A method and system for online estimation of installation angle based on combination navigation of Beidou PPP-B2b

Abstract

This invention discloses an online estimation method and system for the installation angle of a combined navigation system based on BeiDou PPP-B2b, belonging to the field of navigation and positioning services. The method includes: constructing a high-precision position increment using carrier phase time difference observations and converting it into velocity observations; introducing a combined navigation filtering model and combining it with non-holonomic constraints to achieve online estimation and dynamic compensation of the installation angle; designing a sliding window data storage and convergence state discrimination method to analyze the convergence of the installation angle; and correcting the original IMU data after updating. This effectively improves the positioning accuracy and continuity in environments with limited or intermittent GNSS availability, such as tunnels, mountains, and deserts. This invention is applicable to vehicle navigation and unmanned system positioning scenarios in areas with no network coverage or weak infrastructure, effectively overcoming the limitations of traditional methods in terms of strong dependence on installation angle, constraint failure, and insufficient adaptability to weak infrastructure environments.

Description

Technical Field

[0001] This invention belongs to the field of navigation and positioning services, and specifically relates to an online estimation method and system for the installation angle of a combined navigation system based on BeiDou PPP-B2b. Background Technology

[0002] The integrated navigation technology combining BeiDou Precision Point Positioning (PPP-B2b) and a low-cost inertial navigation system (INS) enables continuous positioning even without ground communication infrastructure support, and has gained widespread attention in recent years for applications such as vehicle navigation in remote areas and unmanned system operations. However, under low-cost device conditions, traditional integrated navigation methods relying on vehicle kinematic constraints (such as non-integrity constraints) are prone to constraint failure due to the prevalent installation angle error between the inertial measurement unit (IMU) and the vehicle coordinate system, leading to decreased positioning accuracy. This problem is particularly pronounced in environments where GNSS signals are limited or interrupted. Therefore, accurately estimating and compensating for installation angle errors during actual operation has become a key issue in improving the performance of integrated navigation.

[0003] Existing methods for estimating installation angles are mainly divided into two categories: (1) Post-calibration methods. These methods typically collect data by pre-designing specific motion trajectories (such as straight and turning combination paths) and calibrate the installation angle using offline processing. This method can achieve high accuracy, but it relies on a specially designed calibration process, making it difficult to adapt to the application requirements of frequent carrier changes or temporary deployments. It has poor flexibility and cannot support online applications in actual operation; (2) Real-time estimation methods. These methods typically obtain the installation angle parameters based on the velocity information of the carrier coordinate system output by the integrated navigation system through trigonometric function relationships. This method does not require a special calibration process and has online estimation capabilities, but its performance is highly dependent on the convergence and accuracy of the integrated navigation solution results. When there is an initial alignment error in the inertial navigation system or poor GNSS observation quality, the velocity and attitude calculation errors will be directly transmitted to the installation angle estimation process, which can easily lead to divergence or instability in the estimation results.

[0004] Furthermore, low-cost receivers suffer from significant Doppler noise, making it difficult to provide stable and reliable velocity observations and limiting the applicability of online installation angle estimation methods. To address these issues, there is an urgent need for an online installation angle estimation method that can provide high-precision velocity observations at low cost and reduce dependence on integrated navigation solution results, thereby improving the system's adaptability and stability in environments with weak GNSS or no infrastructure. Summary of the Invention

[0005] This invention addresses the problems of motion constraint failure caused by installation angle errors in low-cost inertial devices under non-fixed installation conditions, making it difficult to guarantee navigation accuracy and continuity in environments lacking ground communication infrastructure or with intermittent GNSS signal availability. It also addresses the challenges of traditional non-holonomic constraints relying on strict alignment between the inertial device coordinate system and the vehicle coordinate system, and the difficulty in online estimation of the installation angle. The invention provides an online estimation method for the installation angle based on BeiDou PPP-B2b integrated navigation. This method constructs a high-precision position increment using carrier phase time difference observations and converts it into velocity observations. An integrated navigation filtering model is introduced and combined with non-holonomic constraints to achieve real-time estimation and dynamic compensation of the installation angle. Simultaneously, a sliding window data storage and convergence status judgment method is designed to analyze the convergence of the installation angle. After updating, the original IMU data is corrected, effectively improving the positioning accuracy and continuity of the system in environments with limited or intermittent GNSS availability, such as tunnels, mountainous areas, and deserts.

[0006] According to one aspect of this specification, an online estimation method for the installation angle of a combined navigation system based on BeiDou PPP-B2b is provided, comprising:

[0007] Based on the acquired GNSS observation data, IMU observation data, and precise correction information, a precise single-point positioning observation model is constructed, and GNSS solution results are obtained.

[0008] The vehicle's position and velocity are initialized based on GNSS calculation results, inertial navigation mechanical arrangement is performed based on IMU observation data, and a combined navigation state prediction model including vehicle position, velocity and attitude is constructed.

[0009] Carrier phase cycle slip detection is performed based on GNSS observation data. Satellites with cycle slips are marked as anomalies and removed. The velocity calculation mode is selected according to the number of satellites removed, and the velocity observation is obtained according to different velocity calculation modes.

[0010] Velocity observations are introduced into the integrated navigation state prediction model, and observation equations including the installation angle are constructed by combining non-integrity constraints.

[0011] Based on the observation equation of the installation angle, when using the carrier phase time difference velocity, the installation angle is estimated at a preset frequency;

[0012] By performing statistical analysis on the historical installation angle estimation results through a sliding window, the stability index of the installation angle is calculated. If the standard deviation of the historical installation angle estimation results within the window is less than the stability index of the installation angle, it is determined that the installation angle estimation has converged; otherwise, the installation angle estimation continues.

[0013] As a further technical solution, when carrier phase time difference velocity is not used, the installation angle estimation is paused, and only non-integrity constraint updates and covariance adjustments are retained.

[0014] As a further technical solution, the method also includes carrier navigation state initialization:

[0015] The carrier's position and velocity are initialized by calculating the BeiDou and GPS observations. The stationary state is determined by the statistics of the gyroscope and accelerometer. The roll angle and pitch angle are initialized when the vehicle is stationary. The heading angle is initialized by combining carrier phase time difference observations under the condition of straight vehicle motion.

[0016] As a further technical solution, a velocity calculation mode is selected based on the number of satellites removed, and velocity observations are obtained based on different velocity calculation modes, including:

[0017] When the number of effective satellites meets a preset threshold, the position increment is constructed using the carrier phase time difference and converted into a velocity observation.

[0018] When the number of effective satellites does not meet the preset threshold, Doppler observations are used to replace velocity observations.

[0019] As a further technical solution, after stopping the installation angle estimation and freezing the currently converged installation angle, the following is also included:

[0020] IMU observation data is corrected based on the converged installation angle, and the non-integrity constraint update frequency is increased to make the update frequency consistent with the inertial navigation recursion frequency, and the integrated navigation results are output; the integrated navigation results include the position, velocity and attitude of the vehicle.

[0021] As a further technical solution, the expression for the precise single-point positioning observation model is:

[0022]

[0023] in, Indicates pseudorange, Indicates carrier phase, Indicates Doppler observations, subscript Indicates receiver, superscript This indicates a satellite; the superscript SYS indicates GPS or BeiDou. Indicates the distance between the satellite and the receiver. Represents the speed of light. This represents the sum of the receiver clock bias and the system bias between different satellites. Indicates satellite clock bias, Indicates ionospheric delay, Indicates tropospheric delay; Indicates receiver hardware delay. Indicates satellite hardware delay, The wavelength representing the carrier phase observation value, Indicates the original carrier phase. The ambiguity representing the carrier phase. This represents the rate of change of the distance between the satellite and the receiver. This represents the rate of change of the receiver clock bias. Indicates the rate of change of satellite clock bias. This indicates the difference between epochs. This indicates the noise term.

[0024] As a further technical solution, the method also includes:

[0025] When the environment in which the current carrier is located is determined to be obstructed or weak signal environment by using the accuracy attenuation factor and the current velocity calculation mode, the installation angle estimation is stopped and the currently converged installation angle is frozen; wherein, the velocity calculation mode includes carrier phase time difference velocity and Doppler velocity.

[0026] According to one aspect of this specification, an online estimation system for installation angle of integrated navigation based on BeiDou PPP-B2b is provided, comprising:

[0027] The first processing module is used to construct a precise single-point positioning observation model based on the acquired GNSS observation data, IMU observation data, and precise correction information, and to obtain the GNSS solution results.

[0028] The second processing module is used to initialize the position and velocity of the carrier based on the GNSS calculation results, perform inertial navigation mechanical arrangement based on IMU observation data, and construct a combined navigation state prediction model that includes the carrier's position, velocity and attitude.

[0029] The third processing module is used to detect carrier phase cycle slips based on GNSS observation data, mark and remove satellites that have cycle slips, select the velocity calculation mode according to the number of satellites removed, and obtain the velocity observations according to different velocity calculation modes.

[0030] The fourth processing module is used to introduce velocity observations into the integrated navigation state prediction model and construct observation equations including the installation angle by combining non-integrity constraints.

[0031] The fifth processing module is used to estimate the installation angle at a preset frequency when using the carrier phase time difference velocity based on the observation equation.

[0032] The sixth processing module is used to perform statistical analysis on the historical installation angle estimation results through a sliding window, calculate the stability index of the installation angle, and determine that the installation angle estimation has converged when the standard deviation of the historical installation angle estimation results within the window is less than the stability index of the installation angle; otherwise, the installation angle estimation continues.

[0033] According to one aspect of the present invention, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the online estimation method for the installation angle of the integrated navigation based on BeiDou PPP-B2b.

[0034] According to one aspect of the present invention, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the online estimation method for the installation angle of the integrated navigation system based on BeiDou PPP-B2b.

[0035] Compared with the prior art, the beneficial effects of the present invention are:

[0036] 1. This invention proposes to combine time-difference carrier phase with non-integrity constraints for online estimation of the installation angle, realizing high-precision online estimation and dynamic compensation of the installation angle in a low-cost integrated navigation system. It achieves stable convergence and adaptive control of the estimation process under varying GNSS observation quality conditions, and completes the continuous transition from parameter estimation under good observation conditions to parameter compensation under obstructed conditions. By utilizing BeiDou PPP-B2b service, the dependence of high-precision positioning on communication facilities is reduced, while improving the continuity and stability of positioning.

[0037] 2. This invention uses low-frequency triggering of installation angle estimation under favorable GNSS observation conditions to reduce the computational burden of non-integrity constraints. In environments with limited or interrupted GNSS, it utilizes the estimated installation angle for high-frequency compensation, achieving synergistic optimization of computational efficiency and navigation performance. While ensuring the feasibility of a low-cost system, it effectively improves the robustness and continuity of integrated navigation in complex environments. It is suitable for vehicle navigation and unmanned system positioning scenarios in areas without network coverage or with weak infrastructure, effectively overcoming the limitations of traditional methods in terms of strong installation angle dependence, constraint failure, and insufficient adaptability to weak infrastructure environments. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 The flowchart illustrates the online estimation method for the installation angle of the integrated navigation system based on BeiDou PPP-B2b, as provided in this embodiment of the invention. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] like Figure 1 As shown in the figure, this embodiment of the invention proposes an online estimation method for the installation angle of integrated navigation based on BeiDou PPP-B2b, including: Step 1, acquiring GNSS observation data and IMU observation data; wherein the GNSS observation data includes pseudorange, carrier phase and Doppler observation values, and the IMU observation data includes angular velocity and acceleration; simultaneously receiving the precise correction information broadcast by BeiDou PPP-B2b, correcting the broadcast ephemeris, and restoring the precise ephemeris, precise satellite clock error and differential code deviation. Step 1: A precise single-point positioning observation model is constructed using GNSS observation data, IMU observation data, and precise correction information, and GNSS calculation results are obtained. Step 2: Position and velocity are initialized based on the GNSS calculation results, and roll and pitch angles are initialized using a stationary state. Initial heading values ​​are obtained by combining carrier phase time difference observations under straight-line vehicle motion conditions. After initialization, inertial navigation mechanical orchestration is performed based on IMU observation data to achieve continuous recursive calculation of attitude, velocity, and position, constructing a combined navigation state prediction model. Step 3: Based on the results of carrier phase cycle slip detection, satellites experiencing cycle slips are anomaly-marked and removed. The velocity calculation mode is selected based on the number of available satellites: when the number of effective satellites meets a preset threshold, the position increment is constructed using carrier phase time difference and converted into velocity observations; when the number of effective satellites does not meet the preset threshold, Doppler observations are used. Step 4: Introduce the velocity observation into the integrated navigation filter model, and establish an observation equation including the installation angle by combining non-holonomic constraints; during the filter update process, the installation angle is used as an extended state variable for joint estimation to achieve coupled solution of the installation angle and navigation state; Step 5: Based on the observation equation of the installation angle, determine the installation angle estimation trigger condition by the velocity observation quality. When using TDCP velocity, perform installation angle estimation at a preset frequency; when not using TDCP velocity, pause installation angle estimation, retain only non-holonomic constraint updates and adjust the covariance to suppress error propagation; Step 6: Construct a sliding window, perform statistical analysis on historical installation angle estimation results, and calculate the stability index of the installation angle; when the standard deviation of the parameters of the historical installation angle estimation results within the window is less than the stability index of the installation angle, it is determined that the installation angle estimation has converged; otherwise, continue to perform online installation angle estimation.

[0042] Specifically, embodiments of the present invention provide the following: acquiring observation data, correction information, and constructing a precise single-point positioning observation model:

[0043] Raw observation data, including GNSS and IMU observations, is acquired through a GNSS receiver and inertial measurement unit. The GNSS observation data includes pseudorange, carrier phase, and Doppler observations, while the IMU observation data includes angular velocity and acceleration. Preprocessing of the GNSS observation data includes preliminary cycle slip detection and observation filtering. Preliminary cycle slip detection uses a Loss of Lock Indicator (LLI) flag, Melbourne-Wübbena (MW) combined observations, and Geometry-Free combination (GF) observations to detect cycle slips in the carrier phase and remove abnormal satellites. Simultaneously, precise correction information broadcast by BeiDou PPP-B2b is received to correct the broadcast ephemeris, restoring the BeiDou PPP-B2b precise ephemeris, precise satellite clock bias, and differential code bias (DCB) correction information. A precise point positioning observation model is constructed based on the GNSS, IMU, and BeiDou PPP-B2b precise correction information.

[0044] Specifically, the construction of a precise single-point positioning model includes a positioning model and a velocity measurement model. Pseudorange and carrier phase are used for positioning, while Doppler and TDCP are used for velocity measurement.

[0045] (1)

[0046] in, Indicates pseudorange, Represents the carrier phase in meters. Indicates Doppler observations, subscript Indicates receiver, superscript This indicates a satellite; the superscript SYS indicates GPS or BeiDou. This indicates the distance between the satellite and the receiver, where the satellite's position is obtained from the broadcast ephemeris or the PPP-B2b precise ephemeris, and the receiver's position is a parameter to be estimated. Represents the speed of light. This represents the sum of the receiver clock bias and the system bias between different satellites, used as a parameter estimate. The satellite clock bias is indicated by broadcast ephemeris or PPP-B2b precision satellite clock bias. Indicates ionospheric delay, The tropospheric delay, ionospheric delay, and tropospheric delay are first corrected using the model, and then the residual terms are estimated. Indicates receiver hardware delay. This indicates satellite hardware delay, with the subscripts P and L distinguishing between pseudorange and carrier phase hardware delay. The pseudorange hardware delay is corrected for by DCB, while the carrier phase hardware delay is very small and is not processed during parameter estimation for real-time positioning. The wavelength representing the carrier phase observation value, Indicates the original carrier phase. The ambiguity representing the carrier phase. This represents the rate of change of the distance between the satellite and the receiver. This represents the rate of change of the receiver clock bias. Indicates the rate of change of satellite clock bias. This indicates the difference between epochs. This indicates the noise term.

[0047] Specifically, the content for restoring the BeiDou PPP-B2b precise ephemeris, precise satellite clock error, and differential code bias (DCB) correction information is as follows:

[0048] The BeiDou broadcast ephemeris refers to the orbital parameters broadcast by each satellite. PPP-B2b requires corrections to obtain precise ephemeris and precise satellite clock biases. The satellite's position in the geocentric coordinate system is calculated based on the BeiDou broadcast ephemeris.

[0049] (2)

[0050] in, This indicates the satellites calculated from the broadcast ephemeris. The position vector in the geocentric-fixed coordinate system; It is a three-dimensional vector, including components in the X, Y, and Z directions, which are respectively... The subscript brd indicates broadcast ephemeris, and the superscript T indicates matrix transpose.

[0051] Orbit corrections broadcast in PPP-B2b are typically given in the satellite radial, tangential, and normal (RAC) coordinate system, denoted as:

[0052] (3)

[0053] in, This represents the PPP-B2b orbital correction vector for satellite s. , , These represent the radial, tangential, and normal orbital corrections, respectively, where r, a, and c represent the radial, tangential, and normal orbital corrections, respectively.

[0054] The satellite velocity vector can be obtained from the broadcast ephemeris:

[0055] (4)

[0056] in, This represents the velocity vector of satellite s. express The three components correspond to the velocities in the X, Y, and Z directions in the geocentric coordinate system, respectively.

[0057] Next, construct three unit vectors for the satellite's RAC coordinate system, where RAC stands for Radial, Along-track, and Cross-track:

[0058] (5)

[0059] (6)

[0060] (7)

[0061] in, The unit vector representing the radial direction of satellite s' orbit. The unit vector representing the normal direction of the satellite's orbit s. Let represent the unit vector of the tangential direction of the satellite's orbit s. From this, we construct the transformation matrix from the RAC coordinate system to the geocentric Earth-fixed coordinate system:

[0062] (8)

[0063] The orbital correction of satellite s in the geocentric coordinate system can be expressed as:

[0064] (9)

[0065] in, This is the transformation matrix from the RAC coordinate system to the geocentric Earth-fixed coordinate system. This represents the PPP-B2b orbital correction vector for satellite s.

[0066] Therefore, the restored precise satellite position for:

[0067] (10)

[0068] Right now:

[0069] (11)

[0070] In addition, the PPP-B2b precision satellite clock error recovery process is as follows:

[0071] (12)

[0072] in, The precise satellite clock bias of satellite s It is the satellite clock bias obtained from broadcast ephemeris. It is the precision satellite clock error correction value broadcast by PPP-B2b. It's the speed of light.

[0073] The DCB broadcast by PPP-B2b is used for pseudorange observation correction, as follows:

[0074] (13)

[0075] Where the subscript j represents frequency, This represents the corrected pseudorange. This represents the original pseudorange observation at frequency j. This represents the DCB correction value for frequency j broadcast by PPP-B2b. This section only describes the recovery process for PPP-B2b precision products, a necessary step for real-time high-precision positioning using PPP-B2b services.

[0076] Specifically, embodiments of the present invention provide the content for constructing an integrated navigation state prediction model:

[0077] In the precise point positioning observation model, the satellite's position and clock bias are provided by BeiDou PPP-B2b, and DCB is used to correct the hardware delay of pseudorange. This precise point positioning observation model uses a commonly used Kalman filter method to calculate the vehicle's position using carrier phase and pseudorange observations, and uses TDCP and Doppler to calculate the vehicle's velocity. The vehicle's position and velocity are GNSS calculation results, which are input into the measurement update of the subsequent integrated navigation state prediction model. After initialization, IMU observation data is used for vehicle position, velocity, and attitude prediction. First, the vehicle's navigation state is initialized, i.e., the vehicle's position, velocity, and attitude are initialized. Position and velocity initialization can be obtained from BeiDou and GPS observations, while attitude includes roll, pitch, and heading angles. The stationary state is determined by statistical measurements of gyroscope and accelerometer readings within 5 seconds. In the stationary state, roll and pitch angles are initialized, with these statistics including the mean and standard deviation. Under conditions of straight vehicle motion, heading angle initialization is performed using carrier phase time difference observations. After initialization, a combined navigation state prediction model is constructed based on IMU observation data, and inertial navigation mechanical arrangement is performed to realize continuous recursive calculation of attitude, velocity and position, and predict the position, velocity and attitude of the vehicle.

[0078] Specifically, embodiments of the present invention provide content on switching between cycle slip detection and velocity calculation modes:

[0079] In GNSS observation data processing, carrier phase cycle slip detection is first performed, and satellites exhibiting cycle slips are anomaly-marked and removed. Based on this, the velocity calculation mode is selected according to the number of available satellites: when the number of available satellites after removing cycle-slipping satellites meets a preset threshold (no less than 6), the TDCP method is used to calculate the position increment between adjacent epochs and convert it into velocity observations for installation angle estimation; when the number of available satellites is insufficient or TDCP is unavailable, Doppler observations are used for velocity calculation, and installation angle estimation is not performed. Through this method, adaptive switching between velocity observation accuracy and availability is achieved.

[0080] Specifically, this embodiment of the invention introduces velocity observations into the integrated navigation filtering model, and establishes an observation equation including the installation angle by combining non-holonomic constraints (NHC). During the filtering update process, the installation angle is used as an extended state variable for joint estimation, realizing the coupled solution of the installation angle and navigation state. This embodiment of the invention assumes that the vehicle has good GNSS observations at startup. When obtaining high-precision velocity observations using the TDCP method, it is determined that the current GNSS velocity accuracy is high. Based on this, a non-holonomic constraint model including the installation angle is established by combining non-holonomic constraints (NHC), constructing an observation equation including the installation angle, and updating the constraints at a frequency of 1Hz to achieve online estimation of the installation angle. Based on the velocity observation quality, the installation angle estimation trigger condition is determined. When using TDCP velocity, the installation angle is updated at a preset frequency; when not using carrier phase time difference velocity, the installation angle estimation is paused, only the constraint update is retained, and the covariance is adjusted to suppress error propagation, realizing adaptive control of the estimation process. At the same time, the convergence state is judged by combining sliding window data. When the standard deviation of the installation angle estimate within the window is less than 0.3°, convergence is considered to be complete; otherwise, estimation should continue. When TDCP velocity is unavailable, it is assumed that the current observation conditions are insufficient to support a reliable estimate. Therefore, installation angle updates are paused, only the non-integrity constraint process is retained, and the covariance is amplified to prevent error propagation. Velocity observation quality can be understood as follows: the accuracy of carrier phase observation is inherently higher than that of Doppler; ideally, using TDCP velocity would result in even higher accuracy. Since velocity has only two sources—Doppler and TDCP—it is assumed that the velocity quality can be distinguished simply by determining the velocity calculation method.

[0081] Specifically, the process of constructing the observation equation for the installation angle in this embodiment of the invention is as follows:

[0082] Carrier coordinate system ( Under the condition that the lateral and vertical velocities are approximately zero, it can be expressed as:

[0083] (14)

[0084] in, express The speed of the tether, Indicates that the vehicle is in The forward velocity of the system, Indicates that the vehicle is in Lateral velocity under the system, Indicates that the vehicle is in The vertical velocity under the system. Ideally, the lateral and vertical velocities should both be zero.

[0085] Since the IMU has an unknown installation angle, projecting the vehicle velocity observed in the navigation coordinate system onto the v-frame can be expressed as:

[0086] (15)

[0087] in, This represents the rotation matrix projected from the IMU coordinate system (b-frame) to the v-frame. Let represent the rotation matrix projected from frame b to frame n. Let represent the rotation matrix projected from the n-frame to the v-frame. The velocity of the n-series downloader is primarily derived from two sources: TDCP velocity and velocity obtained recursively from the IMU under occlusion conditions. The rotation matrix is ​​an attitude representation, equivalent to the roll, pitch, and yaw values ​​mentioned earlier.

[0088] By performing error disturbance analysis on the above equation and considering the installation angle parameter, a speed error model of the vehicle carrier can be constructed:

[0089] (16)

[0090] in, This represents the velocity error in the n-system. This represents the carrier attitude error among the parameters to be estimated. The antisymmetric matrix representing the velocity of the n-system download. This indicates the installation angle error to be estimated.

[0091] Finally, the measurement equation can be constructed:

[0092] (17)

[0093] in, This represents the information of the filter. This represents the design matrix at the corresponding position. This represents measurement noise. Since only the lateral and vertical dimensions are observable, a measurement extraction matrix is ​​constructed. Then the above information and design matrix can be expressed as:

[0094] (18)

[0095] (19)

[0096] (20)

[0097] (twenty one)

[0098] The difference between estimating the installation angle and not is that Whether the parameters at the corresponding location were included in the estimation. Since Kalman filtering requires the covariance of a given observation when updating measurements, NHC is also updated as a virtual observation here. The strategy for constructing the covariance is as follows:

[0099] (twenty two)

[0100] To satisfy the characteristics of vehicle motion, a covariance selection scheme is proposed here, namely:

[0101] (twenty three)

[0102] (twenty four)

[0103] (25)

[0104] in, and The covariance value is designed based on experience, for example , It is a regulating factor. This is the reading of the gyroscope's Z-axis. This indicates taking the absolute value. The covariance values ​​for the lateral and vertical directions are different here, with the vertical covariance value being larger. The lateral covariance value is determined by the absolute value of the gyroscope's Z-axis reading to determine the severity of the turn; the more severe the turn, the larger the corresponding covariance setting.

[0105] Specifically, in this embodiment of the invention, a sliding window is constructed to perform statistical analysis on historical estimation results and calculate the stability index of the installation angle, i.e., when the standard deviation of the installation angle estimation within the window is less than 0.3°; when the standard deviation of the parameter within the window is less than the stability index of the installation angle, it is determined that the installation angle estimation has converged; otherwise, the online estimation process of the installation angle continues to be executed.

[0106] Specifically, this embodiment of the invention also provides a method to determine whether the current carrier is in an environment with obstruction or weak signal by using the DOP value (DOP refers to the accuracy attenuation factor, a dimensionless value that measures the impact of the currently observed satellite spatial geometric distribution on positioning accuracy) and the current velocity calculation mode. The velocity calculation mode includes TDCP velocity and Doppler velocity. A smaller DOP value indicates a more dispersed satellite distribution and better geometric conditions; a larger DOP value indicates a more concentrated satellite distribution and worse geometric conditions. When the real-time calculated DOP value exceeds a set threshold, it indicates that the carrier may be in an environment with obstruction or weak signal, such as tall buildings or mountains. Under normal circumstances, the high-precision TDCP velocity is used first. When a decrease in satellite signal quality, frequent cycle slips, or a decrease in the DOP value is detected, the velocity calculation mode is switched to ensure the continuity of velocity output, for example, by degrading from TDCP velocity to Doppler velocity, which also indicates a possible obstruction or weak signal environment.

[0107] When an environment is identified as obstructed or with weak signal, the installation angle estimation process is stopped, and the currently converged installation angle is frozen. If no obstructed or weak signal environment is identified, installation angle estimation continues. The converged installation angle is used for IMU observation data correction or system state model correction to compensate for installation errors between the sensor and the carrier; simultaneously, the non-integrity constraint update frequency is increased to match the inertial navigation recursion frequency, maintaining the stability and continuity of navigation solutions in the absence of GNSS. The combined navigation results after installation angle compensation are output, including position, velocity, and attitude information, achieving continuous high-precision navigation and positioning in complex environments.

[0108] Specifically, by monitoring the GNSS signal status, the system is determined to have entered an obstructed environment when a satellite signal loss or a significant decrease in observation quality is detected. In this state, the installation angle estimation process is halted, and the converged installation angle is applied to IMU observation data correction or state equation correction to compensate for installation errors between the inertial devices and the carrier. Simultaneously, the non-integrity constraint update frequency is increased from 1Hz to match the inertial navigation recursion frequency, ensuring that constraints continuously apply to the inertial navigation system at a high frequency, thereby maintaining the stability and accuracy of navigation calculations even in the absence of GNSS.

[0109] The implementation of the various embodiments of the present invention is based on programmed processing by a device with processor functionality. Therefore, in practical engineering, the technical solutions and functions of the various embodiments of the present invention are encapsulated into various modules. Based on this reality, and building upon the above embodiments, the embodiments of the present invention provide an online estimation system for the installation angle of integrated navigation based on BeiDou PPP-B2b. This system is used to execute an online estimation method for the installation angle of integrated navigation based on BeiDou PPP-B2b from the above method embodiments.

[0110] The system includes: a first processing module for constructing a precise point positioning observation model based on acquired GNSS observation data, IMU observation data, and precise correction information, and obtaining GNSS solution results; a second processing module for initializing the carrier's position and velocity based on the GNSS solution results, performing inertial navigation mechanical orchestration based on IMU observation data, and constructing a combined navigation state prediction model including the carrier's position, velocity, and attitude; and a third processing module for carrier phase cycle slip detection based on GNSS observation data, marking and removing satellites with cycle slips, and selecting velocity solutions based on the number of removed satellites. The system consists of six modules: a first module for calculating velocity observations, a second module for calculating velocity using different velocity calculation modes, a third module for introducing velocity observations into the integrated navigation state prediction model, and a fourth module for constructing observation equations including the installation angle based on incompleteness constraints, a fifth module for estimating the installation angle at a preset frequency when using carrier phase time difference velocity, and a sixth module for statistically analyzing historical installation angle estimation results through a sliding window, calculating the stability index of the installation angle, and determining that the installation angle estimation has converged when the standard deviation of the historical installation angle estimation results within the window is less than the stability index of the installation angle; otherwise, the installation angle estimation continues.

[0111] This invention provides an online installation angle estimation system for integrated navigation based on BeiDou PPP-B2b. Addressing the problems of motion constraint failure caused by installation angle errors in low-cost inertial devices under non-fixed installation conditions, the system struggles to guarantee navigation accuracy and continuity in environments lacking ground communication infrastructure or with intermittent GNSS signal availability. It also addresses the challenges of traditional non-holonomic constraints relying on strict alignment between the inertial device coordinate system and the vehicle coordinate system, which makes online installation angle estimation difficult. The system employs carrier phase time difference observations to construct a high-precision position increment, converting it into a velocity observation. This is then introduced into an integrated navigation filtering model, combined with non-holonomic constraints, to achieve real-time estimation and dynamic compensation of the installation angle. A sliding window data storage and convergence status judgment method is designed to analyze the convergence of the installation angle. After updating, the original IMU data is corrected, effectively improving the positioning accuracy and continuity of the system in environments with limited or intermittent GNSS availability, such as tunnels, mountainous areas, and deserts.

[0112] Based on the same inventive concept as the foregoing embodiments, this embodiment of the invention also provides an electronic device, including a memory and a processor. The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the online estimation method for the installation angle of the integrated navigation based on BeiDou PPP-B2b proposed in the above embodiments.

[0113] This invention also provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program constructs a high-precision position increment using carrier phase time difference observations, converts it into velocity observations, introduces a combined navigation filter model, and combines it with non-holonomic constraints to achieve real-time estimation and dynamic compensation of the installation angle.

[0114] Finally, it should be noted that the above specific embodiments are merely representative examples of the present invention. Obviously, the present invention is not limited to the above specific embodiments and many variations are possible. Any simple modifications, equivalent changes, and alterations made to the above specific embodiments based on the technical essence of the present invention should be considered within the protection scope of the present invention.

Claims

1. A method for online estimation of installation angle based on combination navigation of Beidou PPP-B2b, characterized in that, include: Based on the acquired GNSS observation data, IMU observation data, and precise correction information, a precise single-point positioning observation model is constructed, and GNSS solution results are obtained. The expression for the precise single-point positioning observation model is: ， in, Indicates pseudorange, Indicates carrier phase, Indicates Doppler observations, subscript Indicates receiver, superscript This indicates a satellite; the superscript SYS indicates GPS or BeiDou. Indicates the distance between the satellite and the receiver. Represents the speed of light. This represents the sum of the receiver clock bias and the system bias between different satellites. Indicates satellite clock bias, Indicates ionospheric delay, Indicates tropospheric delay; Indicates receiver hardware delay. Indicates satellite hardware delay, The wavelength representing the carrier phase observation. Indicates the original carrier phase. The ambiguity representing the carrier phase. This represents the rate of change of the distance between the satellite and the receiver. This represents the rate of change of the receiver clock bias. Indicates the rate of change of satellite clock bias. This indicates the difference between epochs. Indicates the noise term; The vehicle's position and velocity are initialized based on GNSS calculation results, inertial navigation mechanical arrangement is performed based on IMU observation data, and a combined navigation state prediction model including vehicle position, velocity and attitude is constructed. Carrier phase cycle slip detection is performed based on GNSS observation data. Satellites with cycle slips are marked as anomalies and removed. The velocity calculation mode is selected according to the number of satellites removed, and the velocity observation is obtained according to different velocity calculation modes. Velocity observations are introduced into the integrated navigation state prediction model, and observation equations including the installation angle are constructed by combining non-integrity constraints. Based on the observation equation of the installation angle, when using the carrier phase time difference velocity, the installation angle is estimated at a preset frequency; The historical installation angle estimation results are statistically analyzed using a sliding window to calculate the stability index of the installation angle. If the standard deviation of the historical installation angle estimation results within the window is less than the stability index of the installation angle, the installation angle estimation is considered to have converged; otherwise, the installation angle estimation continues. If the environment in which the current carrier is located is determined to be obstructed or weak signal environment by using the accuracy attenuation factor and the current velocity calculation mode, the installation angle estimation is stopped and the currently converged installation angle is frozen. The velocity calculation mode includes carrier phase time difference velocity and Doppler velocity.

2. The method according to claim 1, wherein, When carrier phase time difference velocity is not used, installation angle estimation is paused, and only non-integrity constraint updates and covariance adjustments are retained.

3. The method according to claim 1, wherein, The method further includes carrier navigation state initialization: The carrier's position and velocity are initialized by calculating the BeiDou and GPS observations. The stationary state is determined by the statistics of the gyroscope and accelerometer. The roll angle and pitch angle are initialized when the vehicle is stationary. The heading angle is initialized by combining carrier phase time difference observations under the condition of straight vehicle motion.

4. The method according to claim 1, wherein, The velocity calculation mode is selected based on the number of satellites removed, and velocity observations are obtained based on different velocity calculation modes, including: When the number of effective satellites meets a preset threshold, the position increment is constructed using the carrier phase time difference and converted into a velocity observation. When the number of effective satellites does not meet the preset threshold, Doppler observations are used to replace velocity observations.

5. The method according to claim 1, wherein, After stopping installation angle estimation and freezing the currently converged installation angles, the following steps are also included: IMU observation data is corrected based on the converged installation angle, and the non-integrity constraint update frequency is increased to make the update frequency consistent with the inertial navigation recursion frequency, and the integrated navigation results are output; the integrated navigation results include the position, velocity and attitude of the vehicle.

6. An online estimation system for installation angle of a combined navigation system based on BeiDou PPP-B2b, characterized in that, A method for online estimation of installation angle for integrated navigation based on BeiDou PPP-B2b, as described in any one of claims 1-5, includes: The first processing module is used to construct a precise single-point positioning observation model based on the acquired GNSS observation data, IMU observation data, and precise correction information, and to obtain the GNSS solution results. The second processing module is used to initialize the position and velocity of the carrier based on the GNSS calculation results, perform inertial navigation mechanical arrangement based on IMU observation data, and construct a combined navigation state prediction model that includes the carrier's position, velocity and attitude. The third processing module is used to detect carrier phase cycle slips based on GNSS observation data, mark and remove satellites that have cycle slips, select the velocity calculation mode according to the number of satellites removed, and obtain the velocity observations according to different velocity calculation modes. The fourth processing module is used to introduce velocity observations into the integrated navigation state prediction model and construct observation equations including the installation angle by combining non-integrity constraints. The fifth processing module is used to estimate the installation angle at a preset frequency when using the carrier phase time difference velocity based on the observation equation. The sixth processing module is used to perform statistical analysis on the historical installation angle estimation results through a sliding window, calculate the stability index of the installation angle, and determine that the installation angle estimation has converged when the standard deviation of the historical installation angle estimation results within the window is less than the stability index of the installation angle; otherwise, the installation angle estimation continues. 7.An electronic device comprising a memory and a processor, the memory storing a computer program, wherein, When the processor executes the computer program, it implements the steps of the online estimation method for the integrated navigation installation angle based on BeiDou PPP-B2b as described in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the online estimation method for the integrated navigation installation angle based on BeiDou PPP-B2b as described in any one of claims 1 to 5.

Images