CN-122017915-A - Timing sequence RTK/INS integrated navigation method based on double differences between GNSS epochs

CN122017915ACN 122017915 ACN122017915 ACN 122017915ACN-122017915-A

Abstract

The invention relates to a time sequence RTK/INS integrated navigation method based on double differences between GNSS epochs, and belongs to the technical field of navigation positioning and integrated navigation. The method comprises the steps of carrying out single-point positioning and error correction on GNSS original observation data to construct an inter-epoch double-difference observation value, carrying out pre-integration on an IMU data in a Liqun/manifold space, estimating a carrier state and assisting cycle slip detection, finally, based on a factor graph optimization framework, fusing multiple constraint factor modeling solutions, and carrying out fixed integer ambiguity detection and feedback optimization through an LAMBDA method and Ratio. The invention can realize centimeter-level high-precision relative positioning, and can improve the positioning continuity, robustness and resolving precision in complex environments.

Inventors

Zheng Shucan
JIN RONGHE
LU XUESONG
LI SHITONG
LI MENGYAO
ZHU WENJIA
XU MENGLING
LIU YIFAN

Assignees

黄冈师范学院

Dates

Publication Date: 20260512
Application Date: 20260123

Claims (5)

1. The time sequence RTK/INS integrated navigation method based on the double differences between GNSS epochs is characterized by comprising the following steps: S1, GNSS data preprocessing and inter-epoch double difference observation value construction, namely, carrying out single-point positioning on received GNSS original observation data to obtain an initial position, clock difference and troposphere delay initial value, carrying out observation quality control based on an initial solution to determine satellites, carrying out ionosphere, troposphere, antenna phase center and relative clock difference error model correction on the determined satellites, calculating and constructing the inter-epoch double difference observation value of the GNSS on a time sequence to form a relative constraint factor; S2, pre-integrating the accelerometer and gyroscope data of the IMU in a cluster/manifold space to obtain a relative motion constraint factor between epochs and estimate a carrier state at any moment, and carrying out satellite geometric distance prediction by using the estimated carrier state to construct a cycle slip decision variable, and assisting a GNSS to observe cycle slip detection and provide a primary value; And S3, factor graph optimization solving and ambiguity fixing, namely uniformly modeling a GNSS epoch double-difference observation factor, an IMU pre-integral factor, a priori factor, a troposphere relative constraint factor, a GNSS carrier wave and a pseudo range absolute constraint factor based on a factor graph optimization framework, constructing a time sequence RTK/INS combined navigation factor graph, solving a combined navigation state through a nonlinear least square or incremental optimization algorithm, adopting a partial ambiguity resolution method based on least square ambiguity decorrelation adjustment, screening a robust integer ambiguity solution by combining a Ratio test, feeding back the integer ambiguity as a strong constraint to the factor graph, and updating the navigation state to improve the resolution precision and stability.
2. The method for integrated navigation of time-series RTK/INS based on double differences between GNSS epochs according to claim 1, wherein the steps of preprocessing GNSS data and constructing double-difference observations between epochs in step S1 are as follows: the GNSS observation equation is: ; in the superscript Representing satellites, subscripts Representing the moment; representing the geometric distance of the satellite to the receiver; Representing the speed of light in vacuum; And Respectively representing receiver and satellite clock differences; Representing satellite orbit errors; representing tropospheric delay; representing ionospheric delay; representing other correction terms including at least phase wrapping effect, antenna phase center offset and phase center variation, relativistic effect and earth rotation effect; And The sum of the measured noise and multipath error representing the pseudo-range and carrier phase, respectively; representing pseudorange observations; representing carrier phase observation wavelength; representing carrier phase observations; representing carrier phase ambiguity; GNSS-related state The method comprises the following steps: Wherein, the Is the location of the receiver and, Is the clock-difference of the receiver and, Is the tropospheric delay and the delay of the layer, Is the carrier phase ambiguity for each satellite, Each satellite corresponds to a cycle slip; Single-point positioning is carried out through observation data of a GNSS observation equation, a receiver position, a clock error and a troposphere delay initial value are obtained, and then the following factors are constructed according to relative constraints of the receiver position, the clock error and the troposphere delay initial value: ; Wherein, the As a priori factor, Is a tropospheric random process factor; observing an initial value of data for a GNSS observation equation; meanwhile, pseudo-range and carrier phase absolute constraint factors in traditional precise single-point positioning are introduced: ; Wherein, the And Pseudo-range and carrier phase absolute observation factors are respectively; And The pseudo range and carrier phase residual functions of the original observation are respectively from a GNSS observation equation; is a corresponding jacobian matrix; according to the base station-mobile station double difference mode in real-time dynamic positioning technology, at different moments 、 The double difference is carried out, and the expression is as follows: ; ; In the formula, Is a double difference operator, superscript And Respectively represent the first And Satellite particles, subscript And Respectively represent time of day And ; Representing pseudorange observations; representing carrier phase observations; representing carrier phase ambiguity; Representing double difference noise; After double difference, eliminating receiver end error, satellite end error and atmospheric delay, and finally remaining double difference distance, double difference ambiguity and double difference noise item, forming inter-epoch double difference relative constraint factor according to the double difference distance, double difference ambiguity and double difference noise item: ; In the formula, And Respectively double differential pseudo-ranges and carrier factors; And Respectively double differential pseudo-range and carrier phase residual functions; is the corresponding jacobian matrix.
3. The method for integrated navigation of time sequence RTK/INS based on double differences between GNSS epochs according to claim 1, wherein the IMU pre-integration and auxiliary cycle slip detection in step S2 are as follows: (1) IMU pre-integration Set IMU in The measured value of the moment is the angular velocity Acceleration of From adjacent moments 、 The pre-integral between is calculated as follows: ; In the formula, 、、 The rotation, speed and position variation between adjacent epochs calculated by pre-integration are respectively calculated; And Zero offset of the gyroscope and zero offset of the accelerometer are respectively carried out; Is that 、 At any time in between the two times, Is that 、 A speed variation amount therebetween; Is that 、 Is a time interval of (2); Status related to IMU Comprising the following steps: ; Wherein, the As a function of the position of the object, In order to be able to achieve a speed, Is the attitude angle of the object to be processed, Zero offset of the gyroscope and zero offset of the accelerometer are respectively carried out; the IMU connects two state quantities , The IMU pre-integration factor of (c) is expressed as: ; In the formula, 、 Representing the earth coordinate system and the IMU body coordinate system respectively to The system is an inertial system; 、 Respectively represent And Time B; the gravity acceleration vector is under the E system; And Respectively is Zero offset of the moment gyro and zero offset of the accelerometer; 、 Respectively is Zero offset of the moment gyro and zero offset of the accelerometer; (2) IMU assisted GNSS cycle slip detection Paired satellites The observation at three continuous moments is carried out, and the three moments are respectively 、 And The following double difference observables were constructed: ; In the formula, And Respectively represent the first 、 And (b) 、 Time satellite Geometric distance differences to the receiver; And Respectively represent 、 Cycle slip at two moments; 、 And Respectively represent the first 、、 Time satellite Is a carrier phase observation of (a); representing double-difference carrier phase observation noise; and calculating the distance between the user receiver and the reference station between two continuous epochs by using the position information of inertial navigation, wherein the calculation formula is as follows: ; In the formula, Representing the distance between the user receiver and the reference station between two consecutive epochs; 、 representing satellite-to-receiver geometric distance differences calculated using the IMU pre-integral dead reckoned receiver position; representing carrier phase observations noise calculated using IMU pre-integration; cycle slip decision variables are constructed as follows: ; In the formula, Representing cycle slip decision variables; Cycle slip decision variable For detecting the occurrence of cycle slips and for detecting the number of cycle slips.
4. The method for integrated navigation of time sequence RTK/INS based on double differences between GNSS epochs according to claim 1, wherein the steps of factor graph optimization solving and ambiguity fixing in step S3 are as follows: (1) Factor graph structure GNSS state And IMU state Merging, the total state is: ; Wherein, the , Respectively representing a dynamic state and a static state; the factors in the factor graph structure comprise prior factors, troposphere relative constraint factors, GNSS carrier and pseudo-range absolute constraint factors, double difference factors between GNSS carrier and pseudo-range epochs and IMU pre-integration factors, and the overall optimization function is as follows: ; (2) Ambiguity fixed constraints After obtaining a floating point ambiguity estimation value through FGO, a partial ambiguity resolution method based on least square ambiguity decorrelation adjustment is adopted to improve ambiguity solving performance and fixation rate; When the ambiguity is solved by a least square ambiguity decorrelation adjustment method, integer ambiguity constraint is generated, and the integer ambiguity constraint is applied to floating point estimation, namely: ; In the formula, Is the factor of the degree of ambiguity, For the integer ambiguity to be resolved, Floating ambiguity obtained by FGO; Integer ambiguity The constraint is incorporated into the estimator, and the states are updated to obtain a high-precision solution with carrier phase ambiguity resolution.
5. The method of claim 1, wherein the integer ambiguity fixing condition in step S3 is that when the floating ambiguity is resolved into the same integer in 5 consecutive epochs, it is determined that the ambiguity is correctly resolved and the determined ambiguity is incorporated into the estimator as an integer ambiguity constraint.

Description

Timing sequence RTK/INS integrated navigation method based on double differences between GNSS epochs Technical Field The invention relates to a time sequence RTK/INS integrated navigation method based on double differences between GNSS epochs, and belongs to the technical field of navigation positioning and integrated navigation. Background In modern unmanned, automatic driving and intelligent traffic systems, a Global Navigation Satellite System (GNSS) and an Inertial Navigation System (INS) are core sensors for realizing high-precision positioning navigation, and have good complementarity, wherein the GNSS can provide high-precision positions under a global coordinate system, but the sampling rate is lower, signals are easy to be interfered by environments, and the INS can provide high-sampling rate data and is not limited by the environments, but errors can be accumulated during long-term working. In order to obtain a high-precision and high-sampling rate integrated navigation result with environmental robustness, a GNSS/INS integrated navigation technology is widely researched and applied. The GNSS technology is mainly divided into two types of precise single point positioning (PPP) and real-time dynamic positioning (RTK), and the corresponding combined navigation method has the obvious defects that only one GNSS receiver is needed for PPP/INS combined navigation, the cost is low, the static PPP is converged to a centimeter level and needs more than half an hour, the dynamic PPP is converged for a longer time or even is not converged, the combined navigation convergence time is long, the RTK/INS combined navigation convergence speed is high, the centimeter level precision can be quickly achieved, an additional reference station is needed to be arranged nearby a mobile station, synchronous observation is needed, the cost is high, and the scene is limited. In addition, regardless of PPP/INS or RTK/INS integrated navigation, an Extended Kalman Filtering (EKF) method is often adopted, epoch correlation of GNSS observation values is ignored, when GNSS signals are absent, INS can cause error accumulation through long-time recursion, and the system still needs time to be re-converged after the GNSS signals are recovered. The existing double-difference RTK positioning between GNSS epochs can provide higher plane precision, but has poor stability and continuity under signal shielding or dynamic environment, while INS has short-time autonomous navigation capability but accumulated errors. Disclosure of Invention In order to solve the problems in the prior art, the invention provides a time sequence RTK/INS integrated navigation method based on double differences among GNSS epochs, which can realize centimeter-level high-precision relative positioning and can improve the positioning continuity, robustness and resolving precision in complex environments. In order to achieve the above purpose, the technical scheme provided by the invention is that a time sequence RTK/INS integrated navigation method based on double differences among GNSS epochs, comprising the following steps: S1, GNSS data preprocessing and inter-epoch double difference observation value construction, namely, carrying out single-point positioning on received GNSS original observation data to obtain an initial position, clock difference and troposphere delay initial value, carrying out observation quality control based on an initial solution to determine satellites, carrying out ionosphere, troposphere, antenna phase center and relative clock difference error model correction on the determined satellites, calculating and constructing the inter-epoch double difference observation value of the GNSS on a time sequence to form a relative constraint factor; S2, pre-integrating the accelerometer and gyroscope data of the IMU in a cluster/manifold space to obtain a relative motion constraint factor between epochs and estimate a carrier state at any moment, and carrying out satellite geometric distance prediction by using the estimated carrier state to construct a cycle slip decision variable, and assisting a GNSS to observe cycle slip detection and provide a primary value; And S3, factor graph optimization solving and ambiguity fixing, namely uniformly modeling a GNSS epoch double-difference observation factor, an IMU pre-integral factor, a priori factor, a troposphere relative constraint factor, a GNSS carrier wave and a pseudo range absolute constraint factor based on a factor graph optimization framework, constructing a time sequence RTK/INS combined navigation factor graph, solving a combined navigation state through a nonlinear least square or incremental optimization algorithm, adopting a partial ambiguity resolution method based on least square ambiguity decorrelation adjustment, screening a robust integer ambiguity solution by combining a Ratio test, feeding back the integer ambiguity as a strong constraint to the factor graph, and updating the nav