INS辅助GPS模糊度快速恢复与周跳探测技术研究

发布时间：2019-06-26 16:16

【摘要】：高精度GPS动态定位通常采用载波相位差分定位的方式，需要解决的关键问题是周跳探测和整周模糊度固定。然而，在城市、丛林等障碍物较多的地区进行动态测量时，GPS卫星信号容易失锁，多路径等噪声因素影响也很严重，这会导致周跳的频繁发生和整周模糊度不再可靠。因此，GPS失锁以后整周模糊度的快速恢复以及周跳的可靠探测就成了GPS高精度动态定位的重要问题，也是研究的热点和难点。本文基于GPS精密动态相对定位技术，系统研究了INS辅助GPS动态定位的两项技术，即INS辅助整周模糊度的解算和INS辅助周跳探测。文中包含了以下工作内容和研究成果： 1.简要介绍了GPS动态定位的数学模型和INS的原理特点，分析了影响动态定位精度的误差因素及改正方法。 2.系统研究了GPS动态定位的两项关键技术，即周跳探测与处理和整周模糊度解算，并给出了利用自编的相对定位程序进行GPS动态定位解算的结果。自编程序采用双差观测模型，广播星历计算卫星位置，组合方法探测周跳，Kalman滤波方法估计模糊度浮点解，最小二乘模糊度去相关平差法(Least-square Ambiguity Decorrelation Adjustment，LAMBDA)搜索整周模糊度以及Ratio值检验法确认整周模糊度。解算结果表明，相对定位程序对单频L1模糊度固定成功率在80%左右，宽巷模糊度固定成功率达到99%，模糊度固定以后单频L1解算定位误差的RMS小于两个厘米，说明程序具备了较高的精度及可靠性。 3.研究了INS辅助GPS整周模糊度解算技术。文中给出了三种辅助的方法，即直接法、两步法和全组合法，并对两步法和全组合法的等价性进行了推导。介绍了理论上评价模糊度搜索空间及解算精度的两个指标，即模糊度精度衰减因子(Ambiguity Dilution ofPrecision， ADOP)和模糊度解算精度提高的百分率（相对于仅有GPS数据的情况下，模糊度解算精度提高的百分率），并以此为基础，通过算例分析了不同INS位置精度和辅助方法下，模糊度搜索空间及解算精度变化情况。结果表明，当INS位置精度较高时，如优于0.2m时，模糊度搜索空间将明显缩小，解算精度也会明显提高，随着INS位置精度的下降，三种方法的辅助效果都会变差，且直接法辅助效果下降更快。 4.给出了GPS不同失锁时间及三种辅助方法下，单频L1模糊度固定所需的时间情况。结果表明，在GPS短时间失锁情况下，如失锁时间小于30s，INS辅助模糊度解算能够缩短模糊度的固定时间，，从而快速地恢复整周模糊度。 5.研究了INS辅助GPS周跳探测技术。文中基于Kalman滤波新息检验理论构造周跳检测量，运用χ~2检验的方法进行周跳检验，并利用最小可探测偏差(Minimum Detectable Bias,MDB)进行周跳的识别定位。算例分析了周跳检测量和MDB值的特性，并运用该方法对仿真的周跳进行了探测。结果表明，该方法不仅能够探测单周跳，而且能够同时探测多个周跳的情况，在GPS失锁时间为2秒时，对小到1周的周跳也能正确地探测出来。
[Abstract]:High-precision GPS dynamic positioning usually adopts carrier phase difference positioning mode, and the key problem that needs to be solved is that the round-hop detection and the whole-cycle ambiguity are fixed. However, when dynamic measurement is carried out in areas such as cities, jungles and the like, the influence of GPS satellite signals on noise factors such as lock-out and multi-path is also very serious, which can lead to the frequent occurrence of the cycle and the ambiguity of the whole-cycle ambiguity. Therefore, the fast recovery of the whole-cycle ambiguity and the reliable detection of the cycle jump are the important problems of GPS high-precision dynamic positioning after the GPS is lost, and it is also a hot point and a difficult point of the research. In this paper, based on the technology of GPS precise dynamic relative location, the two techniques of INS-assisted GPS dynamic positioning are studied, that is, the solution of the whole-cycle ambiguity of the INS and the auxiliary cycle of the INS. The following work contents and research are included in this paper. 1. The mathematical model of GPS dynamic positioning and the principle of INS are briefly introduced. The error factors and the change of the dynamic positioning accuracy are analyzed. In this paper, two key technologies of GPS dynamic positioning, namely, the weekly-hop detection and processing and the whole-cycle ambiguity resolution, are studied in this paper, and the dynamic positioning of GPS is given by using the self-designed relative positioning program. The results of the solution are as follows: The self-designed program uses the two-difference observation model, the broadcast ephemeris to calculate the satellite position, the combination method detects the cycle-hop, the Kalman filter method estimates the ambiguity-floating-point solution, the least-square-degree-of-least-square-degree-to-correlation level difference method (LAMDA) searches for the whole-cycle ambiguity, and the Ratio value test method confirms The results show that the fixed success rate of single-frequency L1 ambiguity is about 80%, the fixed success rate of the wide-lane ambiguity is 99%, and the RMS of the single-frequency L1 solution positioning error after the ambiguity is fixed is less than two centimeters. 3. The whole time of INS-assisted GPS is studied. In this paper, three kinds of auxiliary methods, namely, direct method, two-step method and all-group method, are given in this paper, and the two-step method and all-group method are also given. In this paper, the value of the fuzzy degree search space and the resolution precision are derived. The fuzzy degree precision attenuation factor (ADOP) and the percentage of the ambiguity resolution accuracy (with respect to the GPS data only) are introduced. Based on this, a numerical example is given to analyze the search space and solution of fuzzy degree with different INS position accuracy and auxiliary method. The results show that when the position accuracy of INS is high, for example, the position accuracy of INS is better than 0.2m, the fuzzy degree search space will be reduced obviously, and the resolution precision can be obviously improved. With the decrease of the INS position accuracy, the auxiliary effect of the three methods will be deteriorated, and the direct method and auxiliary method 4. The single-frequency L1 ambiguity is given under the different time of the GPS and the three kinds of auxiliary methods. The results show that the INS-assisted ambiguity resolution can shorten the fixed time of the fuzzy degree in the case of short-time lock-out of the GPS, such as the time of the lock-out time is less than 30s, so that the time is fast. Fast recovery of full-cycle ambiguity.5. Study of INS In this paper, based on the theory of Kalman filter new-interest test, the detection of the cycle-hop is constructed, and the method of 1-2 test is used to check the cycle-hop, and the minimum detectable bias (MD) is used. B) The identification and location of the peripheral hop are carried out. An example is given to analyze the characteristics of the week-hop detection and the value of the MDB, and the method is applied. The results show that the method not only can detect the single-cycle jump, but also can detect the condition of multiple cycles at the same time. When the time of the GPS lock-out time is 2 seconds, it is as small as 1 week.
【学位授予单位】：解放军信息工程大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：P228.4

【参考文献】