| Currently, the services based on the high accuracy positioning technology and the navigation methods indoor and outdoor have been applied to many domains, such as in intelligent cities and transportations, smart wearable equipment etc. Generally, the Global Positioning System (GPS) is used as the main technology for the high accuracy positioning due to its capability of positioning in globalism, all-weather, and all-time. Usually, there are three major GPS positioning modes namely Standard Point Positioning (SPP), Real-time Kinematic (RTK), and Precise Point Positioning (PPP), respectively. As is well known, the SPP technology is only can provide meter level even tens of meter level position accuracy by using single GPS receiver due to the lower quality of GPS pseudorange observations. However the implementation of the SPP is simple. Therefore, usually the SPP is applied widely in the fields without high accuracy requirement. The RTK technology based on the double difference carrier-phase observations between the rover station and the base station is used as a high accuracy positioning tool in nowadays. The disadvantages of the RTK are that the requirement of the base station will lead to more cost, and its positioning accuracy degrades clearly along with the increasing of the distance between the rover station and the base station. Fortunately, with the significant accuracy improvement of GPS precise orbit and clock products in recent years, the PPP technology inherits the advantages of both SPP and RTK and can provide centimeter position accuracy using single receiver observation. In this paper, according to the mathematic model of conventional PPP, the theory model of the Un-differenced Un-combined PPP with the ionospheric constraint and the receiver hardware delay constraint will be studied and derived.However, the positioning performance (such as the positioning accuracy, continuity, and reliability) of the SPP, RTK, and PPP can be influenced obviously by the users' observing conditions, especially in the complex environment (for example in urban city). The direct way to improve the GPS performance in such conditions is to launch more navigation satellites, which will increase the available satellite number and improve the performance of GPS in these conditions. Luckily, some countries and the Union have developed their own Global Navigation Satellite System (GNSS) independently owning to the irreplaceable values in the economic domain and the national defense. With the rapid development of GNSSs, there are now four global systems namely GPS from America, GLONASS from Russia, BeiDou from China, and Galileo from the European Union, respectively. Besides, there are another two regional systems namely the Quasi Zenith Satellite System (QZSS) from Japan and the India Regional Navigation Satellite System (IRNSS) from India. And the available satellite numbers will increase from about 7-14 of GPS-only to more than 40 by using the multi-constellation GNSS (multi-GNSS). Then, the positioning accuracy, continuity, and availability of GNSS will be enhanced reasonably. Thus, based on the model of Un-differenced Un-combined PPP, the algorithm, the parameter modelization, and parameter estimation methods of multi-GNSS PPP will also been researched in this paper.Although the application of multi-GNSS can significantly enhance the positioning performance in most of complex environments, both single-and multi-GNSS cannot work in the areas where signals of most even all satellite could be disturbed or lost such as in the shade, the tunnel etc. In such conditions, to obtain the position information, the GNSS should integrate with some other active navigation systems. The Inertial Navigation System (INS), which can provide continuous position, velocity, attitude, and some other navigation parameters with high rate and high accuracy in short-term by just processing the velocity information and angular information from Inertial Measurement Unit (IMU), degrades its accuracy rapidly over time due to the accumulating character of IMU, especially for the Micro-Electro-Mechanical System (MEMS) IMU. Therefore, the integration of GNSS and INS can overcome the drawbacks of each individual system to form a new system which can provide better accuracy, continuity, reliability, and high rate in both open sky and urban canyon environment. According to the information used in the integration system, it can be divided into three modes, called the loosely coupled integration, the tightly coupled integration, and the ultra-tightly (or deeply) coupled integration, respectively. Meanwhile, the kernel integration algorithms of the deeply coupled integration is the loosely/tightly coupled integration. Due to the reasons that (1) the tightly coupled integration is more effective than the loosely coupled integration during the poor satellite visibility periods, (2) most of the researches are about the integration of RTK and INS, (3) there are little researches on the conventional PPP/INS tightly coupled integration, and (4) there are no researches on the tightly coupled integration of multi-GNSS Un-differenced Un-combined PPP (IC PPP) and INS, thus the algorithm of multi-GNSS IC PPP/INS tightly coupled integration and its augmentation models will be the introduced as the kernel content in this paper.According to the theories and mathematic models derived in this paper, a software has been designed and developed independently with the major functions of multi-GNSS conventional PPP/INS tightly coupled integration, multi-GNSS IC PPP/INS tightly coupled integration, single frequency multi-GNSS IC PPP/INS tightly coupled integration, GNSS/INS loosely coupled integration, and multi-GNSS PPP. Based on such software, the performance of the multi-GNSS PPP and the multi-GNSS PPP/INS tightly coupled integration, the influences of INS on PPP data processing and the effects of different grades IMU on PPP performance are evaluated in this paper. Furthermore, the performance of the real-time multi-GNSS PPP/INS tightly coupled integration, the performance of single frequency multi-GNSS PPP/INS tightly coupled integration, and the possibility of adopting the PPP/INS tightly coupled integration to detect the track irregularity are also validated in this paper. And the major works of this paper are:1, The model of the Un-differenced Un-combined PPP with the ionospheric delay constraint in the temporal and spatial and the receiver hardware delay constraint in the temporal is derived based on the GNSS raw observation functions. Then, the models to deal with the time delay of signal, inter-system bias, and inter-frequency bias are introduced. Finally, the unified PPP model is proposed.2, Based on the IC PPP algorithm, the tightly coupled integration IC PPP and INS is proposed. Then, the theory, the mathematic model, and the parameter modelization about the IC PPP/INS tightly coupled integration are derived. Meanwhile, the augmentation models including the Non-Holomomic Constraints (NHC), the Zero Velocity Update (ZUPT), the Odometer augmentation, the Robust-Adaptive Kalman filter (RAKF), and the Variance Component Estimation (VCE) are also studied.3, According to the models and algorithms in this paper, we designed and developed a software independently. In this software, both multi-GNSS PPP functions and the tightly coupled integration of multi-GNSS PPP and INS functions can work.4, By processing the static GPS observations from IGS and the Crustal Movement Observation Network of China, the influence of the accuracy of ionospheric model and the effect of receiver hardware delay on the convergence time and the initialization position accuracy of IC PPP are evaluated. Then, a new ionospheric model based on the IC PPP is provided finally.5, Based on the developed software, we also analyzed the effect of INS on the convergence of PPP, INS aided un-differenced carrier phase cycle slip detection, the effect of different grades IMU on the performance of the PPP/INS tightly coupled integration, and the influence of PPP on the calibration of INS hardware errors.6, By processing the GPS+GLONASS+BeiDou three-system data using the ultra-precise satellite orbit and clock products, we validated the performance of multi-GNSS PPP/INS tightly coupled integration in real-time. After that, the possibility of using PPP/INS tightly coupled integration to detect the track irregularity is evaluated.In summary, we proposed the tightly coupled integration between multi-GNSS IC PPP and INS technology based on the multi-GNSS IC PPP algorithm and GNSS/INS integration theory, and the corresponding software is also developed in this paper. According to the results of experiments, they indicate that(1) Compared to the conventional PPP model, the IC PPP with high accuracy ionospheric model can provide better accuracy in both open sky and urban canyon environment; (2) With the aiding of INS, the performance of both conventional PPP/INS tightly coupled integration and IC PPP/INS tightly coupled integration can be further improved visibly; (3) The performance of both PPP and PPP/INS tightly coupled integration can be enhanced by using the multi-GNSS observations comparing with the solutions using GPS only; (4) The performance of PPP/INS tightly coupled integration can also be improved significantly after adopting the augmentation algorithms (the NHC, ZUPT, RAKF, and VCE); (5) It is possible to apply centimeter level position accuracy by using the ultra-precise satellite orbit and clock products in the multi-GNSS PPP/INS tightly coupled integration mode; (6) Due to the high accuracy (0.3 mm) in equivalent distance measurement, the PPP/INS tightly coupled integration technology can be used to detect the track irregularity; (7) the final positioning accureacy of single frequency multi-GNSS PPP/INS tightly coupled integration can be 20 cm for all of the three components. Finally, based on the researches results of this paper, the multi-GNSS PPP/INS tightly coupled integration can be widely used in the engineering fields like the Mobile Mapping System (MMS), the Unmanned Control System (UCS) etc. |
PDF Full Text Request