Study On The Compensation Technology Of Measurement Error Of Rotary-wing UAV Magnetic Detection System

In the aspect of aeromagnetic survey,rotary-wing UAV(unmanned aerial vehicle)magnetic survey system has a good application prospect.compensation method of the carrier magnetic interference is one of the key techniques to improve the measurement accuracy of the system.In this paper,the sources of measurement error and compensation method in the rotary-wing UAV magnetic survey system are studied.A measurement error model is established and the model parameter estimation algorithm is proposed.The proposed algorithm is verified by simulation and experiment.Firstly,the two main sources of measurement error of the rotary-wing UAV magnetic system are analyzed,namely the measurement error of three-axis magnetometer and the interference magnetic field of UAV,and mathematical models are established for each error source.According to analysis of the two established error models,a comprehensive compensation scheme and a two-step compensation scheme for implementing compensation of magnetic interference of the system are proposed.Secondly,aiming at the problem that traditional ellipsoid fitting algorithm do not achieving component calibration of three-axis magnetometer,a two-step calibration method for the calibration of the component error is proposed.The singularity problem of compensating flight data in interference compensation is analyzed,and a ridge estimation algorithm for parameter estimation is proposed.The proposed algorithms were verified by simulation analysis and experimental verification.Finally,experiments for implementing the compensation of interferential magnetic field of UAV are designed and carried out.A magnetic measurement system was constructed and the interferential magnetic field of the system was tested.The two compensation schemes and parameter estimation algorithms proposed in this paper were used to compensate the interferential magnetic field of the system.The standard deviation of measurement error was reduced from 76.57 nT to 4.41 nT.The results of experiments verify the effectiveness of the proposed compensation schemes and parameter estimation algorithms.