| As the core navigation device of the surface ships,the high-precision GNSS/INS integrated navigation system based on ring laser gyro(RLG)can make the best use of properties of each other,and construct a long-term stable,high reliability,high accuracy,and high output frequency position-velocity-attitude system.Due to the performance enchantment and kind's increment of shipborne weaponry,the higher requirements for the output with high accuracy of GNSS/INS integration are also put forward.Considering this,in this paper,the precise point positioning(PPP)technique is integrating with high-precision RLG INS.Besides this,the mechanism of gravity error has also been analyzed,which aims to improve the performance of GNSS/INS integration further.The detailed research works are shown as follows:(1)The theoretical modeling and experimental verification of PPP/INS integration.Due to the local base station is hard to set up for shipborne application in the remote sea,the PPP technique is put forward to integrate with high-precision RLG INS.Firstly,the error correction with high accuracy for GNSS raw measurement is performed,then the Kalman filtering model for the PPP technique has been also constructed.Subsequently,the principle of loosely-coupled and tightly-coupled PPP/INS integration is introduced,besides this,the Kalman filtering model in the navigation frame(n-frame)is derived.At last,the static semi-physical simulation and the shipborne marine test are conducted.Compared to the traditional Standard Point Position(SPP),the simulation and experimental results shows the positioning accuracy can be improved obviously by the usage of the PPP technique,besides,the attitude accuracy can be improved slightly.While in the case that GNSS satellites are in the preferable geometric distribution,for the measurement of loosely-coupled and tightly-coupled GNSS/INS integration are homologous,the performance of them shows slightly difference.(2)The PPP technique is put forward for the shipborne DOV measurement based on GNSS/INS integration.Firstly,the principle of the Attitude Observation Method(AOM)for DOV measurement is performed,where the AOM is put forward in previous research by our team.The PPP technique is integrating with RLG INS,which aims to obtain the attitude information with high-accuracy.Subsequently,the simulation has been designed,where the tracking error between the horizontal attitude error and DOV characterized as an index to verify the advantage of PPP.At last,the shipborne marine test is conducted,compared to the traditional SPP,the experimental results show that the PPP technique can improve the accuracy of AOM.(3)The analysis and modeling of the error sources in GNSS/INS integration.Firstly,the navigation update principle of RLG INS is performed,and then the mechanism of error sources and its corresponding mathematical models are shown in detail.After that,the research on the systematic calibration for RLG IMU has been carried out,whereas the simulation and experimental results show the error sources of inertial sensors can be calibrated precisely.Subsequently,the uncalibrated gravity disturbance can dominate the residual errors and become the main error sources of GNSS/INS integration.(4)The effect of gravity anomaly to position estimation in high-precision GNSS/INS integration under the conventional error model has been analyzed in detail,subsequently,the model optimization has also been conducted.In the loosely-couple case,the qualitative analysis has been carried out based on INS's error propagation functions and Kalman filter theory.Then,the corresponding simulations are designed,which shows the seriously change gravity anomaly along the track can introduce the height estimation error in the high-precision INS cases.Considering this,several optimization methods of conventional model have been put forward,for example,the compensation or modeling of gravity anomaly,the adjustment of Q parameter and so on,to attenuate the effect of gravity anomaly to height estimation in this cases.At last,the real shipborne marine test is organized,which confirms the effectiveness of such optimization methods.For the tightly-couple case,the feedback of Kalman filter under conventional GNSS/INS integration model has been derived,subsequently,the qualitative analysis has been conducted.Then,static semi-physical simulation has been designed,which shows that seriously change gravity anomaly can introduce the position estimation error both in the horizontal and vertical axis.Similar to the loosely-coupled cases,the real shipborne marine test has been conducted,which also shows the effectiveness of those above optimization methods in the tightly-coupled cases.(5)The effect of DOV to attitude estimation in high-precision GNSS/INS integration has been analyzed in detail.Firstly,the analytical expressions of pitch and roll estimation are derived based on the INS's velocity error propagation function.Then,the yaw estimation error caused by DOV is also obtained by usage of the INS's attitude error propagation and basic Kalman filter theory.Subsequently,the corresponding simulations were designed,which show that the attitude estimation in high-precision GNSS/INS integration can be affected by the DOV indeed.At last,the shipborne marine test has been organized,which the experimental results confirm the concept further.(6)The effect of gravity anomaly to attitude estimation in high-precision GNSS/INS integration under the overturning cases has been deeply analyzed.Based on the analysis of the effect of DOV,the analytical expression of attitude estimation error in the case that INS rotates around the X-axis and Y-axis are derived respectively.Then,the corresponding simulations are designed,which aims to verify the effect of gravity anomaly to attitude estimation under the overturning cases.At last,the static experiment is conducted,the experimental results confirm the concept that gravity anomaly can introduce the attitude estimation error indeed. |
PDF Full Text Request