Identification And Compensation Of Inertial Instrument Error

Dynamically Tuned Gyro (DTG) is a kind of2-DOF (degree of freedom) inertial sensorwith a specially flexible joint. Because of its advantages of simple structure, small size andlow cost, DTG is widely used in military and civil fields, such as guidance and orientation ofaviation, spaceflight, navigation, land vehicle or exploitation of oil mines. Since gyro is themost important component of the inertial system, the precision of carrier’s attitude andposition data provided by the system are closely related to gyros’ performances.In order to improve the control quality and test efficiency of DTG, some methods andtechnology are proposed and researched, which are listed as follows:Concerning that DTG is a two-input two-output system, the decoupling problem of DTGrebalance loop is resolved firstly. Prepositive compensation decoupling methods (includingmatrix inverse decoupling method, invariance decoupling method and converse decouplingmethod) in frequency domain and state variable feedback decoupling method in time domainare proposed to transform the DTG equation into an ideal SISO model.Based on decoupled model of DTG, the auto-disturbance rejection control theory and H∞robust control theory are studied and applied in the compensation of the rebalance loop inorder to improve the system robustness when the loop model changed and externaldisturbance inserted. At last, results of simulative experiments in different assumed conditionsare shown and compared with PID control theory.It is the fact that noises are the main error that affects the precision of inertial instrumenttest data and navigation system. So wavelets de-noise methods for DTG output signalprocessing are analyzed. After comparing de-noise results between soft/hard thresholdmethod and their improved method (interpolation threshold, eclectic threshold and coefficientsquare threshold), the best suitable wavelet function and wavelet coefficient estimationmethod for DTG signal processing are chosen.Based on advanced virtual instrument technology, the procedure of design anddevelopment for DTG test system are also introduced in detail in this dissertation. The testsystem has functions of gyro tune frequency setting, analog/digital rebalance loop torquersinput current sample and gyro test data wavelet de-noise. Finally, the gyro static drift model parameters estimation method is deduced, while thegyro orientation error angle and torquers’ non-vertical error angle are concerned. Using8-position force feedback test data, which are obtained in DTG tuned and out-tuned states,gyro static drift model parameters are identified and analyzed, when error angles consideredor not.