A virtual coupling train speed convergence cooperative control method based on a multi-agent system

Abstract

The present invention belongs to the field of computer science and control technology, and discloses a virtual coupling train speed convergence cooperative control method based on a multi-agent system. The specific steps of the control method are as follows: S1: System initialization and parameter setting; S2: Distributed observer design; S3: Adaptive control algorithm construction; S4: Cooperative control algorithm implementation; S5: Global stability analysis; S6: Train formation dynamic formation; S7: Formation operation and dissolution. In the train formation system of the present invention, the distributed observer cleverly solves the communication problem, collects the states of adjacent trains in real time, dynamically compensates the data, ensures accurate mastery of the formation state. The adaptive control algorithm is based on the barrier Lyapunov function, solidifies the spacing requirements into hard constraints, combines the nonlinear control theory to cope with parameter uncertainties, and stabilizes the formation operation. The cooperative control algorithm is like a lubricant, promoting seamless cooperation between trains, synchronous convergence of speeds, and safe and stable spacing.

Description

Technical Field

[0001] The present invention belongs to the field of computer science and control technology, and specifically relates to a virtual coupling train speed convergence cooperative control method based on a multi-agent system. Background Art

[0002] As an urban transportation artery, rail transit faces the challenges of increasing passenger flow and capacity bottlenecks. The traditional moving block system is approaching saturation, and it is difficult to cope with peak passenger flows by increasing the number of trains. Newly built lines are limited by high costs and space resources. Under this background, train formation technology has become the key to improving transportation capacity. This technology abandons physical coupling and relies on advanced inter-train communication and automatic operation technologies to achieve coordinated train operation, effectively shortening the headway between trains and enhancing line capacity. However, formation cooperative control is still complex in the speed coordination stage and requires precise regulation to ensure safety and efficiency. At the same time, external disturbances and model uncertainties have become non-negligible factors affecting control stability and performance, and further research and development of more robust control strategies are needed to fully unleash the potential of train formation technology, flexibly match passenger flow demands, and optimize the operation quality of urban rail transit. Summary of the Invention

[0003] The purpose of the present invention is to provide a virtual coupling train speed convergence cooperative control method based on a multi-agent system to solve the problems mentioned in the above background art.

[0004] To achieve the above purpose, the present invention provides the following technical solution: A virtual coupling train speed convergence cooperative control method based on a multi-agent system. The specific steps of this control method are as follows:

[0005] S1: System initialization and parameter setting: The initial state of the train includes position, speed, and dynamic parameters. Let R1 be the communication distance and R2 be the safety distance to ensure safe communication and train operation.

[0006] S2: Design of distributed observer: Design a distributed observer to estimate the trajectory and interference in real time, and update using the states of adjacent trains to ensure accurate and real-time information.

[0007] S3: Construction of adaptive control algorithm: Use the barrier Lyapunov to maintain the spacing, design an adaptive control algorithm, and calculate the input based on the error, parameters, and interference estimation.

[0008] S4: Implementation of cooperative control algorithm: The cooperative control algorithm promotes the speed synchronization of trains, maintains stable spacing, and improves the operation efficiency and safety of the formation.

[0009] S5: Global stability analysis: Prove stability by global Lyapunov, the algorithm is effective, robust, and guarantees stable operation under various conditions of the system.

[0010] S6: Dynamic formation of train formation: The train i near the end car R1 of the formation applies for communication. If the link is established, it is incorporated into the formation, and the process loops until the entire train is incorporated.

[0011] S7: Formation operation and dissolution: After formation, distributed observation is carried out, and the speed and distance are adaptively controlled; when dissolving, the communication is interrupted, and it can respond dynamically and flexibly.

[0012] Preferably, the system initialization and parameter setting in S1 refer to that in the railway operation system, the initial state of each train is crucial. The specific position, initial running speed of the train, and the dynamic model parameters used to describe its motion characteristics ensure the safety and efficiency of train operation. We set two key distance parameters: the maximum communication distance R1 and the minimum safety distance R2. R1 ensures that effective communication connections can be maintained between trains, enabling real-time exchange of driving information and preventing potential conflicts; while R2 serves as a safety barrier to ensure that the distance between trains is sufficient to implement braking in case of emergency and avoid collisions.

[0013] Preferably, the design of the distributed observer in S2 aims to enable each train to independently and real-time predict its reference running trajectory and external potential disturbances. This observer innovatively incorporates the state information shared in real-time by neighboring trains and continuously updates its own state through data fusion and cross-validation.

[0014] Preferably, the specific steps for constructing the adaptive control algorithm in S3 are as follows:

[0015] Step 1: Use the barrier Lyapunov function to achieve the analysis of the hard spacing constraint and design the barrier function: According to the safety requirements of the train spacing (such as the minimum safety distance R2), design a barrier Lyapunov function V(z), where z is the state variable or error variable related to the train spacing. This function should satisfy that when the train spacing approaches or is less than R2, the value of V(z) increases sharply, indicating a violation of the spacing constraint.

[0016] Embed the constraint condition: Combine the barrier function V(z) with the control objective of the train to ensure that the hard spacing constraint is considered in the control strategy. Usually, this is achieved by adding a term related to V(z) to the control law to adjust the control input when approaching the spacing limit.

[0017] Step 2: Online estimate uncertain parameters and disturbances

[0018] Establish a parameter estimation model: For the uncertain parameters and external disturbances affecting train operation, establish corresponding estimation models.

[0019] Design estimators: Use filtering techniques or observer design methods to design a suitable estimator for each parameter or disturbance to be estimated. These estimators accurately and real-time track the changes of parameters and disturbances.

[0020] Step 3: Design an adaptive control algorithm

[0021] Construct a control strategy: Based on the speed error, the estimated values of uncertain parameters, and the estimated value of the disturbance, design an adaptive control algorithm that can dynamically adjust the control input to compensate for the impact of uncertainties and disturbances on train operation and meet the hard spacing constraint;

[0022] Optimize control parameters: Optimize the parameters in the control algorithm through optimization methods or model-based design methods to improve the performance and stability of the system;

[0023] Verification and testing: Verify and test the designed adaptive control algorithm in a simulation environment to evaluate its performance under different working conditions. Adjust and optimize according to the test results to ensure that the algorithm can operate stably and reliably in practical applications.

[0024] Preferably, in step S4, the implementation of the cooperative control algorithm adopts an advanced cooperative control algorithm. The train system can make full use of the real-time information of neighboring trains to achieve precise speed synchronization and rapid convergence, ensuring that the train spacing is always maintained within the preset safe range, and effectively reducing the energy consumption and delays caused by speed fluctuations.

[0025] Preferably, in step S5, the global stability analysis is performed by selecting a global Lyapunov function V to analyze the system stability. Evaluate the stability of the system dynamic behavior through the change trend of this function over time. Further, based on the analysis results of this function, strictly prove that the designed control algorithm is not only effective under ideal conditions, but also shows strong robustness in the face of complex working conditions such as parameter changes and external disturbances.

[0026] Preferably, the specific steps for the dynamic formation of train formations in step S6 are as follows:

[0027] Step 1: Detect the distance and request communication

[0028] Detect the distance: Train i continuously monitors the distance between itself and the last train in the current formation. The distance is obtained in real time through sensors or communication systems on the train;

[0029] Judgment condition: When train i detects that the distance between it and the last train in the formation is less than or equal to the preset threshold R1, it is considered that train i is in a suitable position to join the formation;

[0030] Request communication: After meeting the conditions, train i sends a request to the last train in the formation via wireless communication, requesting to establish a direct communication link to join the formation;

[0031] Step 2: Establish a communication link and join the formation

[0032] Response to the request: After the last train in the formation receives the request from Train i, it verifies its identity and the validity of the request.

[0033] Establish a link: After verification, a stable communication link is established between the last train in the formation and Train i. This link will be used for subsequent data exchange and transmission of control instructions.

[0034] Join the formation: After the communication link is successfully established, Train i officially joins the formation, and its operating state will be uniformly managed and scheduled by the formation control system.

[0035] Step 3: Repeat the process until all trains have joined.

[0036] Continue monitoring: After Train i joins the formation, the system continues to monitor the distance between other trains that have not joined the formation and the last train in the formation.

[0037] Repeat the request and joining: Whenever a new train enters the range of R1 and meets the conditions for joining the formation, repeat the processes of Step 1 and Step 2, that is, request to establish a communication link and join the formation after success.

[0038] Complete the formation: This process will continue until all trains have successfully joined the formation in sequence, forming a complete train formation and operating efficiently and safely under unified control.

[0039] Preferably, the specific steps for formation operation and disbanding in S7 are as follows:

[0040] Step 1: Speed synchronization and spacing control after formation.

[0041] Deploy distributed observers: After the train formation is formed, distributed observers are deployed on each train. These observers can collect the status information of neighboring trains in real time and estimate the reference trajectory and external disturbances of the trains based on this information.

[0042] Apply the adaptive control algorithm: According to the information provided by the distributed observers, the adaptive control algorithm on each train will calculate appropriate control inputs to achieve synchronous convergence of train speeds and stable control of spacings.

[0043] Step 2: Continuous monitoring and adjustment.

[0044] Real-time feedback and adjustment: During the operation of the train formation, the distributed observers and the adaptive control algorithm will continue to work, providing real-time feedback on the operating state of the trains and adjusting the control strategy as needed.

[0045] Exception handling: If any abnormal situation is detected, the system will immediately trigger the corresponding emergency handling mechanism to ensure the safe and stable operation of the train formation;

[0046] Step 3: Formation disassembly and communication link disconnection

[0047] Disassembly decision: When the operation ends or other situations requiring formation disassembly occur, the system will make a disassembly decision based on the preset disassembly conditions or the operator's instructions;

[0048] Disconnect the communication link: After the disassembly decision is determined, the system will disconnect the communication link between the trains through a specific communication protocol, which is the key to ensuring the safe and orderly disassembly of the train formation;

[0049] Independent operation: After the communication link is disconnected, each train will resume the independent operation state and continue to move forward according to its respective operation plan and control strategy.

[0050] Preferably, 1.

[0051] The beneficial effects of the present invention are as follows:

[0052] In the train formation system of the present invention, the distributed observer cleverly solves the communication problem, collects the states of neighboring trains in real time, dynamically compensates the data, ensures accurate grasp of the formation state, and the adaptive control algorithm is based on the barrier Lyapunov function, solidifies the spacing requirement into a hard constraint, combines the non-linear control theory, copes with parameter uncertainties, stabilizes the formation operation, and the cooperative control algorithm is like a lubricant, promotes seamless cooperation between trains, the speed synchronously converges, the spacing is safe and stable, this system demonstrates excellent robustness and adaptability, operates stably in complex environments, and provides a solid guarantee for the safe and efficient operation of rail transit. Brief description of the drawings

[0053] Figure 1 It is a schematic diagram of the overall process of the present invention;

[0054] Figure 2 It is a model diagram of the train formation operation of the present invention;

[0055] Figure 3 It is a block diagram of the cooperative control algorithm of the present invention;

[0056] Figure 4 It is a diagram of the dynamic formation process of the train formation of the present invention;

[0057] Figure 5 It is a diagram of the disassembly process of the train formation of the present invention;

[0058] Figure 6 It is a schematic diagram of the speed convergence process of the train formation of the present invention. Detailed implementation manners

[0059] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0060] As Figures 1 to 6 shown, the embodiment of the present invention provides a virtual coupling train speed convergence cooperative control method based on a multi-agent system. The specific steps of this control method are as follows:

[0061] S1: System initialization and parameter setting: The initial state of the train includes position, speed, and dynamic parameters. Set the communication distance R1 and the safety distance R2 to ensure safe communication and train operation.

[0062] S2: Design of distributed observer: Design a distributed observer to estimate the trajectory and interference in real time, and update using the states of adjacent trains to ensure accurate and real-time information.

[0063] S3: Construction of adaptive control algorithm: Use the barrier Lyapunov to maintain the spacing, design an adaptive control algorithm, and calculate the input according to the error, parameters, and interference estimation.

[0064] S4: Implementation of cooperative control algorithm: The cooperative control algorithm promotes the speed synchronization of trains, ensures stable spacing, and improves the operation efficiency and safety of the formation.

[0065] S5: Global stability analysis: Prove stability by global Lyapunov, the algorithm is effective, robust, and ensures stable operation under various conditions of the system.

[0066] S6: Dynamic formation of train formation: Train i applies for communication inside the maximum communication distance R1 from the last car of the formation. If the link is established, it joins the formation, and this process repeats until all trains join the formation.

[0067] S7: Formation operation and dissolution: After formation, perform distributed observation and adaptively control the speed and spacing; when dissolving, cut off the communication and respond dynamically and flexibly.

[0068] Among them, the system initialization and parameter setting in S1 refer to that in the railway operation system, the initial state of each train is crucial. The specific position, initial running speed, and dynamic model parameters describing its motion characteristics of the train ensure the safety and efficiency of train operation. We set two key distance parameters: the maximum communication distance R1 and the minimum safety distance R2. R1 ensures that trains can maintain effective communication connections with each other, so as to exchange driving information in real time and prevent potential conflicts; while R2 serves as a safety barrier to ensure that the spacing between trains is sufficient to implement braking in case of emergency and avoid collisions. Such settings ensure that trains operate within a safe and efficient communication range.

[0069] Among them, the design of the distributed observer in S2 aims to enable each train to independently and real - time predict its reference running trajectory and external potential interferences. This observer innovatively integrates the state information shared in real - time by neighboring trains and continuously updates its own state through data fusion and cross - verification;

[0070] The state update equation of the observer is:

[0071]

[0072] where μ is a designed positive number, is the state estimate value of the i - th train, and a ij is an element of the adjacency matrix of the communication topology.

[0073] Thus, it significantly improves the accuracy and timeliness of the estimation. This process ensures the intelligence and safety of train operation control and promotes the smooth operation of the overall transportation system.

[0074] Among them, the specific steps for constructing the adaptive control algorithm in S3:

[0075] Step 1: Use the barrier Lyapunov function to realize the analysis of hard spacing constraints and design the barrier function: According to the safety requirements of the train spacing (such as the minimum safety distance R2), design a barrier Lyapunov function V(z), where z is a state variable or error variable related to the train spacing. This function should satisfy that when the train spacing approaches or is less than R2, the value of V(z) increases sharply as an indication of violating the spacing constraint.

[0076] Embed the constraint conditions: Combine the barrier function V(z) with the control objective of the train to ensure that the hard spacing constraints are considered in the control strategy. Usually, this is achieved by adding a term related to V(z) to the control law so as to adjust the control input when approaching the spacing limit;

[0077] Step 2: Online estimate uncertain parameters and interferences

[0078] Establish parameter estimation models: For the uncertain parameters (such as unknown parameters in the model) and external interferences (such as environmental factors) affecting train operation, establish corresponding estimation models; these models should be able to update the estimated values of the parameters based on the real - time state and operation data of the train.

[0079] Design estimators: Use filtering techniques (such as Kalman filters, particle filters, etc.) or observer design methods to design a suitable estimator for each parameter or interference to be estimated. These estimators should be able to accurately and real - time track the changes of parameters and interferences;

[0080] Step 3: Design the adaptive control algorithm

[0081] Construct the control strategy: Based on the speed error (i.e., the difference between the actual train speed and the desired speed), the estimated values of the uncertain parameters, and the estimated values of the disturbances, design an adaptive control algorithm that can dynamically adjust the control input (such as traction force or braking force) to compensate for the impact of uncertainties and disturbances on train operation and meet the hard spacing constraint;

[0082] Optimize the control parameters: Through optimization methods (such as genetic algorithms, particle swarm optimization, etc.) or model-based design methods (such as LQR, MPC, etc.), tune the parameters in the control algorithm to improve the performance and stability of the system;

[0083] Verification and testing: Verify and test the designed adaptive control algorithm in a simulation environment to evaluate its performance under different working conditions, and make adjustments and optimizations according to the test results to ensure that the algorithm can operate stably and reliably in practical applications.

[0084] The control input is:

[0085]

[0086] where, is the estimated value of the uncertain parameter, e i is the speed error K i , is the control gain, and is the estimated value of the disturbance.

[0087] Calculation of speed error:

[0088] Define the speed error as:

[0089]

[0090] where, v i is the current train speed, is the speed of the reference trajectory,

[0091] Parameter estimation:

[0092] Update the control input through the parameter estimation of the train's dynamic model:

[0093]

[0094] The advantages of constructing the adaptive control algorithm are as follows: ensuring the hard spacing constraint through the barrier Lyapunov function, enhancing the operation safety; online estimating the uncertain parameters and disturbances, strengthening the system robustness; dynamically adjusting the control strategy, optimizing the control parameters, improving the operation efficiency and stability; simulation verification ensuring the reliability of the algorithm, providing an intelligent, flexible and safe control solution for the train formation system.

[0095] Among them, the implementation of the cooperative control algorithm in S4 adopts an advanced cooperative control algorithm. The train system can make full use of the real-time information of adjacent trains to achieve precise synchronization and rapid convergence of speeds, ensuring that the train spacing is always maintained within the preset safe range, and effectively reducing energy consumption and delays caused by speed fluctuations. This significantly improves the overall operation efficiency of the train formation and the safety of passengers' rides, bringing revolutionary optimization and upgrading to the modern rail transit system.

[0096] The cooperative control algorithm helps optimize the train formation: precisely synchronize speeds, rapidly converge, ensure safe spacing, reduce energy consumption and delays. This algorithm makes full use of the real-time information of adjacent trains, improves the operation efficiency and safety of the formation, brings a smoother and more punctual travel experience for passengers, reduces operating costs at the same time, and promotes the intelligent and green development of the rail transit system.

[0097] Among them, in S5, the global stability analysis is carried out by selecting the global Lyapunov function V to analyze the system stability. The stability of the system's dynamic behavior is evaluated through the change trend of this function over time. Further, based on the analysis results of this function, it is strictly proved that the designed control algorithm is not only effective under ideal conditions, but also shows strong robustness in the face of complex working conditions such as parameter changes and external disturbances. This ensures that the entire train formation system can maintain a stable, efficient and safe operating state under various operating conditions.

[0098] Global Lyapunov analysis ensures the stability of the train formation system: evaluate the system dynamics through the V function, and strictly prove the effectiveness and robustness of the control algorithm. Under complex working conditions, such as parameter changes and external disturbances, the system can still operate stably, improving the overall operation safety and reliability, optimizing the passenger experience, and providing a solid guarantee for the safe and efficient operation of the rail transit.

[0099] Among them, the specific steps of the dynamic formation of the train formation in S6 are as follows:

[0100] Step 1: Detect the distance and request communication

[0101] Detect the distance: Train i continuously monitors the distance between itself and the last train in the current formation. The distance is obtained in real time through sensors or communication systems on the train;

[0102] Judgment condition: When train i detects that the distance between it and the last train in the formation is less than or equal to the preset threshold R1, it is considered that train i is in a suitable position to join the formation;

[0103] Request communication: After meeting the conditions, train i sends a request to the last train in the formation through wireless communication, requesting to establish a direct communication link to join the formation;

[0104] Step 2: Establish a communication link and join the formation

[0105] Respond to the request: After the last train in the formation receives the request from Train i, it confirms the identity and validity of the request;

[0106] Establish the link: After confirmation, a stable communication link is established between the last train in the formation and Train i, which will be used for subsequent data exchange and transmission of control instructions;

[0107] Join the formation: After the communication link is successfully established, Train i officially joins the formation, and its operating state will be uniformly managed and scheduled by the formation control system;

[0108] Step 3: Repeat the process until all trains have joined

[0109] Continue to monitor: After Train i joins the formation, the system continues to monitor the distance between other trains that have not joined the formation and the last train in the formation;

[0110] Repeat the request and join: Whenever a new train enters the range of R1 and meets the conditions for joining the formation, repeat the processes of Step 1 and Step 2, that is, request to establish a communication link and join the formation after success;

[0111] Complete the formation: This process will continue until all trains have successfully joined the formation in sequence, forming a complete train formation and operating efficiently and safely under unified control.

[0112] The advantages of the dynamic train formation process are significant: Through real-time distance monitoring and wireless communication, trains can autonomously request to join the formation, ensuring an efficient and flexible formation process; The construction of a stable communication link ensures the smooth transmission of data and control instructions; Repeating the process ensures that all trains join in an orderly manner to form a complete formation, improving the overall operating efficiency and safety, and providing passengers with a more stable and comfortable travel experience.

[0113] Among them, the specific steps of formation operation and dissolution in S7 are as follows:

[0114] Step 1: Speed synchronization and spacing control after formation

[0115] Deploy distributed observers: After the train formation is formed, distributed observers are deployed on each train. These observers can collect the status information of neighboring trains (such as position, speed, acceleration, etc.) in real time, and estimate the reference trajectory and external interference of the trains based on this information;

[0116] Apply the adaptive control algorithm: Based on the information provided by the distributed observer, the adaptive control algorithm on each train calculates the appropriate control inputs (such as traction or braking force) to achieve the synchronous convergence of train speeds and the stable control of the spacing; this algorithm can dynamically adjust the control parameters to cope with various uncertainties and disturbances; ensuring the stability and safety of the overall operation of the train formation.

[0117] Step 2: Continuous monitoring and adjustment

[0118] Real-time feedback and adjustment: During the operation of the train formation, the distributed observer and the adaptive control algorithm continuously work, providing real-time feedback on the train operation status and adjusting the control strategy as needed; this helps to ensure that the train maintains the best performance in a complex and changing operating environment.

[0119] Exception handling: If any abnormal situations (such as train failures, communication interruptions, etc.) are detected, the system will immediately trigger the corresponding emergency handling mechanism to ensure the safe and stable operation of the train formation;

[0120] Step 3: Formation disassembly and communication link disconnection

[0121] Disassembly decision: When the operation ends or other situations requiring formation disassembly occur, the system makes a disassembly decision based on the preset disassembly conditions or the operator's instructions;

[0122] Disconnect the communication link: After the disassembly decision is determined, the system disconnects the communication link between the trains through a specific communication protocol, which is the key to ensuring the safe and orderly disassembly of the train formation;

[0123] Independent operation: After the communication link is disconnected, each train will resume its independent operation state and continue to move forward according to its respective operation plans and control strategies.

[0124] The optimization of the formation operation and disassembly process brings multiple benefits: Through distributed observation and adaptive control, it ensures the stable and efficient operation of the train formation; real-time monitoring and flexible adjustment to cope with complex environments; the exception handling mechanism guarantees safety; the disassembly process is orderly, the communication link is immediately disconnected, promoting the rapid restoration of the trains to independent operation, enhancing the overall operation efficiency and flexibility, and ensuring the safety and smooth travel of passengers.

[0125] Figure 2 Shows the operation model of the train formation, where each train is regarded as an agent, and through the communication link between the trains, neighboring trains can obtain each other's status information (such as speed and position).

[0126] Figure 3 Shows the block diagram of the cooperative control algorithm. By receiving the status information of neighboring trains, it calculates the control inputs to ensure the synchronous convergence of train speeds and the stability of the spacing between trains.

[0127] Figure 4 It shows the dynamic formation process of train formation. Assume that train i does not belong to the formation and its distance from the last train in the formation is within the range of (R2, R1 - γ). Train i requests to establish a communication link with the last train in the formation. After successful establishment, it joins the formation. Repeat this process until all trains have joined the formation.

[0128] Figure 5 It shows the dissolution process of train formation. If the distance between train i and the train in front exceeds the maximum communication distance R1 or communication is lost, and a communication link cannot be re-established within the specified time, train i and the subsequent trains withdraw from the formation.

[0129] Figure 6 It shows a schematic diagram of the speed convergence process of train formation. Through communication between trains, the trains within the formation obtain the status information of the trains in front and synchronize their speeds through a cooperative control algorithm while maintaining a preset inter-train distance.

[0130] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or device.

[0131] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A virtual coupled train speed convergence cooperative control method based on a multi-agent system, characterized by: The specific steps of the control method are as follows: S1: System initialization and parameter setting: The initial state of the train includes position, speed, and dynamic parameters. Set R1 communication distance and R2 safety distance to ensure safe communication and driving; S2: Distributed observer design: Design a distributed observer to estimate trajectories and interference in real time, and use neighboring vehicle status updates to ensure accurate and real-time information; S3: Adaptive control algorithm construction: Use barrier Lyapunov to maintain spacing, design an adaptive control algorithm, and calculate input based on error, parameter and interference estimates; S4: Implementation of cooperative control algorithm: The cooperative control algorithm promotes train speed synchronization, ensures stable spacing, and improves train operation efficiency and safety; S5: Global stability analysis: The global Lyapunov proof is stable, the algorithm is effective and robust, and ensures the stability of the system under all conditions; S6: Dynamic formation of train marshaling: Train i applies for communication when it is close to the last car R1 of the marshaling group. If the link is established, it will be incorporated, and the cycle will continue until all trains are incorporated; S7: marshaling operation and disbanding: distributed observation after marshaling, adaptive speed control distance; disbanding communication, dynamic and flexible response; The specific steps of constructing the adaptive control algorithm in S3 are as follows: Step 1: Use barrier Lyapunov function to implement spacing hard constraint analysis and design barrier function: According to the safety requirements of train spacing, design a barrier Lyapunov function V(z); Embed constraints: The barrier function V(z) is combined with the control objective of the train to ensure that the hard spacing constraint is taken into account in the control strategy. This is achieved by adding a term related to V(z) in the control law to adjust the control input when approaching the spacing limit. Step 2: Online estimation of uncertain parameters and disturbances Establish parameter estimation model: Establish corresponding estimation model for uncertain parameters and external interference that affect train operation; Design estimator: Using filtering technology and observer design methods, a suitable estimator is designed for each parameter or disturbance to be estimated. The estimator accurately and in real time tracks the changes of parameters and disturbances. Step 3: Design an adaptive control algorithm Build control strategy: Based on the speed error, the estimated values ​​of the uncertain parameters and the estimated values ​​of the disturbance, design an adaptive control algorithm that can dynamically adjust the control input to compensate for the effects of uncertainty and disturbance on train operation and meet the spacing hard constraint; Optimize control parameters: Tune the parameters in the control algorithm through optimization methods and model-based design methods to improve system performance and stability; Verification and testing: Verify and test the designed adaptive control algorithm in a simulation environment to evaluate its performance under different working conditions, and adjust and optimize it according to the test results to ensure that the algorithm can run stably and reliably in actual applications.

2. According to claim 1, a virtual coupled train speed convergence cooperative control method based on a multi-agent system is characterized by: The system initialization and parameter setting in S1 refers to the initial state of each train, the specific position of the train, the initial running speed and the dynamic model parameters used to describe its motion characteristics, to ensure the safety and efficiency of train operation, and set two key distance parameters: the maximum communication distance R1 and the minimum safety distance R2. R1 ensures that the trains can maintain effective communication, exchange driving information in real time, and prevent potential conflicts; while R2 acts as a safety barrier to ensure that the distance between trains is sufficient to brake in an emergency and avoid collisions.

3. The method for virtual coupled train speed convergence cooperative control based on a multi-agent system according to claim 1, characterized in that: The distributed observer in S2 is designed to enable each train to independently and in real time predict its reference running trajectory and external potential interference. This observer innovatively incorporates the real-time shared status information of adjacent trains and continuously updates its own status through data fusion and cross-validation.

4. The method for virtual coupled train speed convergence cooperative control based on a multi-agent system according to claim 1, characterized in that: The collaborative control algorithm in S4 is implemented using an advanced collaborative control algorithm. The train system can make full use of the real-time information of adjacent trains to achieve accurate synchronization and rapid convergence of speed, ensuring that the distance between trains is always maintained within a preset safety range, and effectively reducing energy consumption and delays caused by speed fluctuations.

5. The virtual coupled train speed convergence cooperative control method based on a multi-agent system according to claim 1 is characterized by: The global stability analysis in S5 is to perform system stability analysis by selecting the global Lyapunov function V, and evaluate the stability of the system dynamic behavior by the changing trend of this function over time. Furthermore, based on the analysis results of this function, it is strictly proved that the designed control algorithm is not only effective under ideal conditions, but also exhibits strong robustness when facing complex working conditions such as parameter changes and external interference.

6. The virtual coupled train speed convergence cooperative control method based on a multi-agent system according to claim 1, characterized in that: The specific steps of the dynamic formation of train formation in S6 are as follows: Step 1: Detect distance and request communication Detection distance: Train i continuously monitors the distance between itself and the last train in the current marshaling. The distance is obtained in real time through sensors or communication systems on the train; Judgment condition: When train i detects that the distance between it and the train at the end of the formation is less than or equal to the preset threshold R1, train i is considered to be in a suitable position to join the formation; Request communication: After the conditions are met, train i sends a request to the last train of the formation through wireless communication, requesting to establish a direct communication link in order to join the formation; Step 2: Establish a communication link and join the group Respond to the request: After receiving the request from train i, the train at the end of the marshaling group confirms its identity and the validity of the request; Establishing a link: After confirmation, a stable communication link is established between the last train of the marshaling and train i. The link will be used for subsequent data exchange and transmission of control instructions; Joining the marshaling: After the communication link is successfully established, train i officially joins the marshaling, and its running status will be uniformly managed and dispatched by the marshaling control system; Step 3: Repeat the process until all trains are added Continue monitoring: After train i joins the formation, the system continues to monitor the distance between other trains that have not joined the formation and the train at the end of the formation; Repeated request and joining: Whenever a new train enters the R1 range and meets the conditions for joining the formation, the process of steps 1 and 2 is repeated, that is, requesting to establish a communication link and joining the formation after success; Complete marshaling: This process will continue until all trains are successfully joined in sequence to form a complete train marshaling, and run efficiently and safely under unified control.

7. The method for virtual coupled train speed convergence cooperative control based on a multi-agent system according to claim 1, characterized in that: The specific steps of marshaling operation and disbanding in S7 are as follows: Step 1: Speed ​​synchronization and spacing control after formation Deployment of distributed observers: After the train formation is formed, distributed observers are deployed on each train. These observers can collect status information of neighboring trains in real time and estimate the train's reference trajectory and external interference based on this information; Applying adaptive control algorithms: Based on the information provided by the distributed observers, the adaptive control algorithm on each train will calculate the appropriate control input to achieve synchronous convergence of train speed and stable control of spacing; Step 2: Continuous monitoring and adjustment Real-time feedback and adjustment: During the train marshaling process, the distributed observer and adaptive control algorithm will continue to work, provide real-time feedback on the train's operating status, and adjust the control strategy as needed; Abnormal handling: If any abnormal situation is detected, the system will immediately trigger the corresponding emergency handling mechanism to ensure the safe and stable operation of the train formation; Step 3: Disbanding the group and releasing the communication link Disbanding decision: When the operation ends or other situations occur that require the disbanding of the group, the system will make a disbanding decision based on the preset disbanding conditions or the operator's instructions; Release the communication link: After the disbanding decision is made, the system will release the communication link between trains through a specific communication protocol, which is the key to ensure that the train formation can be disbanded safely and orderly; Independent operation: After the communication link is released, each train will resume independent operation and continue to move forward according to its own operation plan and control strategy.

Images