INS-assisted GNSS PPP precise dynamic navigation positioning method and system

Abstract

The invention discloses an INS-assisted GNSS PPP precise dynamic navigation and positioning method, including constructing a GNSS PPP / INS tight combination observation equation and a state equation, and linearizing the equation; constructing an INS virtual observation equation, and establishing a random model of the INS virtual observation equation through a derived INS random error transmission process; constructing GNSS PPP wide lane and ultra-wide lane combined observation values, and performing externally constrained ionospheric correction, differential code deviation correction and decimal deviation correction; through multiple intermediate processes, firstly using the LAMBDA method to calculate the ambiguity fixed solutions of the wide lane and ultra-wide lane of the reference star, and then calculating the ambiguity fixed solutions of the narrow lanes of each frequency point; estimating the ambiguity fixed solutions of other satellites, and obtaining precise dynamic navigation and positioning results. The invention starts from constructing a suitable GNSS PPP / INS tight combination filtering model and a combined ambiguity fixed solution, constructs INS virtual observation values, proposes an improved function model and a random model for the mixed integer least squares estimation of the GNSS PPP / INS combined system, and realizes precise dynamic navigation and positioning.

Description

Technical Field

[0001] The present invention belongs to the technical field of GNSS / INS combined navigation, and in particular relates to an INS-assisted GNSS PPP precise dynamic navigation and positioning technology. Background Art

[0002] GNSS Precise Point Positioning (PPP) technology can achieve centimeter-level high-precision positioning of a single receiver by relying on externally provided precise orbit, clock error and bias correction products. It is not limited by distance and region and has been widely used in scientific research and commercial applications. Ambiguity fixation is the premise of GNSS PPP precision navigation and positioning. Due to the limitations of GNSS vulnerability, especially in dynamic navigation and positioning, it is easy to be blocked and interfered, resulting in the inability to maintain or even fix the ambiguity. After losing the lock, PPP needs a considerable time to converge again, affecting its navigation and positioning performance. The Inertial Navigation System (INS) measures and outputs the angular velocity and acceleration of the carrier, and uses a recursive method to integrate and calculate the carrier state information. It has the advantage of high precision in the short term and can provide direct auxiliary information such as position and attitude for GNSS dynamic precision positioning; however, INS has drift errors, and GNSS can provide correction for INS. The two have excellent complementary characteristics. In the GNSS PPP / INS combined dynamic navigation and positioning system, the prior position accuracy and GNSS geometric observation conditions determine the performance of the combined system's ambiguity fixation. When the INS recursive position calculation accuracy is high enough, it can directly meet the GNSS cycle slip detection and ambiguity fixation requirements, but its cost is quite expensive, and low-cost INS position recursion will pollute the GNSS high-precision positioning results. How to achieve the ambiguity fixation performance of INS-assisted GNSS dynamic precision positioning without significantly increasing the cost is a problem that needs to be solved urgently.

[0003] The combined system usually uses Kalman filtering to estimate parameters. Accurate function models and random models are the prerequisites for obtaining high-precision parameter solutions. The widely used GNSS PPP / INS combined system state space model fails to take into account the impact of INS drift error after GNSS loss of lock. The random model needs to be frequently debugged and established, which brings difficulties to the re-fixation of PPP ambiguity after loss of lock. High-reliability function models and random models as well as high-reliability ambiguity fixing algorithms are the bottlenecks restricting the application and development of high-precision GNSS PPP / INS combined dynamic navigation and positioning systems, and are currently hot and difficult issues. Summary of the invention

[0004] To solve the above problems, based on previous research, the present invention constructs INS virtual observation values ​​from two key points: constructing a suitable GNSS PPP / INS tight combination filtering model and a combined ambiguity fixed solution. For the mixed integer least squares estimation algorithm of the GNSS PPP / INS combined system, an improved function model and a random model are proposed to realize precise dynamic navigation and positioning.

[0005] The technical solution proposed by the present invention is an INS-assisted GNSS PPP precise dynamic navigation positioning method, comprising the following steps:

[0006] Step 1: construct the GNSS PPP / INS tightly combined observation equation and state equation, and linearize the equation;

[0007] Step 2, construct the INS virtual observation equation, and establish the random model of the INS virtual observation equation through the derived INS random error transmission process;

[0008] Step 3: construct GNSS PPP wide lane and ultra-wide lane combined observation values, and perform externally constrained ionospheric correction, differential code bias correction, and fractional bias correction;

[0009] Step 4: Through multiple intermediate processes, the LAMBDA method is first used to calculate the ambiguity fixed solutions of the wide lane and ultra-wide lane of the reference star, and then the ambiguity fixed solutions of the narrow lane of each frequency point are calculated;

[0010] Step 5: Estimate the ambiguity fixed solutions of other satellites to obtain precise dynamic navigation positioning results.

[0011] Moreover, in step 1, the GNSS PPP / INS tight combination state equation is constructed as follows,

[0012]

[0013] In the formula,

[0014] is the differential of the position error in the Earth-centered Earth-fixed coordinate system, δv e is the velocity error in the Earth-centered Earth-fixed system, ξ r is the random noise of position error;

[0015] is the differential of velocity error in the Earth-centered Earth-fixed system, G is the gravity tensor, δr e is the position error in the Earth-centered Earth-fixed coordinate system, is the projection of the angular velocity of the inertial system to the Earth-centered Earth-fixed system, f e is the specific force in the Earth-centered Earth-fixed system, φ is the misalignment angle, is the rotation matrix from b to e, a bis the accelerometer bias, ξ v is the random noise of velocity error,

[0016] is the differential of the position error in the Earth-centered Earth-fixed coordinate system, ε b is the gyro bias, ξ φ is the random noise of the misalignment angle;

[0017] ξ a is the random noise of the accelerometer; ξ ε is the gyroscope random noise; ξ t is the random noise of the receiver clock error; ξ N is the random noise of the blur; is the random noise of tropospheric delay.

[0018] Moreover, in step 1, the GNSS PPP / INS tight combination observation equation is as follows,

[0019]

[0020] In the formula, v ρ,j 、v L,j 、v D,j They represent the residuals of pseudorange, phase, and Doppler observations, respectively; j represents the satellite frequency; n is the coefficient matrix of the satellite observation equation; l represents the arm error correction; μ is the ionospheric delay error coefficient, I is the ionospheric delay, and l e × is the antisymmetric matrix of the arm error correction for the earth-centered earth-fixed system, μ j is the ionospheric delay error coefficient at frequency j; m w is the projection function of the tropospheric delayed wet component, δI j is the ionospheric delay error at frequency j, δT w is the tropospheric delayed wet component error, δN j is the ambiguity error of frequency point j.

[0021] Moreover, in step 2, three new INS position virtual observations are added as follows:

[0022]

[0023] In the formula, x, y, and z represent the three-dimensional position coordinates of the carrier, respectively. I ,y I 、z I They represent the three-dimensional position coordinates of the carrier calculated by INS, e I represents the random error of the newly added virtual observations;

[0024] Based on the INS error state transfer process, the random model of INS virtual observation equation is derived as follows:

[0025]

[0026]

[0027] Where t and t0 represent the current moment and the last INS error correction moment respectively, and T is the time between the two moments; Indicates the variance of the solution result of the north position at the current moment. Indicates the variance of the solution result of the east position at the current moment, Indicates the variance of the solution result at the previous moment in the north direction. Indicates the variance of the solution result at the previous moment in the east direction, Indicates the variance of the north velocity solution at the last moment, Indicates the variance of the eastward velocity solution at the last moment, Indicates the variance of the solution result of the east attitude at the last moment, Indicates the variance of the solution result of the north attitude at the last moment, Indicates the variance of the solution result of the sky attitude at the last moment; g is the normal gravity value, It represents the projection of the north angular velocity from the inertial system to the navigation system in the navigation system. It represents the projection of the eastward angular velocity from the inertial system to the navigation system in the navigation system;

[0028] Select the above results and The maximum value is taken as the variance of the INS position solution at the current moment. The simplified random model is:

[0029]

[0030] Among them, I 3×3 is a 3×3 dimensional identity matrix; Σ XI (k) is the established stochastic model of INS position solution results.

[0031] Moreover, in step 4, the wide lane combined observations and the original observations are corrected using the GNSS PPP precise positioning ambiguity integer solution products to restore their integer characteristics. The variance-covariance matrix of the INS-aided ambiguity estimate is calculated by the INS virtual observation random model, and the ambiguity search space size V is determined as follows:

[0032]

[0033] Where χ is a constant controlling the size of the search space, n is the fuzzy dimension, and U n represents the volume of the unit sphere of the n-dimensional fuzzy search sphere, represents the ambiguity estimation variance-covariance matrix;

[0034] A step-by-step solution approach is adopted. First, more than four satellites with the highest elevation angles are selected. The LAMBDA method is used to fix and round the ambiguities of ultra-wide lanes and wide lanes, and then the narrow lanes are fixed and rounded.

[0035] On the other hand, the present invention provides an INS-assisted GNSS PPP precise dynamic navigation and positioning system, which is used to implement the INS-assisted GNSS PPP precise dynamic navigation and positioning method as described above.

[0036] Furthermore, the following modules are included,

[0037] The first module is used to construct the GNSS PPP / INS tightly combined observation equation and state equation and linearize the equation;

[0038] The second module is used to construct the INS virtual observation equation. Through the derived INS random error transmission process, the random model of the INS virtual observation equation is established;

[0039] The third module is used to construct GNSS PPP wide lane and ultra-wide lane combined observation values, and perform externally constrained ionospheric correction, differential code bias correction and fractional bias correction;

[0040] The fourth module is used to calculate the ambiguity fixed solutions of the wide lane and ultra-wide lane of the reference star by using the LAMBDA method through multiple intermediate processes, and then calculate the ambiguity fixed solutions of the narrow lane of each frequency point;

[0041] The fifth module is used to estimate the ambiguity fixed solutions of other satellites and obtain precise dynamic navigation positioning results.

[0042] Alternatively, it includes a processor and a memory, the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute the INS-assisted GNSS PPP precise dynamic navigation and positioning method as described above.

[0043] Alternatively, it includes a readable storage medium having a computer program stored thereon, and when the computer program is executed, the INS-assisted GNSS PPP precise dynamic navigation and positioning method as described above is implemented.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] (1) The satellite-to-ground distance is calculated by inverse calculation based on the INS position information, and the INS pseudorange and pseudorange rate are used for calculation to obtain the GNSS PPP ambiguity floating point solution;

[0046] (2) In order to avoid the influence of INS error drift on the GNSS PPP ambiguity fixed solution after long-term loss of lock, the INS virtual observation equation is constructed. The random error characteristics of the virtual observation equation are derived and calculated through the INS component error model, and the INS is given a suitable variance-covariance matrix to avoid the influence of INS error on the GNSS ambiguity estimation accuracy after long-term loss of lock, so as to ensure its effective assistance;

[0047] (3) In order to reduce the requirements of GNSS PPP ambiguity fixation on INS position accuracy, wide lane and ultra-wide lane combinations are constructed through multi-frequency GNSS observations, and external ionospheric constraints are added to correct the amplified ionospheric delay errors in the wide lane and ultra-wide lane combinations. The differential code bias and fractional bias correction of satellite hardware delay are considered, and the systematic error part is removed to facilitate ambiguity rounding and fixation.

[0048] (4) Select more than four satellites with the highest elevation angles of each GNSS system as reference satellites, use pseudorange observations and INS virtual observations to estimate the floating-point ambiguity solutions of the wide-lane and ultra-wide-lane combined estimates of the reference satellites, calculate the variance-covariance matrix of the floating-point ambiguity estimates through the constructed random model, use the LAMBDA method to search and fix the ambiguity, and then inversely calculate the ambiguity of each frequency point;

[0049] (5) The GNSS receiver coordinates are estimated using the fixed reference satellite ambiguity, and the other satellite ambiguity parameters are directly rounded and fixed.

[0050] The solution of the present invention is simple and convenient to implement, has strong practicality, solves the problems of low practicality and inconvenience in actual application existing in related technologies, can improve user experience, and has important market value. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a flow chart of INS-assisted GNSS PPP precise dynamic navigation positioning ambiguity according to an embodiment of the present invention. DETAILED DESCRIPTION

[0052] The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0053] The present invention is different from the traditional combined filtering model. It takes the INS position parameters as virtual observation values ​​and solves the GNSS PPP ambiguity by adding INS coordinate constraints. The INS error is related to the inertial component error, the carrier motion state and environmental factors, and it is difficult to establish its random model. Therefore, the INS position error expression is derived from the three single-channel INS error equations of east, north and sky, and a random model of INS virtual observation values ​​is established based on this. This function model and random model can adjust the weights of each system in real time according to factors such as observation residuals, INS device level, INS correction interval time, etc., which can avoid the INS drift after loss of lock affecting the accuracy of GNSS PPP ambiguity solution.

[0054] Secondly, INS provides additional redundant observations for GNSS, but the normal equations for solving fixed ambiguities are still rank-deficient, so the essence of ambiguity fixation of the combined system is still to solve integer least squares estimates. The LAMBDA algorithm can be used to transform floating-point ambiguities into integers, reduce the correlation between ambiguity parameters, and narrow the search range. Variance and covariance are crucial in the LAMBDA algorithm and are important references for ambiguity fixation. With the help of INS virtual observation equations and their random models, they are directly added to the ambiguity fixation solution, adding coordinate constraints, which can significantly improve the accuracy of ambiguity fixation.

[0055] Finally, the characteristics of each GNSS PPP model and its combination with INS were studied specifically, the functional model and random model of the combination of GNSS PPP inter-satellite single-difference ionosphere-free and INS system were derived, and the ambiguity fixing algorithm process and verification standard of the user-side combined system were given.

[0056] See also Figure 1 The process of an INS-assisted GNSS PPP precise dynamic navigation positioning ambiguity method proposed in an embodiment of the present invention is as follows:

[0057] (1) Construct the GNSS PPP / INS tightly combined observation equation and state equation, and linearize the equation:

[0058] The GNSS PPP / INS tight combination state equation is constructed as follows:

[0059]

[0060] In the formula, the superscripts i, e, and b represent the inertial coordinate system, the Earth-centered Earth-fixed coordinate system, and the carrier coordinate system, respectively; δr, δv, and φ represent the position error, velocity error, and misalignment angle, respectively; the superscript · represents the differential, for example are the differentials of δr, δv, and φ respectively;

[0061] is the angular velocity of the earth's rotation; G is the gravity tensor; is the rotation matrix from b to e; f e is the specific force in the e system; a b , ε b are the accelerometer and gyroscope bias respectively;

[0062] τ a , τ ε are the zero bias correlation time of the accelerometer and gyroscope respectively; δt is the GNSS receiver clock error; δN is the ambiguity error; δT w is the delay of the zenith wet component of the troposphere; ξ represents random noise, the subscripts correspond to the state parameters, and the corresponding spectral density is extracted by Allan variance analysis or given by empirical values.

[0063] Right now,

[0064] is the differential of the position error in the Earth-centered Earth-fixed coordinate system, δv e is the velocity error in the Earth-centered Earth-fixed system, ξ r is the random noise of position error;

[0065] is the differential of velocity error in the Earth-centered Earth-fixed system, G is the gravity tensor, δr e is the position error in the Earth-centered Earth-fixed coordinate system, is the projection of the angular velocity of the inertial system to the Earth-centered Earth-fixed system, f e is the specific force in the Earth-centered Earth-fixed system, φ is the misalignment angle, is the rotation matrix from b to e, a b is the accelerometer bias, ξ v is the random noise of velocity error,

[0066] is the differential of the position error in the Earth-centered Earth-fixed coordinate system, ε b is the gyro bias, ξ φ is the random noise of the misalignment angle;

[0067] ξ a is the random noise of the accelerometer; ξ ε is the gyroscope random noise; ξ t is the random noise of the receiver clock error; ξ N is the random noise of ambiguity; ξ Tw is the random noise of tropospheric delay.

[0068] The GNSS PPP / INS tight combination observation equation is as follows:

[0069]

[0070] Wherein the subscripts ρ, L, and D represent pseudorange, phase, and Doppler observations, respectively; v ρ,j 、v L,j 、v D,j They represent the residuals of pseudorange, phase, and Doppler observations, respectively; j represents the satellite frequency; n is the coefficient matrix of the satellite observation equation; l represents the arm error correction; μ is the ionospheric delay error coefficient, I is the ionospheric delay, and l e × is the antisymmetric matrix of the arm error correction for the earth-centered earth-fixed system, μ j is the ionospheric delay error coefficient at frequency j; m w is the projection function of the tropospheric delayed wet component, δI j is the ionospheric delay error at frequency j, δT w is the tropospheric delayed wet component error, δN j is the ambiguity error of frequency point j.

[0071] (2) Construct the INS virtual observation equation. By deducing the INS random error transmission process, the random model of the INS virtual observation equation is established:

[0072] Three new INS position virtual observation values ​​are added as follows:

[0073]

[0074] In the formula, x, y, and z represent the three-dimensional position coordinates of the carrier, respectively. I ,y I 、z I They represent the three-dimensional position coordinates of the carrier calculated by INS, e I It represents the random error of the newly added virtual observation value. The accuracy of the random model plays a decisive role in the ambiguity fixation. A reliable random model of the virtual observation equation is the premise for maximizing the INS-assisted GNSS PPP ambiguity fixation performance. Therefore, based on the INS error state transfer process, a reasonable random model of the INS virtual observation equation is derived and determined as follows:

[0075]

[0076]

[0077] Among them, σ 2 represents variance, subscripts X, V, They represent position, velocity, and attitude respectively. The subscripts E, N, and U represent the east, north, and sky directions respectively. t and t0 represent the current time and the last INS error correction time respectively. T is the time between the two times, that is, Indicates the variance of the solution result of the north position at the current moment. Indicates the variance of the solution result of the east position at the current moment, Indicates the variance of the solution result at the previous moment in the north direction. Indicates the variance of the solution result at the previous moment in the east direction, Indicates the variance of the north velocity solution at the last moment, Indicates the variance of the eastward velocity solution at the last moment, Indicates the variance of the solution result of the east attitude at the last moment, Indicates the variance of the solution result of the north attitude at the last moment, represents the variance of the solution result of the sky attitude at the last moment; g is the normal gravity value; ω is the angular velocity, and the upper and lower scripts i and n represent the inertial coordinate system and the navigation coordinate system respectively, that is, It represents the projection of the north angular velocity from the inertial system to the navigation system in the navigation system. It represents the projection of the eastward angular velocity from the inertial system to the navigation system.

[0078] In actual data processing, the results of the above two calculations can be selected and The maximum value is taken as the variance of the INS position solution at the current moment. The simplified random model is:

[0079]

[0080] Among them, I 3×3 is a 3×3 dimensional identity matrix; This is the stochastic model of the established INS position solution, where the subscript I represents INS.

[0081] By using this virtual observation random model, the variance of the INS position result can be adjusted as the lock-loss time increases, which can effectively avoid the impact of large INS drift on the GNSS PPP ambiguity fixation after a long period of lock-loss.

[0082] (3) Construct GNSS PPP wide-lane and ultra-wide-lane combined observation values, and perform externally constrained ionospheric correction, differential code bias correction, decimal bias correction, etc.:

[0083] Construct wide-lane observations to amplify the wavelength, relax the accuracy requirements for the prior station coordinates, and use combined observations for positioning to improve the accuracy of the station coordinates. Adopt the method of solving the super-wide lane and wide lane step by step. First, fix the super-wide lane ambiguity to provide the prior station coordinates with better accuracy for the next round of positioning, and finally complete the positioning work through several intermediate transitions.

[0084] For dual-frequency observations, the satellite-to-ground distance ρ is calculated from the INS virtual observations. I, combined with the wavelengths λ1 and λ2 of the two frequencies, the (1, -1) combined INS-aided ambiguity estimation formula is first given:

[0085]

[0086] in, is the ambiguity estimate, and the subscript WL represents the wide lane combination, that is, represents the (1, -1) combined INS-aided ambiguity, Φ WL11 Represents the (1, -1) combined observation value.

[0087] If the navigation accuracy of the INS equipment of the combined system is poor, a wide-lane combination with a longer wavelength such as (-3, 4), (4, -5), (-7, 9) can be constructed in the same way to further reduce the requirements for INS position accuracy. However, the ionospheric delay error is also magnified at the same time. In this case, ionospheric delay constraint compensation is required. Take the (-3, 4) combination as an example:

[0088]

[0089] in, represents the (-3, 4) combined INS-aided ambiguity, Φ WL34 Represents the (-3, 4) combination observation.

[0090] If triple-frequency GNSS observations are used, the INS-aided GNSS triple-frequency ultra-wide term combination ambiguity fixed formula can be constructed in the same way, but it is necessary to ensure that the triple-frequency observation combination can utilize the phase fractional deviation correction product in PPP precise positioning.

[0091] (4) Through multiple intermediate processes, the LAMBDA method is first used to calculate the ambiguity fixed solutions of the wide lane and ultra-wide lane of the reference star, and then the ambiguity fixed solutions of the narrow lane of each frequency point are calculated:

[0092] Using the GNSS PPP precise positioning ambiguity integer solution products released by IGS, the wide lane combined observations and the original observations are corrected to restore their integer characteristics. The variance-covariance matrix of the INS auxiliary ambiguity estimate is calculated based on the INS virtual observation random model calculated in the previous step. They are the variance-covariance matrices of the ambiguity estimation of the wide lane combination and the narrow lane combination, respectively. The size of the ambiguity search space V is determined as follows:

[0093]

[0094] Where χ is a constant controlling the size of the search space, n is the fuzziness dimension, U is the volume of the unit sphere, and U n represents the volume of the unit sphere of the n-dimensional fuzzy search sphere, Represents the variance-covariance matrix of ambiguity estimation. A step-by-step solution is adopted. First, the four or more satellites with the highest elevation angle are selected. The LAMBDA method is used to fix and round the ambiguities of the ultra-wide lane and wide lane, and then the narrow lane is fixed and rounded.

[0095] (5) Estimate the ambiguity fixed solutions of other satellites:

[0096] The receiver coordinates are calculated using the fixed reference satellite ambiguity estimates. This coordinate is used to replace the INS virtual observation value to estimate the remaining satellite ambiguities, which can now be directly rounded and fixed. After all ambiguities are fixed, the observation equations are back-substituted for observation updates to obtain high-precision state parameters to correct the INS, completing the epoch solution process.

[0097] In specific implementation, the above principles can be used to establish a software platform for GNSS PPP / INS combined high-precision dynamic navigation and positioning, and actual data can be used for verification.

[0098] In specific implementation, the INS-assisted GNSS PPP precise dynamic navigation positioning method proposed in the technical solution of the present invention can be automatically operated by those skilled in the art using computer software technology. The system devices for implementing the method, such as a computer-readable storage medium storing the corresponding computer program of the technical solution of the present invention and a computer device and server running the corresponding computer program, should also be within the protection scope of the present invention.

[0099] In some possible embodiments, an INS-assisted GNSS PPP precise dynamic navigation and positioning system is provided, including a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute an INS-assisted GNSS PPP precise dynamic navigation and positioning method as described above.

[0100] In some possible embodiments, an INS-assisted GNSS PPP precise dynamic navigation and positioning system is provided, comprising a readable storage medium having a computer program stored thereon, and when the computer program is executed, an INS-assisted GNSS PPP precise dynamic navigation and positioning method as described above is implemented.

[0101] The specific embodiments described herein are merely examples of the spirit of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in similar ways, but they will not deviate from the spirit of the present invention or exceed the scope defined by the appended claims.

Claims

1. An INS-assisted GNSS PPP precise dynamic navigation and positioning method, characterized by: The following steps are included: Step 1: construct the GNSS PPP / INS tightly combined observation equation and state equation, and linearize the equation; Step 2, construct the INS virtual observation equation, and establish the random model of the INS virtual observation equation through the derived INS random error transmission process; Step 3: construct GNSS PPP wide lane and ultra-wide lane combined observation values, and perform externally constrained ionospheric correction, differential code bias correction, and fractional bias correction; Step 4: Through multiple intermediate processes, the LAMBDA method is first used to calculate the ambiguity fixed solutions of the wide lane and ultra-wide lane of the reference star, and then the ambiguity fixed solutions of the narrow lane of each frequency point are calculated; Step 5: estimate the ambiguity fixed solutions of other satellites to obtain precise dynamic navigation positioning results; In step 1, the GNSS PPP / INS tight combination state equation is constructed as follows: In the formula, is the differential of the position error in the Earth-centered Earth-fixed coordinate system, δv e is the velocity error in the Earth-centered Earth-fixed system, ξ r is the random noise of position error; is the differential of velocity error in the Earth-centered Earth-fixed system, G is the gravity tensor, δr e is the position error in the Earth-centered Earth-fixed coordinate system, is the projection of the angular velocity of the inertial system to the Earth-centered Earth-fixed system, f e is the specific force in the Earth-centered Earth-fixed system, φ is the misalignment angle, is the rotation matrix from b to e, a b is the accelerometer bias, ξ v is the random noise of velocity error, is the differential of the position error in the Earth-centered Earth-fixed coordinate system, ε b is the gyro bias, ξ φ is the random noise of the misalignment angle; ξ a is the random noise of the accelerometer; ξ ε is the gyroscope random noise; ξ t is the random noise of the receiver clock error; ξ N is the random noise of the blur; is the random noise of tropospheric delay; In step 1, the GNSS PPP / INS tight combination observation equation is as follows, In the formula, v ρ,j 、v L,j 、v D,j They represent the residuals of pseudorange, phase, and Doppler observations respectively; j represents the satellite frequency; n is the coefficient matrix of the satellite observation equation; l represents the arm error correction; μ is the ionospheric delay error coefficient, I is the ionospheric delay, l e × is the antisymmetric matrix of the arm error correction for the earth-centered earth-fixed system, μ j is the ionospheric delay error coefficient at frequency j; m w is the projection function of the tropospheric delayed wet component, δI j is the ionospheric delay error at frequency j, δT w is the tropospheric delayed wet component error, δN j is the ambiguity error of frequency point j; In step 2, three new INS position virtual observations are added as follows: In the formula, x, y, and z represent the three-dimensional position coordinates of the carrier, respectively. I ,y I 、z I They represent the three-dimensional position coordinates of the carrier calculated by INS, e I represents the random error of the newly added virtual observations; Based on the INS error state transfer process, the random model of INS virtual observation equation is derived as follows: Where t and t0 represent the current moment and the last INS error correction moment respectively, and T is the time between the two moments; Indicates the variance of the solution result of the north position at the current moment. Indicates the variance of the solution result of the east position at the current moment, Indicates the variance of the solution result at the previous moment in the north direction. Indicates the variance of the solution result at the previous moment in the east direction, Indicates the variance of the north velocity solution at the last moment, Indicates the variance of the eastward velocity solution at the last moment, Indicates the variance of the solution result of the east attitude at the last moment, Indicates the variance of the solution result of the north attitude at the last moment, Indicates the variance of the solution result of the sky attitude at the last moment; g is the normal gravity value, It represents the projection of the north angular velocity from the inertial system to the navigation system in the navigation system. It represents the projection of the eastward angular velocity from the inertial system to the navigation system in the navigation system; Select the above results and The maximum value is taken as the variance of the INS position solution at the current moment. The simplified random model is: Among them, I 3×3 is a 3×3 dimensional identity matrix; The stochastic model of the established INS position solution.

2. The INS-assisted GNSS PPP precise dynamic navigation and positioning method according to claim 1, characterized in that: In step 4, the wide lane combined observations and the original observations are corrected using the GNSS PPP precise positioning ambiguity integer solution products to restore their integer characteristics. The variance-covariance matrix of the INS-aided ambiguity estimate is calculated using the INS virtual observation random model, and the ambiguity search space size V is determined as follows: Where χ is a constant controlling the size of the search space, n is the fuzzy dimension, and U n represents the volume of the unit sphere of the n-dimensional fuzzy search sphere, represents the ambiguity estimation variance-covariance matrix; A step-by-step solution approach is adopted. First, more than four satellites with the highest elevation angles are selected. The LAMBDA method is used to fix and round the ambiguities of ultra-wide lanes and wide lanes, and then the narrow lanes are fixed and rounded.

3. An INS-assisted GNSS PPP precise dynamic navigation and positioning system, characterized by: Used to implement an INS-assisted GNSS PPP precise dynamic navigation positioning method as described in any one of claims 1-2.

4. The INS-assisted GNSS PPP precise dynamic navigation and positioning system according to claim 3 is characterized by: Includes the following modules, The first module is used to construct the GNSS PPP / INS tightly combined observation equation and state equation and linearize the equation; The second module is used to construct the INS virtual observation equation. Through the derived INS random error transmission process, the random model of the INS virtual observation equation is established; The third module is used to construct GNSS PPP wide lane and ultra-wide lane combined observation values, and perform externally constrained ionospheric correction, differential code bias correction and fractional bias correction; The fourth module is used to calculate the ambiguity fixed solutions of the wide lane and ultra-wide lane of the reference star by using the LAMBDA method through multiple intermediate processes, and then calculate the ambiguity fixed solutions of the narrow lane of each frequency point; The fifth module is used to estimate the ambiguity fixed solutions of other satellites and obtain precise dynamic navigation positioning results.

5. An INS-assisted GNSS PPP precise dynamic navigation and positioning device, characterized by: The invention comprises a processor and a memory, wherein the memory is used to store program instructions, and the processor is used to call the stored instructions in the memory to execute an INS-assisted GNSS PPP precise dynamic navigation and positioning method as described in any one of claims 1 to 2.

6. A readable storage medium, characterized in that: The readable storage medium stores a computer program, and when the computer program is executed, it implements the INS-assisted GNSS PPP precise dynamic navigation positioning method as described in any one of claims 1-2.

Images