Train positioning processing method and device, electronic equipment and storage medium

Abstract

The invention provides a train positioning processing method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining ODO measurement data of a train in a GNSS signal interruption process in the GNSS signal interruption process; determining a pseudo GNSS position of the train in the GNSS signal interruption process based on the ODO measurement data; the ODO measurement data and the pseudo GNSS position are fused, and navigation information of the train in the GNSS signal interruption process is determined. According to the method, the pseudo GNSS position of the train in the GNSS signal interruption process is determined based on the ODO measurement data in the GNSS signal interruption process, then the ODO measurement data and the pseudo GNSS position are fused, the navigation information of the train in the GNSS signal interruption process is determined, error compensation is carried out on a navigation system when no signal exists in the GNSS, error divergence is restrained, and the navigation accuracy is improved. And the high-precision positioning of the railway train is improved.

Description

Technical Field

[0001] The present invention relates to the field of train control technology, and in particular to a train positioning processing method, device, electronic equipment and storage medium. Background Art

[0002] High-precision train positioning technology is an important research field in modern heavy-duty railway systems, which is related to the safety, efficiency and intelligent management of trains. With the rapid development of urbanization and industrialization, the demand for railway transportation is increasing, and higher requirements are placed on the real-time monitoring and precise control of train operation. High-precision positioning technology uses a variety of sensors and systems, such as the global navigation satellite system (GNSS), inertial navigation system (INS), odometer (ODO) and various communication technologies. This technology can provide continuous and high-precision location information, which has a significant effect on optimizing train scheduling, increasing line capacity, improving driving safety and reducing operation and maintenance costs.

[0003] However, achieving high-precision positioning faces many challenges, including the fusion processing of multiple sensor data, signal interference, and positioning accuracy in dynamic environments. In addition, the accuracy and reliability of positioning are particularly important in areas where the GNSS signal is weak or absent, such as tunnels, mountainous areas, or densely built-up areas in cities. Due to the weak penetration of satellite signals, satellite signals will be interrupted when the train is in a tunnel, canyon, or other unfavorable environment. At this time, the train lacks satellite positioning signals, which will result in missing information fusion, and the positioning error will quickly diverge, seriously reducing the positioning accuracy. Summary of the invention

[0004] In view of the problems existing in the prior art, the present invention provides a train positioning processing method, device, electronic equipment and storage medium.

[0005] The present invention provides a train positioning processing method, comprising: During the GNSS signal interruption, obtain the train ODO measurement data during the GNSS signal interruption; Determine a pseudo GNSS position of the train during GNSS signal interruption based on the ODO measurement data; The ODO measurement data and the pseudo GNSS position are fused to determine the navigation information of the train during the GNSS signal interruption process.

[0006] According to a train positioning processing method provided by the present invention, the method of determining the pseudo GNSS position of the train during the GNSS signal interruption process based on the ODO measurement data includes: Determine the GNSS position increment at each time point during the GNSS signal interruption process based on the ODO measurement data; Get the train's position before the GNSS signal is interrupted; The pseudo GNSS position of the train during the GNSS signal interruption is determined based on the train's position before the GNSS signal interruption and the train's GNSS position increment at each time point during the GNSS signal interruption.

[0007] According to a train positioning processing method provided by the present invention, the determining of the GNSS position increment at each time point during the GNSS signal interruption process based on the ODO measurement data comprises: Inputting the ODO measurement data into a position prediction model to obtain the GNSS position increment of the train at each time point during the GNSS signal interruption process output by the position prediction model; Among them, the position prediction model is a model that uses the ODO measurement data and GNSS position of the train during the process of GNSS signal availability as input and is obtained through machine learning training, and is used to output the GNSS position increment of the train during the process of GNSS signal interruption.

[0008] According to a train positioning processing method provided by the present invention, the position prediction model adopts the model structure of the neural network AT-LSTM.

[0009] According to a train positioning processing method provided by the present invention, the method determines the pseudo GNSS position of the train during the GNSS signal interruption process based on the position of the train before the GNSS signal interruption and the GNSS position increment of the train at each time point during the GNSS signal interruption process, including: Determine the accumulated amount of the GNSS position increment at each time point according to the GNSS position increment of the train at each time point during the GNSS signal interruption process; Based on the position of the train before the GNSS signal interruption and the accumulated amount, the pseudo GNSS position of the train during the GNSS signal interruption is determined using the following calculation formula;

[0010] in, Represents the pseudo GNSS position during GNSS signal interruption. Represents the position before the GNSS signal is interrupted. is the GNSS position increment at the i-th time point during the GNSS signal interruption process, and k represents the duration of the GNSS signal interruption process.

[0011] According to a train positioning processing method provided by the present invention, a Kalman filter is used to fuse the ODO measurement data and the pseudo GNSS position.

[0012] According to a train positioning processing method provided by the present invention, the method also includes: when the GNSS signal is available, collecting ODO measurement data and GNSS position, and optimizing the position prediction model based on the ODO measurement data and GNSS position.

[0013] The present invention also provides a train positioning processing device, comprising: An acquisition module is used to acquire the ODO measurement data of the train during the GNSS signal interruption process; A determination module, configured to determine a pseudo GNSS position of the train during a GNSS signal interruption process based on the ODO measurement data; The positioning module is used to fuse the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption process.

[0014] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned train positioning processing methods is implemented.

[0015] The present invention also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, any of the above-mentioned train positioning processing methods is implemented.

[0016] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements any of the above-mentioned train positioning processing methods.

[0017] The present invention provides a train positioning processing method, device, electronic device and storage medium, which determine the pseudo GNSS position of the train during the GNSS signal interruption based on ODO measurement data during the GNSS signal interruption, and then fuse the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption, thereby realizing error compensation for the navigation system when there is no GNSS signal, suppressing error divergence, and improving the high-precision positioning of railway trains. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1It is a flow chart of the train positioning processing method provided by the present invention.

[0020] Figure 2 It is a schematic diagram of the unit structure of the LSTM provided by the present invention.

[0021] Figure 3 It is a schematic diagram of the AT-LSTM model structure provided by the present invention.

[0022] Figure 4 It is a structural schematic diagram of the train positioning processing device provided by the present invention.

[0023] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] Combine the following Figure 1-Figure 5 The present invention describes a train positioning processing method, device, electronic equipment and storage medium.

[0026] Figure 1 A flow chart of a train positioning method provided by the present invention is shown. Figure 1 , the method comprises the following steps: Step 11, during the GNSS signal interruption process, obtaining the ODO measurement data of the train during the GNSS signal interruption process; Step 12: determining the pseudo GNSS position of the train during the GNSS signal interruption process based on the ODO measurement data; Step 13: Fuse the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption process.

[0027] Regarding the above steps 11 to 13, it should be noted that due to the weak penetration of satellite signals, when the train is in a tunnel, canyon or other unfavorable environment, the satellite signal will be interrupted. At this time, the error of train positioning will diverge rapidly and the positioning accuracy will be seriously reduced. For this reason, this method is intended to be adapted to the train positioning processing method of the global navigation satellite system (GNSS) / odometer system (ODO) integrated navigation function. The train positioning processing method can compensate for the positioning error that occurs during the train positioning process, so as to provide the train with a more accurate positioning service during the GNSS signal interruption process.

[0028] In the present invention, since it is an integrated positioning and navigation service of the global navigation satellite system (GNSS) / odometer system (ODO), it is necessary to collect GNSS position signals and ODO measurement data when the GNSS signal is available. In the process of GNSS signal interruption, the ODO measurement data of the train during the GNSS signal interruption is obtained.

[0029] In the present invention, since only ODO measurement data can be collected during the GNSS signal interruption process, but the GNSS position needs to be compensated, it is necessary to determine the "position" of the train during the GNSS signal interruption process based on the collected ODO measurement data. Since this position is not the actual GNSS position, it is used as a pseudo GNSS position.

[0030] Due to the existence of pseudo-GNSS position, at this time, under the integrated navigation function of the global navigation satellite system (GNSS) / odometer system (ODO), the information is complete, so the ODO measurement data and the pseudo-GNSS position are fused. The fused information can generate navigation information for the train during the GNSS signal interruption process.

[0031] The train positioning processing method provided by the present invention determines the pseudo GNSS position of the train during the GNSS signal interruption process based on ODO measurement data during the GNSS signal interruption process, and then fuses the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption process, thereby realizing error compensation for the navigation system when there is no GNSS signal, suppressing error divergence, and improving the high-precision positioning of railway trains.

[0032] In a further method of the above method, the processing process of determining the pseudo GNSS position of the train during the GNSS signal interruption based on the ODO measurement data is mainly explained as follows: Determine the GNSS position increment at each time point during the GNSS signal interruption based on the ODO measurement data; Get the train's position before the GNSS signal is interrupted; The pseudo GNSS position of the train during the GNSS signal interruption is determined based on the train's position before the GNSS signal interruption and the train's GNSS position increment at each time point during the GNSS signal interruption.

[0033] In this regard, it should be noted that in the present invention, the change in position has a time attribute. Therefore, during the GNSS signal interruption process, the GNSS position at each time point should be obtained in units of time. The change in position increases with time, so there are different position changes between each time point. The position change here is taken as the increment of each change, that is, the GNSS position increment at each time point during the GNSS signal interruption process is determined based on the ODO measurement data.

[0034] Since it is necessary to calculate the position after the GNSS signal interruption process, it is necessary to first obtain the train position before the GNSS signal interruption, and then combine the GNSS position increments of the train at each time point during the GNSS signal interruption process to deduce the pseudo GNSS position of the train during the GNSS signal interruption process.

[0035] A further method of the present invention is to estimate the increase of the position from a more detailed calculation perspective by dividing the GNSS position at each time point during the GNSS signal interruption process, thereby improving the prediction accuracy of the pseudo-GNSS position and achieving improved navigation accuracy.

[0036] In a further method of the above method, the processing process of determining the GNSS position increment at each time point during the GNSS signal interruption process based on the ODO measurement data is mainly explained as follows: The ODO measurement data is input into the position prediction model to obtain the GNSS position increment of the train at each time point during the GNSS signal interruption process output by the position prediction model; Among them, the position prediction model is a model that uses the ODO measurement data and GNSS position of the train when the GNSS signal is available as input and is trained through machine learning. It is used to output the GNSS position increment of the train during the GNSS signal interruption process.

[0037] In this regard, it should be noted that in the present invention, the train collects ODO measurement data and GNSS position when the GNSS signal is available. In order to reflect the correlation between the ODO measurement data and the GNSS position, the data obtained by fusing the ODO measurement data and the GNSS position can be used as training data for the model to train a position prediction model, so as to establish a nonlinear relationship model between the ODO measurement value and the GNSS position increment.

[0038] The model can quickly output pseudo GNSS positions when GNSS signals are interrupted, and then use the pseudo GNSS signals to compensate the integrated navigation system, ensuring that the navigation system can maintain high accuracy and stability even when GNSS signals are unavailable. Since the model makes the ODO measurement data and GNSS positions closely related, the model predicts the GNSS position increment, which can ensure fast output and higher accuracy and stability of the output results.

[0039] In the further method of the above method, the location prediction model mainly adopts the model structure of the neural network AT-LSTM. AT-LSTM (Attention-Based LSTM) is a neural network model that combines LSTM (Long Short-Term Memory) and attention mechanism. The specific description is as follows: Long short-term memory network (LSTM) is a special type of recurrent neural network (RNN) used to handle long-term dependency problems. It saves historical information through a specific structure - memory unit, and uses three key gate structures to control the flow of information: input gate: determines which information will be updated; forget gate: determines which information will be discarded; output gate: determines which information will be used for output.

[0040] The unit structure of LSTM is as follows Figure 2 As shown, , is the input and output data at time t, i, f and o are the input layer, forgetting layer and output layer respectively, is the state value of the memory unit at time t.

[0041] The output gate is used to control the output of the memory cell state value. The sigmoid function is used to calculate the output part, and the memory cell state processed by the tanh function is compared with the result. Multiply them together to get the output of the LSTM unit at time t .

[0042]

[0043] LSTM solves the problems of gradient explosion and gradient vanishing that are prone to occur when RNN processes long sequences through the gating mechanism, which makes the model perform better in predicting the position of moving objects.

[0044] The attention mechanism is integrated on the basis of the LSTM model. The main purpose of integrating the attention mechanism into the LSTM (Long Short-Term Memory Network) is to enhance the model's ability to process complex and long sequence data. LSTM itself effectively manages the long-term storage and access of information through its unique gating structure (forget gate, input gate, and output gate), solving the gradient vanishing problem that is prone to occur in traditional recurrent neural networks (RNNs) when dealing with long-term dependencies. Although LSTM has performed well in processing time series data, in some cases, it still has difficulty capturing and prioritizing the most critical information in the sequence.

[0045] The attention mechanism can assign different weights to the input of the model from a large number of chaotic input sequences, which is used to indicate that the model strengthens important feature information and suppresses non-important feature information for the input. By focusing on the basic parts with high weight selection and ignoring other irrelevant or less content, the transmission of low information to key information sequences is achieved, which has better scalability and robustness. Because the output of the LSTM network depends more on the feature information from the nearest neurons, while ignoring the feature information of neurons in the long-term sequence. Therefore, introducing the attention mechanism into LSTM can eliminate the dependency between input and output and improve the ability of the LSTM network to capture long-term sequence feature information.

[0046] The AT-LSTM model structure is as follows Figure 3 As shown in the figure, the feature vector is input into the LSTM model, and the hidden layer state vector h is obtained through iterative calculation. It contains the temporal correlation between the output of the LSTM layer at different time steps. In the process of training the LSTM network with input data, as the information propagates forward, valuable information will be constantly forgotten. The introduced attention mechanism extracts this information in the hidden state and can adjust the weights connected to each hidden layer output. The calculation formula of the attention layer is as follows:

[0047] It is the attention score, which indicates the degree of attention paid to the hidden state at a certain moment.

[0048] and They are weight matrices, representing the weights learned by the network, and are used to calculate the attention score.

[0049] It is a hidden state, which indicates the feature representation of the input data after passing through the LSTM network at the kth moment.

[0050] is a bias term used to adjust the calculation results.

[0051] Represents the hyperbolic tangent function, which is used to introduce nonlinearity. This function compresses the values ​​so that they fall within the range of −1 to 1 to increase the nonlinear expression ability of the model.

[0052] The main function of this formula is to Perform weighted calculation to generate attention score .

[0053]

[0054] is the attention weight coefficient (Attention Weight), which represents the importance weight of the current time step k among all time steps.

[0055] is the attention score calculated by the previous formula.

[0056] yes An exponential function is used to convert the attention scores into non-negative numbers.

[0057] The denominator is to perform exponential normalization on the attention scores of all time steps so that all The sum of is 1, forming a probability distribution.

[0058] This formula normalizes the attention score through the Softmax function to obtain the attention weight for each time step , used to indicate the importance of each time step

[0059] is the output at the kth moment, representing the prediction of the GNSS position increment.

[0060] is the attention weight at the i-th time step.

[0061] is the attention score corresponding to the hidden state at the i-th time step.

[0062] The meaning of this formula is to convert the attention weights of each time step and the corresponding hidden state score Perform weighted summation to get the final output , which is the GNSS position increment of the current time step.

[0063] First, calculate the attention score , which is the core part of the attention module, where and is the attention weight matrix, is the bias term; then, the Softmax function is used to normalize the attention weights to obtain the attention weight coefficient vector ; Finally, according to the attention weight coefficient For hidden state A weighted sum is performed to output the GNSS position increment at time k.

[0064] The network design and parameter adjustment process of the long short-term memory network based on the attention mechanism are strictly protected to compensate for GNSS signal interruption and ensure the stable operation of the navigation system.

[0065] In a further method of the above method, a processing process of determining a pseudo GNSS position of a train during a GNSS signal interruption is mainly explained based on the position of the train before the GNSS signal interruption and the GNSS position increment of the train at each time point during the GNSS signal interruption, as follows: Determine the accumulated amount of the GNSS position increment at each time point according to the GNSS position increment of the train at each time point during the GNSS signal interruption process; Based on the position of the train before the GNSS signal interruption and the accumulated amount, the pseudo GNSS position of the train during the GNSS signal interruption is determined using the following calculation formula;

[0066] in, Represents the pseudo GNSS position during GNSS signal interruption. Represents the position before the GNSS signal is interrupted. is the GNSS position increment at the i-th time point during the GNSS signal interruption process, and k represents the duration of the GNSS signal interruption process.

[0067] The model will be trained when GNSS is available. ODO data, namely timestamp, distance and speed, are used as input to the model. The output of the model is the GNSS position increment. When the GNSS signal is not available, the integrated navigation system will work independently of the ODO system, and the positioning error will gradually increase over time. The trained model is used to output a pseudo-GNSS signal to correct the ODO system. Starting from the first period after GNSS is unavailable, the position increments output by the model are accumulated to obtain a pseudo-GNSS position in each signal interruption period. Furthermore, when the GNSS signal is available, ODO measurement data and GNSS position are collected, and the position prediction model is optimized based on the ODO measurement data and GNSS position.

[0068] In a further method of the above method, a Kalman filter is used to fuse the ODO measurement data and the pseudo GNSS position.

[0069] The Kalman Filter (KF) method is used for data fusion. KF is an effective state estimation method for linear dynamic systems, which can estimate the state variables of dynamic systems in the presence of noise.

[0070] Assuming the system state is x and the measurement value is z, the two main steps of KF are prediction and update: Prediction steps:

[0071] : This is the predicted state estimate (the predicted state). The superscript "-" indicates that this is the predicted value before the update.

[0072] : The state transition matrix is ​​used to describe how the system state is transferred from the previous time k−1 to the current time k.

[0073] : This is the estimated value of the state at the previous moment.

[0074] : Control input matrix, used to describe the impact of control input on system state.

[0075] : This is the control input at time k−1.

[0076] This formula describes the prediction process in Kalman filtering, predicting the system state at the current time k

[0077] : This is the covariance matrix of the forecast, which represents the uncertainty of the forecast state.

[0078] : The state covariance matrix of the previous time k−1.

[0079] : state transfer matrix.

[0080] : Transpose of the state transfer matrix.

[0081] : The process noise covariance matrix represents the influence of random noise inside the system on state prediction.

[0082] This formula is used to update the uncertainty of the predicted state, predicting the uncertainty of the system at time k.

[0083] Update steps:

[0084] : Kalman gain, which represents the weighted fusion ratio of the measured value and the predicted value.

[0085] : The predicted covariance matrix, from the previous prediction step.

[0086] : Observation matrix, used to map state variables to measurement space.

[0087] : Transpose of the observation matrix.

[0088] : Observation noise covariance matrix, which indicates the noise size in the measurement process.

[0089] : This part represents the weighted average of measurement uncertainty and prediction uncertainty, which is used to calculate the Kalman gain.

[0090] The Kalman gain K determines the trade-off between predictions and measurements. A higher Kalman gain means that measurements have a greater influence on the final estimate, while a lower Kalman gain means that predictions have a greater influence.

[0091] : Updated state estimate.

[0092] : The predicted state estimate, from the prediction step.

[0093] : Kalman gain.

[0094] : The actual measured value of k at the current moment.

[0095] : The measurement value calculated from the predicted state estimate.

[0096] This formula represents the predicted state Make corrections and update the state estimate by incorporating the effects of the measurements.

[0097]

[0098] : The updated covariance matrix, indicating the updated uncertainty.

[0099] : Identity matrix, used to keep the matrix dimensions consistent.

[0100] : Kalman gain.

[0101] : Observation matrix.

[0102] : The predicted covariance matrix.

[0103] This formula is used to update the state covariance matrix, reflecting the uncertainty in the system state after the measurements are introduced.

[0104] A further method of the present invention utilizes a Kalman filter (KF) to efficiently fuse data from GNSS and ODO sensors to provide the most accurate and reliable navigation information.

[0105] The train positioning processing device provided by the present invention is described below. The train positioning processing device described below and the train positioning processing method described above can be referenced to each other.

[0106] Figure 4 A schematic diagram of the structure of a train positioning processing device provided by the present invention is shown. Figure 3 The device includes an acquisition module 41, a determination module 42 and a positioning module 43, wherein: An acquisition module 41 is used to acquire ODO measurement data of the train during the GNSS signal interruption process; A determination module 42, configured to determine a pseudo GNSS position of the train during a GNSS signal interruption process based on the ODO measurement data; The positioning module 43 is used to fuse the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption process.

[0107] In a further device of the above device, the determining module is specifically used for: Determine the GNSS position increment at each time point during the GNSS signal interruption based on the ODO measurement data; Get the train's position before the GNSS signal is interrupted; The pseudo GNSS position of the train during the GNSS signal interruption is determined based on the train's position before the GNSS signal interruption and the train's GNSS position increment at each time point during the GNSS signal interruption.

[0108] In a further device of the above device, the determination module is specifically used in the process of determining the GNSS position increment at each time point during the GNSS signal interruption process based on the ODO measurement data: Inputting the ODO measurement data into a position prediction model to obtain the GNSS position increment of the train at each time point during the GNSS signal interruption process output by the position prediction model; Among them, the position prediction model is a model that uses the ODO measurement data and GNSS position of the train during the process of GNSS signal availability as input and is obtained through machine learning training, and is used to output the GNSS position increment of the train during the process of GNSS signal interruption.

[0109] In a further device of the above device, the location prediction model adopts the model structure of the neural network AT-LSTM.

[0110] In a further device of the above device, the determination module is specifically used in the process of determining the pseudo GNSS position of the train during the GNSS signal interruption based on the position of the train before the GNSS signal interruption and the GNSS position increment of the train at each time point during the GNSS signal interruption to: Determine the accumulated amount of the GNSS position increment at each time point according to the GNSS position increment of the train at each time point during the GNSS signal interruption process; Based on the position of the train before the GNSS signal interruption and the accumulated amount, the pseudo GNSS position of the train during the GNSS signal interruption is determined using the following calculation formula;

[0111] in, Represents the pseudo GNSS position during GNSS signal interruption. Represents the position before the GNSS signal is interrupted. is the GNSS position increment at the i-th time point during the GNSS signal interruption process, and k represents the duration of the GNSS signal interruption process.

[0112] In a further embodiment of the above device, a Kalman filter is used to fuse the ODO measurement data and the pseudo GNSS position.

[0113] In a further device of the above device, the determination module is also used to: collect ODO measurement data and GNSS position when the GNSS signal is available, and optimize the position prediction model based on the ODO measurement data and GNSS position.

[0114] The present invention provides a train positioning processing device, which determines the pseudo GNSS position of the train during the GNSS signal interruption process based on ODO measurement data during the GNSS signal interruption process, and then fuses the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption process, thereby realizing error compensation for the navigation system when there is no GNSS signal, suppressing error divergence, and improving the high-precision positioning of railway trains.

[0115] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5 As shown, the electronic device may include: a processor 51, a communication interface 52, a memory 53 and a communication bus 54, wherein the processor 51, the communication interface 52 and the memory 53 communicate with each other through the communication bus 54. The processor 51 may call the logic instructions in the memory 53 to execute the train positioning processing method, which includes: during the GNSS signal interruption process, obtaining the ODO measurement data of the train during the GNSS signal interruption process; determining the pseudo GNSS position of the train during the GNSS signal interruption process based on the ODO measurement data; fusing the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption process.

[0116] In addition, the logic instructions in the above-mentioned memory 53 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0117] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the train positioning processing method provided by the above-mentioned methods, which includes: during the GNSS signal interruption, obtaining the ODO measurement data of the train during the GNSS signal interruption; determining the pseudo-GNSS position of the train during the GNSS signal interruption based on the ODO measurement data; and fusing the ODO measurement data and the pseudo-GNSS position to determine the navigation information of the train during the GNSS signal interruption.

[0118] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it is implemented to execute the train positioning processing method provided by the above-mentioned methods. The method includes: during a GNSS signal interruption, obtaining ODO measurement data of the train during the GNSS signal interruption; determining a pseudo GNSS position of the train during the GNSS signal interruption based on the ODO measurement data; and fusing the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption.

[0119] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0120] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A train positioning processing method, characterized in that: include: During the GNSS signal interruption, obtain the train ODO measurement data during the GNSS signal interruption; Determine a pseudo GNSS position of the train during GNSS signal interruption based on the ODO measurement data; The ODO measurement data and the pseudo GNSS position are fused to determine the navigation information of the train during the GNSS signal interruption process.

2. The train positioning processing method according to claim 1, characterized in that: The determining, based on the ODO measurement data, a pseudo GNSS position of the train during a GNSS signal interruption process comprises: Determine the GNSS position increment at each time point during the GNSS signal interruption process based on the ODO measurement data; Get the train's position before the GNSS signal is interrupted; The pseudo GNSS position of the train during the GNSS signal interruption is determined based on the train's position before the GNSS signal interruption and the train's GNSS position increment at each time point during the GNSS signal interruption.

3. The train positioning processing method according to claim 1, characterized in that: The determining of the GNSS position increment at each time point during the GNSS signal interruption process based on the ODO measurement data includes: Inputting the ODO measurement data into a position prediction model to obtain the GNSS position increment of the train at each time point during the GNSS signal interruption process output by the position prediction model; Among them, the position prediction model is a model that uses the ODO measurement data and GNSS position of the train during the process of GNSS signal availability as input and is obtained through machine learning training, and is used to output the GNSS position increment of the train during the process of GNSS signal interruption.

4. The train positioning processing method according to claim 3, characterized in that: The location prediction model adopts the model structure of the neural network AT-LSTM.

5. The train positioning processing method according to claim 2, characterized in that: The method of determining the pseudo GNSS position of the train during the GNSS signal interruption process based on the position of the train before the GNSS signal interruption and the GNSS position increments of the train at each time point during the GNSS signal interruption process comprises: Determine the accumulated amount of the GNSS position increment at each time point according to the GNSS position increment of the train at each time point during the GNSS signal interruption process; Based on the position of the train before the GNSS signal interruption and the accumulated amount, the pseudo GNSS position of the train during the GNSS signal interruption is determined using the following calculation formula; in, Represents the pseudo GNSS position during GNSS signal interruption. Represents the position before the GNSS signal is interrupted. is the GNSS position increment at the i-th time point during the GNSS signal interruption process, and k represents the duration of the GNSS signal interruption process.

6. The train positioning processing method according to claim 1, characterized in that: A Kalman filter is used to fuse the ODO measurement data and the pseudo GNSS position.

7. The train positioning processing method according to claim 3, characterized in that: The method further includes: when the GNSS signal is available, collecting ODO measurement data and GNSS position, and optimizing the position prediction model based on the ODO measurement data and GNSS position.

8. A train positioning processing device, characterized in that: include: An acquisition module is used to acquire the ODO measurement data of the train during the GNSS signal interruption process; A determination module, configured to determine a pseudo GNSS position of the train during a GNSS signal interruption process based on the ODO measurement data; The positioning module is used to fuse the ODO measurement data and the pseudo GNSS position to determine the navigation information of the train during the GNSS signal interruption process.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the train positioning processing method as described in any one of claims 1-7 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the train positioning processing method as described in any one of claims 1 to 7 is implemented.

Images