Train locomotive high-precision positioning method and system based on GNSS and INS loose combination mode

Abstract

The invention discloses a high-precision train locomotive positioning method and system based on a GNSS (Global Navigation Satellite System) and INS (Inertial Navigation Satellite System) loose combination mode, and the method realizes the precise positioning of a train by combining the data of an IMU (Inertial Measurement Unit) and a GNSS (Global Navigation Satellite System). The method comprises the following steps: firstly, acquiring IMU data, performing error correction, and outputting corrected acceleration and angular velocity; meanwhile, a GNSS signal is obtained and a pseudo-range measurement value is extracted. Then, inertial navigation calculation is carried out, including attitude calculation, position and speed calculation and sensor deviation updating; through Kalman filtering fusion, state estimation and error correction are carried out by using complementary characteristics of the INS and the GNSS, so that the positioning precision is improved. In the initialization stage, the normal work of the system is ensured through the installation and parameter setting of the IMU and the GNSS. Finally, through Kalman filtering processing, accurate train position, speed, attitude and sensor deviation data are output, and reliable data support is provided for a train control system.

Description

Technical Field

[0001] The present invention belongs to the technical field of train positioning, and particularly relates to a high-precision positioning method and system for train locomotives based on the loose integration mode of GNSS and INS. Background Art

[0002] Existing train positioning technologies mainly rely on a single GNSS or INS system, and there are the following problems:

[0003] Limitations of GNSS: It is easy to lose lock in areas where satellite signals are blocked (such as tunnels and building clusters), resulting in positioning interruption.

[0004] INS error accumulation: The inertial navigation system will generate drift errors during long-term operation, and the positioning accuracy decreases as the driving distance increases.

[0005] Complex calculation of the tight integration mode: Existing GNSS / INS tight integration schemes require high-precision receivers and complex algorithms, with high costs and demanding hardware performance requirements.

[0006] Therefore, there is an urgent need for a train integrated navigation scheme with low cost, high reliability and adaptability to complex environments. Summary of the Invention

[0007] In order to solve the problems of easy interference of GNSS signals, INS error accumulation and complex calculation of the tight integration mode in the prior art, the present invention aims to provide a high-precision positioning method and system for train locomotives based on the loose integration mode of GNSS and INS. High-precision positioning of train locomotives based on the loose integration mode of the Global Navigation Satellite System (GNSS) and the Inertial Navigation System (INS) is applicable to the real-time positioning and navigation of train locomotives in complex environments (such as coal mines, tunnels, etc.).

[0008] In order to solve the technical problems, the technical solution of the present invention is:

[0009] A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS, the method includes:

[0010] IMU data acquisition and GNSS data acquisition, inputting the data of the accelerometer and gyroscope, correcting the collected data, and outputting the corrected acceleration and angular velocity; inputting the satellite signals received by the GNSS receiver, extracting the pseudorange measurement values, and outputting the pseudorange measurement values;

[0011] Inertial navigation solution: First, perform attitude solution. Input the corrected angular velocity, update the quaternion according to the gyroscope angular velocity to obtain attitude information, and convert it from the vehicle coordinate system to the navigation coordinate system through the rotation matrix, outputting the quaternion and the rotation matrix. Then, perform position and velocity solution. Input the corrected acceleration data, gravitational acceleration, and rotation matrix, calculate the train position and velocity based on the acceleration data and the rotation matrix, use gravity compensation, and output the current train position and velocity. Finally, perform sensor bias update. Input the noise of the accelerometer and gyroscope, update the sensor bias using random noise, and output the updated sensor bias value;

[0012] Kalman filter fusion: First, perform time update. Input the state estimate at the current moment, predict the state vector and error covariance matrix at the next moment through the state equation, and output the predicted state vector and covariance matrix. Then, perform measurement update. Input the difference between the GNSS measurement value and the INS measurement value, calculate the Kalman gain, and update the state vector and error covariance matrix through the Kalman filter formula, outputting the corrected state vector and the updated error covariance matrix.

[0013] Furthermore, before the data acquisition stage, the method further includes: Hardware installation and initialization, specifically including:

[0014] Hardware installation

[0015] IMU installation: Install the accelerometer and gyroscope, and the IMU sensor is ready to start collecting acceleration and angular velocity data; GNSS receiver installation: Install the GNSS receiver antenna, ensure there is no obstruction, and the GNSS receiver starts to receive satellite signals, providing position and pseudorange measurement data; Data connection: Connect the IMU, GNSS to the data processing unit, and transmit the IMU and GNSS sensor data to the data processing unit;

[0016] Parameter setting

[0017] According to the technical manual of the sensor and the previous calibration results, determine the key parameters of the inertial navigation system, such as the accelerometer zero bias, gyroscope zero bias, and related matrix parameters;

[0018] Set the update interval T of the system to determine the time frequency of data processing and state update;

[0019] Determine the relevant matrices in the Kalman filter algorithm, including the state transition matrix A, noise drive matrix B, measurement matrix H, and process noise covariance matrix Q and measurement noise covariance matrix R;

[0020] Initial state determination

[0021] Initialize the GNSS system: Start the GNSS receiver to obtain the initial position information;

[0022] Set the initial parameters of the Kalman filter: including the state vector and the error covariance matrix;

[0023] The state vector x0: includes position, velocity, attitude, accelerometer bias, and gyroscope bias;

[0024] The error covariance matrix P0: represents the uncertainty of the initial state.

[0025] Further, after the Kalman filter fusion, the method further includes: inputting the state estimation result after the Kalman filter fusion, sorting and formatting the fused data, and outputting the final accurate position, velocity, attitude parameters, and sensor bias data to the train control system to finally obtain the accurate train motion state.

[0026] Further, perform error correction on the collected data and output the corrected acceleration and angular velocity, including:

[0027] The acceleration f measured by the accelerometer after error correction:

[0028] The angular velocity ω measured by the gyroscope after error correction:

[0029]

[0030] Among them, represents the specific force output of the accelerometer, b a represents the zero bias of the acceleration, w a is the noise interference vector in the accelerometer; represents the angular velocity output of the gyroscope, b g represents the zero bias of the gyroscope, w g is the noise interference received in the gyroscope sensor measurement. The zero bias described in the formula includes both static zero bias and dynamic zero bias.

[0031] Further, the pseudorange measurement value represents the distance from the locomotive to the satellite. Extracting the pseudorange measurement value specifically includes:

[0032] Assume that the time when the receiver acquires the GNSS signal is t r , considering the deviation of the receiver clock, use Δt r to represent this unknown deviation amount. If the actual time of the current system is t, the relationship between the time when the receiver acquires the signal and the actual time can be expressed as follows:

[0033] t r = t + Δt r (3)

[0034] Similarly, there is also a certain deviation Δt in the atomic clock of the GNSS satellites , so the time when the GNSS satellite sends a signal can be expressed as:

[0035] t s = t + Δt s (4)

[0036] Then, according to the definition of the pseudorange, the pseudorange observation value d':

[0037] d' = c * τ' = c * (t r - t s ) + c * (Δt r - Δt s )(5)

[0038] In the actual navigation process, the user simultaneously observes more than four satellites through the receiver. At the same time, the clock biases of the satellites are known. Assuming the coordinates of satellite i are (x i , y i , z i ), and the current coordinates of the receiver are (x, y, z), the pseudorange observation value of each satellite can be rewritten as:

[0039]

[0040] The current position information of the receiver and the clock bias of the receiver are obtained by solving the above equations.

[0041] Further, in the inertial navigation solution, the attitude solution specifically includes:

[0042] Using the corrected gyroscope angular velocity value, the attitude quaternion is updated and calculated according to formula (7),

[0043]

[0044] The updated attitude quaternion is normalized to ensure that it satisfies the properties of the quaternion; the rotation matrix from the body coordinate system to the navigation coordinate system is calculated through formula (8)

[0045]

[0046] The position and velocity solution specifically includes:

[0047] According to the rotation matrix The corrected accelerometer specific force value And the gravitational acceleration in the navigation coordinate system, the position and velocity of the train at time k are calculated using formula (9) and formula (10), and further the position and velocity at time k + 1 are predicted according to formula (11) and formula (12);

[0048]

[0049] r n,k+1 = r n,k + v n,k ·T(11)

[0050]

[0051] The sensor bias update specifically includes: updating the biases of the accelerometer and gyroscope according to formulas (13-14), where w f,k and w ω,k are the random noises of the accelerometer and gyroscope respectively;

[0052]

[0053] Further, in the Kalman filter fusion, the time update specifically includes:

[0054] Based on the state estimate value at time k, predicting the state vector at time k+1 using the state equation Error covariance matrix

[0055]

[0056]

[0057] The measurement update specifically includes:

[0058] Calculating the measurement residual y k+1 , that is, the difference between GNSS and INS:

[0059]

[0060] Calculating the Kalman gain K k+1 :

[0061]

[0062] Updating the state vector

[0063]

[0064] Updating the error covariance matrix P k+1 :

[0065]

[0066] A high-precision system for train locomotives based on the loose integration mode of GNSS and INS, the system includes: a hardware installation and initialization module, an inertial navigation solution module, a Kalman filter fusion module, and an output module;

[0067] Hardware Installation and Initialization Module: It includes a hardware installation, parameter setting, and initial state determination sub-module;

[0068] Inertial Navigation Solution Module: First, perform attitude solution. Input the corrected angular velocity, update the quaternion according to the gyroscope angular velocity to obtain attitude information, and convert it from the vehicle coordinate system to the navigation coordinate system through the rotation matrix, outputting the quaternion and the rotation matrix; Then, perform position and velocity solution. Input the corrected acceleration data, gravitational acceleration, and rotation matrix, calculate the train position and velocity based on the acceleration data and the rotation matrix, use gravity compensation, and output the current train position and velocity; Finally, perform sensor bias update. Input the noise of the accelerometer and gyroscope, update the sensor bias using random noise, and output the updated sensor bias value;

[0069] Kalman Filter Fusion Module: First, perform time update. Input the state estimate at the current moment, predict the state vector and error covariance matrix at the next moment through the state equation, and output the predicted state vector and covariance matrix; Then, perform measurement update. Input the difference between the GNSS measurement value and the INS measurement value, calculate the Kalman gain, and update the state vector and error covariance matrix through the Kalman filter formula, outputting the corrected state vector and the updated error covariance matrix;

[0070] Output Module, used to input the state estimation result after Kalman filter fusion, organize and format the fused data, and output the final accurate position, velocity, attitude parameters, and sensor bias data to the train control system, finally obtaining the accurate train motion state;

[0071] Hardware Installation Sub-module:

[0072] IMU Installation: Install the accelerometer and gyroscope, and the IMU sensor is ready to start collecting acceleration and angular velocity data; GNSS Receiver Installation: Install the GNSS receiver antenna, ensure there is no obstruction, and the GNSS receiver starts receiving satellite signals to provide position and pseudorange measurement data; Data Connection: Connect the IMU, GNSS, and the data processing unit, and transmit the IMU and GNSS sensor data to the data processing unit;

[0073] Parameter Setting Sub-module:

[0074] Determine the key parameters of the inertial navigation system according to the technical manual of the sensor and the previous calibration results, such as the accelerometer zero bias, gyroscope zero bias, and related matrix parameters;

[0075] Set the update interval T of the system to determine the time frequency of data processing and state update;

[0076] Determine the relevant matrices in the Kalman filter algorithm, including the state transition matrix A, the noise driving matrix B, the measurement matrix H, as well as the process noise covariance matrix Q and the measurement noise covariance matrix R;

[0077] IMU data acquisition and GNSS data acquisition. Input the accelerometer and gyroscope data, correct the errors of the acquired data, and output the corrected acceleration and angular velocity; Input the satellite signals received by the GNSS receiver, extract the pseudorange measurement values, and output the pseudorange measurement values;

[0078] Initial state determination sub-module:

[0079] Initialize the GNSS system: Start the GNSS receiver to obtain the initial position information;

[0080] Set the initial parameters of the Kalman filter: including the state vector and the error covariance matrix;

[0081] State vector x0: including position, velocity, attitude, accelerometer bias, and gyroscope bias;

[0082] Error covariance matrix P0: represents the uncertainty of the initial state.

[0083] A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the high-precision method for train locomotives based on the loose combination mode of GNSS and INS described in any one of the above.

[0084] A computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the high-precision method for train locomotives based on the loose combination mode of GNSS and INS described in any one of the above.

[0085] Compared with the prior art, the advantages of the present invention are:

[0086] High-precision positioning: By combining the GNSS and INS systems and fusing through Kalman filtering, it can effectively improve the measurement accuracy of the train's position, velocity, and attitude.

[0087] Strong real-time performance: Through real-time data acquisition and processing, it ensures the efficiency and accuracy of train positioning and adapts to complex track environments.

[0088] High redundancy: The complementarity of the INS and GNSS systems enhances the reliability of the positioning system. Even when the GNSS signal is weak or lost, the INS can still provide effective positioning compensation.

[0089] Flexibility and Adaptability: The system supports the fusion of multiple sensors and measurement data, and can adjust parameters according to actual needs to adapt to positioning requirements in different environments. Description of the Drawings

[0090] Figure 1 , Schematic Diagram of Integrated Navigation and Positioning;

[0091] Figure 2 , Flow Chart of Integrated Navigation and Positioning;

[0092] Figure 3 , Schematic Diagram of the Navigation System Coordinate System;

[0093] Figure 4 , Schematic Diagram of Strapdown Inertial Navigation;

[0094] Figure 5 , Process Diagram of Kalman Filtering;

[0095] Figure 6 , Running Trajectory Diagram in Integrated Navigation Mode;

[0096] Figure 7, System Speed and Position Information in Two Integrated Navigation Modes (a - Speed Information, b - Position Information);

[0097] Figure 8, System Errors in Two Integrated Navigation Modes (a - Speed Error, b - Position Error). Detailed Implementation Modes

[0098] The following describes the specific implementation modes of the present invention in combination with embodiments:

[0099] It should be noted that the structures, ratios, sizes, etc. shown in this specification are only used to cooperate with the content disclosed in the specification for those skilled in this technology to understand and read, and are not used to limit the limited conditions for the implementation of the present invention. Any modification of the structure, change of the ratio relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope covered by the technical content disclosed in the present invention.

[0100] At the same time, the terms such as "upper", "lower", "left", "right", "middle", and "one" cited in this specification are only for the convenience of clear narration, and are not used to limit the scope of implementation of the present invention. The change or adjustment of their relative relationships, without substantial change in the technical content, should also be regarded as the scope of implementation of the present invention.

[0101] Embodiment 1:

[0102] This Embodiment 1 provides the technical solution of the high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS of the present invention, which is summarized as follows:

[0103] Design of the Integrated Navigation System:

[0104] Provide a combined navigation system, including an Inertial Navigation System (INS) and a Global Navigation Satellite System (GNSS), for achieving high-precision positioning of locomotives in complex environments.

[0105] The INS includes accelerometers and gyroscopes for measuring the acceleration and angular velocity of the locomotive.

[0106] The GNSS includes a satellite signal receiving antenna and a receiver for obtaining satellite positioning information.

[0107] Combined navigation algorithm:

[0108] Provide a combined navigation algorithm based on the Kalman filtering algorithm for fusing the data of INS and GNSS to improve positioning accuracy.

[0109] Data fusion is achieved by establishing a locomotive movement model and combining the inertial navigation results of INS and the positioning results of GNSS.

[0110] The quaternion method is used to describe the rotational motion of the carrier, simplifying the calculation process and improving the real-time performance and accuracy of the navigation system.

[0111] Coordinate system transformation:

[0112] Provide a coordinate system transformation method for transforming the data of INS and GNSS into a unified navigation coordinate system.

[0113] It includes the transformation of the Earth-Centered Earth-Fixed coordinate system (ECEF), the inertial coordinate system (i-system), the local navigation coordinate system (n-system), and the body coordinate system (b-system).

[0114] Error compensation:

[0115] Provide an error compensation method for compensating the errors in INS and GNSS systems, including the zero biases of accelerometers and gyroscopes, noise interference, etc.

[0116] By establishing an error model, the measured data is compensated to improve the accuracy and stability of the navigation system.

[0117] Simulation and experimental verification:

[0118] Provide a simulation and experimental verification method for verifying the performance of the combined navigation system.

[0119] Through the MATLAB simulation toolbox and combined with actual operation data, verify the positioning accuracy and reliability of the combined navigation system.

[0120] Embodiment 2:

[0121] The high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS specifically includes:

[0122] 1. System initialization

[0123] Hardware installation:

[0124] Reasonably install an inertial measurement unit (IMU) on the train locomotive to ensure that sensors such as accelerometers and gyroscopes are firmly installed and accurately positioned to accurately collect the motion data of the train.

[0125] Install the GNSS receiver antenna, select a location with good signal reception and no obstruction to ensure its effective reception of satellite signals.

[0126] Reliably connect the IMU and GNSS receiver to the data processing unit to ensure the stability and accuracy of data transmission.

[0127] Parameter setting:

[0128] According to the technical manual of the sensor and the previous calibration results, determine the key parameters of the inertial navigation system, such as accelerometer zero bias, gyroscope zero bias, and related matrix parameters.

[0129] Set the update interval T of the system to determine the time frequency of data processing and status update.

[0130] Determine the relevant matrices in the Kalman filter algorithm, including the state transition matrix A, the noise drive matrix B, the measurement matrix H, as well as the process noise covariance matrix Q and the measurement noise covariance matrix R.

[0131] Initial state determination

[0132] Initialize the GNSS system: Start the GNSS receiver to obtain the initial position information.

[0133] Set the initial parameters of the Kalman filter: including the state vector and the error covariance matrix.

[0134] State vector x0: including position, velocity, attitude (quaternion), accelerometer bias, and gyroscope bias.

[0135] Error covariance matrix P0: representing the uncertainty of the initial state.

[0136] 2. Data acquisition

[0137] Collect INS data:

[0138] Accelerometer and gyroscope data: Collect accelerometer and gyroscope data through the IMU and calculate the error-corrected locomotive acceleration and angular velocity.

[0139] The accelerometer after error correction measures the acceleration f of the locomotive. (Formula 1 in 2.2)

[0140] The gyroscope after error correction measures the angular velocity ω of the locomotive. (Formula 2 in 2.2)

[0141] Collect GNSS data:

[0142] Pseudorange measurement value: The satellite signal is received by the GNSS receiver to obtain the pseudorange measurement value z of the locomotive k .

[0143] The pseudorange measurement value represents the distance from the locomotive to the satellite. (Formulas 10 - 12 in 3.1)

[0144] 3. Inertial navigation solution

[0145] Attitude solution:

[0146] Using the corrected gyroscope angular velocity value, the attitude quaternion is updated and calculated according to formula (23). The updated attitude quaternion is normalized to ensure that it satisfies the properties of the quaternion

[0147] Calculate the rotation matrix from the body coordinate system to the navigation coordinate system through formula (28)

[0148] Position and velocity solution:

[0149] According to the rotation matrix The corrected accelerometer specific force value And the gravitational acceleration in the navigation coordinate system, use formulas (25) and (26) to calculate the position and velocity of the train at time k.

[0150] Further predict the position and velocity at time k + 1 according to formulas (29) and (30).

[0151] Sensor bias update:

[0152] Update the biases of the accelerometer and gyroscope according to formulas (31 - 32), where w f,k and w ω,k Are the random noises of the accelerometer and gyroscope respectively.

[0153] 4. Kalman filter fusion

[0154] Time update (prediction):

[0155] Based on the state estimate at time k, use the state equation to predict the state vector at time k + 1 (Formula 16 in 3.2)

[0156] Error covariance matrix (3.2 Formula 17)

[0157] Measurement update (calibration):

[0158] Calculate the measurement residual y k+1 (Difference between GNSS and INS): (3.3 Formula 38)

[0159] Calculate the Kalman gain K k+1 : (3.2 Formula 18)

[0160] Update the state vector (3.2 Formula 19)

[0161] Update the error covariance matrix P k+1 : (3.2 Formula 20)

[0162] 5. Result output

[0163] Output the state estimation value at time k + 1 obtained after Kalman filter fusion, including accurate position information (longitude, latitude, altitude), real-time speed (magnitude and direction), attitude parameters (heading angle, pitch angle, roll angle), and sensor bias data, etc., to the train control system.

[0164] Combined with Embodiment 1, this embodiment provides a detailed theoretical method and a description of the technical solution of the present invention, specifically including:

[0165] 1 Locomotive combined positioning principle

[0166] In response to the requirements of coal mine enterprises for "low cost and high reliability", the present invention selects the combination of inertial navigation and global navigation satellite system as a relatively common combined positioning method. By utilizing the complementary characteristics of each other's information, it can maintain the positioning of the locomotive through dead reckoning when the satellite signal is interfered. The advantages of combined navigation compared to satellite navigation mainly include: the combined navigation data refresh rate is higher, which is more meaningful for fast-moving scenarios, and the combined navigation can provide information such as heading, acceleration, and angular velocity. Since there are various sources of data in the combined navigation system, an accurate and stable data fusion algorithm is also an essential part.

[0167] 1.1 Combined navigation positioning principle

[0168] In the combined navigation mode, due to the problem of satellite signal occlusion in the complex operating environment of shunting locomotives, it cannot fully meet the requirements of coal mine enterprises for locomotive positioning in terms of accuracy, continuity, availability, and safety integrity. To ensure the adaptability of the shunting locomotive positioning function to the complex and changeable operating environment and signal reception conditions, the GNSS system is used to assist inertial navigation for combined positioning. The locomotive combined positioning principle based on GNSS / INS is as Figure 1as shown

[0169] 1.2 Flow Chart of Integrated Navigation Algorithm

[0170] The integrated positioning algorithm adopted by this system first establishes a locomotive movement model based on inertial navigation; secondly, based on the Kalman Filter (KF), the inertial navigation results are fused with the GNSS positioning results for integrated positioning; the integrated positioning process is as Figure 2 shown

[0171] 1.3 Commonly Used Coordinate Systems in Integrated Navigation

[0172] The GNSS navigation system measures by obtaining the relative position and velocity information between the satellite and the ground satellite receiver. The results measured by the inertial measurement unit in the INS system are based on the inertial coordinate system. The results required for GNSS / INS integrated navigation positioning adopted by the present invention are the motion state of the carrier relative to the earth. Different navigation methods and data results require different coordinate systems, such as Figure 3 shown in the following description of several commonly used coordinate systems of the navigation system:

[0173] (1) Earth-Centered Earth-Fixed Coordinate System

[0174] The Earth-Centered Earth-Fixed coordinate system is also called the e-system (ECEF). The e-system remains unchanged relative to the motion of the earth. This coordinate system takes the center of the earth as the origin. The Xe axis points to the prime meridian of the earth, and the Ze axis coincides with the earth's axis of rotation. Therefore, the Earth-Centered Earth-Fixed coordinate system has the same angular velocity of rotation as the earth compared to the inertial coordinate system.

[0175] (2) Inertial Coordinate System

[0176] The inertial coordinate system is also called the i-system. Geographically, the inertial navigation coordinate system generally refers to a coordinate system with the center of the earth as the origin, the earth's axis of rotation as the Z axis, and the line connecting the center of the earth and the vernal equinox as the Xi axis. Among them, the Zi axis points to the north pole of the earth, and the Yi axis is perpendicular to the plane where XZ is located and points to the equator. The XYZ axes satisfy the right-hand rule in the inertial coordinate system.

[0177] (3) Local Navigation Coordinate System

[0178] The local navigation coordinate system is also called the n-system. The origin of this coordinate system is located on the earth's reference surface, and it is also called the north-east-up coordinate system. Its X axis points to the north, the Y axis points to the east, and the Z axis is perpendicular to the XY plane and points to the sky. In addition, the position of the navigation coordinate system relative to the inertial coordinate system can be further calculated through longitude and latitude. In the present invention, since the height change of the locomotive in the z-axis direction can usually be ignored, it can be considered that the carrier moves in the XY plane.

[0179] (4) Body Coordinate System

[0180] The vehicle coordinate system is also called the b - system. The origin of this coordinate system is located at the center of the vehicle. The X - axis points in the forward direction of the vehicle, the Y - axis is perpendicular to the X - axis and points to the right of the X - axis direction, and the Z - axis is perpendicular to the XY plane and points from top to bottom.

[0181] 2 Inertial Navigation System Modeling

[0182] 2.1 Composition of INS Navigation System

[0183] An inertial navigation system usually consists of an inertial measurement unit, a calculation unit, a display DMI, etc. The inertial measurement unit includes accelerometers and gyroscopes. Among them, a 3 - axis gyroscope is used to measure the angular change of the target; a 3 - axis accelerometer is used to measure the acceleration change of the target. The calculation unit then performs calculations based on the measured data and outputs the dead - reckoning result. According to the installation method of the inertial measurement unit on the target to be measured, it is divided into a platform - type inertial navigation system (the inertial measurement unit is installed on the platform body of the inertial platform) and a strap - down inertial navigation system (the inertial measurement unit is directly installed on the vehicle). The latter omits the platform, the working conditions of the instruments are not good (affecting accuracy), and the calculation workload is large. The comparison between the strap - down and platform - type inertial navigation systems is shown in Table 1.

[0184] Table 1 - Comparison of Various Indexes between Strap - down and Platform - type Inertial Navigation

[0185]

[0186] Nowadays, with the development of measurement technology and technology, low - cost MEMS devices have been widely used in the navigation field. The strap - down inertial navigation system composed of MEMS inertial devices has the advantages of small size, low cost, and simple architecture. In integrated navigation, the strap - down inertial navigation integrated system is mainly composed of multiple gyroscope sensors and accelerometers. The inertial navigation solution unit on the vehicle processes the data, processes the original detection data output by sensors such as accelerometers and gyroscopes, and obtains the current position, traveling speed, and vehicle body attitude of the vehicle. Therefore, an inertial navigation system usually consists of an IMU (gyroscope, accelerometer, etc.) and a solution unit. The basic principle of strap - down inertial navigation is as Figure 4 shown.

[0187] In integrated navigation, the attitude solution of the vehicle is an indispensable step. The inertial navigation system decomposes the original specific force value output by the accelerometer into the corresponding navigation coordinate system through the direction cosine matrix.

[0188] 2.2 INS Navigation System Error Model

[0189] During the running of the vehicle body, due to noise interference and changes in the working environment, interference will be introduced into the measurement results of the IMU sensor, resulting in an error between the actual measured value and the true value. The error can be divided into a constant error and a random error. Usually, in an inertial navigation system, the error effects of the gyroscope and the accelerometer can be summarized as follows:

[0190]

[0191] Among them, represents the specific force output of the accelerometer, and b a represents the zero bias of the acceleration, and w a is the noise interference vector in the accelerometer; represents the angular velocity output of the gyroscope, and b g represents the zero bias of the gyroscope, and w g is the noise interference received in the measurement of the gyroscope sensor. The zero bias described in the formula includes the static zero bias and the dynamic zero bias. The static zero bias is usually marked when the sensor leaves the factory, while the dynamic zero bias is affected by environmental factors such as temperature and repeatability.

[0192] 2.3 Strapdown Inertial Navigation Equation

[0193] To facilitate the description of the running process of the vehicle body, the present invention optionally selects the Earth-centered Earth-fixed coordinate system as the reference coordinate system for the movement of the vehicle body. Considering the influence of the Earth's rotation, the projection of the running speed of the vehicle body onto the inertial coordinate system can be expressed as:

[0194]

[0195] Among them, is the projection of the current position of the vehicle body onto the i-system, represents the angular velocity of the vehicle body relative to the i-system. By differentiating the above formula, the acceleration of the vehicle body relative to the i-system can be obtained:

[0196]

[0197] In inertial navigation, this acceleration value can be considered as the sum of the gravitational acceleration g at the current position and the specific force value output by the acceleration sensor. Under the condition that the Earth's angular velocity remains unchanged (i.e., ), the following can be obtained:

[0198]

[0199] If the vehicle body travels a large range around the Earth, the relative movement between the coordinate axes will be considered. However, usually, each driving distance of the locomotive can be considered as a small-range scenario. Therefore, relative to the Earth-centered Earth-fixed coordinate system, the acceleration of the vehicle body in the navigation coordinate system can be expressed as:

[0200]

[0201] Similarly, the angular velocity of the vehicle body can be obtained as follows:

[0202]

[0203] wherein, represents the rotational angular velocity of the vehicle body relative to the n system; is the measured value of the gyroscope of the vehicle body relative to the inertial coordinate system.

[0204] In addition to the changes in position and velocity, the inertial navigation system also takes into account the attitude changes of the vehicle body. The rotational angular velocity can be measured by a gyroscope. When the rotational angular velocity changes between coordinate systems and the influence of the Earth's rotation are not considered, the attitude change of the vehicle body can be expressed as:

[0205]

[0206] 3 Modeling of Integrated Navigation System

[0207] 3.1 Composition and Measurement Principle of GNSS System

[0208] (1) Basic Composition of GNSS System

[0209] The GNSS system mainly consists of three parts: the space segment, the user segment, and the control segment. Among them, the space segment is composed of multiple satellites distributed near the Earth's orbit. In the GNSS navigation system, there are a total of 24 satellites moving on 6 nearly circular orbits with a radius of 26,580 kilometers, which enables ground observation equipment to obtain signals from more than 4 satellites simultaneously.

[0210] The control segment mainly includes monitoring stations, uplink injection stations, and master control stations. The main function of the monitoring stations is to use the receiving equipment and computing components they are equipped with to obtain the observation data of GNSS satellites, preprocess the data through a computer, and finally send the processing results to the control station. The main function of the control station is to analyze the status of the satellites in the current space segment based on the data downloaded from the monitoring stations. If the satellites need to be attitude-adjusted, the control station needs to send the control information to the injection stations, and the injection stations will organize the information and send it to the satellite body to achieve the active control function of the satellites.

[0211] The user segment mainly obtains the positioning data of GNSS through receiving equipment. The GNSS signal reaches the receiver through the antenna. The receiver preprocesses the obtained ephemeris data, which also includes information such as pseudorange, and finally transmits the data to the navigation processing unit. The navigation unit uses algorithms to calculate the position, velocity, and other information of the current user segment.

[0212] (2) Measurement Principle of GNSS System

[0213] Currently, there are two methods for GNSS navigation systems to measure distance. The first method uses pseudorange information for measurement, and the second method uses the carrier phase method for measurement. Among them, pseudorange measurement calculates using the clock time difference. This is because there is usually no high-precision atomic clock in the receiving device. At the same time, when the satellite signal passes through the atmosphere and clouds, it will be affected by various factors and cause a certain time delay error. Multiplying this time error by the speed of light can obtain the pseudorange. This method has a low calculation cost and high real-time performance. Carrier phase measurement calculates the distance between the satellite and the receiver by comparing the signal phase difference between them. This method has low real-time performance but high positioning accuracy. In this design, considering the high real-time requirement of the locomotive positioning for the system, the present invention mainly elaborates on the pseudorange measurement method. Assume that the time when the receiver acquires the GNSS signal is t r , considering the deviation of the receiver clock, use Δt r to represent this unknown deviation amount. The actual time of the current system is t. Then the relationship between the time when the receiver acquires the signal and the actual time can be expressed as follows:

[0214] t r = t + Δt r (9)

[0215] Similarly, there is also a certain deviation Δt s in the atomic clock of the GNSS satellite. Therefore, the time when the GNSS satellite sends the signal can be expressed as

[0216] t s = t + Δt s (10)

[0217] Then according to the definition of pseudorange, the pseudorange observation value d':

[0218] d' = c * τ' = c * (t r - t s ) + c * (Δt r - Δt s )(11)

[0219] In the actual navigation process, the user can simultaneously observe at least four satellites through the receiver. At the same time, the clock deviation of the satellite is known. Assume that the coordinates of satellite i are (x i , y i , z i ), and the current coordinates of the receiver are (x, y, z). The pseudorange observation value of each satellite can be rewritten as:

[0220]

[0221] By solving the above equations, the current position information of the receiver and the clock deviation of the receiver can be obtained.

[0222] 3.2 Kalman Filter Algorithm

[0223] As early as the 1960s, Rudolph E. Kalman proposed a method of using recursive iteration to solve the state estimation problem in linear systems. This algorithm is also known as the Kalman filter (KF). The Kalman filter takes the minimum state mean square error as the estimation principle and generates the state estimate at the current moment based on the state measurement value at the current moment and the state estimate value at the previous moment. Compared with previous algorithms such as Wiener filtering, this filter requires less data volume and reduces the computational cost of the filtering system. Therefore, it has been widely used in engineering.

[0224] When satellite signals pass through the ionosphere or troposphere, certain delays will occur. At the same time, due to the incomplete alignment of the clocks of the receiver and the satellite, there is a certain clock error. These errors affect the measurement accuracy of GNSS. In addition, during the driving process of a locomotive, especially when passing through sections blocked by buildings or trees, the number of satellites that GNSS can identify will decrease and the signal will be attenuated by electromagnetic interference. In more serious cases, the GNSS signal will be completely lost. Therefore, the data fusion of multiple sensors can achieve complementary advantages. For example, the INS system has a relatively fast update speed but is prone to error accumulation, which may lead to the divergence of the navigation result. The GNSS signal has a slow update rate, but the positioning result can remain stable for a long time. Therefore, the combination of the inertial navigation system and GNSS can achieve long-term, fast and stable positioning effects.

[0225] Each cycle of the discrete Kalman filter can be divided into two processes: time update and measurement update, as Figure 5 shown. In the time update process, the estimate value at the previous moment is used as the input, and the one-step prediction value at the current moment is calculated through the state space model, including one-step state prediction and error covariance matrix prediction.

[0226] In the measurement update, the measurement value of the sensor at the current moment is used for update. The estimation gain is calculated through the measurement matrix and the noise vector, and the state and error covariance matrix are updated.

[0227] Consider the following linear system:

[0228] X k = A·X k-1 + B·W k (13)

[0229] Z k = H·X k + V k (14)

[0230] where A and B are the state transition matrix and the noise driving matrix of the system respectively; H and V represent the measurement matrix and the measurement noise vector respectively.

[0231]

[0232] It is usually assumed that the system noise W and the measurement noise V are Gaussian white noises and satisfy the following relationship:

[0233]

[0234] where Q is the process noise variance matrix and R is the measurement noise variance matrix.

[0235] In summary, the basic equations of the Kalman filter are as follows:

[0236] The time update is:

[0237]

[0238] The measurement update:

[0239]

[0240] 3.3 GNSS / INS integrated navigation filtering algorithm based on quaternion

[0241] In this system, quaternions are used to describe the rotational motion of the carrier, where represents the projection of the rotational angular velocity of the carrier relative to the navigation coordinate system in the carrier coordinate system. At the same time, during the actual navigation process, the angular velocity of the Earth's rotation may be contaminated by the noise of the sensors, making it difficult for the inertial navigation sensors to detect. Therefore, in order to simplify the derivation process, the present invention will no longer consider the influence of the Earth's centripetal force and the Coriolis force. Thus:

[0242]

[0243] where represents the projection of the rotational angular velocity relative to the inertial coordinate system. (Note that usually, the original vectors such as the angular velocity in the present invention are all selected with the carrier coordinate system as the reference vector).

[0244]

[0245] Therefore, the above formula can be further simplified to:

[0246]

[0247] Considering the position update and the velocity update, the inertial navigation equation can be written as follows:

[0248]

[0249] where g n represents the gravitational acceleration in the n - system (navigation coordinate system); is the rotation matrix, representing the transformation relationship between the vehicle body coordinate system and the navigation coordinate system, which can be expressed as follows:

[0250]

[0251] Considering that the inertial sensors may be affected by noise in the actual process, therefore, usually, the angular velocity and specific force returned by the accelerometers and gyroscopes are compensated before processing. As described above, the update speed of the inertial sensors is much faster than that of the GNSS signals. The state update rate of the INS is usually in the millisecond level. Therefore, when the update rate is relatively fast, assuming the update interval is T, the state update equation of the INS can be written as follows:

[0252] r n,k+1 = r n,k + v n,k ·T(29)

[0253]

[0254] The state vector of the integrated navigation system is selected as:

[0255]

[0256] where, represents the three - dimensional position vector, represents the three - dimensional velocity vector, represents the attitude vector; is the accelerometer bias; is the gyroscope measurement bias.

[0257] The present invention adopts the extended Kalman nonlinear filtering algorithm. After Taylor expansion of the state estimation, the first - order linearized equation is obtained:

[0258] δx k+1 = A·δx k + w k (34)

[0259] Expanding the discretized state - space equation of the above formula, the system state error model of the loosely - coupled integrated navigation can be obtained:

[0260]

[0261] In the above formula, represents the attitude rotation matrix at time k.

[0262] In the tightly - coupled integrated navigation, the receiver clock bias cδt needs to be added to the state vector kand clock drift error Its corresponding state error space model is as follows:

[0263]

[0264] As can be seen from the above, in the GNSS system, the measurement model of the receiver is

[0265]

[0266] where represents the pseudorange measurement value of the j-th satellite. (x j , y j , z j ) represents the three-dimensional position information of the j-th satellite in the Earth-centered Earth-fixed coordinate system.

[0267] In the loose integration, usually the three-dimensional position and velocity are used as GNSS measurement values, and the difference is taken with the predicted position and velocity provided by the INS system, and the deviation is used as the measurement update input of the filter. It can be seen that the measurement model of the loose integration navigation is simpler and the computational load is smaller, but there is also the disadvantage that the minimum number of satellites is four. The measurement model of the loose integration can be expressed as follows:

[0268]

[0269] In the tight integration navigation, due to the consideration of the pseudorange information, its measurement model usually uses the pseudorange measurement value and the pseudorange estimate value of the INS as the measurement update input of the filter. At the same time, the number of satellites in the tight integration navigation system is variable. Its advantage is that the coupling degree of this combination method is high, and the positioning accuracy has been greatly improved compared with the loose integration, but the subsequent computational load also increases. The measurement model of the tight integration navigation is expressed as follows:

[0270]

[0271] 4 Simulation experiments of integrated navigation system

[0272] Through the MATLAB toolbox, combined with the simulation data and the working principle of the GNSS / INS integrated navigation system, the inertial navigation is calculated using the designed simulation gyroscope and accelerometer information; the tight integration navigation and the loose integration navigation are calculated according to the satellite position, pseudorange, and pseudorange rate. The trajectories obtained by solving under the two combination modes are as Figure 6 shown. The blue and red lines in the figure represent the trajectories under the two combination modes respectively, and the yellow line represents the standard reference trajectory.

[0273] In order to quantitatively analyze the errors of the navigation results of the two combination modes, the errors of the velocity information and the position information as well as the position and velocity are plotted as shown in Figures 7 and 8.

[0274] As shown in Figure 8, due to the additional clock error in the tight combination, the error will be slightly larger at the beginning of the combined navigation. After the clock error converges, the errors of the tight combination and the loose combination are basically the same. This shows that the navigation results of the two combined mode systems are not much different. Considering that the structure of the tight combination system is relatively complex, the computing performance requirements of the system are too high, and if the tight combination system wants to be perfectly implemented, an RTK-level satellite positioning module must be used, which increases the cost of the overall system equipment. In summary, the present invention selects a loose combination mode with a relatively simple structure.

[0275] It can be understood that: the present invention establishes a positioning model for a GNSS / INS combined navigation system, first introduces the working principle and algorithm flow of the combined navigation system, and conducts an overall design of the combined positioning system; explains the four common coordinate systems in the navigation system, laying the foundation for subsequent modeling; then describes the composition of the inertial navigation system, selects a strapdown inertial navigation platform by comparing the advantages and disadvantages of the inertial navigation installation method, and conducts navigation and error modeling for the strapdown inertial navigation system; then, the composition and measurement principle of the satellite navigation system are introduced in detail. Finally, according to the Kalman filtering algorithm, a filtering algorithm for INS / GNSS combined navigation based on quaternions is proposed, and the derivation of the navigation system measurement model under the two data fusion modes is completed, and the position and velocity errors under the two modes are compared and analyzed through simulation data.

[0276] Embodiment 3:

[0277] This embodiment provides a terminal device, which includes a processor and a memory, wherein the memory is used to store a computer program, the computer program includes program instructions, and the processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, which is suitable for implementing one or more instructions, and is specifically suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function; the processor described in the embodiment of the present invention can be used for the operation of a high-precision positioning method for train locomotives based on a loose combination mode of GNSS and INS, including the following steps:

[0278] IMU data acquisition and GNSS data acquisition, input the accelerometer and gyroscope data, correct the errors of the acquired data, and output the corrected acceleration and angular velocity; input the satellite signals received by the GNSS receiver, extract the pseudorange measurement values, and output the pseudorange measurement values;

[0279] Inertial navigation solution. First, perform attitude solution. Input the corrected angular velocity, update the quaternion according to the gyroscope angular velocity to obtain the attitude information, and convert it from the body coordinate system to the navigation coordinate system through the rotation matrix, and output the quaternion and the rotation matrix; then perform position and velocity solution. Input the corrected acceleration data, gravitational acceleration, and rotation matrix, calculate the train position and velocity based on the acceleration data and the rotation matrix, use gravity compensation, and output the current train position and velocity; finally, perform sensor bias update. Input the noise of the accelerometer and gyroscope, update the sensor bias using the random noise, and output the updated sensor bias value;

[0280] Kalman filter fusion. First, perform time update. Input the state estimate at the current moment, predict the state vector and error covariance matrix at the next moment through the state equation, and output the predicted state vector and covariance matrix; then perform measurement update. Input the difference between the GNSS measurement value and the INS measurement value, calculate the Kalman gain, and update the state vector and error covariance matrix through the Kalman filter formula, and output the corrected state vector and the updated error covariance matrix.

[0281] Example 4:

[0282] This embodiment provides a storage medium, specifically a computer-readable storage medium (Memory). The computer-readable storage medium is the memory device in the terminal device and is used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, and this storage space stores the operating system of the terminal. And, one or more instructions suitable for being loaded and executed by the processor are also stored in this storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory.

[0283] One or more instructions stored in a computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the method for high-precision positioning of a train locomotive based on the loose combination mode of GNSS and INS in the above embodiments; the one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps:

[0284] IMU data acquisition and GNSS data acquisition, input the accelerometer and gyroscope data, correct the acquired data, and output the corrected acceleration and angular velocity; input the satellite signals received by the GNSS receiver, extract the pseudorange measurement values, and output the pseudorange measurement values;

[0285] Inertial navigation solution. First, perform attitude solution, input the corrected angular velocity, update the quaternion according to the gyroscope angular velocity to obtain the attitude information, and convert from the body coordinate system to the navigation coordinate system through the rotation matrix, and output the quaternion and the rotation matrix; then perform position and velocity solution, input the corrected acceleration data, gravitational acceleration, and rotation matrix, calculate the train position and velocity based on the acceleration data and the rotation matrix, use gravity compensation, and output the current train position and velocity; finally, perform sensor bias update, input the noise of the accelerometer and gyroscope, use the random noise to update the sensor bias, and output the updated sensor bias value;

[0286] Kalman filter fusion. First, perform time update, input the state estimate at the current moment, predict the state vector and the error covariance matrix at the next moment through the state equation, and output the predicted state vector and the covariance matrix; then perform measurement update, input the difference between the GNSS measurement value and the INS measurement value, calculate the Kalman gain, and update the state vector and the error covariance matrix through the Kalman filter formula, and output the corrected state vector and the updated error covariance matrix.

[0287] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0288] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and combinations of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to produce a machine, such that the instructions executed by the processor of the computer or other programmable data processing device generate a means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or in multiple blocks.

[0289] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including an instruction means that implements the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or in multiple blocks.

[0290] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or in multiple blocks.

[0291] The preferred embodiments of the present invention have been described in detail above, but the present invention is not limited to the above embodiments. Within the knowledge scope of those of ordinary skill in the art, various changes can be made without departing from the gist of the present invention.

[0292] Many other changes and modifications can be made without departing from the concept and scope of the present invention. It should be understood that the present invention is not limited to specific embodiments, and the scope of the present invention is defined by the appended claims.

Claims

1. A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS, characterized in that, The method includes: IMU data acquisition and GNSS data acquisition. Input the accelerometer and gyroscope data, correct the errors of the acquired data, and output the corrected acceleration and angular velocity. Input the satellite signals received by the GNSS receiver, extract the pseudorange measurement values, and output the pseudorange measurement values. Inertial navigation solution. First, perform attitude solution. Input the corrected angular velocity, update the quaternion according to the gyroscope angular velocity to obtain the attitude information, and convert it from the body coordinate system to the navigation coordinate system through the rotation matrix, and output the quaternion and the rotation matrix. Then, perform position and velocity solution. Input the corrected acceleration data, gravitational acceleration, and rotation matrix, calculate the train position and velocity based on the acceleration data and the rotation matrix, use gravity compensation, and output the current train position and velocity. Finally, perform sensor bias update. Input the noise of the accelerometer and gyroscope, update the bias of the sensor using the random noise, and output the updated sensor bias value. Kalman filter fusion. First, perform time update. Input the state estimate at the current moment, predict the state vector and error covariance matrix at the next moment through the state equation, and output the predicted state vector and covariance matrix. Then, perform measurement update. Input the difference between the GNSS measurement value and the INS measurement value, calculate the Kalman gain, and update the state vector and error covariance matrix through the Kalman filter formula, and output the corrected state vector and the updated error covariance matrix.

2. A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS according to claim 1, characterized in that Before the data acquisition stage, the method further includes: hardware installation and initialization, specifically including: Hardware installation IMU installation: Install the accelerometer and gyroscope, and the IMU sensor is ready to start collecting acceleration and angular velocity data. GNSS receiver installation: Install the GNSS receiver antenna to ensure no obstruction, and the GNSS receiver starts to receive satellite signals and provides position and pseudorange measurement data. Data connection: Connect the IMU, GNSS, and the data processing unit, and transmit the IMU and GNSS sensor data to the data processing unit. Parameter setting According to the technical manual of the sensor and the previous calibration results, determine the key parameters of the inertial navigation system, such as the accelerometer zero bias, gyroscope zero bias, and related matrix parameters. Set the update interval T of the system to determine the time frequency of data processing and state update. Determine the relevant matrices in the Kalman filter algorithm, including the state transition matrix A, the noise drive matrix B, the measurement matrix H, and the process noise covariance matrix Q and the measurement noise covariance matrix R. Initial state determination Initialize the GNSS system: Start the GNSS receiver to obtain the initial position information. Set the initial parameters of the Kalman filter: including the state vector and the error covariance matrix. State vector x0: including position, velocity, attitude, accelerometer bias, and gyroscope bias. Error covariance matrix P0: represents the uncertainty of the initial state.

3. A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS according to claim 1, characterized in that After Kalman filter fusion, the method further includes: inputting the state estimation result after Kalman filter fusion, sorting and formatting the fused data, and outputting the final accurate position, velocity, attitude parameters, and sensor deviation data to the train control system to finally obtain the accurate train motion state.

4. A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS according to claim 1, characterized in that, Performing error correction on the collected data and outputting the corrected acceleration and angular velocity, including: The accelerometer measures the acceleration f of the locomotive after error correction: The gyroscope measures the angular velocity ω of the locomotive after error correction: Among them, represents the specific force output of the accelerometer, and b a represents the bias of the acceleration, and w a is the noise interference vector in the accelerometer; represents the angular velocity output of the gyroscope, and b g represents the bias of the gyroscope, and w g is the noise interference suffered in the measurement of the gyroscope sensor. The bias described in the formula includes the static bias and the dynamic bias.

5. A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS according to claim 1, characterized in that, The pseudorange measurement value represents the distance from the locomotive to the satellite. Extracting the pseudorange measurement value specifically includes: Assume that the time when the receiver acquires the GNSS signal is t r , considering the deviation of the receiver clock, use Δt r to represent this unknown deviation amount. If the actual time of the current system is t, the relationship between the time when the receiver acquires the signal and the actual time can be expressed as follows: t r = t + Δt r (3) Similarly, there is also a certain deviation Δt in the atomic clocks of GNSS satellites. s , so the time when the GNSS satellite sends a signal can be expressed as: t s = t + Δt s (4) Then, according to the definition of the pseudorange, the pseudorange observation value d': d′ = c * τ′ = c * (t r - t s ) + c * (Δt r - Δt s ) (5) In the actual navigation process, the user observes more than four satellites simultaneously through the receiver, and the clock biases of the satellites are known. Assume that the coordinates of satellite i are (x i , y i , z i ), the current coordinates of the receiver are (x, y, z), and the pseudorange observation value of each satellite can be rewritten as: Obtaining the current position information of the receiver and the clock deviation of the receiver by solving the above equations.

6. A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS according to claim 1, characterized in that, In the inertial navigation solution, the attitude solution specifically includes: Using the corrected gyroscope angular velocity value, performing the update calculation of the attitude quaternion according to formula (7), Normalize the updated attitude quaternion to ensure that it satisfies the properties of quaternions; calculate the rotation matrix from the vehicle coordinate system to the navigation coordinate system through formula (8). The position and velocity solution specifically includes: According to the rotation matrix The corrected accelerometer specific force value and the gravitational acceleration in the navigation coordinate system, use equations (9) and (10) to calculate the position and velocity of the train at time k, and further predict the position and velocity at time k+1 according to equations (11) and (12); r n,k+1 = r n,k + v n,k · T (11) The sensor bias update specifically includes: updating the biases of the accelerometer and gyroscope according to formulas (13-14), where w f,k and w ω,k are the random noises of the accelerometer and gyroscope respectively; 7. A high-precision positioning method for train locomotives based on the loose integration mode of GNSS and INS according to claim 1, characterized in that In the Kalman filter fusion, the time update specifically includes: Based on the state estimation value at time k, use the state equation to predict the state vector at time k+1 Error covariance matrix The measurement update specifically includes: Calculate the measurement residual y k+1 , which is the difference between GNSS and INS: Calculate the Kalman gain K k+1 : Update status vector Update the error covariance matrix P k+1 :

8. A high-precision system for train locomotives based on the loose integration mode of GNSS and INS, characterized in that, The system includes: a hardware installation and initialization module, an inertial navigation solution module, a Kalman filter fusion module, and an output module; The hardware installation and initialization module: includes a hardware installation, parameter setting, and initial state determination sub-module; The inertial navigation solution module: First, perform attitude solution, input the corrected angular velocity, update the quaternion according to the gyroscope angular velocity to obtain the attitude information, and convert from the body coordinate system to the navigation coordinate system through the rotation matrix, outputting the quaternion and the rotation matrix; then perform position and velocity solution, input the corrected acceleration data, gravitational acceleration, and rotation matrix, calculate the train position and velocity based on the acceleration data and the rotation matrix, use gravity compensation, and output the current train position and velocity; finally, perform sensor deviation update, input the noise of the accelerometer and the gyroscope, update the deviation of the sensor using the random noise, and output the updated sensor deviation value; The Kalman filter fusion module: First, perform time update, input the state estimation at the current moment, predict the state vector and error covariance matrix at the next moment through the state equation, and output the predicted state vector and covariance matrix; then perform measurement update, input the difference between the GNSS measurement value and the INS measurement value, calculate the Kalman gain, and update the state vector and error covariance matrix through the Kalman filter formula, outputting the corrected state vector and the updated error covariance matrix; The output module is used to input the state estimation result after Kalman filter fusion, sort and format the fused data, and output the final accurate position, velocity, attitude parameters, and sensor deviation data to the train control system to finally obtain the accurate train motion state; The hardware installation sub-module: IMU Installation: Install the accelerometer and gyroscope, and the IMU sensor is ready to start collecting acceleration and angular velocity data; GNSS Receiver Installation: Install the GNSS receiver antenna to ensure no obstruction, and the GNSS receiver starts receiving satellite signals to provide position and pseudorange measurement data; Data Connection: Connect the IMU, GNSS to the data processing unit, and transmit the IMU and GNSS sensor data to the data processing unit; Parameter Setting Sub-module: Determine the key parameters of the inertial navigation system according to the technical manual of the sensor and the previous calibration results, such as accelerometer zero bias, gyroscope zero bias, and related matrix parameters; Set the update interval T of the system to determine the time frequency of data processing and status update; Determine the relevant matrices in the Kalman filter algorithm, including the state transition matrix A, the noise drive matrix B, the measurement matrix H, as well as the process noise covariance matrix Q and the measurement noise covariance matrix R; IMU Data Acquisition and GNSS Data Acquisition, input the accelerometer and gyroscope data, correct the errors of the collected data, and output the corrected acceleration and angular velocity; input the satellite signals received by the GNSS receiver, extract the pseudorange measurement values, and output the pseudorange measurement values; Initial State Determination Sub-module: Initialize the GNSS system: Start the GNSS receiver to obtain the initial position information; Set the initial parameters of the Kalman filter: including the state vector and the error covariance matrix; State vector x0: including position, velocity, attitude, accelerometer bias, and gyroscope bias; Error covariance matrix P0: represents the uncertainty of the initial state.

9. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements a high-precision method for train locomotives based on the loose combination mode of GNSS and INS according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium. When the program is executed by the processor, it implements a high-precision method for train locomotives based on the loose combination mode of GNSS and INS according to any one of claims 1 to 7.

Images