Train control method and device

Abstract

The application relates to a train control method and device. The method comprises the following steps: in the process of running a virtual marshalling train, determining the safety distance of a slave train according to the actual parameters of the slave train at a current control time, and receiving the predicted speed curves sent by each preceding train of the slave train; wherein the actual parameters comprise track parameters, position parameters and running parameters; the predicted speed curve of each preceding train is the speed curve of the preceding train in a future preset time length predicted at the current control time; determining the current control instruction of the slave train at the current control time according to the predicted speed curves, the safety distance and the vehicle state of the slave train; and controlling the slave train to run according to the current control instruction. The method can improve the driving safety of the virtual marshalling train.

Description

Technical Field

[0001] This application relates to the field of train control technology, and in particular to a train control method and device. Background Technology

[0002] With the development of train control technology, virtual coupling technology has emerged. This technology allows multiple trains to be assembled into virtual train formations without relying on physical couplers. Through wireless communication, precise positioning, and coordinated control, they can synchronously start, stop, accelerate, and brake, forming a "virtual soft connection." This enables dynamic coordination and safe, close following between trains. Each train in the virtual formation can independently control its own traction and braking functions while maintaining strict relative position and speed matching with other trains, thus forming a dynamic and flexible formation system. This method overcomes the limitations of physical connections through virtual coupling operation, achieving the goal of improving transportation flexibility and track throughput capacity.

[0003] In traditional technology, in order to ensure the safe operation of virtual train formations, a static braking model based on "the rear of the preceding train as a safety protection point" is usually used to control the operation of the virtual train formations. The aforementioned static braking model requires that, assuming the speed of the preceding train is zero, the following train maintains a sufficient safe distance from the preceding train to ensure that the following train has enough distance to stop when the preceding train suddenly brakes.

[0004] However, since the core objective of virtual coupling technology is to achieve synchronous operation of the front and rear cars in a virtual train with extremely short intervals, the application of the above-mentioned traditional technology may cause the rear car to rear-end the front car due to insufficient dynamic braking distance during the operation of the virtual train, which seriously affects the driving safety of the virtual train. Summary of the Invention

[0005] Therefore, it is necessary to provide a train control method and device to address the aforementioned technical problems, which can improve the driving safety of virtual train formations.

[0006] In a first aspect, this application provides a train control method applied to each slave car in a virtual train formation, comprising:

[0007] During the operation of the virtual train formation, the safe distance of the slave train is determined based on the actual parameters of the slave train at the current control time, and the predicted speed curves sent by each preceding train of the slave train are received; the actual parameters include track parameters, position parameters and operating parameters; the predicted speed curve of each preceding train is the speed curve of the preceding train within a preset future time period predicted at the current control time.

[0008] Based on the predicted speed curves, as well as the safe distance and vehicle status of the slave vehicle, determine the current control command of the slave vehicle at the current control moment;

[0009] Control the operation of the slave vehicle according to the current control command.

[0010] In one embodiment, determining the safe distance of the slave vehicle based on its actual parameters at the current control moment includes: predicting the target track adhesion coefficient of the slave vehicle at the current control moment based on its actual parameters at the current control moment; and determining the safe distance of the slave vehicle based on the current environmental parameters, the target track adhesion coefficient, the current speed of the slave vehicle, and the current acceleration of the slave vehicle.

[0011] In one embodiment, predicting the target track adhesion coefficient of the slave vehicle at the current control moment based on the actual parameters of the slave vehicle at the current control moment includes: predicting the initial track adhesion coefficient of the slave vehicle at the current control moment based on the control parameters of the slave vehicle at the current control moment and the target track adhesion coefficient of the slave vehicle at the previous control moment; determining the prediction error based on the initial track adhesion coefficient and the actual parameters of the slave vehicle at the current control moment; and correcting the initial track adhesion coefficient based on the prediction error to obtain the target track adhesion coefficient of the slave vehicle at the current control moment.

[0012] In one embodiment, the current control command of the slave vehicle at the current control moment is determined based on each predicted speed curve, the safe distance of the slave vehicle, and the vehicle status. This includes: if the prediction error is less than a first error threshold, determining the current control command of the slave vehicle at the current control moment based on each predicted speed curve, the safe distance of the slave vehicle, and the vehicle status.

[0013] In one embodiment, determining the safe distance of the slave vehicle based on current environmental parameters, target track adhesion coefficient, current speed and current acceleration of the slave vehicle includes: updating the target track adhesion coefficient to a preset track adhesion coefficient when the prediction error is not less than a first error threshold; and determining the safe distance of the slave vehicle based on current environmental parameters, preset track adhesion coefficient, current speed and current acceleration of the slave vehicle.

[0014] In one embodiment, the train control method further includes: acquiring the sensing parameters of the slave train at the current control moment when the prediction error is not less than a first error threshold; inputting the sensing data into a preset data reconstruction model to obtain sensing reconstruction data corresponding to the sensing parameters; determining the reconstruction error between the sensing parameters and the sensing reconstruction data; and triggering preset response measures for data drift when the reconstruction error is not less than a second error threshold.

[0015] In one embodiment, triggering a preset response to data drift includes sending fault information to other trains in the virtual train formation, excluding the slave train; wherein the fault information is used to trigger other trains to reassemble the trains in the virtual train formation, excluding the slave train.

[0016] In one embodiment, receiving predicted speed curves sent by each preceding vehicle of the slave vehicle includes: receiving speed curve information carrying the digital signature and identity certificate of each preceding vehicle of the slave vehicle; wherein the speed curve information is obtained by encrypting the predicted speed curve; verifying the identity certificate for authorization, and verifying the digital signature if the identity certificate verification is successful; and decrypting the speed curve information if the digital signature verification is successful to obtain the predicted speed curve sent by the preceding vehicle.

[0017] In one embodiment, the train control method further includes: when the slave car is not the last car, predicting the predicted speed curve of the slave car within a preset time period in the future, and sending the predicted speed curve of the slave car to each of the cars following the slave car; wherein the predicted speed curve of the slave car is used to determine the current control command of each of the cars following the slave car.

[0018] Secondly, this application also provides a train control device, applied to each slave car in a virtual train formation, comprising:

[0019] The parameter determination module is used to determine the safe distance of the slave train based on the actual parameters of the slave train at the current control time during the operation of the virtual train formation, and to receive the predicted speed curves sent by each preceding train of the slave train; wherein, the actual parameters include track parameters, position parameters and operating parameters; the predicted speed curve of each preceding train is the speed curve of the preceding train within a preset future time period predicted at the current control time;

[0020] The instruction determination module is used to determine the current control instruction of the slave vehicle at the current control moment based on each predicted speed curve, as well as the safe distance and vehicle status of the slave vehicle.

[0021] The operation control module is used to control the operation of the slave vehicle according to the current control command.

[0022] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the various method embodiments provided in the first aspect above.

[0023] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the various method embodiments provided in the first aspect above.

[0024] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the various method embodiments provided in the first aspect above.

[0025] In the aforementioned train control method and apparatus, during the operation of a virtual train formation, each slave car in the virtual train formation can determine its safe distance based on its track parameters, position parameters, and operating parameters at the current control moment, and receive the predicted speed curves sent by each preceding car. Therefore, the slave car can determine its current control command at the current control moment based on the received predicted speed curves, its safe distance, and vehicle status, and control its operation according to the current control command. In this way, on the one hand, by acquiring the track parameters, position parameters, and operating parameters of the slave train, a digital twin of all elements of the train-track-environment can be constructed. By integrating the multi-source actual parameters of the slave train, the safe distance of the slave train can be determined. This safe distance is the minimum distance that ensures the slave train avoids rear-ending the preceding train during braking and stopping when the preceding train suddenly brakes. This provides a dynamic safety boundary constraint for predictive cooperative control between trains and resolves the contradiction between static safety models and dynamic control mechanisms in traditional technologies. On the other hand, by combining its own vehicle status with the predicted speed curves of each preceding train, the slave train determines the current control command and adjusts its traction / braking strategy in advance. This enables the cooperative operation of each train in a virtual train formation, ensuring the coordination and consistency between the control mechanism and safety protection. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 A schematic flowchart illustrating a train control method provided in some embodiments of this application;

[0028] Figure 2 A flowchart illustrating the process of determining a safe distance provided for some embodiments of this application;

[0029] Figure 3 A flowchart illustrating the prediction of target orbit adhesion coefficient provided in some embodiments of this application;

[0030] Figure 4 A flowchart illustrating the triggering of preset response measures provided in some embodiments of this application;

[0031] Figure 5 A flowchart illustrating the reception prediction speed curves provided in some embodiments of this application;

[0032] Figure 6 A schematic flowchart illustrating a train control method provided in other embodiments of this application;

[0033] Figure 7 Structural block diagrams of train control devices provided in some embodiments of this application;

[0034] Figure 8 Internal structural diagrams of a computer device provided in some embodiments of this application;

[0035] Figure 9 Internal structural diagrams of a computer device provided for other embodiments of this application. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0037] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments or any combination of multiple embodiments.

[0038] In traditional technology, in order to ensure the safe operation of virtual train formations, a static braking model based on "the rear of the preceding train as a safety protection point" is usually used to control the operation of the virtual train formations. The aforementioned static braking model requires that, assuming the speed of the preceding train is zero, the following train maintains a sufficient safe distance from the preceding train to ensure that the following train has enough distance to stop when the preceding train suddenly brakes.

[0039] However, since the core objective of virtual coupling technology is to achieve synchronous operation of the front and rear cars in a virtual train with extremely short intervals, the application of the above-mentioned traditional technology may cause the rear car to rear-end the front car due to insufficient dynamic braking distance during the operation of the virtual train, which seriously affects the driving safety of the virtual train.

[0040] To address the aforementioned technical problems, in one exemplary embodiment, a train control method is provided, which is applied to each slave car in a virtual train formation.

[0041] Virtual train formations consist of a master control car and slave control cars. Typically, the first train in the forward direction during a virtual train formation is called the master control car, and the other trains are called slave control cars.

[0042] In one optional embodiment, the train control method can be applied to the computer equipment corresponding to each slave car. The computer equipment can be a server or a terminal. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services; the terminal can be an onboard terminal of each slave car, etc.

[0043] In one exemplary embodiment, such as Figure 1 As shown, a train control method is provided, and the method is illustrated using an example of its application to each slave car in a virtual train formation. The method includes the following steps:

[0044] S101, during the operation of the virtual train formation, determines the safe distance of the slave train based on the actual parameters of the slave train at the current control time, and receives the predicted speed curves sent by each preceding train of the slave train.

[0045] The actual parameters include track parameters, position parameters, and operating parameters; the predicted speed curve of each preceding vehicle is the speed curve of the preceding vehicle within a preset future time period predicted at the current control moment.

[0046] Optionally, during the operation of the virtual train formation, the operation control of each train in the virtual train formation is performed periodically. Therefore, the control cycle of the virtual train formation can be preset. According to the time interval corresponding to the control cycle, each time interval marks a control moment. The time difference between two adjacent control moments is the aforementioned time interval. For example, if the train's control cycle is 0.01 seconds, then a control moment is reached every 0.01 seconds, with a 0.01-second interval between two adjacent control moments. Each time a control moment is reached, that control moment is the current control moment.

[0047] In this way, during the operation of the virtual train formation, when the current control time is reached, each slave train can obtain the actual parameters of the slave train at the current control time, and determine the safe distance of the slave train based on the actual parameters. Optionally, the safe distance is the minimum distance that ensures the slave train avoids rear-ending the train in front of it when the train in front of it suddenly brakes.

[0048] The slave train can be equipped with various sensors to collect various parameters during the virtual train formation operation. For example, the slave train is equipped with lidar, Beidou high-precision positioning module, on-board IMU (Inertial Measurement Unit), fiber optic grating (FBG) sensor, acoustic emission (AE) sensor, odometer, speed sensor, etc.

[0049] Optionally, track parameters include track surface roughness, used to reflect the microscopic morphological changes of the wheel-rail contact surface between the slave vehicle's wheels and the track; position parameters include the train position of the slave vehicle and the gradient of that position, used to correct for deviations in the track adhesion coefficient caused by terrain undulations; and operating parameters include the slave vehicle's speed information, such as acceleration and angular velocity, as input to the slave vehicle's dynamic model. Optionally, the track surface roughness can be obtained by lidar detection, the train position and gradient information of the slave vehicle can be obtained by BeiDou high-precision positioning module detection, and the speed information of the slave vehicle can be obtained by onboard IMU measurement.

[0050] Optionally, since the acquisition frequencies of different actual parameters may be different, the actual parameters whose acquisition time is closest to the current control time are obtained and used as the actual parameters of the slave vehicle at the current control time. The actual parameters are then time-synchronized and format-standardized to eliminate the problems of data timing deviation and format inconsistency between different actual parameters. Thus, the safe distance of the slave vehicle is determined based on the processed actual parameters.

[0051] Furthermore, upon reaching the current control time, each train, except for the last car in the virtual train formation (i.e., the last train in the forward direction during the virtual train formation's operation), predicts its speed curve for a future preset duration starting from the current control time, thus obtaining a predicted speed curve, and transmits this predicted speed curve to each train following it (i.e., each subsequent car). Optionally, each train, except for the last car in the virtual train formation, generates its speed curve for the future preset duration based on its actual parameters at the current control time via a PCC (Predictive Cooperative Controller). Optionally, each train, except for the last car in the virtual train formation, broadcasts the obtained predicted speed curve to each subsequent car, or transmits the obtained predicted speed curve to each subsequent car via a 5G-A (5th Generation Mobile Communication Technology-Advanced) network. The aforementioned future preset duration can be set based on empirical values, experimental values ​​from multiple trials, and application requirements in actual applications; no specific limitation is imposed, for example, the aforementioned future preset duration is 5 seconds.

[0052] Based on this, upon reaching the current control time, the slave train can also receive predicted speed curves from each train preceding it (i.e., each preceding train). Correspondingly, in an optional embodiment, the train control method further includes, if the slave train is not the last train, predicting the predicted speed curve of the slave train within a preset future time period, and sending the predicted speed curve of the slave train to each following train; wherein the predicted speed curve of the slave train is used to determine the current control command for each following train.

[0053] S102, based on the predicted speed curves, the safe distance of the slave vehicle, and the vehicle status, determine the current control command of the slave vehicle at the current control moment.

[0054] After obtaining the aforementioned safe distance and predicted speed curves, the slave train can use the safe distance as a dynamic safety constraint boundary, combined with the predicted speed curves and the train's vehicle status, to adjust its traction or braking strategy. This includes adjusting the train's speed, acceleration, and other speed information to ensure it maintains a sufficient safe distance during operation, preventing rear-end collisions and other accidents, thereby achieving coordinated operation of trains in the virtual train formation. Furthermore, the slave train can determine its current control command at the current control moment based on the predicted speed curves, its safe distance, and its vehicle status.

[0055] Optionally, the vehicle status mentioned above may include vehicle operating status, such as speed, acceleration, load, current remaining braking distance (i.e., the distance to the nearest preceding vehicle); further optionally, the vehicle status may also include vehicle environmental status, such as gradient information, target track adhesion coefficient, etc.

[0056] S103 controls the operation of the slave vehicle according to the current control command.

[0057] Among them, the current control command is a control command used to control the operation of the slave vehicle at the current control moment. Thus, the slave vehicle can be controlled to run at the current control moment according to the current control command.

[0058] In one optional embodiment, the current control command may be an air brake control command or a traction control command. Optionally, the current control command may include at least one of the following data: speed, acceleration, air brake force increment, traction force increment, etc.

[0059] In the above-mentioned train control method, during the operation of the virtual train formation, each slave train in the virtual train formation can determine its safe distance based on its track parameters, position parameters, and operating parameters at the current control time, and receive the predicted speed curves sent by each preceding train. Thus, the slave train can determine its current control command at the current control time based on the received predicted speed curves, its safe distance, and vehicle status, and control its operation according to the current control command. In this way, on the one hand, by acquiring the track parameters, position parameters, and operating parameters of the slave train, a digital twin of all elements of the train-track-environment can be constructed. By integrating the multi-source actual parameters of the slave train, the safe distance of the slave train can be determined. This safe distance is the minimum distance that ensures the slave train avoids rear-ending the preceding train during braking and stopping when the preceding train suddenly brakes. This provides a dynamic safety boundary constraint for predictive cooperative control between trains and resolves the contradiction between static safety models and dynamic control mechanisms in traditional technologies. On the other hand, by combining its own vehicle status with the predicted speed curves of each preceding train, the slave train determines the current control command and adjusts its traction / braking strategy in advance. This enables the cooperative operation of each train in a virtual train formation, ensuring the coordination and consistency between the control mechanism and safety protection.

[0060] Based on the above embodiments, in an exemplary embodiment, the determination of the safety distance in S101 is further refined. Optionally, as... Figure 2 As shown, it includes the following steps:

[0061] S201, based on the actual parameters of the slave vehicle at the current control moment, predict the target track adhesion coefficient of the slave vehicle at the current control moment.

[0062] The so-called track adhesion coefficient refers to the ratio of the maximum adhesion force between the wheel and the rail to the normal load of the wheelset (axle load / adhesive weight). In essence, it is the dynamic equivalent friction coefficient of near-static friction between the wheel and the rail.

[0063] Upon reaching the current control moment, the slave train can first predict the target track adhesion coefficient of the slave train at the current control moment based on the actual parameters of the slave train at the current control moment.

[0064] Optionally, the actual parameters of the slave vehicle at the current control moment can be input into a preset track adhesion coefficient prediction model to obtain the target track adhesion coefficient output by the track adhesion coefficient model. This target track adhesion coefficient model is trained based on historical actual parameters and historical true track adhesion coefficients.

[0065] Optionally, Kalman filtering can be used to predict the target track adhesion coefficient of the slave vehicle at the current control moment based on the actual parameters of the slave vehicle at the current control moment.

[0066] S202. Based on the current environmental parameters, the target track adhesion coefficient, the current speed and current acceleration of the slave vehicle, determine the safe distance of the slave vehicle.

[0067] After obtaining the target track adhesion coefficient, current environmental parameters, such as current wind speed and gradient of the current train position of the slave train, can be further obtained to determine the safe distance of the slave train based on the current environmental parameters, the target track adhesion coefficient, and the current speed and acceleration of the slave train.

[0068] Optionally, the safe distance from the controlled vehicle can be determined using the following formula:

[0069]

[0070] in, To maintain a safe distance from the vehicle, To control the vehicle's current speed, To control the current acceleration of the vehicle, The target orbital adhesion coefficient, The gradient should be determined from the position of the train where the control unit is located. This represents the current wind speed.

[0071] Based on the above embodiments, in an exemplary embodiment, the prediction of the target orbital adhesion coefficient in S201 is further refined. Optionally, such as Figure 3 As shown, it includes the following steps:

[0072] S301, based on the control parameters of the slave train at the current control moment and the target track adhesion coefficient of the slave train at the previous control moment, predict the initial track adhesion coefficient of the slave train at the current control moment.

[0073] In one optional embodiment, the track adhesion coefficient is used as a dynamic state variable. Based on the train dynamics model of the slave vehicle, a state transition equation is constructed. Using this state transition equation, the initial track adhesion coefficient of the slave vehicle at the current control time is predicted according to the control parameters of the slave vehicle at the current control time and the target track adhesion coefficient of the slave vehicle at the previous control time.

[0074] S302, based on the initial track adhesion coefficient and the actual parameters of the slave train at the current control moment, determine the prediction error.

[0075] Optionally, based on the initial track adhesion coefficient, the predicted observation parameters of the slave vehicle at the current control moment are determined, and then the difference between the above-mentioned observation parameters and the actual parameters is determined as the prediction error.

[0076] S303, based on the prediction error, the initial track adhesion coefficient is corrected to obtain the target track adhesion coefficient of the slave vehicle at the current control moment.

[0077] Optionally, based on the initial track adhesion coefficient, the corresponding prior error covariance matrix is ​​determined. Then, based on the aforementioned prior error covariance matrix, a preset observation matrix, and the transformation matrix of the aforementioned observation matrix, a preset observation noise covariance matrix is ​​simultaneously formed. The gain coefficient is obtained by solving the optimal weights. The product of the aforementioned gain coefficient and the aforementioned prediction error is determined, and then the initial track adhesion coefficient is corrected based on this product to obtain the target track adhesion coefficient of the slave vehicle at the current control moment. For example, adding or subtracting the aforementioned product from the initial track adhesion coefficient yields the target track adhesion coefficient.

[0078] Optionally, if the prediction error is less than a first error threshold, the initial track adhesion coefficient is corrected based on the prediction error to obtain the target track adhesion coefficient of the slave vehicle at the current control moment. Specifically, after obtaining the prediction error, to ensure the accuracy of the final target track adhesion coefficient, the relationship between the prediction error and the first error threshold can be compared firstly. Then, if the prediction error is less than the first error threshold, the initial track adhesion coefficient is corrected based on the prediction error to obtain the target track adhesion coefficient of the slave vehicle at the current control moment.

[0079] In one optional embodiment, the method for determining the safe distance in S202 above includes updating the target track adhesion coefficient to a preset track adhesion coefficient when the prediction error is not less than a first error threshold; and determining the safe distance of the slave vehicle based on the current environmental parameters, the preset track adhesion coefficient, the current speed and current acceleration of the slave vehicle.

[0080] In this embodiment, if the prediction error is not less than the first error threshold, it can be considered that the prediction of the target track adhesion coefficient of the slave train at the current control moment, based on the actual parameters of the slave train at the current control moment, is no longer accurate. Therefore, the slave train can switch to a preset conservative control mode, making the control of the slave train more cautious to ensure its operational safety. In the conservative control mode, it is necessary to increase the safety distance between adjacent trains, i.e., increase the safety distance of the slave train. As mentioned earlier, the safety distance of the slave train is determined based on its track adhesion coefficient. Therefore, a more conservative (i.e., smaller) predicted track adhesion coefficient can be used to determine the safety distance of the slave train. In this way, even if the actual track adhesion coefficient is greater than the predicted track adhesion coefficient, the slave train can have a sufficient safety distance.

[0081] Based on this, when the prediction error is not less than the first error threshold, the target track adhesion coefficient is updated to a preset track adhesion coefficient. This allows for the determination of the safe distance for the slave vehicle based on current environmental parameters, the preset track adhesion coefficient, and the current speed and acceleration of the slave vehicle. The first error threshold and the preset track adhesion coefficient can be set based on empirical values, experimental values ​​from multiple tests, and application requirements in practical applications. No specific limitations are imposed, as long as sufficient time and space (safety distance) are provided for the slave vehicle to brake or avoid obstacles in the event of abnormal prediction errors. Optionally, when the prediction error is not less than the first error threshold, the acceleration of the slave vehicle can be reduced and / or its maximum speed limited to reduce reliance on the target track adhesion coefficient prediction and improve the operational stability of the slave vehicle.

[0082] Optionally, if the prediction error is less than a first error threshold, the safe distance of the slave vehicle is determined based on the current environmental parameters, the target track adhesion coefficient, and the current speed and acceleration of the slave vehicle; or, if the prediction error is not less than the first error threshold, the safe distance of the slave vehicle is determined based on the current environmental parameters, the preset track adhesion coefficient, and the current speed and acceleration of the slave vehicle. In this way, if the prediction error is not less than the first error threshold, it is unnecessary to determine the target track adhesion coefficient, saving computation time and resources and improving control efficiency.

[0083] Optionally, at each control moment, the target orbit adhesion coefficient can be predicted based on filtering methods such as extended Kalman filtering, unscented Kalman filtering, and particle filtering. The extended Kalman filtering method will be used as an example for explanation.

[0084] Treating the track adhesion coefficient as a dynamic state variable, and based on the train dynamics model of the slave-controlled vehicle, the state transition equation is constructed as follows:

[0085]

[0086] in, Let the initial track adhesion coefficient of the train at the current control moment be denoted as . Here is the state transition matrix. To control the input matrix, The control parameters (such as speed, acceleration, etc.) of the slave vehicle at the current control moment. For process noise, The viscosity coefficient of the target track at the previous control moment is given by the train control. For the current control moment, This is the previous control moment.

[0087] The observation equation is constructed by using the actual parameters mentioned above to create the observation variables, as shown in the following equation:

[0088]

[0089] in, For observed variables, For the observation matrix, To observe the noise, the above actual parameters are mapped to... middle.

[0090] The prior state estimate is calculated as follows:

[0091]

[0092] in, For the above The initial track adhesion coefficient of the train at the current control moment; For the above The adhesion coefficient of the target track at the previous control moment.

[0093] The prior error covariance is calculated as shown in the following formula;

[0094]

[0095] in, To predict the uncertainty of the current control moment, This represents the uncertainty after correction at the previous control time. Let be the process noise covariance.

[0096] The Kalman gain is calculated as follows:

[0097]

[0098] in, The weights used to balance prediction and observation at the current control moment. To observe the noise covariance.

[0099] The corrected state estimate is shown in the following equation:

[0100]

[0101] in, The target track adhesion coefficient of the slave vehicle at the current control moment;

[0102] and then, To determine the predictive observation parameters based on the initial track adhesion coefficient of the slave vehicle at the current control moment, This represents the aforementioned prediction error.

[0103] The error covariance is updated as shown in the following equation:

[0104]

[0105] in, This represents the uncertainty after correction at the current control moment.

[0106] Based on the above embodiments, in an exemplary embodiment, this application further provides a triggering method for a preset response measure when the prediction error is not less than a first error threshold. Optionally, such as Figure 4 As shown, it includes the following steps:

[0107] S401, if the prediction error is not less than the first error threshold, acquire the sensing parameters of the slave vehicle at the current control moment.

[0108] If the prediction error is not less than the first error threshold, data augmentation can be triggered to address the prediction error anomaly, thereby first obtaining the sensing parameters of the slave vehicle at the current control moment.

[0109] Optionally, the aforementioned sensing parameters include: speed information of the slave vehicle measured by the IMU, such as acceleration and angular velocity, to monitor the motion state of the slave vehicle in real time; strain and temperature physical quantities collected by fiber optic grating sensors to provide health status information of the track and slave vehicle structure; and acoustic signals collected by acoustic detection sensors during the operation of the slave vehicle to detect abnormalities such as track cracks and slave vehicle wheel failures.

[0110] Optionally, the initial parameters corresponding to the current control moment are acquired from various sensors, and these initial parameters are preprocessed. For example, digital filters are used to remove noise from the initial parameters and improve data quality; the initial parameters acquired by the IMU are filtered to eliminate high-frequency noise and vibration interference; the initial parameters acquired by the FBG and AE sensors are calibrated to improve the accuracy of the measurement results. Feature extraction is then performed on the preprocessed initial parameters, extracting acceleration and angular velocity from the preprocessed IMU parameters, strain and temperature features from the preprocessed FBG parameters, and acoustic signals from the preprocessed AE sensor parameters. Furthermore, the extracted features are fused to obtain a more comprehensive feature representation. For example, a feature layer fusion method can be used to splice or weightedly combine features from different sensors to obtain a more comprehensive feature representation. Based on the fused features, a sensing matrix is ​​constructed as the sensing parameters. The sensing matrix is ​​a multidimensional array, where each element represents a specific element (such as track state, train attitude, acoustic event, etc.) and its state (such as normal, abnormal, fault, etc.).

[0111] S402, input the perception parameters into the preset data reconstruction model to obtain the perception reconstruction data corresponding to the perception parameters.

[0112] S403, determine the reconstruction error between the sensing parameters and the sensing reconstruction data.

[0113] The aforementioned perception parameters are input into a preset data reconstruction model to obtain the perception reconstruction data corresponding to the perception parameters, and the reconstruction error between the perception parameters and the perception reconstruction data is further determined.

[0114] Optionally, an autoencoder (i.e., a data reconstruction model) can be trained using historical sensor data to learn feature representations of normal sensor data. During training, the autoencoder learns to compress input data into a latent representation (encoding), and then reconstructs the original input data from the latent representation (decoding). By minimizing the reconstruction error, the autoencoder can capture the main features of the input data. Then, real-time acquired sensor data (i.e., sensing parameters) is input into the trained autoencoder. The autoencoder encodes and decodes the input data, generates reconstructed data, and outputs it. This allows the determination of the difference (i.e., reconstruction error) between the input data (i.e., sensing parameters) and the autoencoder's reconstructed data (the sensing reconstructed data corresponding to the sensing parameters output by the sensing reconstruction model).

[0115] S404, if the reconstruction error is not less than the second error threshold, trigger the preset response measures for data drift.

[0116] Based on the aforementioned reconstruction error, it can be determined whether data drift exists. If data drift exists, preset response measures for data drift can be triggered, such as data correction, sensor replacement, or model update, to ensure the high accuracy of the aforementioned actual parameters and the reliability of the current control command decision.

[0117] Based on this, after obtaining the relationship between the reconstruction error and the second error threshold, and further determining that data drift exists when the reconstruction error is not less than the second error threshold, a preset response measure for data drift is triggered.

[0118] Optionally, the above actual parameters can be corrected to correct the actual drift parameters. Then, based on the corrected actual parameters, the safe distance for vehicle control can be re-established, and the current control command can be re-determined.

[0119] Optionally, the system can detect the cause of data drift and, if the cause is a sensor malfunction, output alarm information about the sensor malfunction, such as voice prompts, illuminating warning lights on the sensor, or displaying text prompts on the screen, to remind technicians to troubleshoot and replace the sensor.

[0120] In one optional embodiment, the triggering method of the preset response measure in 404 above includes sending fault information to other trains in the virtual train formation except for the slave control car; wherein, the fault information is used to trigger other trains to reassemble the trains in the virtual train formation except for the slave control car.

[0121] In this embodiment, if the reconstruction error is not less than the second error threshold, it can be determined that the slave train has malfunctioned. Therefore, the slave train can send fault information, such as its identity information, location information, and vehicle status, to other trains in the virtual train formation. This allows other trains to reach a consensus on the soft isolation decision for the slave train based on the fault information. Consequently, the PCCs of other trains can automatically "soft-isolate" the slave train from the virtual train formation, marking it as faulty, and reconstruct the formation of all trains in the virtual train formation except for the slave train. This updates the formation control strategy of the virtual train formation, ignoring various information released by the faulty slave train, such as predicted speed curves and control commands, thus enhancing the safety and reliability of the control mechanism in dealing with emergencies. Optionally, each train in the virtual train formation runs a distributed consensus algorithm to reach a consensus on the soft isolation decision for the slave train.

[0122] Optionally, when any train in the virtual trainset detects a braking system failure, that train (or the lead train in the trainset) broadcasts the failure information via a communication link. This allows other trains in the virtual trainset, excluding that train, to reach a consensus on a soft isolation decision for that train based on the failure information. Consequently, the PCCs of other trains can automatically "soft isolate" that train from the virtual trainset, mark it as faulty, and reconfigure the trainset for all other trains in the virtual trainset to update the trainset control strategy and ignore control commands and other information issued by the faulty train.

[0123] Optionally, each train in the virtual trainset can collect data such as network traffic and command content from the communication link as input for intrusion detection; perform preprocessing operations such as cleaning and format conversion on the data collected by the sensors to improve the accuracy of subsequent analysis based on the data collected by the sensors; extract command frequency, packet size, source and destination address features related to intrusion detection; and use machine learning and deep learning algorithms (such as isolated forest and autoencoder) to build an anomaly detection model to identify abnormal communication behavior in real time, so as to achieve decentralized intrusion detection (DID) and ensure the communication security between different trains.

[0124] Based on the above embodiments, in an exemplary embodiment, the reception of the predicted velocity curve in S101 is further refined. Optionally, such as Figure 5 As shown, it includes the following steps:

[0125] S501 receives speed curve information from each preceding vehicle of the controlled vehicle, carrying the digital signature and identity certificate of the preceding vehicle.

[0126] Among them, the velocity curve information is obtained by encrypting the predicted velocity curve.

[0127] Each train, except the last car in a virtual train formation, can encrypt its predicted speed curve for a predetermined future timeframe after obtaining the predicted speed curve. This encrypted speed curve information is then sent to each of the train's successor cars, carrying the train's digital signature and identity certificate. Based on this, upon reaching the current control time, the slave train can receive speed curve information from each preceding car, with that information carrying the preceding car's digital signature and identity certificate.

[0128] S502 verifies the permissions of the identity certificate and, if the identity certificate verification is successful, verifies the digital signature.

[0129] S503, after the digital signature verification is successful, decrypts the speed curve information to obtain the predicted speed curve sent by the preceding vehicle.

[0130] For each speed curve information sent by the preceding train, the slave train first verifies the authorization of the identity certificate carried in the speed curve information to determine whether the preceding train sending the speed curve information is an authorized train, or whether the preceding train is a faulty train. If the identity certificate verification is successful, the preceding train sending the speed curve information can be determined to be an authorized train; if it is determined that the preceding train is not a faulty train, then, if the identity certificate verification is successful, the digital signature carried in the speed curve information is verified to ensure the integrity and source authentication of the speed curve information. Furthermore, if the digital signature verification is successful, the authenticity and integrity of the speed curve information can be determined, and the speed curve information can then be decrypted to obtain the predicted speed curve sent by the preceding train.

[0131] Optionally, when any train in a virtual trainset receives a control command from another train carrying the other train's digital signature and identity certificate, it first verifies the identity certificate's authorization. If the identity certificate verification is successful, it then verifies the digital signature. Subsequently, if the digital signature verification is successful, it executes the control command.

[0132] Optionally, blockchain nodes can be deployed on each train of the virtual trainset to form a distributed network. These blockchain nodes are responsible for verifying and storing communication commands, ensuring the immutability and transparency of the data. Consensus algorithms suitable for the onboard environment, such as Practical Byzantine Fault Tolerance (PBFT) and Raft (Raft Consensus Algorithm), can be employed to achieve rapid consensus and reduce computational overhead. Integrating the blockchain module with existing train-to-train communication protocols (such as 5G-A) enables secure and efficient transmission of information (predicted speed curves, control commands, etc.). Optionally, each train in the virtual trainset can use blockchain-supported encryption algorithms, such as AES (Advanced Encryption Standard) and RSA (Rivest-Shamir-Adleman), to encrypt information (predicted speed curves, control commands, etc.).

[0133] Based on the above embodiments, in an exemplary embodiment, the train control method is applied to each slave car in a virtual train formation, such as... Figure 6 As shown, it includes the following steps:

[0134] S601, during the operation of a virtual train formation, predicts the initial track adhesion coefficient of the slave train at the current control moment based on the control parameters of the slave train at the current control moment and the target track adhesion coefficient of the slave train at the previous control moment.

[0135] S602. Based on the initial track adhesion coefficient and the actual parameters of the slave vehicle at the current control moment, determine the prediction error, and correct the initial track adhesion coefficient based on the prediction error to obtain the target track adhesion coefficient of the slave vehicle at the current control moment.

[0136] S603, if the prediction error is less than the first error threshold, determine the safe distance of the slave vehicle based on the current environmental parameters, the target track adhesion coefficient, the current speed of the slave vehicle, and the current acceleration; or if the prediction error is not less than the first error threshold, update the target track adhesion coefficient to the preset track adhesion coefficient, and determine the safe distance of the slave vehicle based on the current environmental parameters, the preset track adhesion coefficient, the current speed of the slave vehicle, and the current acceleration.

[0137] S604 receives speed curve information from each preceding vehicle of the controlled vehicle, carrying the digital signature and identity certificate of the preceding vehicle, and performs authorization verification on the identity certificate.

[0138] S605 verifies the digital signature if the identity certificate verification is successful; and decrypts the speed curve information to obtain the predicted speed curve sent by the preceding vehicle if the digital signature verification is successful.

[0139] S606, when the controlled vehicle is not the last vehicle, predicts the predicted speed curve of the controlled vehicle within a preset time period in the future, and sends the predicted speed curve of the controlled vehicle to each vehicle following the controlled vehicle.

[0140] S607 determines the current control command of the slave vehicle at the current control moment based on the predicted speed curves, the safe distance of the slave vehicle, and the vehicle status.

[0141] S608 controls the operation of the slave vehicle according to the current control command.

[0142] S609: Under the condition that the prediction error is not less than the first error threshold, the sensing parameters of the slave vehicle at the current control moment are obtained.

[0143] S610: Input the perception parameters into the preset data reconstruction model to obtain the perception reconstruction data corresponding to the perception parameters.

[0144] S611, determine the reconstruction error between the sensing parameters and the sensing reconstruction data, and if the reconstruction error is not less than the second error threshold, send fault information to other trains in the virtual train group except for the slave train.

[0145] The specific implementation methods of S601-S611 are the same as those in the above method embodiments, and will not be repeated here.

[0146] It should be understood that although the steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0147] Based on the same inventive concept, this application also provides a train control device for implementing the train control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more train control device embodiments provided below can be found in the limitations of the train control method described above, and will not be repeated here.

[0148] In one exemplary embodiment, such as Figure 7 As shown, a train control device is provided, applied to each slave car in a virtual train formation, including: a parameter determination module 710, an instruction determination module 720, and a running control module 730, wherein:

[0149] The parameter determination module 710 is used to determine the safe distance of the slave train based on the actual parameters of the slave train at the current control time during the operation of the virtual train formation, and to receive the predicted speed curves sent by each preceding train of the slave train; wherein, the actual parameters include track parameters, position parameters and operating parameters; the predicted speed curve of each preceding train is the speed curve of the preceding train within a future preset time period predicted at the current control time.

[0150] The instruction determination module 720 is used to determine the current control instruction of the slave vehicle at the current control moment based on each predicted speed curve, as well as the safe distance and vehicle status of the slave vehicle.

[0151] The operation control module 730 is used to control the operation of the slave vehicle according to the current control command.

[0152] In an exemplary embodiment, the parameter determination module 710 includes: a coefficient determination unit, used to predict the target track adhesion coefficient of the slave vehicle at the current control time based on the actual parameters of the slave vehicle at the current control time; and a distance determination unit, used to determine the safe distance of the slave vehicle based on the current environmental parameters, the target track adhesion coefficient, the current speed and current acceleration of the slave vehicle.

[0153] In an exemplary embodiment, the coefficient determination unit includes: a coefficient prediction subunit, configured to predict the initial track adhesion coefficient of the slave vehicle at the current control time based on the control parameters of the slave vehicle at the current control time and the target track adhesion coefficient of the slave vehicle at the previous control time; an error determination subunit, configured to determine the prediction error based on the initial track adhesion coefficient and the actual parameters of the slave vehicle at the current control time; and a coefficient determination subunit, configured to correct the initial track adhesion coefficient based on the prediction error to obtain the target track adhesion coefficient of the slave vehicle at the current control time.

[0154] In an exemplary embodiment, the coefficient determination subunit is specifically used to: update the target track adhesion coefficient to a preset track adhesion coefficient when the prediction error is not less than a first error threshold; and determine the safe distance of the slave vehicle based on the current environmental parameters, the preset track adhesion coefficient, the current speed of the slave vehicle, and the current acceleration.

[0155] In an exemplary embodiment, the train control device further includes: a parameter acquisition module, used to acquire the sensing parameters of the slave train at the current control moment when the prediction error is not less than a first error threshold; a data reconstruction module, used to input the sensing parameters into a preset data reconstruction model to obtain sensing reconstruction data corresponding to the sensing parameters; an error determination module, used to determine the reconstruction error between the sensing parameters and the sensing reconstruction data; and a measure triggering module, used to trigger preset response measures for data drift when the reconstruction error is not less than a second error threshold.

[0156] In an exemplary embodiment, the measure triggering module is specifically used to: send fault information to other trains in the virtual train formation except for the slave control car; wherein the fault information is used to trigger other trains to reassemble the trains in the virtual train formation except for the slave control car.

[0157] In an exemplary embodiment, the parameter determination module 710 is specifically configured to: receive speed curve information carrying the digital signature and identity certificate of the preceding vehicle sent by each preceding vehicle of the controlled vehicle; wherein the speed curve information is obtained by encrypting the predicted speed curve; verify the identity certificate for authorization, and if the identity certificate verification is successful, verify the digital signature; if the digital signature verification is successful, decrypt the speed curve information to obtain the predicted speed curve sent by the preceding vehicle.

[0158] In an exemplary embodiment, the train control device further includes: a speed prediction module, used to predict the predicted speed curve of the slave car within a preset time period in the case that the slave car is not the last car, and to send the predicted speed curve of the slave car to each of the following cars of the slave car; wherein the predicted speed curve of the slave car is used to determine the current control command of each of the following cars of the slave car.

[0159] Each module in the aforementioned train control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0160] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores data such as actual parameters of the controlled train. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network. When the computer program is executed by the processor, it implements a train control method.

[0161] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a train control method. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0162] Those skilled in the art will understand that Figure 8 and Figure 9The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0163] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the various method embodiments of the train control method described above.

[0164] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the various method embodiments of the above-described train control method.

[0165] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the various method embodiments of the train control method described above.

[0166] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0167] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0168] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A train control method, characterized in that, The method is applied to each slave car in a virtual train formation; the method includes: During the operation of the virtual train formation, the safe distance of the slave train is determined based on the actual parameters of the slave train at the current control time, and the predicted speed curves sent by each preceding train of the slave train are received; wherein, the actual parameters include track parameters, position parameters and operating parameters; the predicted speed curve of each preceding train is the speed curve of the preceding train within a future preset time period predicted at the current control time; Based on the predicted speed curves, the safe distance of the slave vehicle, and the vehicle status, the current control command of the slave vehicle at the current control moment is determined. The slave vehicle is controlled to operate according to the current control command.

2. The method according to claim 1, characterized in that, Determining the safe distance of the slave vehicle based on its actual parameters at the current control moment includes: Based on the actual parameters of the slave vehicle at the current control moment, predict the target track adhesion coefficient of the slave vehicle at the current control moment; The safe distance of the slave vehicle is determined based on the current environmental parameters, the adhesion coefficient of the target track, the current speed and acceleration of the slave vehicle.

3. The method according to claim 2, characterized in that, The step of predicting the target track adhesion coefficient of the slave vehicle at the current control moment based on the actual parameters of the slave vehicle at the current control moment includes: Based on the control parameters of the slave vehicle at the current control moment and the target track adhesion coefficient of the slave vehicle at the previous control moment, the initial track adhesion coefficient of the slave vehicle at the current control moment is predicted. The prediction error is determined based on the initial track adhesion coefficient and the actual parameters of the slave vehicle at the current control moment; Based on the prediction error, the initial track adhesion coefficient is corrected to obtain the target track adhesion coefficient of the slave vehicle at the current control moment.

4. The method according to claim 3, characterized in that, Determining the safe distance of the slave vehicle based on current environmental parameters, the target track adhesion coefficient, and the current speed and acceleration of the slave vehicle includes: If the prediction error is not less than the first error threshold, the target orbit adhesion coefficient is updated to a preset orbit adhesion coefficient. The safe distance of the slave vehicle is determined based on the current environmental parameters, the preset track adhesion coefficient, the current speed and current acceleration of the slave vehicle.

5. The method according to claim 3, characterized in that, The method further includes: If the prediction error is not less than the first error threshold, the sensing parameters of the slave vehicle at the current control moment are obtained. The sensing parameters are input into a preset data reconstruction model to obtain the sensing reconstruction data corresponding to the sensing data. Determine the reconstruction error between the perceived data and the perceived reconstructed data; If the reconstruction error is not less than the second error threshold, a preset response measure for data drift is triggered.

6. The method according to claim 5, characterized in that, The preset response measures for triggering data drift include: The fault information is sent to all trains in the virtual train formation except for the slave car; wherein the fault information is used to trigger the other trains to reassemble the trains in the virtual train formation except for the slave car.

7. The method according to any one of claims 1-6, characterized in that, Receiving the predicted speed curves sent by each preceding vehicle from the slave vehicle includes: The system receives speed curve information sent by each preceding vehicle of the slave vehicle, carrying the digital signature and identity certificate of the preceding vehicle; wherein the speed curve information is obtained by encrypting the predicted speed curve. The identity certificate is verified for permissions, and if the identity certificate is verified, the digital signature is verified. If the digital signature verification is successful, the speed curve information is decrypted to obtain the predicted speed curve sent by the preceding vehicle.

8. The method according to any one of claims 1-6, characterized in that, The method further includes: When the slave vehicle is not the last vehicle, the predicted speed curve of the slave vehicle is predicted within the preset future time period, and the predicted speed curve of the slave vehicle is sent to each of the following vehicles of the slave vehicle; wherein, the predicted speed curve of the slave vehicle is used to determine the current control command of each of the following vehicles of the slave vehicle.

9. A train control device, characterized in that, The device, applied to each slave car in a virtual train formation, includes: The parameter determination module is used to determine the safe distance of the slave train based on the actual parameters of the slave train at the current control time during the operation of the virtual train formation, and to receive the predicted speed curves sent by each preceding train of the slave train; wherein, the actual parameters include track parameters, position parameters and operating parameters; the predicted speed curve of each preceding train is the speed curve of the preceding train within a future preset time period predicted at the current control time; The instruction determination module is used to determine the current control instruction of the slave vehicle at the current control moment based on each predicted speed curve, as well as the safe distance and vehicle status of the slave vehicle. The operation control module is used to control the operation of the slave vehicle according to the current control command.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images