Key Problems For Implementation Of External Angular Rate Sensing With Gyrowheel System

Gyrowheel is an integrated attitude control and determination instrument.The angular momentum is varied by controlling the gyrowheel rotor at non-zero tilts and detuned speeds,so that the gyrowheel can provide three-axis control torques and the capability to measure two-axis angular rates.In such a manner,the integration and efficiency of attitude control system(ACS)can be improved,which offers the potential to meet the mass and power requirements of small spacecrafts.The Gyrowheel is sill in the early stage of laboratory research.Unlike the traditional gyroscopes,it should always operate at non-zero tilts and de-tuned speeds according to the demand of the spacecraft ACS.Hence,it is more complicated to measure external angular rates with a gyrowheel.The traditional rate sensing method has obvious approximate errors under large tilt conditions.In addition,the inaccurate knowledge of dynamic parameters and mechanical errors significantly influence the rate sensing accuracy,which restrict the practical application of the gyrowheel.This study takes the attitude measurement of small spacecrafts with the gyrowheel as the research background.Motivated by the above facts,this study attempts to realize measurement and improve the measurement accuracy by developing a novel rate sensing method and performing calibration tests for the gyrowheel.Main contents and contributions are as follows:By introducing generalized coordinates and applying the Lagrange equations of the second kind,a complete dynamic model of the gyrowheel system is established considering the angular motion of the case.The model has been validated by simulations using SimMechanics and experiments on a gyrowheel prototype.On this basis,the angular rate sensing principle with the gyrowheel is presented,and the main factors influencing the rate sensing accuracy are analyzed,which lay the foundation for the following work.A novel external rate sensing method is developed.In an effort to avoid the linearization error in the traditional method,the low-frequency dynamic model is obtained by employing the Jacobi-Anger identity,which can preserve the nonlinear dynamic characteristics of the gyrowheel system.Then the complicated torque terms in the model is simplified through quantitative analysis.The online calculation of the Bessel functions is realized by polynomial fitting,thereby a nonlinear algorithm is developed,which is capable of estimating the external angular rates within the whole operating range of gyrowheel.The computational efficiency of the proposed method is similar to that of the traditional one,moreover,the angular rate sensing accuracy is significantly enhanced.An offline parameter identification method for the gyrowheel system is proposed.The identification principle is presented by deriving an inverse dynamic model of the gyrowheel,and the ill-conditioned problem is analyzed.On the basis of the analysis,a series of special tests are designed,and a rough estimation of the parameters is obtained by an approximate principal component transformation.Then the prior knowledge of the parameters is introduced as a form of regularization to guide and constraint the illconditioned problem,by which an exact and reliable identification result of the parameters can be given.Compared with the design values,the model output with the identification values fits the system output better,thus the identified parameters can significantly enhance the angular rate sensing accuracy of the gyrowheel.Error calibration and compensation method is investigated to deal with the drift errors induced by the non-ideal mechanical factors.A complete error model of the gyrowheel is given according to error analysis.On the basis of significance test method,the regression effect of the complete model is validated,and a practical error model is established by stepwise regression,which provides a better description of the drift error characteristics of the gyrowheel.Then by utilizing the multi-objective optimization method,an improved multi-position calibration scheme is proposed,which meets the requirements of D-optimal and well-conditioned criteria.The optimal calibration scheme is efficiency and effective,thus it can make error compensation more effectively.Finally,the problems related to the ground tests on a gyrowheel prototype are investigated.To extract the useful information in the gyrowheel signals,a hybrid denoising method is proposed according to the noise characteristics,which ensures the performance of the prototype tests.Besides,an experimental platform is set up based on the gyrowheel prototype and the precision turntable.Experiments are carried out to calibrate the drift errors and identify the parameters of the prototype,and the effectiveness of the propose rate sensing method is also verified.