GNSS (Global Navigation Satellite System) and INS (Inertial Navigation System) integrity monitoring method and system based on 5G and multilayer fusion mechanism

Abstract

The invention relates to the technical field of multi-source data fusion analysis and monitoring, in particular to a GNSS (Global Navigation Satellite System) and INS (Inertial Navigation System) integrity monitoring method and system based on 5G and a multilayer fusion mechanism, and the method comprises the steps: obtaining multi-source data; preprocessing the data based on the data type of the multi-source data, and performing multi-source hierarchical fusion processing on the preprocessed multi-source data to obtain uncertainty information; based on the uncertainty information, adopting a statistical method to calculate a protection level; performing consistency evaluation on the multi-source data to obtain an evaluation result, and calculating a positioning confidence coefficient according to the uncertainty information output by the neural network model and the evaluation result; and obtaining multi-dimensional related data, carrying out analysis learning on the multi-dimensional related data by using the neural network, and outputting a risk prediction value. The invention provides a three-layer fusion mechanism, from first-layer fusion and second-layer fusion to introduction of the neural network for deep feature extraction and fusion, the fusion precision and reliability are gradually improved, and the positioning navigation performance is improved.

Description

Technical Field

[0001] The present invention relates to the field of multi-source data fusion analysis and monitoring technology, and in particular to a GNSS and INS integrity monitoring method and system based on 5G and a multi-layer fusion mechanism. Background Art

[0002] In positioning and navigation scenarios with high safety requirements, such as autonomous driving, aerospace and other fields, accurate and reliable positioning information is crucial. The global GNSS navigation satellite system and the INS inertial navigation system are commonly used positioning methods, but they each have limitations; GNSS signals are easily affected by obstruction and interference, resulting in reduced positioning accuracy or even positioning failure; INS has the problem of error accumulation over time; the development of 5G technology has brought new opportunities for positioning and navigation. It has the characteristics of high bandwidth, low latency, and large connections, and can provide high-precision positioning information.

[0003] Existing technologies make it difficult to fully tap the potential value of multi-source data. Therefore, there is an urgent need for a GNSS and INS integrity monitoring solution based on 5G and a multi-layer fusion mechanism, aiming to give full play to the advantages of 5G, GNSS and INS and improve the accuracy and reliability of positioning and navigation. Summary of the Invention

[0004] The purpose of the present invention is to address the problems existing in the background technology and propose a GNSS and INS integrity monitoring method and system based on 5G and multi-layer fusion mechanism.

[0005] The technical solution of the present invention is a GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism, comprising the following steps:

[0006] Acquire GNSS position data, INS position data, and 5G position data to obtain multi-source data;

[0007] Preprocess the data based on the data type of multi-source data, and perform multi-source layered fusion processing on the preprocessed multi-source data to obtain uncertainty information;

[0008] Based on uncertainty information, statistical methods are used to calculate the protection level;

[0009] Conduct consistency assessment on multi-source data to obtain assessment results, and calculate the positioning reliability based on the uncertainty information output by the neural network model and the assessment results;

[0010] It is used to obtain multi-dimensional correlation data of GNSS signals, multi-dimensional correlation data of INS signals and multi-dimensional correlation data of 5G signals, use neural network multi-dimensional correlation data for analysis and learning, and output risk prediction values.

[0011] Preferably, the GNSS position data includes at least GNSS position information and speed information; the INS position data includes at least INS position information and speed information; and the 5G position data includes at least initial position information.

[0012] Preferably, the method for performing multi-source layered fusion processing on the pre-processed multi-source data includes:

[0013] The weighted average algorithm is used to perform the first-layer fusion of the pre-processed GNSS position and velocity information with the INS position and velocity information to obtain the first fusion result;

[0014] The extended Kalman filter algorithm is used to take the initial position information as observation information and the first fusion result as state prediction information. The state is estimated and updated based on the observation information and the state prediction information to obtain the second fusion result.

[0015] Deep feature extraction and fusion are performed based on GNSS position data, INS position data and the second fusion result to output more accurate position information, speed information and attitude information, and integrate them into uncertainty information.

[0016] Preferably, the first layer fusion specifically includes determining the position vector P according to the GNSS position information and the speed information. GNSS and V GNSS , determine P according to INS position information and speed information INS and V INS , and fuse the position vector and velocity vector based on the following formula:

[0017] P fusion =w GNSS ×P GNSS +w INS ×P INS ;

[0018] V fusion =w GNSS ×V GNSS +w INS ×V INS ;

[0019] Where w GNSS and w INS are GNSS weight and INS weight, respectively, which are dynamically adjusted based on the measurement accuracy of GNSS and INS.

[0020] Preferably, the second layer fusion specifically includes taking the position and velocity information after the first layer fusion as the initial value of the state vector, and setting the state vector X = [x, y, z, v1, v2, v3] T , where (x, y, z) represents the position coordinates and (v1, v2, v3) represents the corresponding velocity components;

[0021] The initial position based on 5G is used as the observation vector G = [x 5G ,y 5G ,z 5G ] T , where (x 5G ,y 5G ,z 5G )The initial position coordinates obtained by 5G positioning.

[0022] Preferably, the second layer fusion further includes constructing a state transfer matrix based on a uniform motion model of the carrier;

[0023] According to the transfer matrix and the state vector at the previous moment, the state vector at the current moment is predicted and marked as the predicted state vector. The prediction calculation formula is as follows:

[0024] X k|k-1 =ZJ·X k-1 Where, X k|k-1 is the predicted state vector at the current moment; X k-1 is the state vector of the previous moment; k is the moment number based on the timestamp; ZJ is the state matrix.

[0025] Preferably, the second layer fusion also includes obtaining the state vector obtained after the first layer fusion at the previous moment and the predicted state vector, and calculating the error covariance matrix Z at the previous moment. k-1 , predict the error covariance matrix Z at the current moment k|k-1 , the calculation formula is as follows:

[0026] Z k|k-1 =ZJ·Z k-1 +Q; where Q is the process noise covariance matrix, which is calculated based on the predicted process noise data;

[0027] According to the error covariance matrix Z at the current moment k|k-1 Calculate the Kalman gain K k , the calculation formula is as follows:

[0028] K k =Z k|k-1 ·H T ·(H·Z k|k-1 ·H T +R) -1 ;

[0029] Where H is the observation matrix; R is the observation noise covariance matrix.

[0030] Preferably, the second layer fusion also includes the Kalman gain K k , observation vector G = [x 5G ,y5G ,z 5G ] T and the predicted state vector X k|k-1 , update the state vector X k , the calculation formula is as follows:

[0031] X k =X k|k-1 +K k (G k -H·X k|k-1 );

[0032] According to the Kalman gain K k and the error covariance matrix Z at the current moment k|k-1 Update the error covariance matrix Z k , the calculation formula is as follows:

[0033] Z k =(IK k ·H)·Z k|k-1 ;Where, I is the unit matrix;

[0034] The state vector X of the position and velocity information after the second layer fusion k Output as the second fusion result;

[0035] Deep feature extraction and fusion are performed based on GNSS position data, INS position data and the second fusion result to output more accurate position information, speed information and attitude information, and integrate them into uncertainty information.

[0036] Preferably, deep feature extraction and fusion includes input layer setting, hidden layer architecture, output layer setting and neural network training;

[0037] The input layer setting includes obtaining the number of features of GNSS position data, INS position data, and 5G position data and the second fusion result, and summing the number of features to determine the number of input neurons;

[0038] The hidden layer architecture adopts a multi-hidden layer deep neural network structure. Each hidden layer uses the ReLU activation function to introduce nonlinear factors and enhance the network's ability to express complex data relationships.

[0039] The output layer setting includes three output neurons, which respectively output accurate position information, speed information, and posture information; among them, the output layer uses a linear activation function to directly output the predicted numerical results.

[0040] The present invention also discloses a GNSS and INS integrity monitoring system based on 5G and multi-layer fusion mechanism, which applies the above-mentioned GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism, specifically including:

[0041] Data monitoring and acquisition module, used to obtain GNSS position data, INS position data and 5G position data to obtain multi-source data;

[0042] The data preprocessing and fusion module is used to preprocess the data based on the data type of multi-source data, and perform multi-source layered fusion processing on the preprocessed multi-source data to obtain uncertainty information;

[0043] The protection level definition module calculates the protection level using statistical methods based on uncertainty information;

[0044] The location reliability calculation module is used to evaluate the consistency of multi-source data, obtain the evaluation results, and calculate the location reliability based on the uncertainty information output by the neural network model and the evaluation results;

[0045] The relevant data collection and risk prediction module is used to obtain multi-dimensional relevant data of GNSS signals, multi-dimensional relevant data of INS signals and multi-dimensional relevant data of 5G signals, use neural network multi-dimensional relevant data for analysis and learning, and output risk prediction values.

[0046] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects:

[0047] The present invention proposes a three-layer fusion mechanism, from the first layer fusion, the second layer fusion to the deep fusion, gradually improving the accuracy and reliability of the fusion. Especially in the deep fusion layer, the neural network is introduced for deep feature extraction and fusion, which can better handle the complex relationship between multi-source data and improve positioning and navigation performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a method block diagram of embodiment 1 proposed by the present invention. DETAILED DESCRIPTION

[0049] Example 1, as Figure 1 As shown, the GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism proposed in the present invention includes the following steps:

[0050] Acquire GNSS position data, INS position data, and 5G position data to obtain multi-source data;

[0051] GNSS position data includes at least GNSS position information and velocity information; INS position data includes at least INS position information and velocity information; 5G position data includes at least initial position information; GNSS position data can be obtained by using a multi-frequency, multi-constellation GNSS receiver to receive signals from different satellite constellations and convert and calculate the signals; INS data is obtained by collecting the acceleration and angular velocity information of the carrier through an inertial measurement unit and performing position calculation; 5G data uses a 5G communication module to receive signals from a 5G base station and obtain measurement values ​​such as the base station's position information, signal strength, arrival time, and arrival angle;

[0052] Data preprocessing is performed based on the data type of multi-source data, including denoising GNSS position data using an adaptive filtering algorithm; performing zero bias calibration and scale factor calibration on INS position data; and performing multipath suppression and non-line-of-sight identification on 5G position data. It should be noted that the above data preprocessing methods are all existing technical means and will not be elaborated on here.

[0053] The pre-processed multi-source data is subjected to multi-source layered fusion processing to obtain uncertainty information. The method includes:

[0054] The weighted average algorithm is used to perform the first-layer fusion of the pre-processed GNSS position and velocity information with the INS position and velocity information to obtain the first fusion result;

[0055] The first layer of fusion specifically includes determining the position vector P according to the GNSS position information and velocity information. GNSS and V GNSS , determine P according to INS position information and speed information INS and V INS , and fuse the position vector and velocity vector based on the following formula:

[0056] P fusion =w GNSS ×P GNSS +w INS ×P INS ;

[0057] V fusion =w GNSS ×V GNSS +w INS ×V INS ;

[0058] Where w GNSS and w INSare the GNSS weight and INS weight, respectively, which are dynamically adjusted based on the measurement accuracy of GNSS and INS. For example, when the GNSS signal is good, that is, the satellite signal noise ratio is high and the GDOP value is small, the GNSS weight is increased; when the GNSS signal is blocked or interfered with, such as when the signal noise ratio decreases and the number of visible satellites decreases, the GNSS weight is reduced and the INS weight is increased.

[0059] The extended Kalman filter algorithm is used to take the initial position information as observation information and the first fusion result as state prediction information. The state is estimated and updated based on the observation information and the state prediction information to obtain the second fusion result.

[0060] The second layer of fusion specifically includes taking the position and velocity information after the first layer of fusion as the initial value of the state vector, and setting the state vector X = [x, y, z, v1, v2, v3] T , where (x, y, z) represents the position coordinates and (v1, v2, v3) represents the corresponding velocity components;

[0061] The initial position based on 5G is used as the observation vector G = [x 5G ,y 5G ,z 5G ] T , where (x 5G ,y 5G ,z 5G ) Initial position coordinates obtained by 5G positioning;

[0062] Based on the uniform motion model of the carrier, a state transfer matrix is ​​constructed. It should be noted that under the uniform motion assumption, the state transfer matrix can be expressed as a matrix combination of the state vector and the time interval, which is used to predict the state at the next moment.

[0063] According to the transfer matrix and the state vector at the previous moment, the state vector at the current moment is predicted and marked as the predicted state vector. The prediction calculation formula is as follows:

[0064] X k|k-1 =ZJ·X k-1 Where, X k|k-1 is the predicted state vector at the current moment; X k-1 is the state vector of the previous moment; k is the moment number based on the timestamp; ZJ is the state matrix;

[0065] Get the state vector obtained after the first layer fusion and the predicted state vector at the previous moment, and calculate the error covariance matrix Z at the previous moment k-1 , predict the error covariance matrix Z at the current moment k|k-1 , the calculation formula is as follows:

[0066] Zk|k-1 =ZJ·Z k-1 +Q; where Q is the process noise covariance matrix, which is calculated based on the predicted process noise data;

[0067] According to the error covariance matrix Z at the current moment k|k-1 Calculate the Kalman gain K k , the calculation formula is as follows:

[0068] K k =Z k|k-1 ·H T ·(H·Z k|k-1 ·H T +R) -1 ;

[0069] Where H is the observation matrix; R is the observation noise covariance matrix. It should be noted that the function of the observation matrix H is to map the state vector of the system to the observation space and establish the connection between the state and the observation data. The observation matrix H is based on the geometric relationship between the signal arrival time, arrival angle and position coordinates, and is calculated by linearizing these relationships.

[0070] According to the Kalman gain K k , observation vector G = [x 5G ,y 5G ,z 5G ] T and the predicted state vector X k|k-1 , update the state vector X k , the calculation formula is as follows:

[0071] X k =X k|k-1 +K k (G k -H·X k|k-1 );

[0072] According to the Kalman gain K k and the error covariance matrix Z at the current moment k|k-1 Update the error covariance matrix Z k , the calculation formula is as follows:

[0073] Z k =(IK k ·H)·Z k|k-1 ;Where, I is the unit matrix;

[0074] The state vector X of the position and velocity information after the second layer fusion k Output as the second fusion result;

[0075] Based on GNSS position data, INS position data and the second fusion result, deep feature extraction and fusion are performed to output more accurate position information, speed information and attitude information, and integrate them into uncertainty information;

[0076] Deep feature extraction and fusion including input layer setting, hidden layer architecture, output layer setting and neural network training;

[0077] The input layer setting includes obtaining the number of features of GNSS position data, INS position data, and 5G position data and the second fusion result, and summing the number of features to determine the number of input neurons;

[0078] The hidden layer architecture adopts a multi-hidden layer deep neural network structure. Each hidden layer uses the ReLU activation function to introduce nonlinear factors and enhance the network's ability to express complex data relationships.

[0079] The output layer includes three output neurons, which respectively output accurate position information, velocity information, and posture information. The output layer uses a linear activation function to directly output the predicted numerical results.

[0080] Neural network training involves collecting a large amount of multi-source data and its corresponding real position information, speed information, and posture information in different scenarios as sample data, and dividing the sample data into training set, validation set, and test set;

[0081] The mean square error loss function is used to calculate the average value L of the square of the difference between the predicted value and the true value of each dimension of position, velocity and posture. The calculation formula is as follows:

[0082] Where m is the number of training samples; y c i is the predicted value of the i-th training sample; yi is the true value of the i-th training sample; it should be noted that the mean square error loss function is used to measure the difference between the output of the neural network and the true value;

[0083] The Adam optimization algorithm is used to train the neural network. During the training process, the training data is input into the neural network in batches. Each batch of data is calculated and output through forward propagation. Then, the gradient of the loss function with respect to the parameters of each layer is calculated through the backpropagation algorithm. The Adam optimization algorithm is used to update the network parameters and continuously adjust the network weights and biases to make the loss function converge.

[0084] Evaluate the trained neural network model using the test set, including adjusting and optimizing the neural network's hyperparameters through cross-validation. The hyperparameters of the neural network include, but are not limited to, the number of hidden layers, the number of neurons per layer, and the learning rate. For example, the number of hidden layers is varied, and the model's loss and accuracy on the validation set are observed to determine the hyperparameter combination that optimizes model performance.

[0085] Use the test set to evaluate the trained neural network model, including but not limited to calculating the root mean square error and mean absolute error of the position, velocity, and attitude prediction results to comprehensively evaluate the accuracy and reliability of the model. If the evaluation results do not meet expectations, further adjust the hyperparameters or increase the amount of training data to retrain the model.

[0086] Utilize the trained and optimized neural network model to output accurate position information, accurate speed information, and accurate posture information, and integrate them into a layered fusion result;

[0087] Based on uncertainty information, statistical methods are used to calculate the protection level, including:

[0088] By formula Where α is a constant related to the confidence level; Z xx are the diagonal elements of the position component in the error covariance matrix;

[0089] Perform consistency assessment on multi-source data to obtain assessment results, and calculate the positioning reliability based on the uncertainty information output by the neural network model and the assessment results. The method includes:

[0090] According to the positioning results and uncertainty distribution output by the neural network, the entropy value Q is calculated. The entropy value Q is based on the following formula:

[0091] Where p(xi) is the probability that the positioning result is in state xi; n is the number of possible positioning states;

[0092] Calculate the Euclidean distances of the GNSS position information, INS position information, and initial position information, and perform normalization to obtain the consistency index C.

[0093] The comprehensive entropy value Q and consistency index C are weighted and fused to obtain the location reliability R. The calculation formula is as follows:

[0094] R = w1 × (1-Q) + w2 × C; where w1 and w2 are weights; weights w1 and w2 are adjusted according to actual conditions to balance the effects of entropy and consistency on location reliability;

[0095] By dynamically adjusting constants related to the confidence level, the protection level can be accurately calculated based on actual application scenarios and security requirements. In terms of location confidence calculation, the uncertainty information output by the neural network and the consistency assessment of multi-source data are combined to more comprehensively and accurately reflect the credibility of the positioning results compared to traditional location confidence calculations that rely only on a single data source or simple statistical methods.

[0096] Obtain multi-dimensional correlation data of GNSS signals, multi-dimensional correlation data of INS signals, and multi-dimensional correlation data of 5G signals, use neural network multi-dimensional correlation data for analysis and learning, and output risk prediction values. The method includes:

[0097] Evaluating the signal risk value, based on the following first evaluation model and second evaluation model;

[0098] The expression of the first evaluation model is Where Out1 is the output value of the first evaluation model, P in is the input parameter, Pmin and Pmax are the lower limit and upper limit of the standard parameter range respectively;

[0099] The expression of the second evaluation model is Where Out2 is the output value of the second evaluation model; P thr is the parameter-related threshold;

[0100] Multi-dimensional GNSS signal data includes real-time GNSS signal strength, real-time signal-to-noise ratio, and real-time carrier phase residual. Obtain the corresponding historical data to determine the signal strength range, signal-to-noise ratio range, and carrier phase residual threshold:

[0101] Calculating the signal strength risk value and the signal-to-noise ratio risk value based on the first assessment model, calculating the carrier phase residual risk value based on the second assessment model, and performing a weighted sum operation to obtain the GNSS signal risk value;

[0102] The multi-dimensional data related to the INS signal includes the accelerometer bias drift parameters, gyroscope drift parameters, and integral error. The corresponding historical data is obtained to determine the accelerometer bias drift threshold and gyroscope drift threshold:

[0103] Calculate the accelerometer bias drift risk value and gyroscope drift risk value based on the first evaluation model:

[0104] The INS signal risk value is calculated by weighted summing the accelerometer bias drift risk value and the gyroscope drift risk value;

[0105] Multi-dimensional 5G signal data includes signal arrival time series, arrival angle series, and signal strength series; determining parameter change rates based on sequence data; obtaining corresponding historical data and determining corresponding parameter change rate thresholds;

[0106] Based on the second assessment model, the signal arrival time variation risk value, the arrival angle variation risk value, and the signal strength variation risk value are calculated and weighted summed to obtain the 5G signal risk value;

[0107] Normalize the input risk value; use a multi-layer fully connected neural network structure to extract and transform the input risk value through the ReLU function, and use the Sigmoid activation function to output the risk prediction value;

[0108] By simultaneously processing multi-source risk factors such as GNSS, INS, 5G data, and environmental information, risk prediction is performed by automatically learning the complex relationships between risk factors.

[0109] In the second embodiment, the present invention proposes a GNSS and INS integrity monitoring system based on a 5G and multi-layer fusion mechanism, which is applied to the GNSS and INS integrity monitoring method based on a 5G and multi-layer fusion mechanism proposed in the first embodiment, and specifically includes:

[0110] Data monitoring and acquisition module, used to obtain GNSS position data, INS position data and 5G position data to obtain multi-source data;

[0111] The data preprocessing and fusion module is used to preprocess the data based on the data type of multi-source data, and perform multi-source layered fusion processing on the preprocessed multi-source data to obtain uncertainty information;

[0112] The protection level definition module calculates the protection level using statistical methods based on uncertainty information;

[0113] The location reliability calculation module is used to evaluate the consistency of multi-source data, obtain the evaluation results, and calculate the location reliability based on the uncertainty information output by the neural network model and the evaluation results;

[0114] The relevant data collection and risk prediction module is used to obtain multi-dimensional relevant data of GNSS signals, multi-dimensional relevant data of INS signals and multi-dimensional relevant data of 5G signals, use neural network multi-dimensional relevant data for analysis and learning, and output risk prediction values.

[0115] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism is characterized by: The following steps are involved: Acquire GNSS position data, INS position data, and 5G position data to obtain multi-source data; Preprocess the data based on the data type of multi-source data, and perform multi-source layered fusion processing on the preprocessed multi-source data to obtain uncertainty information; Based on uncertainty information, statistical methods are used to calculate the protection level; Conduct consistency assessment on multi-source data to obtain assessment results, and calculate the positioning reliability based on the uncertainty information output by the neural network model and the assessment results; It is used to obtain multi-dimensional correlation data of GNSS signals, multi-dimensional correlation data of INS signals and multi-dimensional correlation data of 5G signals, use neural network multi-dimensional correlation data for analysis and learning, and output risk prediction values.

2. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 1 is characterized in that: GNSS position data includes at least GNSS position information and speed information; INS position data includes at least INS position information and speed information; 5G position data includes at least initial position information.

3. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 1 is characterized in that: The method of performing multi-source layered fusion processing on the pre-processed multi-source data includes: The weighted average algorithm is used to perform the first-layer fusion of the pre-processed GNSS position and velocity information with the INS position and velocity information to obtain the first fusion result; The extended Kalman filter algorithm is used to take the initial position information as observation information and the first fusion result as state prediction information. The state is estimated and updated based on the observation information and the state prediction information to obtain the second fusion result. Deep feature extraction and fusion are performed based on GNSS position data, INS position data and the second fusion result to output more accurate position information, speed information and attitude information, and integrate them into uncertainty information.

4. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 3 is characterized in that: The first layer of fusion specifically includes determining the position vector P according to the GNSS position information and velocity information. GNSS and V GNSS , determine P according to INS position information and speed information INS and V INS , and fuse the position vector and velocity vector based on the following formula: P fusion =w GNSS ×P GNSS +w INS ×P INS ; V fusion =w GNSS ×V GNSS +w INS ×V INS ; Where w GNSS and w INS are GNSS weight and INS weight, respectively, which are dynamically adjusted based on the measurement accuracy of GNSS and INS.

5. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 4 is characterized in that: The second layer of fusion specifically includes taking the position and velocity information after the first layer of fusion as the initial value of the state vector, and setting the state vector X = [x, y, z, v1, v2, v3] T , where (x, y, z) represents the position coordinates and (v1, v2, v3) represents the corresponding velocity components; The initial position based on 5G is used as the observation vector G = [x 5G ,y 5G ,z 5G ] T , where (x 5G ,y 5G ,z 5G )The initial position coordinates obtained by 5G positioning.

6. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 5 is characterized in that: The second layer of fusion also includes a uniform motion model based on the carrier to construct a state transfer matrix; According to the transfer matrix and the state vector at the previous moment, the state vector at the current moment is predicted and marked as the predicted state vector. The prediction calculation formula is as follows: X k|k-1 =ZJ·X k-1 Where, X k|k-1 is the predicted state vector at the current moment; X k-1 is the state vector of the previous moment; k is the moment number based on the timestamp; ZJ is the state matrix.

7. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 6 is characterized in that: The second layer fusion also includes obtaining the state vector obtained after the first layer fusion at the previous moment and the predicted state vector, and calculating the error covariance matrix Z at the previous moment k-1 , predict the error covariance matrix Z at the current moment k|k-1 , the calculation formula is as follows: Z k|k-1 =ZJ·Z k-1 +Q; where Q is the process noise covariance matrix, which is calculated based on the predicted process noise data; According to the error covariance matrix Z at the current moment k|k-1 Calculate the Kalman gain K k , the calculation formula is as follows: K k =Z k|k-1 ·H T ·(H·Z k|k-1 ·H T +R) -1 ; Where H is the observation matrix; R is the observation noise covariance matrix.

8. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 7 is characterized in that: The second layer fusion also includes the Kalman gain K k , observation vector G = [x 5G ,y 5G ,z 5G ] T and the predicted state vector X k|k-1 , update the state vector X k , the calculation formula is as follows: X k =X k|k-1 +K k (G k -H·X k|k-1 ); According to the Kalman gain K k and the error covariance matrix Z at the current moment k|k-1 Update the error covariance matrix Z k , the calculation formula is as follows: Z k =(IK k ·H)·Z k|k-1 ;Where, I is the unit matrix; The state vector X of the position and velocity information after the second layer fusion k Output as the second fusion result; Deep feature extraction and fusion are performed based on GNSS position data, INS position data and the second fusion result to output more accurate position information, speed information and attitude information, and integrate them into uncertainty information.

9. The GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism according to claim 8 is characterized in that: Deep feature extraction and fusion including input layer setting, hidden layer architecture, output layer setting and neural network training; The input layer setting includes obtaining the number of features of GNSS position data, INS position data, and 5G position data and the second fusion result, and summing the number of features to determine the number of input neurons; The hidden layer architecture adopts a multi-hidden layer deep neural network structure. Each hidden layer uses the ReLU activation function to introduce nonlinear factors and enhance the network's ability to express complex data relationships. The output layer setting includes three output neurons, which respectively output accurate position information, speed information, and posture information; among them, the output layer uses a linear activation function to directly output the predicted numerical results.

10. A GNSS and INS integrity monitoring system based on 5G and multi-layer fusion mechanism, applied to the GNSS and INS integrity monitoring method based on 5G and multi-layer fusion mechanism described in any one of claims 1 to 9, characterized in that: Specifically include: Data monitoring and acquisition module, used to obtain GNSS position data, INS position data and 5G position data to obtain multi-source data; The data preprocessing and fusion module is used to preprocess the data based on the data type of multi-source data, and perform multi-source layered fusion processing on the preprocessed multi-source data to obtain uncertainty information; The protection level definition module calculates the protection level using statistical methods based on uncertainty information; The location reliability calculation module is used to evaluate the consistency of multi-source data, obtain the evaluation results, and calculate the location reliability based on the uncertainty information output by the neural network model and the evaluation results; The relevant data collection and risk prediction module is used to obtain multi-dimensional relevant data of GNSS signals, multi-dimensional relevant data of INS signals and multi-dimensional relevant data of 5G signals, use neural network multi-dimensional relevant data for analysis and learning, and output risk prediction values.

Images