Train benchmarking method, system, electronic equipment, storage medium and program product

Through the real-time positioning and ranging of the train intelligent detection system, combined with the limited time optimal decoupling control strategy, the problem of inaccurate distance during the train jump benchmarking process is solved, and the accuracy and efficiency of benchmarking are improved.

CN118409499BActive Publication Date: 2024-11-15BEIJING URBAN CONSTR INTELLIGENT CONTROL TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410853193.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-06-28
Publication Date
2024-11-15
Estimated Expiration
2044-06-28

AI Technical Summary

Technical Problem

In the prior art, there is a problem of inaccurate jump distance during the jumping benchmarking process of trains, resulting in poor practicality.

Method used

The jump benchmarking method based on the train intelligent detection system is adopted, and through real-time positioning and ranging information, combined with the limited time optimal decoupling control strategy, the train power level and power instructions are accurately controlled to ensure the train's accurate benchmarking.

Benefits of technology

It improves the accuracy and efficiency of train jump targets, reduces the interaction between variables, and achieves more accurate determination of power levels and accurate positioning and ranging.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118409499B_ABST
    Figure CN118409499B_ABST
Patent Text Reader

Abstract

The present invention provides a train alignment method, system, electronic device, storage medium and program product, which belongs to the technical field of vehicle brake control system. Based on the input data set, the first train state vector and the first train control rate vector, the second train state vector is determined, the current power level is obtained, the current power command is determined, and the train is controlled to move to the position of the target parking window based on the current power level and the current power command; based on the current power level and the optimal motion variable, the second train control rate vector is determined, and the current power level and the current power command are updated until the train stops moving. The present invention realizes the precise positioning and ranging of the train. The present invention realizes the optimal decoupling control strategy based on finite time, minimizes the interaction effect between variables, improves the accuracy of determining the second train state vector, improves the accuracy of determining the current power level, and is conducive to improving the accuracy and efficiency of train alignment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle brake control systems, and in particular to a train benchmarking method, system, electronic equipment, storage medium and program product. Background Art

[0002] With the development of fully automatic operation systems, people have higher and higher requirements for the accuracy of fully automatic operation. When a train stops at a station under or over a certain range in the Fixed Block Automatic Block (FAM) and Continuous Automatic Block (CAM) modes, the onboard controller controls the train to move forward or backward and stop at the same mark again, which is called "jump mark".

[0003] In the first existing jump marking process, after the jump distance reaches the preset distance, the traction command is invalid and the traction level is reset to zero, the train is controlled to slow down and stop by outputting the brake after a certain delay. In the second existing jump marking process, the train obtains the current position information of the train through the transponder, and then calculates the jump distance.

[0004] In the existing jump benchmarking process, there is a possibility that the jump distance exceeds expectations, resulting in inaccurate jumps and poor practicality. Summary of the invention

[0005] The present invention provides a train alignment method, system, electronic equipment, storage medium and program product, which are used to solve the defect of inaccurate train jumping in the prior art and improve the accuracy of train jumping.

[0006] The present invention provides a train benchmarking method, which is applied to a train intelligent detection system, comprising the following steps: in response to a start jump instruction sent by a train automatic driving system, based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period, determining a second train state vector of the train in a current time period; the input data set is used to characterize the position of the train relative to a target parking window and information on the track on which the train is located, the first train state vector is used to characterize the benchmarking state of the train during the benchmarking process, the benchmarking state includes a traction state or an idling state or a braking state, and the first train control rate vector is used to characterize the magnitude of the control power required by the train during the benchmarking process; the second train The state vector is converted into the current power level of the train in the current time period, and the current power command is determined based on the input data set of the current time period, and the current power level and the current power command are sent to the train automatic driving system, so that the train automatic driving system controls the train to move to the position of the target parking window based on the current power level and the current power command; based on the current power level and the optimal motion variable, the second train control rate vector of the train in the current time period is determined, and the current time period is used as the previous time period to update the current power level and the current power command until the train stops moving, the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period.

[0007] A train benchmarking method provided by the present invention determines the second train state vector of the train in the current time period based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period, including: determining a state matrix and a control rate matrix based on the input data set of the previous time period, the state matrix representing the relationship between the benchmarking state of the previous time period and the input data set of the previous time period, and the control rate matrix representing the relationship between the control power of the previous time period and the input data set of the previous time period; determining the second train state vector based on the first train state vector, the first train control rate vector, the state matrix and the control rate matrix.

[0008] According to a train benchmarking method provided by the present invention, the optimal motion variable is determined based on the following method: obtaining the average state of the state matrix of the previous time period and the average control rate of the control rate matrix of the previous time period; the average state is the average value of multiple state values ​​in the state matrix, and the average control rate is the average value of multiple control rates in the control rate matrix; based on the average state, the average control rate, the preset weighting matrix and the preset weight matrix, the state covariance matrix of the train is determined; the preset weighting matrix is ​​used to characterize the relationship between the minimum energy consumption of the train, the optimal tracking control variable and the tracking error, the tracking error is the difference between the current power level and the standard power level, and the optimal tracking control variable is the control variable corresponding to the minimum energy consumption; the preset weighting matrix is ​​used to characterize the conversion relationship between the optimal tracking control variable and the tracking error; based on the state covariance matrix, the standard power level, the average state, the average control rate, the preset weighting matrix and the preset weight matrix, the optimal motion variable is determined.

[0009] According to a train benchmarking method provided by the present invention, based on the current power level and the optimal motion variable, the second train control rate vector of the train in the current time period is determined, including: constructing a conversion relationship between the power level, the optimal tracking control variable and the train control rate vector to obtain conversion information, the conversion information includes a first diagonal matrix, a second diagonal matrix, a first characteristic matrix and a second characteristic matrix, and the main diagonal elements of the first characteristic matrix and the second characteristic matrix are both zero; based on the first diagonal matrix, the first characteristic matrix, the second characteristic matrix, the second diagonal matrix, the average control rate, the preset weight matrix, the optimal motion variable and the current power level, the second train control rate vector is determined.

[0010] According to a train benchmarking method provided by the present invention, the average state of the state matrix is ​​obtained, including: determining the state intermediate matrix based on the state matrix, the transposed matrix of the preset conversion matrix and the preset conversion matrix, the preset conversion matrix representing the conversion relationship between the second train state vector and the current power level; obtaining each state value on the diagonal of the state intermediate matrix; and taking the average value of each state value as the average state.

[0011] According to a train alignment method provided by the present invention, an input data set includes the current position of the train, the maximum distance between the train and the target parking window, the minimum distance between the train and the target parking window, the track slope of the track where the train is located, the track friction coefficient of the track where the train is located, and the track obstacle information of the track where the train is located.

[0012] According to a train benchmarking method provided by the present invention, the current power instruction includes a traction instruction, a coasting instruction or a braking instruction, and the current power instruction is determined based on an input data set of a current time period, including: if the minimum distance in the input data set of the current time period is greater than or equal to a preset traction distance, and the minimum distance in the input data set of the current time period is less than a preset coasting distance, the current power instruction is determined to be a traction instruction, and the traction instruction is used to control the train to accelerate; if the minimum distance in the input data set of the current time period is greater than or equal to the preset coasting distance, and the minimum distance in the input data set of the current time period is less than a preset braking distance, the current power instruction is determined to be a coasting instruction, and the coasting instruction is used to control the train to decelerate; if the minimum distance in the input data set of the current time period is greater than or equal to the preset braking distance, the current power instruction is determined to be a braking instruction, and the braking instruction is used to control the train to stop, the preset traction distance is greater than the preset coasting distance, and the preset coasting distance is greater than the preset braking distance.

[0013] According to a train beaconing method provided by the present invention, the start jump instruction includes a start backward jump instruction. Before responding to the start jump instruction sent by the train automatic driving system, the method also includes: after receiving the train stop information sent by the train automatic driving system, obtaining the position relationship between the front of the train and the platform beacon; if the position relationship is that the front of the train exceeds the platform beacon, sending train passing information to the train automatic driving system; receiving the start backward jump instruction sent by the train automatic driving system.

[0014] According to a train beaconing method provided by the present invention, the start jump instruction includes a start forward jump instruction. Before responding to the start jump instruction sent by the train automatic driving system, the method also includes: after receiving the train stop information sent by the train automatic driving system, obtaining the position relationship between the front of the train and the platform beacon; if the position relationship is that the distance between the front of the train and the platform beacon is greater than the set distance, sending train under-beacon information to the train automatic driving system; receiving the start forward jump instruction sent by the train automatic driving system.

[0015] According to a train alignment method provided by the present invention, after the train stops moving, the method also includes: receiving train stop information sent by the train; based on the train stop information, obtaining the position relationship between the front of the train and the platform beacon; and sending the position relationship to the train automatic driving system, so that the train automatic driving system can determine whether to perform alignment again.

[0016] The present invention also provides a train benchmarking method, which is applied to a train automatic driving system, including: sending a start jump instruction to a train intelligent detection system; wherein the start jump instruction is used to trigger the train intelligent detection system to determine a second train state vector of the train in a current time period based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period; the input data set is used to characterize the position of the train relative to a target parking window and information on the track on which the train is located, the first train state vector is used to characterize the benchmarking state of the train during the benchmarking process, the benchmarking state includes a traction state, an idling state, or a braking state, and the first train control rate vector is used to characterize the control power required by the train during the benchmarking process. size; converting the second train state vector into the current power level of the train in the current time period, and determining the current power command based on the input data set of the current time period; sending the current power level and the current power command to the train automatic driving system; based on the current power level and the optimal motion variable, determining the second train control rate vector of the train in the current time period, taking the current time period as the previous time period to update the current power level and the current power command until the train stops moving; the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period; based on the current power level and the current power command, controlling the train to move to the position of the target stop window until the train stops moving.

[0017] The present invention also provides a train benchmarking system, including a train intelligent detection system and a train automatic driving system, wherein: the train intelligent detection system is used to respond to a start jump instruction sent by the train automatic driving system, based on an input data set of a previous time period, a first train state vector of the previous time period and a first train control rate vector of the previous time period, to determine a second train state vector of the train in a current time period, the input data set representing the position of the train relative to a target parking window and information on the track on which the train is located, the first train control rate vector being used to represent the benchmarking state of the train during the benchmarking process, the benchmarking state including a traction state or an idling state or a braking state, the first train control rate vector being used to represent the magnitude of the control power required by the train during the benchmarking process; the first train control rate vector being used to represent the control power required by the train during the benchmarking process The second train state vector is converted into the current power level of the train in the current period, and the current power command is determined based on the input data set of the current time period; the current power level and the current power command are sent to the train automatic driving system; the train automatic driving system is used to control the train to move to the position of the target parking window based on the current power level and the current power command; the train intelligent detection system is also used to determine the second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and use the current time period as the previous time period to update the current power level and the current power command until the train stops moving, the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period.

[0018] A train benchmarking system provided according to the present invention also includes a train automatic protection system, wherein: a train intelligent detection system is used to obtain the position relationship between the front of the train and the platform beacon after receiving the train stop information sent by the train automatic driving system; if the position relationship is that the front of the train exceeds the platform beacon, then the train over-the-mark information is sent to the train automatic driving system; a start backward jump instruction sent by the train automatic driving system is received; the train automatic driving system is used to send a backward jump application to the train automatic protection system after receiving the train over-the-mark information; the train automatic protection system is used to send a permission backward jump instruction to the train automatic driving system after receiving the backward jump application; the train automatic driving system is also used to send a start backward jump instruction to the train intelligent detection system after receiving the permission backward jump instruction.

[0019] 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 alignment methods is implemented.

[0020] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the computer program implements any of the above-mentioned train alignment methods.

[0021] The present invention also provides a computer program product, including a computer program, which implements any of the above-mentioned train alignment methods when executed by a processor.

[0022] The train benchmarking method, system, electronic device, storage medium and program product provided by the present invention determine the second train state vector of the train in the current time period based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period in response to the start jump instruction sent by the train automatic driving system; the input data set is used to characterize the position of the train relative to the target parking window and the information of the track on which the train is located, the first train state vector is used to characterize the benchmarking state of the train during the benchmarking process, the benchmarking state includes the traction state or the idling state or the braking state, and the first train control rate vector is used to characterize the size of the control power required by the train during the benchmarking process; the second train The state vector is converted into the current power level of the train in the current time period, and the current power command is determined based on the input data set of the current time period, and the current power level and the current power command are sent to the train automatic driving system, so that the train automatic driving system can control the train to move to the position of the target parking window based on the current power level and the current power command; based on the current power level and the optimal motion variable, the second train control rate vector of the train in the current time period is determined, and the current time period is used as the previous time period to update the current power level and the current power command until the train stops moving. The optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period. The present invention realizes accurate positioning and ranging of the train by periodically updating the input data set of the train. The second train state vector is determined according to the input data set of the previous time period, the first train state vector and the first train control rate vector, realizing the optimal decoupling control strategy based on finite time, minimizing the interaction effect between variables, and improving the accuracy of determining the second train state vector. By periodically determining the second train state vector, the accuracy of determining the current power level is improved, which is conducive to improving the accuracy and efficiency of train benchmarking. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] 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.

[0024] Figure 1 It is one of the flow charts of the train benchmarking method provided by the present invention.

[0025] Figure 2 This is the second flow chart of the train benchmarking method provided by the present invention.

[0026] Figure 3 This is the third flow chart of the train benchmarking method provided by the present invention.

[0027] Figure 4 This is the fourth flow chart of the train benchmarking method provided by the present invention.

[0028] Figure 5 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0029] 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.

[0030] In the first jump benchmarking process, the Automatic Train Protection (ATP) system sends a jump command to the Automatic Train Operation (ATO) system, and after a delay of a preset first time period, it sends a direction command to the train and ATO at the same time. After receiving the direction instruction, the train completes the jump preparation and sends effective feedback of the direction instruction to the ATO. After receiving the direction instruction and the effective feedback, the ATO delays for a preset second time period and then sends a traction instruction to the traction system of the train. After sending the traction instruction to the traction system of the train, the ATO delays for a preset third time period and then sends an instruction carrying a preset traction level to the traction system of the train, so that the traction system outputs traction force according to the traction instruction and the preset traction level to control the train to jump in the jumping direction. After the jumping distance reaches the preset distance, the traction instruction received by the traction system is invalid and the traction level returns to zero. After delaying for a preset fourth time period, the ATO sends a braking instruction carrying a preset braking level to the braking system of the train, so that the braking system outputs braking force according to the braking instruction to control the train to decelerate to a stop. After the train stops, the ATO delays for a preset fifth time period and then sends a jump completion message to the ATP, so that the ATP determines that the jump is completed after receiving the jump completion message. During the first jump marking process, after the jump distance reaches the preset distance, the traction command is invalid and the traction level is returned to zero. The train is then controlled to slow down and stop by outputting braking after a certain delay. This jump marking method makes the jump distance inflexible and can only jump a fixed distance. If the under-mark or over-mark distance is long, it is likely to require multiple jumps, and there is a possibility that the jump distance will exceed expectations, so it is less practical.

[0031] In the second jump benchmarking process, the jumping direction of the train is determined according to the target parking position and the current parking position of the train, and the traction system of the train is controlled to pull the train in the jumping direction with a preset first traction torque. From the start of traction to the jump control time reaching the maximum jump control time, the current position of the train is periodically obtained. If it is determined that the distance between the current position obtained in this period and the target parking position is less than or equal to the preset parking distance threshold, the braking system of the train is controlled to brake the train with a preset braking torque to slow down the train until it stops. The ultimate goal of jump benchmarking is to solve the problem of accurate benchmarking and parking. However, in the second train benchmarking process, there are errors and delays in the train obtaining the current position information through the transponder, so there are errors in the calculation of the jump distance. In actual jump tests, the problem of exceeding the preset distance often occurs, and in the jump process, traction is applied to control the vehicle to jump forward a certain distance and then directly output braking, resulting in a poor riding experience.

[0032] In view of the problems existing in the existing jump benchmarking schemes, the present invention proposes a jump benchmarking method based on the Intelligent Train Detection (ITD) system. Through the precise positioning and ranging information of the ITD and based on the finite-time optimal decoupling control strategy, the interaction effects between variables can be minimized, and train control decisions can be made more accurately, which can greatly improve the accuracy and efficiency of train jump benchmarking.

[0033] Combine the following Figure 1-Figure 5 The train alignment method, system and electronic equipment of the present invention are described.

[0034] Figure 1 It is one of the flow charts of the train benchmarking method provided by the present invention, such as Figure 1 As shown, the train alignment method includes steps S100 to S300, and each step is specifically described as follows.

[0035] S100: In response to a start jump instruction sent by the train automatic driving system, based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period, the second train state vector of the train in the current time period is determined.

[0036] The input data set is used to characterize the position of the train relative to the target parking window and the information of the track on which the train is located. The first train state vector is used to characterize the alignment state of the train during the alignment process. The alignment state includes the traction state, the idling state, or the braking state. The first train control rate vector is used to characterize the magnitude of the control power required by the train during the alignment process.

[0037] Based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period, the second train state vector of the train in the current time period is determined. Specifically, based on the input data set of the previous time period, the state matrix and the control rate matrix are determined, the state matrix represents the relationship between the benchmark state of the previous time period and the input data set of the previous time period, and the control rate matrix represents the relationship between the control power of the previous time period and the input data set of the previous time period; based on the first train state vector, the first train control rate vector, the state matrix and the control rate matrix, the second train state vector is determined.

[0038] The input data set includes the current position of the train, the maximum distance between the train and the target stop window, the minimum distance between the train and the target stop window, the track slope of the track where the train is located, the track friction coefficient of the track where the train is located, and the track obstacle information of the track where the train is located.

[0039] The intelligent train detection system (ITD) includes a positioning module, a distance measurement module and a decision module. The positioning module is used to detect the current position of the train in real time. The distance measurement module includes a camera, a laser radar, etc., which is used to calculate the maximum distance between the train and the target stop window and the minimum distance between the train and the target stop window. The decision module is used to perform a finite-time optimal decoupling control strategy based on the input data set, make a train control decision (including the power level and power command for controlling the movement of the train) and send it to the ATO. When the automatic train control system (ATO) detects that the train needs to jump to the mark, it sends a start jump command to the intelligent train detection system. ITD measures the current position of the train, the maximum distance between the train and the target stop window, the minimum distance between the train and the target stop window, the track slope of the track where the train is located, the track friction coefficient and the track obstacle information (input data set) at a fixed time period. The ITD period performs a finite-time optimal decoupling control strategy based on the input data set of the current time period, the input data set of the previous time period, the first train state vector of the previous time period, and the first train control rate vector of the previous time period, makes a train control decision (including the current power level and current power command for controlling the movement of the train) and sends it to the ATO.

[0040] The decision module adopts a finite-time optimal decoupling control strategy to determine the second train state vector. The calculation formula of the second train state vector is as follows.

[0041] (1-1);

[0042] in, is the state vector of the second train in the current time period, The data dimension is , is the first train state vector of the previous time period, The data dimension is , is the first train control rate vector of the previous time period, The data dimension is , , is the state matrix of the previous time period, The data dimension is , is the control rate matrix of the previous time period, The data dimension is .

[0043] State Matrix and the control rate matrix It is determined by the input data set of the previous time period.

[0044] The input data set of the train is updated in real time according to the ITD module, which improves the accuracy of the input data set. Through the input data set, the state matrix and the control rate matrix are determined, and then the state vector of the second train is determined, which is conducive to realizing the optimal decoupling control strategy in a limited time and improving the accuracy of determining the current power level.

[0045] S200: Convert the second train state vector into the current power level of the train in the current time period, determine the current power command based on the input data set of the current time period, and send the current power level and the current power command to the train automatic driving system, so that the train automatic driving system can control the train to move to the target parking window based on the current power level and the current power command.

[0046] The current power level is calculated as follows.

[0047] (1-2);

[0048] in, is the current power level, is the preset transformation matrix, reversible, for The transposed matrix of is the state vector of the second train in the current time period.

[0049] The current power instruction includes a traction instruction, a coasting instruction or a braking instruction. The current power instruction is determined based on the input data set of the current time period. Specifically, if the minimum distance in the input data set of the current time period is greater than or equal to the preset traction distance, and the minimum distance in the input data set of the current time period is less than the preset coasting distance, the current power instruction is determined to be a traction instruction, and the traction instruction is used to control the train to accelerate; if the minimum distance in the input data set of the current time period is greater than or equal to the preset coasting distance, and the minimum distance in the input data set of the current time period is less than the preset braking distance, the current power instruction is determined to be a coasting instruction, and the coasting instruction is used to control the train to decelerate; if the minimum distance in the input data set of the current time period is greater than or equal to the preset braking distance, the current power instruction is determined to be a braking instruction, and the braking instruction is used to control the train to stop, the preset traction distance is greater than the preset coasting distance, and the preset coasting distance is greater than the preset braking distance.

[0050] The current power level (including traction level, coasting level and braking level) is used to represent the power level that needs to be input to the train in the current time period. For example, the first traction level means that 80%-100% of the traction power needs to be provided to the train. The preset traction distance, preset coasting distance and preset braking distance are determined based on the input data set of the first time period when the jump starts.

[0051] like Figure 4 As shown in the figure, when the minimum distance between the train and the target stop window is between the preset traction distance (traction point) and the preset coasting distance (coiling point), the ITD periodically sends a traction control decision (including traction level and traction command) to the ATO. When the minimum distance is between the preset coasting distance (coiling point) and the preset braking distance (braking point), the ITD periodically sends a coasting control decision (including coasting level and coasting command) to the ATO. When the minimum distance is less than the preset braking distance, the ITD periodically sends a braking control decision (including braking level and braking command) to the ATO.

[0052] ITD sends the current power level and current power command calculated in the current cycle to ATO. ATO controls the train to move to the target stop window position according to the current power level and current power command.

[0053] By dividing the current power command into traction command, coasting command and braking command, it is helpful to improve the accuracy of train alignment.

[0054] S300: Based on the current power level and the optimal motion variable, determine the second train control rate vector of the train in the current time period, and use the current time period as the previous time period to update the current power level and the current power command until the train stops moving.

[0055] The optimal motion variables are determined based on the standard power level of the train, which is determined based on the input data set for the current time period.

[0056] The optimal motion variable is determined based on the following method: obtaining the average state of the state matrix of the previous time period and the average control rate of the control rate matrix of the previous time period; the average state is the average value of multiple state values ​​in the state matrix, and the average control rate is the average value of multiple control rates in the control rate matrix; based on the average state, the average control rate, the preset weighting matrix and the preset weight matrix, the state covariance matrix of the train is determined; the preset weighting matrix is ​​used to characterize the relationship between the minimum energy consumption of the train, the optimal tracking control variable and the tracking error, the tracking error is the difference between the current power level and the standard power level, and the optimal tracking control variable is the control variable corresponding to the minimum energy consumption; the preset weighting matrix is ​​used to characterize the conversion relationship between the optimal tracking control variable and the tracking error; based on the state covariance matrix, the standard power level, the average state, the average control rate, the preset weighting matrix and the preset weight matrix, the optimal motion variable is determined.

[0057] Based on the current power level and the optimal motion variable, the second train control rate vector of the train in the current time period is determined. Specifically, a conversion relationship between the current power level, the optimal tracking control variable and the second train control rate vector is constructed to obtain conversion information, and the conversion information includes a first diagonal matrix, a second diagonal matrix, a first characteristic matrix and a second characteristic matrix, and the main diagonal elements of the first characteristic matrix and the second characteristic matrix are both zero; based on the first diagonal matrix, the first characteristic matrix, the second characteristic matrix, the second diagonal matrix, the average control rate, the preset weight matrix, the optimal motion variable and the current power level, the second train control rate vector is determined.

[0058] Obtain the average state of the state matrix. Specifically, based on the state matrix, the transposed matrix of the preset conversion matrix and the preset conversion matrix, determine the state intermediate matrix, the preset conversion matrix represents the conversion relationship between the second train state vector and the current power level; obtain each state value on the diagonal of the state intermediate matrix; and take the average value of each state value as the average state.

[0059] Further, the average control rate of the control rate matrix is ​​obtained. Specifically, the control rate intermediate matrix is ​​determined according to the control rate matrix and the preset conversion matrix. Each control rate value on the diagonal of the control rate intermediate matrix is ​​obtained; and the average value of each control rate value is taken as the average control rate.

[0060] In order to design the decoupling controller, the calculation formulas of the current power level and the current power command are converted into differential equations about the output.

[0061] (2);

[0062] in, is the initial power level of the first time period, For the first time period, is the preset transformation matrix, is the initial train state vector for the first time period, is the state matrix of the previous time period, is the control rate matrix of the previous time period, is the state intermediate matrix, is the control rate intermediate matrix, is the power level of the previous time period, is the first train control rate vector of the previous time period, is the current power level, reversible, for The transposed matrix of .

[0063] The system Time period input For Time period input The coupling effect of is considered as a measurable interference and eliminated by the feedforward method, and the feedback method is used to eliminate the interference in the autonomous system ( ) The i-th input train control rate vector is used to control the j-th Therefore, the calculation formula of the current power level and the current power command is converted into the following equation.

[0064] (3);

[0065] in, is the current power level; is the power level of the previous time period; is the first train control rate vector; is the first state matrix, is a diagonal matrix, The values ​​on the diagonal are equal to State values ​​on the main diagonal; The second state matrix with zero main diagonal represents the mutual influence between different variables in the autonomous system; for Each state value on the diagonal of ; is the state intermediate matrix; is the control rate intermediate matrix; is the first control rate matrix, is a diagonal matrix, The values ​​on the diagonal are equal to Control ratio values ​​on the main diagonal; The second control rate matrix with zeros on the main diagonal represents the coupling between different time periods; for Each control rate value on the diagonal of .

[0066] Construct the conversion relationship between the power level, the optimal tracking control variable and the train control rate vector to obtain the conversion information. The conversion relationship is a decoupling conversion relationship. Introduce the control variable and design the inner loop decoupling controller. The algorithm formula of the inner loop decoupling controller is as follows.

[0067] (4);

[0068] in, is the first diagonal matrix, reversible, is the second diagonal matrix, is the first characteristic matrix whose main diagonal elements are zero, K is the second characteristic matrix whose main diagonal elements are zero, is the optimal tracking control variable, For the power level, is the train control rate vector.

[0069] use Multiply by formula (3) and use Multiply by formula (4) to obtain the calculation formula for the current power level.

[0070] (5);

[0071] To achieve the decoupling of formula (2), , The relationship between and K satisfies formula (6) and formula (7).

[0072] (6)

[0073] (7)

[0074] According to formula (6) and formula (7), formula (5) is converted into formula (8).

[0075] (8);

[0076] in, is the current power level; is the power level of the previous time period; is the first train control rate vector; is the first diagonal matrix; is the second diagonal matrix; is the first characteristic matrix whose main diagonal elements are zero; K is the second characteristic matrix whose main diagonal elements are zero; is the first control rate matrix, is the second control rate matrix, is the optimal tracking control variable.

[0077] because and They are all diagonal matrices, so formula (2) achieves complete decoupling.

[0078] The optimal tracking control variables are designed according to formula (8) to minimize the energy consumption of the train during the jump benchmarking process.

[0079] (9);

[0080] in, is the tracking error, , is the power level and power command of the previous time period, For standard power level, is the tracking error when the train consumes the least energy, is the minimum energy consumption of the train, is the preset weighting matrix, is the optimal tracking control variable, For the first time period, is the time period corresponding to the minimum energy consumption of the train, is the preset weight matrix, It is the state transfer matrix corresponding to the minimum energy consumption of the train.

[0081] Determine the optimal tracking control variable, and the calculation formula of the optimal tracking control variable is as follows.

[0082] (10);

[0083] in, , is the average control rate, is the first control rate matrix, is the first diagonal matrix, is the second diagonal matrix, is the optimal tracking control variable, is the preset weight matrix, is the state covariance matrix, is the optimal motion variable, For power level.

[0084] The calculation formulas for the optimal motion variables and state covariance matrix are as follows.

[0085] (11);

[0086] (12);

[0087] (13);

[0088] (14);

[0089] in, is the state covariance matrix, is the optimal motion variable, is the average control rate, is the average state, , is the first state matrix, is the preset weighting matrix, is the preset weight matrix, is the state transfer matrix, For standard power level, for The standard power level at all times, is the time period corresponding to the minimum energy consumption of the train.

[0090] The optimal decoupling control strategy of the present application is composed of an inner loop decoupling controller and an outer loop tracking controller (optimal tracking control variable). According to the optimal decoupling control strategy, a calculation formula for the second train control rate vector is obtained.

[0091] The finite-time optimal decoupling control strategy is adopted to calculate the second train control rate vector of the current time period. The calculation formula of the second train control rate vector is as follows.

[0092] (15);

[0093] in, is the second train control rate vector of the current time period, is the first diagonal matrix, is the first characteristic matrix, K is the second characteristic matrix, is the second diagonal matrix, is the preset weight matrix, is the average control rate, is the optimal motion variable, The current power level.

[0094] Based on the input data set of the current time period, the state matrix of the next time period and the control rate matrix of the next time period are determined. According to formula (1-1), formula (1-2), formula (2) and formula (15), the next power level of the next time period is determined according to the state matrix of the next time period, the control rate matrix of the next time period, the second train state vector and the second train control rate vector, and the current power level is updated to the next power level. The next power command of the next time period is determined according to the input data set of the next time period, and the current power command is updated to the next power command until the train automatic driving system stops moving based on the updated current power level and the updated current power command, and the train alignment is completed.

[0095] The present invention uses the precise positioning and ranging information of ITD and based on the finite-time optimal decoupling control strategy to minimize the interaction effects between variables, more accurately determine the current power level, and greatly improve the accuracy and efficiency of train jump alignment.

[0096] After the train stops moving, the train benchmarking method also includes: receiving train stop information sent by the train; based on the train stop information, obtaining the position relationship between the front of the train and the platform beacon; and sending the position relationship to the train automatic driving system for the train automatic driving system to determine whether to perform benchmarking again.

[0097] like Figure 2 , 3 As shown, ATO controls the train to move toward the target stop window according to the current power level and current power command, and sends the train stop information to ITD after the train stops steadily. ITD obtains the position relationship between the locomotive and the platform beacon. ITD sends the position relationship to ATO. ATO sends the jump completion information to ATP. ATO determines whether the train has passed the mark according to the position relationship or ATO determines whether the train has failed to meet the mark according to the position relationship, and then determines whether to re-mark.

[0098] After the train stops moving, the positional relationship between the train's front and the platform beacon is further obtained through ITD, and then it is determined whether to perform alignment again, which is conducive to improving the accuracy of train alignment.

[0099] The train benchmarking method provided in an embodiment of the present invention is applied to a train intelligent detection system, and determines the second train state vector of the train in the current time period based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period in response to a start jump instruction sent by a train automatic driving system; the input data set is used to characterize the position of the train relative to a target parking window and information on the track on which the train is located, the first train state vector is used to characterize the benchmarking state of the train during the benchmarking process, the benchmarking state includes a traction state or an idling state or a braking state, and the first train control rate vector is used to characterize the size of the control power required by the train during the benchmarking process; the second train state vector is used to characterize ... position of the train relative to a target parking window and information on the track on which the train is located. The state vector is converted into the current power level of the train in the current time period, and the current power command is determined based on the input data set of the current time period, and the current power level and the current power command are sent to the train automatic driving system, so that the train automatic driving system can control the train to move to the position of the target parking window based on the current power level and the current power command; based on the current power level and the optimal motion variable, the second train control rate vector of the train in the current time period is determined, and the current time period is used as the previous time period to update the current power level and the current power command until the train stops moving. The optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period. The present invention realizes accurate positioning and ranging of the train by periodically updating the input data set of the train. The second train state vector is determined according to the input data set of the previous time period, the first train state vector and the first train control rate vector, realizing the optimal decoupling control strategy based on finite time, minimizing the interaction effect between variables, and improving the accuracy of determining the second train state vector. By periodically determining the second train state vector, the accuracy of determining the current power level is improved, which is conducive to improving the accuracy and efficiency of train benchmarking.

[0100] Based on the above embodiment, the start jump instruction includes a start backward jump instruction. Before responding to the start jump instruction sent by the train automatic driving system, the train benchmarking method also includes steps S400 to S600. The details of each step are as follows.

[0101] S400: After receiving the train stop information sent by the train automatic driving system, the position relationship between the train head and the platform beacon is obtained.

[0102] S500: If the position relationship is that the front of the train exceeds the platform beacon, the train passing the beacon information is sent to the train automatic driving system.

[0103] S600: Receive a start backward jump instruction sent by the train automatic driving system.

[0104] like Figure 3 As shown in the figure, in FAM / CAM mode, after ATO controls the train to run to the platform and stop steadily, ATO sends the train stop information to ITD. After ITD receives the train stop information, the positioning module detects the platform beacon for accurate positioning and determines whether the train has stopped past the mark. If the train has stopped past the mark, ITD sends the train pass information to ATO. After receiving the train pass information, ATO sends a backward jump application to ATP. After receiving the backward jump application, ATP determines whether the backward jump conditions are met. After determining that the conditions are met, ATP sends a backward jump permission instruction to ATO. After receiving the backward jump permission instruction, ATO sends a start backward jump permission instruction to ITD. The ITD ranging module detects the platform beacon in real time and calculates the distance between the front of the train and the stop point. After receiving the start jump permission instruction sent by ATO, the decision module calculates the traction point, idling point and braking point of the train jump according to the position information of the train start jump, and determines the current power instruction according to the real-time distance between the front of the train and the target stop window. According to the input data set, the ITD determines the current power level and the current power command based on the finite-time optimal decoupling control strategy, and sends the current power level and the current power command to the ATO.

[0105] ITD periodically sends the current power level and current power command to ATO. ATO controls the train to move toward the target stop window according to the current power level and current power command. After the train stops steadily, it sends the train stop information to ITD.

[0106] Optionally, the start jump instruction includes a start forward jump instruction. Before responding to the start jump instruction sent by the train automatic driving system, the train benchmarking method also includes: after receiving the train stop information sent by the train automatic driving system, obtaining the position relationship between the front of the train and the platform beacon; if the position relationship is that the distance between the front of the train and the platform beacon is greater than the set distance, sending train under-beacon information to the train automatic driving system; and receiving the start forward jump instruction sent by the train automatic driving system.

[0107] like Figure 2 As shown in the figure, in FAM / CAM mode, after ATO controls the train to run to the platform and stop steadily, ATO sends the train stop information to ITD. After ITD receives the train stop information, the positioning module accurately locates by detecting the platform beacon and then determines whether the train stops short of the mark. If the train stops short of the mark, ITD sends the train short of the mark information to ATO. After receiving the train short of the mark information, ATO sends forward jump information to ATP and sends a start jump instruction to ITD. The ITD ranging module detects the platform beacon in real time and calculates the distance between the front of the train and the stop point. After receiving the start jump instruction sent by ATO, the decision module calculates the current power level and current power instruction based on the input data set and the finite time optimal decoupling control strategy, and periodically sends the current power level and current power instruction to ATO. ATO controls the train to move to the target stop window according to the current power level and current power instruction until the train stops steadily, and then sends the train stop information to ITD.

[0108] The embodiment of the present invention determines the train passing mark information through ITD positioning and ranging, with high accuracy and good practicability. Determining the train stopping information and starting the backward jump instruction through ATO is conducive to improving the accuracy and efficiency of the train's backward alignment. Determining the train under-mark information through ITD positioning and ranging is accurate and practical. Determining the train stopping information and starting the forward jump instruction through ATO is conducive to improving the accuracy and efficiency of the train's forward alignment.

[0109] An embodiment of the present invention provides a train benchmarking method, which is applied to a train automatic driving system, including: sending a start jump instruction to a train intelligent detection system; wherein the start jump instruction is used to trigger the train intelligent detection system to determine a second train state vector of the train in a current time period based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period; the input data set is used to characterize the position of the train relative to a target parking window and information on the track on which the train is located, the first train state vector is used to characterize the benchmarking state of the train during the benchmarking process, the benchmarking state includes a traction state, an idling state, or a braking state, and the first train control rate vector is used to characterize the control power required by the train during the benchmarking process. the size of; converting the second train state vector into the current power level of the train in the current time period, and determining the current power command based on the input data set of the current time period; sending the current power level and the current power command to the train automatic driving system; determining the second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and taking the current time period as the previous time period to update the current power level and the current power command until the train stops moving; the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period; based on the current power level and the current power command, the train is controlled to move to the position of the target stop window until the train stops moving.

[0110] The train benchmarking method provided by the embodiment of the present invention is applied to the train automatic driving system, by sending a start jump instruction to the train intelligent detection system; wherein the start jump instruction is used to trigger the train intelligent detection system to determine the second train state vector of the train in the current time period based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period; convert the second train state vector into the current power level of the train in the current time period, and determine the current power instruction based on the input data set of the current time period; send the current power level and the current power instruction to the train automatic driving system; determine the second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and use the current time period as the previous time period to update the current power level and the current power instruction until the train stops moving; control the train to move to the position of the target parking window based on the current power level and the current power instruction until the train stops moving. The present invention realizes accurate positioning and ranging of the train by periodically updating the input data set of the train. The second train state vector is determined based on the input data set of the previous time period, the first train state vector and the first train control rate vector, realizing the optimal decoupling control strategy based on finite time, minimizing the interaction effect between variables and improving the accuracy of determining the second train state vector. By periodically determining the second train state vector, the accuracy of determining the current power level is improved, which is conducive to improving the accuracy and efficiency of train benchmarking.

[0111] An embodiment of the present invention provides a train benchmarking system, including a train intelligent detection system and a train automatic driving system, wherein: the train intelligent detection system is used to respond to a start jump instruction sent by the train automatic driving system, based on an input data set of a previous time period, a first train state vector of the previous time period and a first train control rate vector of the previous time period, to determine a second train state vector of the train in a current time period, the input data set representing the position of the train relative to a target parking window and information on the track on which the train is located, the first train control rate vector is used to represent the benchmarking state of the train during the benchmarking process, the benchmarking state including a traction state or an idling state or a braking state, the first train control rate vector is used to represent the magnitude of the control power required by the train during the benchmarking process; the second train state vector is converted into a current power level of the train in the current period, and a current power instruction is determined based on the input data set of the current time period; the current power level and the current power instruction are sent to the train automatic driving system; the train automatic driving system is used to control the train to move to the position of the target parking window based on the current power level and the current power instruction. The train intelligent detection system is also used to determine the second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and use the current time period as the previous time period to update the current power level and the current power command until the train stops moving. The optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period.

[0112] Based on the above embodiments, the train benchmarking system also includes a train automatic protection system, wherein: a train intelligent detection system is used to obtain the position relationship between the front of the train and the platform beacon after receiving the train stop information sent by the train automatic driving system; if the position relationship is that the front of the train exceeds the platform beacon, then the train over-the-mark information is sent to the train automatic driving system; a start backward jump instruction sent by the train automatic driving system is received; the train automatic driving system is used to send a backward jump application to the train automatic protection system after receiving the train over-the-mark information; the train automatic protection system is used to send a permission backward jump instruction to the train automatic driving system after receiving the backward jump application; the train automatic driving system is also used to send a start backward jump instruction to the train intelligent detection system after receiving the permission backward jump instruction.

[0113] Optionally, the train intelligent detection system is used to obtain the position relationship between the front of the train and the platform beacon after receiving the train stopped information sent by the train automatic driving system; if the position relationship is that the distance between the front of the train and the platform beacon is greater than the set distance, then send train under-beacon information to the train automatic driving system; receive the start forward jump instruction sent by the train automatic driving system; the train automatic driving system is used to send a forward jump application to the train automatic protection system after receiving the train under-beacon information; the train automatic protection system is used to send a forward jump permission instruction to the train automatic driving system after receiving the forward jump application; the train automatic driving system is also used to send a start forward jump instruction to the train intelligent detection system after receiving the forward jump permission instruction.

[0114] The train benchmarking system provided by the embodiment of the present invention determines the second train state vector of the train in the current time period based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period, converts the second train state vector into the current power level of the train in the current time period, determines the current power command based on the input data set of the current time period, and sends the current power level and the current power command to the train automatic driving system, so that the train automatic driving system controls the train to move to the position of the target parking window based on the current power level and the current power command; based on the current power level and the optimal motion variable, determines the second train control rate vector of the train in the current time period, and takes the current time period as the previous time period to update the current power level and the current power command until the train stops moving. The present invention realizes accurate positioning and ranging of the train by periodically updating the input data set of the train. The second train state vector is determined according to the input data set of the previous time period, the first train state vector and the first train control rate vector, and the optimal decoupling control strategy based on finite time is realized, which minimizes the interaction between variables and improves the accuracy of determining the second train state vector. By periodically determining the second train state vector, the accuracy of determining the current power level is improved, which is beneficial to improving the accuracy and efficiency of train alignment.

[0115] Figure 5 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 5As shown, the electronic device may include: a processor 510, a communication interface 520, a memory 530 and a communication bus 540, wherein the processor 510, the communication interface 520 and the memory 530 communicate with each other through the communication bus 540. The processor 510 may call the logic instructions in the memory 530 to execute the train alignment method, which is applied to the train intelligent detection system. The method includes: in response to the start jump instruction sent by the train automatic driving system, based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period, determine the second train state vector of the train in the current time period; the input data set is used to characterize the position of the train relative to the target parking window and the information of the track on which the train is located, the first train state vector is used to characterize the alignment state of the train during the alignment process, the alignment state includes the traction state or the idling state or the braking state, and the first train control rate vector is used to characterize the control power required by the train during the alignment process ; converting the second train state vector into the current power level of the train in the current time period, and determining the current power command based on the input data set of the current time period, sending the current power level and the current power command to the train automatic driving system, so that the train automatic driving system controls the train to move to the position of the target parking window based on the current power level and the current power command; determining the second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, taking the current time period as the previous time period to update the current power level and the current power command until the train stops moving, the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period.

[0116] Alternatively, a train benchmarking method is executed, which is applied to a train automatic driving system, the method comprising: sending a start jump instruction to a train intelligent detection system; wherein the start jump instruction is used to trigger the train intelligent detection system to determine a second train state vector of the train in a current time period based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period; the input data set is used to characterize the position of the train relative to the target parking window and the information of the track on which the train is located, and the first train control rate vector is used to characterize the magnitude of the power of the train during the benchmarking process; the second train state vector is converted into a current power level of the train in the current time period, and a current power command is determined based on the input data set of the current time period; the current power level and the current power command are sent to the train automatic driving system; based on the current power level and the optimal motion variable, the second train control rate vector of the train in the current time period is determined, and the current time period is used as the previous time period to update the current power level and the current power command until the train stops moving; the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period; based on the current power level and the current power command, the train is controlled to move to the position of the target parking window until the train stops moving.

[0117] In addition, the logic instructions in the above-mentioned memory 530 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.

[0118] 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 calibration method provided by the above-mentioned methods, which includes: responding to a start jump instruction sent by a train automatic driving system, based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period, determining a second train state vector of the train in a current time period; the input data set is used to characterize the position of the train relative to the target parking window and information on the track on which the train is located, the first train state vector is used to characterize the calibration state of the train during the calibration process, the calibration state includes a traction state or an idling state or a braking state, and the first train control rate vector is used to characterize the calibration state of the train during the calibration process. A train control rate vector is used to characterize the size of the control power required by the train during the benchmarking process; the second train state vector is converted into the current power level of the train in the current time period, and the current power command is determined based on the input data set of the current time period, and the current power level and the current power command are sent to the train automatic driving system, so that the train automatic driving system can control the train to move to the position of the target parking window based on the current power level and the current power command; based on the current power level and the optimal motion variable, the second train control rate vector of the train in the current time period is determined, and the current time period is used as the previous time period to update the current power level and the current power command until the train stops moving, the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period.

[0119] Alternatively, the method includes: sending a start jump instruction to a train intelligent detection system; wherein the start jump instruction is used to trigger the train intelligent detection system to determine a second train state vector of the train in a current time period based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period; the input data set is used to characterize the position of the train relative to the target parking window and information on the track on which the train is located, and the first train control rate vector is used to characterize the magnitude of the power of the train during the benchmarking process; converting the second train state vector into a current power level of the train in the current time period, and determining a current power command based on the input data set of the current time period; sending the current power level and the current power command to the train automatic driving system; determining a second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and taking the current time period as the previous time period to update the current power level and the current power command until the train stops moving; the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period; controlling the train to move toward the position of the target parking window based on the current power level and the current power command until the train stops moving.

[0120] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the train alignment method provided by the above-mentioned methods, the method comprising: responding to a start jump instruction sent by a train automatic driving system, based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period, determining a second train state vector of the train in a current time period; the input data set is used to characterize the position of the train relative to a target parking window and information on the track on which the train is located, the first train state vector is used to characterize the alignment state of the train during the alignment process, the alignment state including a traction state or an idling state or a braking state, and the first train control rate vector is used to characterize the train The amount of control power required during the benchmarking process; converting the second train state vector into the current power level of the train in the current time period, and determining the current power command based on the input data set of the current time period, and sending the current power level and the current power command to the train automatic driving system, so that the train automatic driving system can control the train to move to the target parking window position based on the current power level and the current power command; based on the current power level and the optimal motion variable, determining the second train control rate vector of the train in the current time period, using the current time period as the previous time period to update the current power level and the current power command until the train stops moving, the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period.

[0121] Alternatively, the method includes: sending a start jump instruction to a train intelligent detection system; wherein the start jump instruction is used to trigger the train intelligent detection system to determine a second train state vector of the train in a current time period based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period; the input data set is used to characterize the position of the train relative to the target parking window and information on the track on which the train is located, and the first train control rate vector is used to characterize the magnitude of the power of the train during the benchmarking process; converting the second train state vector into a current power level of the train in the current time period, and determining a current power command based on the input data set of the current time period; sending the current power level and the current power command to the train automatic driving system; determining a second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and taking the current time period as the previous time period to update the current power level and the current power command until the train stops moving; the optimal motion variable is determined based on the standard power level of the train, and the standard power level is determined based on the input data set of the current time period; controlling the train to move toward the position of the target parking window based on the current power level and the current power command until the train stops moving.

[0122] 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.

[0123] 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.

[0124] 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 benchmarking method, characterized in that: Applied to train intelligent detection system, including: In response to a start jump instruction sent by the train automatic driving system, a second train state vector of the train in the current time period is determined based on an input data set of a previous time period, a first train state vector of the previous time period, and a first train control rate vector of the previous time period; the input data set is used to characterize the position of the train relative to the target parking window and information on the track on which the train is located, the first train state vector is used to characterize the alignment state of the train during the alignment process, the alignment state includes a traction state, a coasting state, or a braking state, and the first train control rate vector is used to characterize the magnitude of the control power required by the train during the alignment process; converting the second train state vector into a current power level of the train in a current time period, determining a current power command based on an input data set of the current time period, and sending the current power level and the current power command to the train automatic driving system, so that the train automatic driving system controls the train to move toward the position of the target stop window based on the current power level and the current power command; determining a second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and using the current time period as the previous time period to update the current power level and the current power command until the train stops moving, wherein the optimal motion variable is determined based on a standard power level of the train and a minimum energy consumption of the train, wherein the standard power level is determined based on an input data set in the current time period; The determining, based on the current power level and the optimal motion variable, a second train control rate vector of the train in the current time period comprises: Constructing a conversion relationship between a power level, an optimal tracking control variable and a train control rate vector to obtain conversion information, wherein the conversion information includes a first diagonal matrix, a second diagonal matrix, a first characteristic matrix and a second characteristic matrix, wherein the main diagonal elements of the first characteristic matrix and the second characteristic matrix are both zero; the optimal tracking control variable is a control variable corresponding to the minimum energy consumption of the train; The second train control rate vector is determined based on the first diagonal matrix, the first characteristic matrix, the second characteristic matrix, the second diagonal matrix, the average control rate of the control rate matrix in the previous time period, the preset weight matrix, the optimal motion variable and the current power level.

2. The train calibration method according to claim 1, characterized in that: The determining of a second train state vector of the train in the current time period based on an input data set in the previous time period, a first train state vector in the previous time period, and a first train control rate vector in the previous time period comprises: Based on the input data set of the previous time period, determine a state matrix and a control rate matrix, wherein the state matrix represents the relationship between the benchmark state of the previous time period and the input data set of the previous time period, and the control rate matrix represents the relationship between the control power of the previous time period and the input data set of the previous time period; The second train state vector is determined based on the first train state vector, the first train control rate vector, the state matrix and the control rate matrix.

3. The train calibration method according to claim 1, characterized in that: The optimal motion variables are determined based on the following method: Obtaining an average state of a state matrix in a previous time period and an average control rate of a control rate matrix in a previous time period; the average state is an average value of a plurality of state values ​​in the state matrix, and the average control rate is an average value of a plurality of control rates in the control rate matrix; Determining a state covariance matrix of the train based on the average state, the average control rate, a preset weighting matrix and a preset weight matrix; The preset weighted matrix is ​​used to characterize the relationship between the minimum energy consumption, the optimal tracking control variable and the tracking error of the train, the tracking error is the difference between the current power level and the standard power level, and the optimal tracking control variable is the control variable corresponding to the minimum energy consumption; The preset weight matrix is ​​used to characterize the conversion relationship between the optimal tracking control variable and the tracking error; The optimal motion variable is determined based on the state covariance matrix, the standard power level, the average state, the average control rate, the preset weighting matrix and the preset weight matrix.

4. The train calibration method according to claim 3, characterized in that: The step of obtaining the average state of the state matrix of the previous time period includes: Determining a state intermediate matrix based on the state matrix, a transposed matrix of a preset conversion matrix, and the preset conversion matrix, wherein the preset conversion matrix represents a conversion relationship between the second train state vector and the current power level; Obtain each state value on the diagonal of the state intermediate matrix; The average value of each of the state values ​​is taken as the average state.

5. The train calibration method according to claim 1, characterized in that: The input data set includes the current position of the train, the maximum distance between the train and the target parking window, the minimum distance between the train and the target parking window, the track slope of the track where the train is located, the track friction coefficient of the track where the train is located, and the track obstacle information of the track where the train is located.

6. The train calibration method according to claim 5, characterized in that: The current power command includes a traction command, a coasting command or a braking command, and the current power command is determined based on the input data set of the current time period, including: If the minimum distance in the input data set of the current time period is greater than or equal to the preset traction distance, and the minimum distance in the input data set of the current time period is less than the preset coasting distance, the current power instruction is determined to be the traction instruction, and the traction instruction is used to control the train to accelerate and move; If the minimum distance in the input data set of the current time period is greater than or equal to the preset coasting distance, and the minimum distance in the input data set of the current time period is less than the preset braking distance, the current power instruction is determined to be the coasting instruction, and the coasting instruction is used to control the train to decelerate and move; If the minimum distance in the input data set of the current time period is greater than or equal to the preset braking distance, the current power instruction is determined to be the braking instruction, and the braking instruction is used to control the train to stop, the preset traction distance is greater than the preset coasting distance, and the preset coasting distance is greater than the preset braking distance.

7. The train calibration method according to claim 1, characterized in that: The start jump instruction includes a start backward jump instruction. Before responding to the start jump instruction sent by the train automatic driving system, the method further includes: After receiving the train stop information sent by the train automatic driving system, obtaining the position relationship between the front of the train and the platform beacon; If the position relationship is that the front of the train exceeds the platform beacon, sending train passing information to the train automatic driving system; Receive the start backward jump instruction sent by the train automatic driving system.

8. The train alignment method according to claim 1, characterized in that: The start jump instruction includes a start forward jump instruction. Before responding to the start jump instruction sent by the train automatic driving system, the method further includes: After receiving the train stop information sent by the train automatic driving system, obtaining the position relationship between the front of the train and the platform beacon; If the position relationship is that the distance between the front of the train and the platform beacon is greater than a set distance, sending train under-beacon information to the train automatic driving system; Receive the start forward jump instruction sent by the train automatic driving system.

9. The train alignment method according to claim 1, characterized in that: After the train stops moving, the method further comprises: Receiving train stop information sent by the train; Based on the train stopping information, obtaining the position relationship between the front of the train and the platform beacon; The position relationship is sent to the train automatic driving system so that the train automatic driving system can determine whether to perform benchmarking again.

10. A train benchmarking method, characterized in that: Applied to train automatic driving systems, including: Send a start jump instruction to the train intelligent detection system; wherein the start jump instruction is used to trigger the train intelligent detection system to determine the second train state vector of the train in the current time period based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period; the input data set is used to characterize the position of the train relative to the target parking window and the information of the track on which the train is located, the first train state vector is used to characterize the alignment state of the train during the alignment process, and the alignment state includes a traction state or an idling state or a braking state, and the first train control rate vector is used to characterize the magnitude of the control power required by the train during the alignment process; convert the second train state vector into the current power level of the train in the current time period, and determine the current power instruction based on the input data set of the current time period; send the current power level and the current power instruction to the train automatic driving system; determine the second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and convert the current power level vector of the train into the current power level of the train in the current time period; The time period is taken as the previous time period to update the current power level and the current power instruction until the train stops moving; the optimal motion variable is determined based on the standard power level of the train and the minimum energy consumption of the train, and the standard power level is determined based on the input data set of the current time period; the second train control rate vector of the train in the current time period is determined based on the current power level and the optimal motion variable, including: constructing a conversion relationship between the power level, the optimal tracking control variable and the train control rate vector to obtain conversion information, the conversion information includes a first diagonal matrix, a second diagonal matrix, a first characteristic matrix and a second characteristic matrix, and the main diagonal elements of the first characteristic matrix and the second characteristic matrix are all zero; the optimal tracking control variable is the control variable corresponding to the minimum energy consumption of the train; based on the first diagonal matrix, the first characteristic matrix, the second characteristic matrix, the second diagonal matrix, the average control rate of the control rate matrix of the previous time period, the preset weight matrix, the optimal motion variable and the current power level, the second train control rate vector is determined; The train is controlled to move toward the target parking window based on the current power level and the current power command until the train stops moving.

11. A train benchmarking system, characterized in that: It includes train intelligent detection system and train automatic driving system, including: The train intelligent detection system is used to respond to the start jump instruction sent by the train automatic driving system, determine the second train state vector of the train in the current time period based on the input data set of the previous time period, the first train state vector of the previous time period and the first train control rate vector of the previous time period, the input data set represents the position of the train relative to the target parking window and the information of the track on which the train is located, the first train state vector is used to represent the alignment state of the train in the alignment process, and the alignment state includes a traction state or an idling state or a braking state, and the first train control rate vector is used to represent the size of the control power required by the train in the alignment process; convert the second train state vector into the current power level of the train in the current time period, and determine the current power instruction based on the input data set of the current time period; send the current power level and the current power instruction to the train automatic driving system; The train automatic driving system is used to control the train to move toward the position of the target parking window based on the current power level and the current power command; The train intelligent detection system is further used to determine a second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable, and to use the current time period as the previous time period to update the current power level and the current power command until the train stops moving, wherein the optimal motion variable is determined based on the standard power level of the train and the minimum energy consumption of the train, and the standard power level is determined based on the input data set of the current time period; The method of determining the second train control rate vector of the train in the current time period based on the current power level and the optimal motion variable includes: constructing a conversion relationship between the power level, the optimal tracking control variable and the train control rate vector to obtain conversion information, wherein the conversion information includes a first diagonal matrix, a second diagonal matrix, a first characteristic matrix and a second characteristic matrix, and the main diagonal elements of the first characteristic matrix and the second characteristic matrix are both zero; the optimal tracking control variable is the control variable corresponding to the minimum energy consumption of the train; and determining the second train control rate vector based on the first diagonal matrix, the first characteristic matrix, the second characteristic matrix, the second diagonal matrix, the average control rate of the control rate matrix of the previous time period, the preset weight matrix, the optimal motion variable and the current power level.

12. The train calibration system according to claim 11, characterized in that: It also includes an automatic train protection system, which: The train intelligent detection system is used to obtain the position relationship between the front of the train and the platform beacon after receiving the train stop information sent by the train automatic driving system; if the position relationship is that the front of the train exceeds the platform beacon, send train over-beacon information to the train automatic driving system; receive the start backward jump instruction sent by the train automatic driving system; The train automatic driving system is used to send a backward jump request to the train automatic protection system after receiving the train passing information; The automatic train protection system is used to send a backward jump permission instruction to the automatic train driving system after receiving the backward jump application; The train automatic driving system is also used to send the start backward jump instruction to the train intelligent detection system after receiving the allow backward jump instruction.

13. 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 calibration method as described in any one of claims 1 to 9 is implemented, or the train calibration method as described in claim 10 is implemented.

14. 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 calibration method according to any one of claims 1 to 9 is implemented, or the train calibration method according to claim 10 is implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the train calibration method according to any one of claims 1 to 9 is implemented, or the train calibration method according to claim 10 is implemented.