A communication processing method for safety signals on a single-track multi-vehicle system

Abstract

This invention relates to the field of rail transit technology and discloses a communication processing method for safety signals of multiple vehicles on a single track. The method first collects communication quality indicators between each guide vehicle and the track-side communication module in real time. Based on the collected communication quality indicators and the estimated link status, a link switching decision model is solved. The guide vehicles are then controlled to switch communication links according to the optimal switching strategy. A pre-connection is established and seamless switching is performed. When switching fails or a link is abnormal, an abnormality handling process is triggered, and the switching strategy is adaptively updated. By introducing a link switching decision model based on dynamic game theory, when the interference factors between vehicles increase, game theory helps to assess the communication impact between each vehicle and other vehicles, and switches to a less interfering link when necessary, avoiding mutual interference between optical communications of multiple RGVs and ensuring the stability of the communication link.

Description

Technical Field

[0001] This invention relates to the field of rail transit technology, specifically to a communication processing method for safety signals of a single-track multi-vehicle system. Background Technology

[0002] With the rapid development of intelligent manufacturing and automated logistics systems, contactless powered monorail guided vehicles (RGVs) are widely used for material handling and production scheduling within workshops. To achieve efficient collaborative operation of multiple RGVs on the same track, the communication system needs to meet the requirements of continuous data exchange in a high-speed moving environment while ensuring real-time performance, reliability, and safety.

[0003] The existing system primarily uses optical communication and WiFi communication for information exchange between the vehicle and the ground control system. However, when three or more RGVs are present on a single track simultaneously, the optical communication link is easily interrupted due to vehicle obstruction, leading to unstable signal transmission and severely affecting the execution of control commands. Meanwhile, limited by the controlled 5G network resources and dense 2.4GHz band equipment in industrial settings, WiFi communication often suffers from channel congestion and severe interference, making it impossible to guarantee communication quality. Furthermore, abnormal situations such as the opening of maintenance doors can also cause momentary interruptions to the optical communication link, further exacerbating communication unreliability.

[0004] Therefore, it is necessary to continue to provide a secure industrial control software to effectively control the entire formation process and achieve the goal of safe operation. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a communication processing method for safety signals of single-track multi-vehicle systems, solving the problems of strong physical link dependence and weak anti-interference capability: optical communication relies on line-of-sight transmission, and factors such as opening maintenance doors and vehicle vibration can easily lead to optical path obstruction or signal attenuation, directly causing communication interruption.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a communication processing method for safety signals on a single-rail multi-vehicle track, comprising the following steps:

[0007] First, collect real-time communication quality indicators between each track guidance vehicle and the track-side communication module;

[0008] Based on the collected communication quality indicators, link state estimation and prediction are performed;

[0009] Based on the link status estimation results, a link switching decision model is constructed, with the switching cost and link interruption risk as optimization objectives.

[0010] The optimal handover strategy is obtained by solving the link handover decision model;

[0011] The control track guidance vehicle switches communication links according to the optimal switching strategy;

[0012] During the handover process, a pre-connection is established and a seamless handover is performed to ensure that the communication link is not interrupted during the handover.

[0013] When a handover fails or a link malfunctions, an exception handling process is triggered and the handover strategy is updated adaptively.

[0014] Preferably, the communication quality indicators include received signal strength indication, packet loss rate, and average communication delay. The received signal strength indication is collected in a fixed periodic manner, and the packet loss rate and average communication delay are obtained by statistical analysis of continuous data packets.

[0015] Preferably, the link state estimation adopts the Bayesian inference method, which updates the posterior probability of the link availability state in real time based on historical observation data and current observation data. The link state is a binary state, indicating whether the link is available or unavailable.

[0016] Preferably, the prediction of the link state is based on a Markov transition model, which infers the state transition probability at the next moment based on the current link state. The state transition matrix is ​​obtained through offline training using historical vehicle operation data or updated online in real time.

[0017] Preferably, the link switching decision model is based on dynamic game theory, and the game participants include the vehicle itself and other interfering vehicles. The optimal switching strategy is obtained by minimizing the weighted value of the vehicle's communication interruption risk and switching energy consumption under the worst interference condition.

[0018] Preferably, the optimal handover strategy, based on link status prediction and combined with the link availability prediction value of the target link, selects the link that minimizes the handover cost and maximizes communication reliability as the handover target, and adjusts the connection status of the vehicle communication module according to the selected handover target.

[0019] Preferably, a delay-capacity trade-off mechanism is set during the handover control process. The delay-capacity trade-off mechanism limits the number of handovers per unit time. When the number of handovers per unit time exceeds a set threshold, it is used to suppress unnecessary link handovers and prevent the average communication latency from being too high.

[0020] Preferably, the switching execution process includes the following sub-steps:

[0021] After the vehicle's rear communication module establishes a handshake connection with the target module in advance;

[0022] The verification target is whether the module link quality meets the switching criteria;

[0023] After the verification is successful, quickly disconnect the original primary link and upgrade the target link to the new primary link.

[0024] Preferably, the switching delay after the switching is completed is ≤ ten milliseconds, and the control command link is not interrupted throughout the entire switching process, so as to ensure the continuity and real-time performance of the safety control signal transmission for the track-guided vehicle.

[0025] Preferably, the exception handling process includes:

[0026] When a handover fails or a link is interrupted, the system automatically reverts to the original link or triggers a safe stop command. After executing the stop command, the abnormal handover record is fed back to the link state estimation module for subsequent link estimation and policy optimization updates.

[0027] This invention provides a communication processing method for safety signals on a single-track multi-vehicle system. It has the following beneficial effects:

[0028] 1. This invention introduces a link switching decision model based on dynamic game theory, realizing a mechanism for intelligently selecting the optimal communication link under different communication environments. When the interference factors between vehicles increase, game theory helps to evaluate the communication impact between each vehicle and other vehicles, and switches to a link with less interference when necessary, avoiding mutual interference between optical communications of multiple RGVs and ensuring the stability of the communication link.

[0029] 2. This invention limits the number of handovers per unit time by designing a delay-capacity trade-off mechanism, preventing unnecessary frequent link handovers and reducing the system burden caused by excessive handovers. The result is an effective reduction in communication latency and system energy consumption, while improving the stability of the communication link and avoiding the problem of excessively high average latency.

[0030] 3. By introducing a short-time window for parallel maintenance of dual links during the switching process, this invention ensures that communication continuity is maintained during link switching. The result is that the switching latency is compressed to less than ten milliseconds, which greatly improves the real-time performance of the switching and the system's response speed, and provides a reliable guarantee for the transmission of safety control signals for track-guided vehicles.

[0031] 4. This invention designs an anomaly handling process that automatically reverts to the original link or triggers a safe stop command when a switchover fails or a link is interrupted. The anomaly switchover record is then fed back to the link state estimation module for subsequent strategy optimization and updates. The result is an enhanced fault tolerance and adaptive capability of the system, ensuring vehicle safety and communication stability under extreme conditions. Attached Figure Description

[0032] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0033] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Please see the appendix Figure 1 This invention provides a communication processing method for safety signals on a single-track multi-vehicle system, comprising the following steps:

[0035] First, collect real-time communication quality indicators between each track guidance vehicle and the track-side communication module;

[0036] Based on the collected communication quality indicators, link state estimation and prediction are performed;

[0037] Based on the link status estimation results, a link switching decision model is constructed, with the switching cost and link interruption risk as optimization objectives.

[0038] The optimal handover strategy is obtained by solving the link handover decision model;

[0039] The control track guidance vehicle switches communication links according to the optimal switching strategy;

[0040] During the handover process, a pre-connection is established and a seamless handover is performed to ensure that the communication link is not interrupted during the handover.

[0041] When a handover fails or a link malfunctions, an exception handling process is triggered and the handover strategy is updated adaptively.

[0042] Communication quality indicators include received signal strength indication, packet loss rate, and average communication delay. Received signal strength indication is collected periodically, while packet loss rate and average communication delay are obtained through statistical analysis of continuous data packets.

[0043] Specifically, the onboard communication module of the track guidance vehicle periodically collects the Received Signal Strength Indication (RSSI) between the vehicle and the track-side communication module. To ensure the timeliness of the data and the accuracy of the link status estimation, the RSSI collection period is set to 100 milliseconds. When the vehicle speed is high or the environment changes drastically, the collection period can be appropriately shortened, for example, set to 50 milliseconds, to cope with rapidly changing link conditions. The packet error rate (PER) is obtained by the ratio of the number of failed transmissions in continuously sent data packets to the total number of transmissions, defined within a continuous M data packet window.

[0044] ;

[0045] in, This represents the number of data packets lost in a series of M consecutive data packets. The value of M can be set between 50 and 200 depending on the vehicle's operating environment. A larger value results in a more stable packet loss rate estimate, while a smaller value leads to a faster system response. The value of M can be dynamically adjusted to adapt to changes in the link speed. The average communication latency is obtained by measuring the round-trip time (RTT) of consecutive data packets and averaging the results. If the number of valid samples within the observation window is set to L, then the formula for calculating the average latency is:

[0046] ;

[0047] in, Let be the round-trip delay value of the i-th data packet. To improve statistical stability, a sliding window mechanism can be used to update the average delay index in real time. That is, as new sampled data enters, the oldest historical samples are simultaneously eliminated. To further improve the robustness of communication quality perception, this embodiment assigns different weights to the three indices RSSI, PER, and Latency to construct a joint link quality score. The link quality score (LQS) can be expressed as:

[0048] ;

[0049] in, , , These are the weighting coefficients for each indicator, satisfying... .function , , These are the standardized processing functions for each communication index. The larger the RSSI, the better the link, so function f can be set to linear normalization; the smaller the PER and Latency, the better the link, so g and h can be set to inverse normalization. In order to simplify the calculation and reduce hardware resource consumption, the standardized function can use piecewise linear approximation instead of complex nonlinear normalization, reducing the real-time calculation burden of the vehicle module. In order to enhance the communication robustness of the system in extreme environments, an abnormal alarm mechanism based on communication quality mutation detection is also introduced.

[0050] When RSSI drops below a preset threshold for several consecutive periods, or PER exceeds the normal range for several consecutive periods, the system triggers an early handover warning and initiates the link handover preparation process in advance. Through real-time perception and statistics of the above communication quality indicators, this invention can provide reliable link status observation data while ensuring real-time performance and accuracy. This not only significantly improves the accuracy of link status inference, but also effectively supports subsequent dynamic game modeling and optimal handover strategy formulation.

[0051] Link state estimation uses a Bayesian inference method, which updates the posterior probability of link availability in real time based on historical and current observation data. The link state is a binary state, indicating whether the link is available or unavailable.

[0052] Specifically, through this module, the system can dynamically assess the current availability of the link between the track guidance vehicle and the track side based on communication quality indicators, and provide data support for subsequent link switching decisions. The link status is abstracted as a binary random variable with a value set of {0,1}, where 1 represents that the link is available and 0 represents that the link is unavailable. The link status cannot be directly observed and can only be inferred based on the collected communication quality indicators. Let the link status be... Communication observation data is Based on Bayes' theorem, the formula for calculating the posterior probability is as follows:

[0053] ;

[0054] in, Given observation data The posterior probability of the link state after that. For the link state is Observational data under conditions The likelihood probability of occurrence This represents the prior probability of the link state. The marginal probabilities of the observed data can be expanded using the law of total probability as follows:

[0055] ;

[0056] Initial prior probability Based on historical communication statistics for the rail area, for a specific rail segment, the percentage of available link time over the past month is used to set the parameters. 80%, The percentage is 20%. Of course, in the initial operation of the track section, a uniform distribution can also be set, i.e. To avoid over-reliance on historical data, likelihood probability In modeling, it is assumed that communication quality indicators (such as RSSI, PER, and Latency) follow different probability distributions in link availability and unavailability states. RSSI approximately follows a normal distribution with a relatively high mean when the link is available, but its mean decreases significantly and its variance increases when the link is unavailable. PER and Latency fluctuate less when the link is available, but exhibit abnormal increases when the link is unavailable. These assumptions can be tested and parameter fitted using actual sampled data to ensure the rationality of the modeling. The joint likelihood function of the communication indicators can be assumed to be independent of each indicator, then:

[0057] ;

[0058] in, , , These represent the conditional probabilities of each communication quality index under link state S. To simplify calculations and improve real-time performance, this embodiment adopts an incremental Bayesian update method. That is, each time a new set of observation data is acquired, a rapid update is immediately performed based on the existing posterior probability, rather than recalculating the entire dataset. The update process is as follows:

[0059] Let the posterior probability at the current time be... The new observational data is The updated posterior probability is:

[0060] ;

[0061] It can significantly reduce computational complexity and is particularly suitable for scenarios with extremely high real-time requirements in rail transit systems. To improve the sensitivity of anomaly detection, a state transition penalty mechanism is introduced in this embodiment. That is, when the link state frequently switches between available and unavailable, the system automatically reduces the switching frequency and suppresses jitter by adjusting the state transition probability. Through the above-mentioned link state Bayesian estimation method, this invention can still accurately perceive the current state of the link in environments with drastic fluctuations in communication quality or complex environments. Compared with the traditional method based on fixed threshold judgment, this scheme has higher adaptability and robustness, and is particularly suitable for complex scheduling scenarios with multiple trains running concurrently on a single track.

[0062] The prediction of link state is based on the Markov transition model. The Markov transition model infers the state transition probability at the next moment based on the current state of the link. The state transition matrix is ​​obtained by offline training through vehicle operation history data or updated online in real time.

[0063] Specifically, by establishing a link state transition probability model, the link availability trend at the next moment can be predicted based on the current observed state, thereby making communication link management decisions in advance. By combining offline historical data training with online real-time learning mechanisms, the transition probability matrix is ​​dynamically adjusted to ensure that the model maintains high accuracy over a long period of time. The link state is also modeled as a binary state, i.e., the link is available (1) or unavailable (0). The link state transition follows the first-order Markov property, i.e., the link state at the current moment is only related to the state at the previous moment and is not related to earlier states. Let the link state sequence be... Then we have:

[0064] ;

[0065] If the system detects strong time correlation in the communication environment, a higher-order Markov chain can be used in the extended implementation. However, this embodiment prioritizes first-order modeling, and the link state transition is defined by the following 2×2 state transition matrix:

[0066] ;

[0067] in: The probability that the link is currently unavailable and will remain unavailable in the next moment;

[0068] The probability that a link is currently unavailable but will become available in the next moment;

[0069] The probability that a link is currently available but becomes unavailable in the next moment;

[0070] The probability that a link is currently available and will remain available in the next moment.

[0071] The basic conditions that need to be met are:

[0072]

[0073] During the offline training phase, a large number of link state observation sequences of the track guidance vehicle in different track areas and time periods are collected. The transition frequency of adjacent state pairs is counted, and a preliminary state transition probability matrix is ​​calculated. If the number of times the link transitions from an available state to an unavailable state in the training data is N10, and the number of times it remains available from an available state is N11, then:

[0074] ;

[0075] The sampling window length is chosen to be greater than tens of thousands of state transitions to ensure statistical confidence. To adapt to dynamic environmental changes, this embodiment also introduces an online real-time update mechanism. During the online update process, the system continuously collects link state change data and updates the state transition matrix using a sliding time window method. Whenever a new state transition sample is collected, the original probability matrix is ​​smoothly updated using the exponentially weighted moving average (EWMA) method. The update formula is as follows:

[0076] ;

[0077] in, The updated transition probabilities;

[0078] For newly observed transition probability samples;

[0079] The probability from the last update;

[0080] The smoothing factor is typically set between 0.01 and 0.1 to balance freshness and stability. To avoid extreme outliers affecting state prediction, this embodiment includes an anomaly detection mechanism. When the deviation between a single observation sample and the current model's statistical result exceeds a preset threshold, the system marks the sample as an anomaly and rejects it from use for transition matrix updates. Anomaly determination can be achieved using Mahalanobis distance or a simple discrimination rule based on the mean and standard deviation. Based on the above prediction model, this embodiment implements look-ahead prediction of link availability, i.e., for a link that is currently available, predicting the probability that the link will become unavailable in the next moment. 10 By setting a threshold, such as 0.7, the link handover preparation process is triggered in advance. Through the link state prediction method based on the Markov transition model, the present invention can effectively alleviate the handover delay problem caused by communication link mutations.

[0081] The link switching decision model is based on dynamic game theory. The game participants include the vehicle itself and other interfering vehicles. The optimal switching strategy is obtained by minimizing the weighted value of the vehicle's communication interruption risk and switching energy consumption under the worst interference condition.

[0082] Specifically, by treating the vehicle and other potentially interfering vehicles as game participants and establishing a competitive and cooperative relationship, this invention seeks the optimal strategy for communication link switching in complex environments. The decision-making objective not only considers the risk of link interruption but also comprehensively considers the energy consumption incurred during the switching process, ensuring a balance between system reliability and economy. Each track guidance vehicle is considered a rational participant in the game, and the game scenario is modeled as a dynamic complete information game. At each moment, each participant makes a decision on whether to switch links based on the observed local link state and interference situation, A = {maintain the current link, switch to another link}. The action set of other vehicles is similarly denoted as A′, and the link state space is... Each state contains information such as link availability and interference intensity. In this embodiment, it is assumed that the interfering vehicle causes random interference to the communication link of this vehicle. The interference intensity can be modeled as a random variable that follows a certain statistical distribution, such as a normal distribution or a Poisson distribution. In the game modeling process, the payoff function (utility function) of each participant is designed as follows:

[0083] ;

[0084] Here, Risk represents the probability of communication interruption or the expected loss;

[0085] Energy represents the energy cost of the switching action;

[0086] , For the weighting coefficients, satisfying This reflects the relative importance of communication reliability and energy efficiency. Communication outage risk estimation can be based on link status prediction results. If the predicted probability of the current link being out of service in the future exceeds a certain threshold, such as 0.7, then the outage risk is given a higher weight. The energy consumption cost is obtained based on the power consumption measurement of the communication module during vehicle handover operations, such as the average increase in energy consumption per handover. This vehicle obtains the handover decision by solving the optimal strategy problem that minimizes the loss of benefits in the worst case. Specifically, it is modeled as a dynamic zero-sum game problem.

[0087] ;

[0088] in, For this vehicle's strategy, A set of strategies to interfere with vehicles. To represent the desired operation, when formulating a switching strategy, this vehicle assumes that other vehicles always adopt the strategy most unfavorable to it (i.e., the most severe interference). Under this assumption, it finds the optimal response strategy for its own benefit. To avoid the computational overhead of large-scale game theory, this embodiment adopts the Approximate Dynamic Programming (ADP) technique. By discretizing the state space and action space, it uses a strategy evaluation method based on Monte Carlo sampling to quickly approximate the optimal strategy. A strategy update mechanism is also introduced, that is, when the track environment changes or the interference situation fluctuates drastically, the vehicle can correct its switching decision strategy online based on the latest observation data, ensuring that the decision system is always in an efficient operating state.

[0089] The optimal handover strategy, based on link status prediction and combined with the target link's link availability prediction, selects the link that minimizes handover cost and maximizes communication reliability as the handover target, and adjusts the connection status of the vehicle communication module according to the selected handover target.

[0090] Specifically, after completing link state prediction, the vehicle communication module maintains a candidate link pool. This pool contains all available trackside communication module connection options that meet the minimum communication quality requirements. Each target link... Corresponding to a link availability prediction value and the handover cost required to switch to the current link. .

[0091] in, Indicates link Maintain the available prediction probability in the next cycle. Indicates switching from the current link to the link. The overall cost includes time consumption, increased energy consumption, and load control. In one possible implementation, a comprehensive evaluation of the metrics for each candidate link can be used to define an objective function. as follows:

[0092] ;

[0093] in, , These are the weighting coefficients for communication reliability and handover cost, respectively, satisfying... System default Set to 0.7, Setting it to 0.3 emphasizes the priority of communication continuity; in scenarios with low battery or energy shortages, the setting can be appropriately increased. The weights are assigned to save energy. Then, the vehicle selects the target link. ,satisfy:

[0094] Then, the vehicle selects the target link. ,satisfy:

[0095] ;

[0096] in, Given the set of all candidate links, to further improve decision robustness, this embodiment sets a lower bound for link availability prediction, i.e., if the predicted availability of the target link is... If the value is below a certain threshold (e.g., 0.6), even if the switching cost is low, it will not be selected as a switching target to avoid frequent switching to unstable links. After determining the optimal switching target, the vehicle communication module will select the link based on the chosen link. Automatically adjusting the connection status involves pre-connection: establishing a low-power handshake connection with the target link in advance; switchover preparation: verifying whether the real-time communication quality of the target link meets the minimum requirements; and execution of the switchover: after successful verification, quickly disconnecting the current main link and promoting the target link to the main link.

[0097] To ensure business continuity during the switching process, dual links are maintained in parallel for a short time window during the switching period until the target link is working stably before disconnecting the original link. This method can effectively reduce the risk of interruption during the switching moment. In rail transit application scenarios, a multi-target switching mechanism is also supported. That is, if the quality of the preferred target link is detected to drop sharply, the system can automatically switch to the next optimal link in the candidate link pool in a short time to ensure the ultimate continuity of communication.

[0098] By implementing the above-mentioned optimal switching strategy, intelligent switching of communication links can be achieved in dynamic and complex environments, which greatly improves the communication stability of rail-guided vehicles in high-density traffic environments. Compared with the traditional mechanism based on a single threshold switching, this invention has significantly improved the switching success rate and the average communication quality of the system, and has excellent overall energy consumption control effect.

[0099] A delay-capacity trade-off mechanism is set during the handover control process. This mechanism limits the number of handovers per unit time. When the number of handovers per unit time exceeds a set threshold, it is used to suppress unnecessary link handovers and prevent excessively high average communication latency.

[0100] Specifically, by setting a threshold limit on the number of link handovers per unit time, the system can effectively suppress unnecessary frequent handovers, thereby avoiding abnormal increases in average communication latency and improving the overall link's Quality of Service (QoS). The statistical time window is set as follows: The unit can be seconds or minutes, and can be flexibly selected according to the communication load of the actual rail transit application. In most scenarios, It can be set to 10 seconds, within each statistical time window. The system records the number of link switching events. . like Exceeding the preset maximum number of switching thresholds If the delay is too high, the system will enter suppression mode, temporarily prohibiting or delaying subsequent non-emergency link handover requests. The determination logic for the delay-capacity tradeoff can be described by the following formula:

[0101] If , then Restrict Switching;

[0102] To preset the upper limit of the number of handovers, it is generally set jointly based on vehicle operating speed, link quality fluctuation frequency, and the importance of communication services. When the communication environment of the track section is stable, Two to three handover attempts are acceptable; in high-interference environments, this can be relaxed to four to five. While suppressing subsequent handovers, the system still allows emergency handovers caused by extremely high risks of link failure or interruption (such as link predicted availability being less than 0.4), prioritizing safety. Non-emergency handover requests are postponed to the next statistical window for reassessment, using the following steps:

[0103] Statistics of the current Internal switching count ;

[0104] contrast and

[0105] like Handle the switch request normally; if This implementation filters and allows only emergency switches, while postponing the rest. To improve the system's adaptability, this embodiment supports dynamic adjustment. Vehicles can adjust their configurations based on real-time changes in communication load, link congestion status, or track segment safety levels. The value of is determined to balance latency and stability under different operating conditions. The latency-capacity trade-off mechanism can significantly reduce the additional overhead caused by switching, especially in scenarios with multiple vehicles operating concurrently and complex communication environments. It effectively suppresses abnormal fluctuations in average latency and ensures the real-time and reliable transmission of scheduling instructions and safety signals.

[0106] The switching execution process includes the following sub-steps:

[0107] After the vehicle's rear communication module establishes a handshake connection with the target module in advance;

[0108] The verification target is whether the module link quality meets the switching criteria;

[0109] After the verification is successful, quickly disconnect the original primary link and upgrade the target link to the new primary link.

[0110] Specifically, before initiating a handshake request, the vehicle's rear-end communication module establishes a handshake connection with the target slave module. This handshake process includes, but is not limited to, the following steps: establishing an initial communication connection between the target slave module and the rear-end communication module; exchanging necessary link quality information, including link latency, bandwidth, and signal strength; executing the handshake process using a low-power protocol to minimize resource consumption; and after completing the handshake connection, the system further verifies the link quality of the target slave module to ensure that the handshake criteria are met. The target link quality criteria include:

[0111] Link availability It meets the preset threshold, which usually needs to be higher than 0.7;

[0112] Link bandwidth Meets the minimum required bandwidth;

[0113] Link Delay Within an acceptable range, such as less than 100ms, the system employs the following decision logic during the verification process:

[0114] If and and , then Switch Appro;

[0115] If the target link meets the switching criteria, the backend communication module will perform a fast switching procedure. Once the verification is successful, the system will quickly disconnect the current main link. This step must ensure that the current link does not interrupt communication services before disconnection. Data caching or cache recovery mechanisms are usually used to avoid data loss. After disconnecting the original link, the target link will be immediately upgraded to the main connection and begin processing new communication data streams. All communication traffic of the target link will be forwarded through the new link to ensure the stability of the link after switching. After the availability of the target link is confirmed, the system will prepare the target link in advance, including checking the link stability and buffering data, so that the original main link can be quickly disconnected during the switch. During the switch, the system will maintain a short-term dual-link parallel working mode, that is, the original main link and the target link transmit data at the same time until the target link is stable before the original main link is completely disconnected.

[0116] It also considers fault tolerance mechanisms in extreme interference or special environments. In the event of temporary signal loss or link interruption during link switching, the system can automatically re-initiate the switching process or switch to the next target link in the backup link pool. This mechanism can ensure that the link switching can be quickly restored when the communication environment is extremely unstable, avoiding communication interruption. Through the above switching execution process, the present invention can minimize switching delay and improve the overall system efficiency while ensuring link stability and communication continuity. Compared with the traditional step-by-step switching method, the execution process of the present invention is faster and more reliable, and can cope with the challenges in highly dynamic and complex environments.

[0117] The switching delay after the switching is completed is ≤ ten milliseconds, and the control command link is not interrupted throughout the entire switching process, which is used to ensure the continuity and real-time performance of the safety control signal transmission for track-guided vehicles.

[0118] Specifically, by controlling the switching delay, the system ensures that the control command link remains uninterrupted throughout the switching process, thereby guaranteeing the continuous transmission of safety control signals for the track guidance vehicle and avoiding potential safety risks caused by delays or interruptions during the switching process. During the link switching execution, the system employs a strict switching delay control mechanism to ensure that the switching execution delay does not exceed a predetermined threshold, typically set to 10 milliseconds (ms). The switching delay includes the following sub-processes:

[0119] Once the quality of the target link is confirmed, the system immediately initiates a handshake connection with the target link. The handshake connection uses low-latency, high-efficiency protocols, such as IEEE 802.11p and other vehicle networking protocols, to ensure that the connection is established in a very short time. After the handshake connection is completed, the system performs a quality check on the target link to ensure that the link latency and bandwidth meet the switching requirements. The target link latency should not exceed 100ms, the bandwidth requirement is not less than 500kbps, and the link stability requirement is to be maintained for at least 30 seconds during the switching period. Once the check passes, the system will quickly disconnect the current main link and promote the target link to the main connection. This process ensures that the target link has been fully established and is in a stable state before the main connection is switched.

[0120] To achieve this delay control, the following technical means were adopted:

[0121] During link switching, the original link and the target link will run in parallel for a short period of time to ensure no data loss. The parallel time window is usually set to within 10ms to ensure a seamless switching process. This mechanism ensures that the safety control signals of the track guidance vehicle are always transmitted by bidirectionally transmitting critical control signals during the switching period. The system uses low-latency protocols to optimize the link switching process, such as Time-Sensitive Networking (TSN) technology, to ensure that the transmission delay of control signals during link switching does not exceed 10ms. The TSN protocol can ensure that the communication network latency is maintained within an extremely low range during multi-link parallel operation or link switching. In addition, the control command link will not be interrupted at any time during the entire switching process. This is achieved through the following steps:

[0122] The system monitors the link quality in real time during the switching process and adjusts the switching timing based on feedback information to ensure that the command link is not interrupted after the switching is completed. The system uses a data caching mechanism during the switching process to buffer the command data transmitted during the switching period and avoid information loss during the link switching process. If a link is disconnected or the quality is substandard, the system can quickly and automatically restore to the original link or select a backup link to ensure the real-time transmission of control commands. This effectively ensures that the transmission of control commands for the track guidance vehicle is not interrupted during the entire link switching process, thereby ensuring the safety and real-time performance of the system.

[0123] The exception handling process includes:

[0124] When a handover fails or a link is interrupted, the system automatically reverts to the original link or triggers a safe stop command. After executing the stop command, the abnormal handover record is fed back to the link state estimation module for subsequent link estimation and policy optimization updates.

[0125] Specifically, during the link switching process, if an abnormal situation such as switching failure or link interruption occurs, the system will immediately take the following measures:

[0126] ;

[0127] At this point, the system will re-establish the connection with the original link to ensure the continuous transmission of control commands and avoid communication interruptions caused by handover failure. In the event of a rollback failure, if the link interruption continues for more than a predetermined threshold (e.g., 3 seconds), the system will automatically trigger a safe stop command to ensure the safe operation of the vehicle. The safe stop command will be executed in the subsequent control system, immediately stopping the vehicle's movement and maintaining the vehicle in its current state until safe communication is restored. Whether the system rolls back to the original link or triggers a stop command, it will automatically record the handover anomaly and feed the anomaly log back to the link state estimation module. The recorded content includes: the handover failure time; the time and duration of the link interruption; the cause analysis of the handover failure; and the recovery measures taken (e.g., rollback or safe stop). The purpose of anomaly recording and feedback is to provide data support for subsequent link state estimation and strategy optimization.

[0128] By analyzing anomaly logs, the link state estimation module can update the link quality prediction model and optimize subsequent link switching decision strategies.

[0129] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A communication processing method for safety signals on a single-track multi-vehicle system, characterized in that, Includes the following steps: First, collect real-time communication quality indicators between each track guidance vehicle and the track-side communication module; Based on the collected communication quality indicators, link state estimation and prediction are performed; Based on the link status estimation results, a link switching decision model is constructed, with the switching cost and link interruption risk as optimization objectives. The optimal handover strategy is obtained by solving the link handover decision model; The control track guidance vehicle switches communication links according to the optimal switching strategy; During the handover process, a pre-connection is established and a seamless handover is performed to ensure that the communication link is not interrupted during the handover. When a handover fails or a link malfunctions, an exception handling process is triggered and the handover strategy is updated adaptively. The link switching decision model is based on dynamic game theory, with the game participants including the vehicle itself and other interfering vehicles. The optimal switching strategy is obtained by minimizing the weighted value of the vehicle's communication interruption risk and switching energy consumption under the worst interference condition.

2. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: The communication quality indicators include received signal strength indication, packet loss rate, and average communication delay. The received signal strength indication is collected in a fixed periodic manner, and the packet loss rate and average communication delay are obtained by statistical analysis of continuous data packets.

3. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: The link state estimation uses a Bayesian inference method, which updates the posterior probability of the link availability state in real time based on historical and current observation data. The link state is a binary state, indicating whether the link is available or unavailable.

4. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: The prediction of the link state is based on a Markov transition model, which infers the state transition probability at the next moment based on the current link state. The state transition matrix is ​​obtained through offline training using historical vehicle operation data or updated online in real time.

5. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: The optimal handover strategy, based on link status prediction and combined with the target link's link availability prediction, selects the link that minimizes handover cost and maximizes communication reliability as the handover target, and adjusts the connection status of the vehicle communication module according to the selected handover target.

6. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: A delay-capacity trade-off mechanism is set during the handover control process. This mechanism limits the number of handovers per unit time. When the number of handovers per unit time exceeds a set threshold, it is used to suppress unnecessary link handovers and prevent excessively high average communication latency.

7. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: The switching execution process includes the following sub-steps: After the vehicle's rear communication module establishes a handshake connection with the target module in advance; The verification target is whether the module link quality meets the switching criteria; After the verification is successful, quickly disconnect the original primary link and upgrade the target link to the new primary link.

8. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: The switching delay after the switching is completed is ≤ ten milliseconds, and the control command link is not interrupted throughout the entire switching process, which is used to ensure the continuity and real-time performance of the safety control signal transmission for track-guided vehicles.

9. The communication processing method for a single-track multi-vehicle safety signal according to claim 1, characterized in that: The exception handling process includes: when a handover failure or link interruption occurs, automatically reverting to the original link or triggering a safe stop command. After executing the stop command, the exception handover record is fed back to the link state estimation module for subsequent link estimation and policy optimization updates.

Images