis attitude measurement system is a commonly used UAV attitude measurement method,its correction technology and attitude calculation technology are not directly applicable to the geophysical exploration drone.Therefore,this paper focuses on the working methods and data processing characteristics of geophysical exploration drones,and the research work and achievements to improve its attitude measurement accuracy and computational efficiency include:(1)Rapid calibration of magnetometer and accelerometerAccelerometers and magnetic sensors in the nine-axis attitude measurement system need to be quickly corrected before the geophysical exploration.Therefore,in the field of limited field equipment conditions,how to complete the higher accuracy of the correction work is to improve the accuracy of attitude measurement of important research content.This article has conducted in-depth research on ellipsoid fitting without external reference.It is pointed out that this method can only correct the error of the sensor’s scale factor,non-orthogonality and constant value,but it cannot correct the cause of misalignment error.In order to solve the problem of independent correction by ellipsoid fitting method,a joint correction method consisting of ellipsoid fitting method and dot product invariant method was established by using the constraint condition that the dot product of gravity and magnetic vectors was invariable.The geophysical prospecting theodolite is used as a horizontal reference to further perfect this joint calibration method,and a three-step joint calibration technique for the magnetic sensor and accelerometer at the geophysical field is formed.Through simulation analysis,it is verified that the ellipsoid fitting method is insufficient for the independent correction of the magnetic sensor and the accelerometer.The calibration experiment results show that this three-step joint correction method improves the accuracy by 62%~69% compared to the ellipsoid fitting method.(2)Attitude measurement method using magnetometer and accelerometer.Magnetic sensors and accelerometers are the key sources of attitude information in attitude measurement systems,but they are susceptible to magnetic environment and motion acceleration.Therefore,how to improve the attitude measurement accuracy of magnetic sensors and accelerometers is an important research issue under the noise characteristics of geophysical exploration drones.In this paper,two common attitude measurement methods for these two kinds of sensors are studied: electronic compass and TRIAD.The attitude accuracy of the two methods under different interference conditions is compared,and it is pointed out that the two methods cannot perform high-precision attitude measurement under the interference characteristics of the geophysical exploration drone.Based on the theory of TRIAD,the optimal and optimal TRIAD method based on reconstruction vector is studied.The method adjusts the parameters of the algorithm based on the characteristics of the dynamic interference of the aircraft geophysical exploration drone,and forms the optimal attitude of the magnetic sensor and accelerometer.Measurement methods.Simulation experiments show that this method can improve the accuracy by up to 50% compared with the electronic compass and TRIAD algorithm.Experimental results show that the optimal TRIAD algorithm has significantly improved the attitude measurement accuracy of different specifications of measurement systems.(3)Real time attitude estimation algorithm for Aero Geophysical unmanned aerial vehicleThe attitude measurement system needs to provide attitude information in real time during the flight of the drone for attitude compensation of the geophysical data.Therefore,the attitude estimation algorithm in this stage must take into account both the computational complexity and the accuracy.This dissertation deeply studies the pose estimation algorithm based on complementary filters and finds that the method cannot adjust the limits of its own cutoff frequency according to the external dynamic interference conditions.Therefore,this paper adopts a pose estimation algorithm combining time-varying complementary filter and optimal TRIAD to perform real-time attitude computation for a geophysical exploration drone.The experimental results show that the attitude estimation algorithm combined with the optimal TRIAD and time-varying complementary filters is basically consistent with the accuracy of the untracked Kalman filter.However,the computational efficiency of this combinatorial algorithm proposed in this paper has been increased by a factor of three and has been adapted to real-time requirements.(4)Post-processing attitude estimation algorithm for Aero Geophysical unmanned aerial vehiclesIn the post-processing stage of airborne geophysical data,the attitude measurement system is required to provide attitude information with as high accuracy as possible.Therefore,how to use post-processing characteristics of attitude data to improve attitude measurement accuracy is an important research content.This paper analyzes the processing flow of aviation geophysical data in depth and points out the post-processing features of the aviation geophysical data.At the same time,the current mainstream unobtrusive Kalman filter pose estimation algorithm is studied deeply.After improving the observation model in the filter algorithm to the optimal TRIAD,it further points out the limitation that this method can not use the geophysical data post processing.In order to improve the attitude information precision in the post-processing stage,an attitude estimation algorithm based on improved combination of untracked Kalman filter and fixed interval inverse smoother(RTS,Rauch-Tung-Striebel)is adopted to perform post-processing of geophysical exploration drone.Gesture calculations.The simulation results show that the unscented Kalman filter with the optimal TRIAD as the observation model and the R-T-S inverse smoother can improve the attitude measurement accuracy by 64%~73.9%.The results of drone flight experiments show that the post-processing algorithm further improves the attitude measurement accuracy by 60.1% compared with the traditional untracked Kalman filter,and improves the original dynamic technology index to meet the requirement of aviation geomagnetic gradient surveying. |
PDF Full Text Request