Lane controller abnormity real-time detection method and system based on multi-modal data fusion

Abstract

The invention discloses a lane controller abnormity real-time detection method and system based on multi-modal data fusion. The method comprises the following steps: acquiring a ground induction coil signal, a DSRC communication state and transaction flow data to obtain multi-modal data; performing feature analysis on the multi-modal data to obtain an analysis result; inputting the analysis result into a decision tree model to judge whether the current state of the lane controller is abnormal or not, and performing classification according to different anomaly types to obtain corresponding anomaly types; executing a corresponding repair strategy according to the exception type to obtain a repair result; and confirming the repair result through a self-checking mode, ensuring that the vehicle is allowed to pass after the abnormity is solved, and updating the health state library. By implementing the method provided by the invention, the accuracy of fault detection can be improved, the repair delay is reduced, and the overall reliability of the system is enhanced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation, and more specifically to a method and system for real-time detection of lane controller anomalies based on multi-modal data fusion. Background Art

[0002] For an ETC (Electronic Toll Collection) lane controller, anomaly detection technology is crucial. It can help the system monitor and identify possible faults or anomalies in real time, thereby preventing toll collection errors or traffic congestion caused by equipment problems. Although current anomaly detection technologies have been applied in ETC lane controllers, there are still many limitations. Among them, most current ETC lane controllers rely on inductive loop pulse counting to determine whether there is a vehicle on the lane. When the induction value deviates by more than 15% due to the interference of metal objects or the aging of the inductive loop, the system may misjudge that there is a vehicle present, resulting in false toll collection or transactions. The existing system cannot effectively distinguish between the breakage of the inductive loop and the problem of a vehicle crossing lanes, resulting in more than 90% of the maintenance work orders being false alarms, increasing the unnecessary maintenance burden. Severe weather is likely to cause the inductive loop to short-circuit, which will directly lead to system paralysis, and the existing system lacks a degradation processing scheme for dealing with these anomalies, reducing the reliability of the system.

[0003] In addition, the current toll controller status monitoring system uses a polling method every 10 seconds. However, when a vehicle passes through a 4-meter detection area at a speed of 30 km / h, the processing window of the system is only 480 milliseconds, and the polling cycle of every 10 seconds causes 12.5% of the anomaly events to be missed, resulting in insufficient real-time performance and affecting the effect of anomaly detection. When a controller anomaly is detected, the system requires manual on-site intervention for reset operation, resulting in an average fault recovery time of 47 minutes. This lagging repair method causes a decrease in the lane's passing capacity and obvious traffic congestion. Moreover, there are an average of 3.2 false interception events per month, increasing the complexity of traffic management; the fault repair speed is slow, and the lane availability is only 92.7%, resulting in an impact on the normal operation of the lane; the existing equipment lacks the ability of adaptive fault handling, resulting in a 37% increase in the equipment's maintenance cost.

[0004] Therefore, it is necessary to design a new method to improve the accuracy of fault detection, reduce repair delays, and enhance the overall reliability of the system. Summary of the Invention

[0005] The purpose of the present invention is to overcome the defects of the prior art and provide a method and system for real-time detection of lane controller anomalies based on multi-modal data fusion.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A real-time detection method for anomalies of a lane controller based on multimodal data fusion, comprising:

[0007] Obtain the ground loop signal, DSRC communication status, and transaction flow data to obtain multimodal data;

[0008] Perform feature analysis on the multimodal data to obtain an analysis result;

[0009] Input the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classify according to different abnormal types to obtain the corresponding abnormal type;

[0010] Execute the corresponding repair strategy according to the abnormal type to obtain a repair result;

[0011] Confirm the repair result through a self-check method, ensure that vehicles are allowed to pass after the anomaly is resolved, and update the health status library.

[0012] A further technical solution thereof is: The performing feature analysis on the multimodal data to obtain an analysis result includes:

[0013] Perform fast Fourier transform analysis on the ground loop signal to extract frequency and amplitude characteristics;

[0014] Calculate the bit error rate of the DSRC communication link corresponding to the DSRC communication status;

[0015] Verify the timestamp and amount in the transaction flow data to obtain a verification result;

[0016] Wherein, the analysis result includes frequency and amplitude characteristics, bit error rate, and verification result.

[0017] A further technical solution thereof is: The inputting the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classify according to different abnormal types to obtain the corresponding abnormal type includes:

[0018] Input the analysis result into a decision tree model, and the decision tree model combines with the status code table to determine whether the current state of the lane controller is abnormal;

[0019] When the current state of the lane controller is abnormal, classify according to different abnormal types to obtain the corresponding abnormal type.

[0020] A further technical solution thereof is: The abnormal type includes at least one of sensor failure, communication timeout, and data anomaly.

[0021] Its further technical solution is: performing a corresponding repair strategy according to the abnormal type to obtain a repair result, including:

[0022] Determining a corresponding repair strategy from a preset security mechanism table according to the abnormal type, and executing the corresponding repair strategy to obtain a repair result.

[0023] Its further technical solution is: the repair strategy includes:

[0024] When a sensor failure is detected, immediately switch to the standby inductive loop to ensure uninterrupted signal acquisition;

[0025] In the case of communication timeout, dynamically adjust the transmission power of the DSRC module and enable the 4G standby channel to ensure smooth communication;

[0026] When abnormal transaction data is detected, call the compensation mechanism, store the abnormal transaction record and generate audit evidence.

[0027] Its further technical solution is: confirming the repair result through a self-checking method to ensure that passing vehicles are allowed after the abnormality is resolved, and updating the health status library, including:

[0028] Confirm the repair result through a self-checking method to ensure that passing vehicles are allowed after the abnormality is resolved, and update the health status library. When there is still an abnormality in the lane controller, trigger an audible and visual alarm.

[0029] Its further technical solution is: confirming the repair result through a self-checking method to ensure that passing vehicles are allowed after the abnormality is resolved, and updating the health status library, including:

[0030] Confirm the repair result through a self-checking method to ensure that passing vehicles are allowed after the abnormality is resolved, and update the current status and repair record to the health status library.

[0031] The present invention also provides a lane controller abnormal real-time detection system based on multi-modal data fusion, including:

[0032] A data acquisition unit for acquiring inductive loop signals, DSRC communication status, and transaction flow data to obtain multi-modal data;

[0033] A feature analysis unit for performing feature analysis on the multi-modal data to obtain an analysis result;

[0034] An abnormal classification unit for inputting the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classifying according to different abnormal types to obtain corresponding abnormal types;

[0035] A repair unit, configured to execute corresponding repair strategies according to the exception type to obtain a repair result;

[0036] An update unit, configured to confirm the repair result through a self-check method, allow passing vehicles after ensuring that the exception is resolved, and update the health status library.

[0037] Its further technical solution is that: the feature analysis unit includes:

[0038] An extraction subunit, configured to perform fast Fourier transform analysis on the ground loop signal to extract frequency and amplitude features;

[0039] An error rate calculation subunit, configured to calculate the error rate of the DSRC communication link corresponding to the DSRC communication status;

[0040] A verification subunit, configured to verify the timestamp and amount in the transaction flow data to obtain a verification result.

[0041] The beneficial effects of the present invention compared with the prior art are as follows: through the multi-modal data fusion technology, the present invention comprehensively combines the ground loop signal, DSRC communication status and transaction flow data to achieve comprehensive monitoring of the lane controller; in-depth feature analysis of these data helps to improve the accuracy of anomaly detection, so as to quickly identify potential faults; the decision tree model can adopt the most appropriate repair strategy for different faults by judging the exception type, reducing misjudgment and repair delay; the repair operation is automatically executed and the effect is confirmed through a self-check method to ensure that the fault is resolved in a timely and effective manner; the updated health status library provides health data for the long-term operation of the system, which helps to enhance the reliability and stability of the overall system.

[0042] The present invention will be further described below with reference to the accompanying drawings and specific embodiments. Description of the Drawings

[0043] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0044] Figure 1 It is a schematic diagram of the application scenario of the method for real-time anomaly detection of a lane controller based on multi-modal data fusion provided by an embodiment of the present invention;

[0045] Figure 2 It is a schematic diagram of the flow of the method for real-time anomaly detection of a lane controller based on multi-modal data fusion provided by an embodiment of the present invention;

[0046] Figure 3 Schematic diagram of a sub - process of the method for real - time detection of lane controller anomalies based on multi - modal data fusion provided by an embodiment of the present invention;

[0047] Figure 4 Schematic diagram of a sub - process of the method for real - time detection of lane controller anomalies based on multi - modal data fusion provided by an embodiment of the present invention;

[0048] Figure 5 Schematic block diagram of a system for real - time detection of lane controller anomalies based on multi - modal data fusion provided by an embodiment of the present invention;

[0049] Figure 6 Schematic block diagram of a feature analysis unit of a system for real - time detection of lane controller anomalies based on multi - modal data fusion provided by an embodiment of the present invention;

[0050] Figure 7 Schematic block diagram of an anomaly classification unit of a system for real - time detection of lane controller anomalies based on multi - modal data fusion provided by an embodiment of the present invention;

[0051] Figure 8 Schematic block diagram of a computer device provided by an embodiment of the present invention. Detailed implementation manners

[0052] 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 some, but not all, of the embodiments of the present invention. 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.

[0053] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprise" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0054] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in this specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms.

[0055] It should be further understood that the term "and / or" used in this specification of the present invention and the appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0056] Please refer to Figure 1 and Figure 2 , Figure 1 which is a schematic diagram of the application scenario of the real-time detection method for lane controller anomalies based on multi-modal data fusion provided by an embodiment of the present invention. Figure 2 which is a schematic flowchart of the real-time detection method for lane controller anomalies based on multi-modal data fusion provided by an embodiment of the present invention. This real-time detection method for lane controller anomalies based on multi-modal data fusion is applied to the station-level server, two lane controllers, and the core switch involved in the toll station. Among them, the core switch is connected to the networking center. One of the lane controllers respectively conducts data interaction with the fee display / barrier machine, the inductive loop, and the RSU, and the other lane controller conducts data interaction with the edge computing layer. The entire method realizes improving the accuracy of fault detection, reducing repair delays, and enhancing the overall reliability of the system.

[0057] Specifically, as Figure 1 in the described architecture, the networking center is connected to the security gateway of the toll station through an IPSec VPN to form a top-level data center. The core switch and the station-level server inside the toll station ensure the reliability of the system through dual-fiber redundancy, while the lane controllers are connected to the core switch through PoE and respectively execute ETC lane control and edge computing tasks.

[0058] The lane controller 1 in the ETC lane part is connected to the RSU through RS485, controls the fee display and the barrier machine through the CAN bus, and accesses the inductive loop through the GPIO interface. The RSU communicates with the on-vehicle unit OBU through the DSRC protocol. The lane controller 2 in the edge computing part communicates with the edge node through the OPC UA protocol, and the edge node then uploads the data to the cloud through the MQTT protocol.

[0059] The lane controller 1 obtains the inductive loop signal in real time through the GPIO interface. The DSRC communication data between the RSU and the OBU is also collected and incorporated into the data input as the source of multi-source data. Through the core switch, the collected data is transmitted to the station-level server for real-time feature analysis to monitor the vehicle status. The anomaly determination is based on the data collected from the lane controller and the edge computing layer to ensure the timeliness and accuracy of the decision-making. When the system detects an anomaly, the controller 1 can control the barrier machine and the fee display through the CAN bus for repair operations to ensure a quick response. Through the controller 2 in the edge computing layer, the system can upload the repaired status information to the cloud to form a complete status monitoring closed loop. This process ensures the efficient operation of the system and the rapid repair of anomalies.

[0060] In this embodiment, the data transmission protocol between the ETC lane controller and the data center aims to ensure the integrity, correctness, and security of data transmission between the ETC (Electronic Toll Collection System) lane controller and the data center. By defining a detailed protocol frame structure, command words, information field data structure, and security mechanism, this protocol can effectively support rapid exception identification and handling, ensure the normal operation of the lane, and prevent the tampering or leakage of transaction data during transmission. It includes the content of the following five tables respectively.

[0061] Table 1. Protocol Frame Structure Table

[0062]

[0063]

[0064] Table 2. Command Word Definition Table

[0065]

[0066] Table 3. Information Field Data Structure Table (Transaction Data)

[0067]

[0068]

[0069] Table 4. Security Mechanism Table

[0070] Security Level Implementation Method Verification Mechanism StatusCode Range Link Layer SM4 Encryption CRC32 Verification 0x00 - 0x7F Application Layer MAC Verification Transaction Serial Number Continuity Check 0x80 - 0xCF Business Layer Amount Second Confirmation License Plate - Vehicle Model Matching Verification 0xD0 - 0xFF

[0071] Table 5. Abnormal Status Code Table

[0072] StatusCode Name Processing Suggestion 0xE1 CRC Verification Failed Check Physical Connection and Resend Data Packet 0xE2 Instruction Timeout Check Network Latency and Retry No More Than 3 Times 0xE3 Hardware Failure Switch to Backup Device and Send Maintenance Alert 0xE4 Data Illegal Verify Data Source Legitimacy and Record Audit Log

[0073] This protocol design takes into account the rapid response and handling mechanisms for various abnormal situations, and at the same time ensures the security of data transmission through multi-level security measures, thus providing a solid technical foundation for the stable operation of the ETC system.

[0074] Figure 2 It is a schematic flow diagram of the method for real-time detection of lane controller anomalies based on multi-modal data fusion provided by the embodiments of the present invention. As Figure 2 shown, this method includes the following steps S110 to S150.

[0075] S110. Obtain the inductive loop signal, DSRC communication status, and transaction flow data to obtain multi-modal data.

[0076] In this embodiment, the multi-modal data includes the inductive loop signal, DSRC communication status, and transaction flow data.

[0077] Specifically, the system is initialized to load necessary driver programs and configuration files to ensure the normal operation of each module. Then, data acquisition is carried out. Among them, the inductive loop detects the presence and movement of vehicles through an induction coil buried on the road. It utilizes the principle of electromagnetic induction. When a vehicle passes by, it will affect the magnetic field around the coil, resulting in a change in the current in the coil. The passing of the vehicle is detected through high-frequency pulse signals, and an ADC (Analog-to-Digital Converter) is used to convert the analog signal into a digital signal. This process is usually sampled at a frequency of 100Hz to accurately capture the time point when the vehicle passes.

[0078] Perform a fast Fourier transform (FFT) on the inductive loop signal to extract its frequency and amplitude characteristics to determine the validity of the signal. This helps to distinguish the actual passing of a vehicle from abnormal signals caused by environmental interference or equipment failures.

[0079] DSRC (Dedicated Short Range Communication) is a technical standard for two-way communication between vehicles and the road, mainly used to achieve automatic vehicle identification and transaction information transmission at high speeds. Its status includes parameters such as Received Signal Strength Indicator (RSSI) and Bit Error Rate (BER).

[0080] Monitor the communication quality with vehicles in real time and record the RSSI and BER values. These indicators can reflect the quality status of the current communication link. Calculate the bit error rate of DSRC communication to evaluate the stability of the communication link. If the bit error rate exceeds the set threshold, it is marked as abnormal, indicating possible communication problems, such as signal attenuation caused by bad weather conditions.

[0081] Transaction flow data refers to all relevant information generated each time a vehicle passes through the ETC lane, including key fields such as timestamp, license plate number, and transaction amount. These data are crucial for verifying the integrity and accuracy of transactions. Extract the latest transaction records from the transaction database to ensure data integrity and consistency. Each record contains detailed transaction information for subsequent analysis and auditing. Perform timestamp and amount verification on the transaction flow data to ensure that there are no duplicate or abnormal transactions. In addition, a checksum (such as crypto_mac) is generated, and data security is ensured through encryption algorithms (such as SM4).

[0082] By integrating three different types of raw data: the ground loop coil waveform, the DSRC bit error rate, and the transaction log, potential faults in the system can be more accurately located, increasing the accuracy of fault diagnosis to 99.7%, far higher than the 85% of traditional solutions. At the same time, the comprehensive analysis of this multi-source data also supports non-disruptive hot repair operations, enabling necessary switching or adjustment to be completed in a very short time in case of hardware failures or communication timeouts, significantly improving the reliability and efficiency of the system. For example, when a sensor failure is detected, the system can immediately switch to a standby ground loop coil to ensure uninterrupted signal acquisition; while in case of a communication timeout, communication can be maintained by dynamically adjusting the transmission power of the RSU module or enabling a standby channel.

[0083] S120. Perform feature analysis on the multi-modal data to obtain an analysis result.

[0084] In this embodiment, the analysis result includes frequency and amplitude characteristics, bit error rate, and verification result.

[0085] The analysis result refers to a comprehensive evaluation based on the following three aspects:

[0086] Frequency and amplitude characteristics: Provide an in-depth understanding of the physical characteristics when a vehicle passes through the ground loop coil, and are the key basis for distinguishing different vehicle types and judging whether there is environmental interference.

[0087] Bit error rate: Reflects the actual working condition of the DSRC communication link, helps to detect and solve potential communication problems in a timely manner, and maintains the efficient operation of the system.

[0088] Verification result: Ensures the authenticity and reliability of the transaction log data, providing a solid guarantee for subsequent financial settlement and user services.

[0089] In one embodiment, please refer to Figure 3 , the above step S120 may include steps S121 to S123.

[0090] S121. Perform a fast Fourier transform analysis on the ground loop coil signal to extract frequency and amplitude characteristics.

[0091] In this embodiment, by applying FFT analysis to the ground loop coil signal, the signal can be transformed from the time domain to the frequency domain, and the main frequency components of the signal and their corresponding amplitudes can be extracted. This helps to identify the specific patterns and intensities generated when a vehicle passes through the ground loop coil, thereby judging the validity of the signal. For example, when a vehicle passes by, a series of waveforms with specific frequencies will be generated, and their amplitudes reflect information such as the weight or type of the vehicle.

[0092] S122. Calculate the bit error rate of the DSRC communication link corresponding to the DSRC communication status.

[0093] In this embodiment, the bit error rate of the DSRC communication link refers to the proportion of errors occurring during data transmission. It measures the quality of the communication link. Especially in rainy, foggy weather or other interference environments, the bit error rate may increase significantly. By calculating and monitoring this indicator, the stability and reliability of the current communication link can be evaluated, and corresponding compensation measures can be taken accordingly, such as dynamically adjusting the transmission power or switching to a backup channel.

[0094] S123. Verify the timestamp and amount in the transaction flow data to obtain a verification result;

[0095] In this embodiment, this step aims to verify the integrity and accuracy of the transaction flow data. The specific operations include:

[0096] Timestamp verification: Ensure that each transaction record is attached with an accurate time stamp for easy tracking and auditing.

[0097] Amount verification: Check whether the transaction amount is reasonable and consistent to prevent duplicate deductions or abnormal amounts.

[0098] Generate a verification code: Use an encryption algorithm (such as SM4) to generate a verification code (crypto_mac) for the transaction data to ensure the security and integrity of the data during transmission.

[0099] By integrating these analysis results, the system can perform a more accurate comprehensive determination of anomalies, and then adopt appropriate repair strategies, such as switching to a backup device, increasing the transmission power, or performing flow compensation storage, etc., and finally achieve an efficient non-interruptive hot repair operation. This multi-level data analysis and processing mechanism greatly improves the robustness and response speed of the entire ETC lane control system.

[0100] S130. Input the analysis results into a decision tree model to determine whether the current state of the lane controller is abnormal, and classify according to different abnormal types to obtain the corresponding abnormal types.

[0101] In this embodiment, the abnormal types include at least one of sensor failure, communication timeout, and data anomaly.

[0102] Specifically, sensor failure refers to the situation where the inductive loop or other sensors used for vehicle detection are physically damaged or their performance deteriorates. This includes but is not limited to the following situations:

[0103] Signal loss: The sensor cannot generate the expected pulse signal, resulting in the inability to correctly detect the presence of a vehicle.

[0104] False alarm: The sensor misidentifies a non-vehicle object as a vehicle, or does not respond to an actual passing vehicle.

[0105] Waveform distortion: Due to reasons such as environmental interference or equipment aging, the signal waveform generated by the sensor is significantly different from the normal situation, affecting the accuracy of subsequent processing.

[0106] Communication timeout means that the DSRC module fails to complete the information exchange with the vehicle or other devices within the specified time. This situation is caused by the following reasons:

[0107] Insufficient signal strength: If the RSSI value is too low, it indicates that the signal received by the receiving end is too weak, which may lead to data transmission failure.

[0108] Channel congestion: Too much data attempts to use the same wireless channel simultaneously, resulting in packet collisions or losses.

[0109] Hardware problems: Such as insufficient transmit power, damaged antennas, etc., may all lead to a decline in communication quality and thus cause timeouts.

[0110] Data anomalies involve the integrity and consistency of transaction flow data. The specific manifestations include:

[0111] Timestamp mismatch: That is, the time stamp in the transaction record does not match other system logs or external reference time sources, suggesting possible tampering or synchronization errors.

[0112] Amount error: The transaction amount does not match the actual situation, which may be caused by software bugs, malicious attacks, or system configuration errors.

[0113] Duplicate transactions: The same transaction is recorded multiple times, usually due to the failure of the confirmation mechanism or retransmission operations caused by network delays.

[0114] In one embodiment, please refer to Figure 4 , the above step S130 may include steps S131 to S132.

[0115] S131. Input the analysis result into the decision tree model, and the decision tree model combines the status code table to determine whether the current state of the lane controller is abnormal;

[0116] S132. When the current state of the lane controller is abnormal, classify it according to different abnormal types to obtain the corresponding abnormal types.

[0117] The analysis results obtained from step S130 - including the waveform characteristics of the inductive loop, the bit error rate of DSRC communication, and the integrity verification results of transaction flow data, etc. - are used as input parameters and fed into the decision tree model. These analysis results represent the working states and performance indicators of each key module of the system.

[0118] The decision tree model will refer to a predefined status code table to evaluate whether the input data is normal. For example, if the error rate of the inductive loop signal exceeds a certain threshold, it may correspond to a specific code in the status code table, indicating a potential fault. By comparing the input data with the patterns in known normal situations, the decision tree can identify which aspects deviate from the expected behavior and thus mark them as anomalies.

[0119] Through the feature analysis of the above three main types of anomalies (such as waveform feature extraction, error rate analysis, and transaction integrity verification), a series of quantitative indicators can be obtained. These indicators will then be input into a pre-trained decision tree model, which uses the status code table as a reference to evaluate whether the current system status is normal. Once an anomaly is determined, the model will further distinguish the specific type of anomaly according to the pattern of the input features, so as to guide the selection of subsequent repair strategies. For example, if it is determined that there is a sensor failure, the standby coil will be activated; if it is a communication timeout, the transmission power will be adjusted or the standby communication channel will be enabled, etc.

[0120] S140. Execute the corresponding repair strategy according to the anomaly type to obtain a repair result.

[0121] In this embodiment, the repair result refers to taking corresponding repair measures for different types of detected anomalies and finally obtaining a repair result. This repair result means whether the system status has returned to normal or at least reached an acceptable state after executing the specific repair strategy.

[0122] Specifically, determine the corresponding repair strategy from a preset security mechanism table according to the anomaly type, and execute the corresponding repair strategy to obtain a repair result.

[0123] Among them, when a sensor failure is detected, immediately switch to the standby inductive loop to ensure that the signal acquisition is not interrupted;

[0124] In the case of communication timeout, dynamically adjust the transmission power of the DSRC module and enable the 4G standby channel to ensure smooth communication;

[0125] When abnormal transaction data is detected, call the compensation mechanism, store the abnormal transaction record and generate audit evidence.

[0126] Specifically, when the system detects a failure of sensors such as inductive loops, it will immediately switch to the standby sensor (such as the standby inductive loop) to ensure that the signal acquisition process is not interrupted. The repair result here means that the system has successfully switched to the standby sensor, and the new sensor can work normally and the collected data is accurate.

[0127] If a communication timeout problem is detected in the DSRC module, the system will dynamically adjust its transmission power and may enable the 4G backup channel to ensure the stability of data transmission. The repair result refers to the improvement of communication quality after these adjustments, with the error rate reduced to an acceptable range and the communication restored to normal or reaching a stable state.

[0128] When abnormal transaction data is found (such as inconsistent timestamps or amounts), the system will initiate a compensation mechanism, save the abnormal transaction records, and generate audit evidence for subsequent review and correction. The repair result here means that the system has securely stored all abnormal transaction information and provided the necessary audit clues for resolving these abnormalities, ensuring data integrity and consistency.

[0129] In summary, the repair result is a confirmation of the effectiveness of the adopted repair strategy. It indicates whether the system can recover from an abnormal state to a normal operating state or at least maintain a state that does not affect the overall function and service level. In addition, such a result usually requires certain verification steps (such as self-check, re-acquisition and analysis of data, etc.) to ensure that the repair measures are indeed effective and the operating state of the system is reliable.

[0130] S150. Confirm the repair result through self-check to ensure that vehicles are allowed to pass after the abnormality is resolved, and update the health status library.

[0131] In this embodiment, the repair result is confirmed through self-check to ensure that vehicles are allowed to pass after the abnormality is resolved, and update the health status library. When there is still an abnormality in the lane controller, an audible and visual alarm is triggered.

[0132] Specifically, the repair result is confirmed through self-check to ensure that vehicles are allowed to pass after the abnormality is resolved, and the current status and repair records are updated to the health status library.

[0133] Specifically, first, the system will restart data acquisition and collect the latest data from the inductive loop, DSRC communication module, and transaction database again.

[0134] Then, the newly acquired data is analyzed, including waveform feature extraction (for inductive loop signals), error rate analysis (for DSRC communication), and transaction integrity verification (checking transaction flow data). This step aims to verify whether there are still previous abnormal problems in the repaired system.

[0135] The above features are input into an improved decision tree model to determine whether the current state has returned to normal. If the model output result indicates that no abnormality is detected, the repair is considered successful; otherwise, if an abnormality is still found, further processing is required.

[0136] When the self-test results show that all modules have resumed normal operation, the system will remove the alarm state and allow subsequent vehicles to pass smoothly. At this time, the system will also record the results of the repair (i.e. the state before the repair, the repair measures taken and the final repair results).

[0137] If problems that cannot be ignored are still found during the self-inspection process or the expected repair results are not achieved, the system will trigger the sound and light alarm mechanism to prompt on-site maintenance personnel to conduct manual intervention and inspection to prevent potential risks from affecting the normal operation of the ETC lane.

[0138] Regardless of whether the repair is successful or not, the system will update the relevant information to the health status library:

[0139] Status update: contains the latest system operation status, such as which components have been repaired and whether there are any remaining issues.

[0140] Repair record preservation: record in detail the time, type, repair method used and its effectiveness of each fault, and provide a reference for subsequent maintenance work.

[0141] Generation of optimization suggestions: Based on the accumulated historical data, the system can analyze common failure modes and high-incidence periods, and then make preventive maintenance suggestions to help reduce future failure rates and operation and maintenance costs.

[0142] In short, this process not only ensures the stability and reliability of the system, but also provides data support for continuous improvement, helping to improve the performance and service level of the entire ETC system.

[0143] The method of this embodiment integrates three data sources: ground sensor coil waveform, DSRC bit error rate and transaction flow, and the fault location accuracy is greatly improved to 99.7%, far exceeding the 85% of the traditional solution. The hot switching technology can complete the seamless switching of coils or communication modules within 50 milliseconds, which is 10 times more efficient than the 500 millisecond restart time required by the traditional method. Based on the improved PID algorithm, the real-time and precise adjustment of the RSU (roadside unit) transmission power is achieved (±0.5dB accuracy), ensuring a stable communication distance of 30±0.5 meters.

[0144] The method of this embodiment solves several problems that have long plagued the industry:

[0145] Previous technologies had difficulty distinguishing the waveform feature differences between vehicle occlusion and equipment failure, resulting in inaccurate diagnosis. In severe weather conditions, the existing system cannot effectively deal with the problem of a sharp increase in bit error rate and lacks the necessary adaptive compensation mechanism. The trading system, sensor system, and communication system operate independently, and lack effective fault correlation analysis tools, which affects the stability and efficiency of the overall system.

[0146] This embodiment combines a variety of advanced technologies and conducts in-depth integrated optimization. By reducing on-site maintenance requirements, it is expected to significantly cut operating costs. The fault recovery time has been shortened from the minute level to the second level, increasing the throughput of the ETC lane by approximately 15%, greatly enhancing the traffic flow management ability and user experience.

[0147] In summary, this embodiment not only solves several key technical problems in the industry, but also its unique solution has extremely high commercial value and broad application prospects. By transforming these technological innovations into actual products and services, it is expected to trigger a major revolution in the field of intelligent transportation and bring significant competitive advantages to related enterprises.

[0148] The above real-time abnormal detection method for lane controllers based on multi-modal data fusion comprehensively monitors the lane controller by integrating inductive loop signals, DSRC communication status, and transaction flow data through multi-modal data fusion technology; in-depth feature analysis of these data helps improve the accuracy of abnormal detection, thus quickly identifying potential faults; the decision tree model can adopt the most appropriate repair strategy for different faults by judging the type of abnormality, reducing misjudgment and repair delay; the repair operation is automatically executed and the effect is confirmed through self-checking to ensure that the fault is resolved in a timely and effective manner; the updated health status library provides the health data for the long-term operation of the system, helping to enhance the reliability and stability of the overall system.

[0149] Figure 5 It is a schematic block diagram of a real-time abnormal detection system 300 for lane controllers based on multi-modal data fusion provided by an embodiment of the present invention. As Figure 5 shown, corresponding to the above real-time abnormal detection method for lane controllers based on multi-modal data fusion, the present invention also provides a real-time abnormal detection system 300 for lane controllers based on multi-modal data fusion. The real-time abnormal detection system 300 for lane controllers based on multi-modal data fusion includes units for executing the above real-time abnormal detection method for lane controllers based on multi-modal data fusion, and this system can be configured in a controller. Specifically, please refer to Figure 5 , the real-time abnormal detection system 300 for lane controllers based on multi-modal data fusion includes a data acquisition unit 301, a feature analysis unit 302, an abnormal classification unit 303, a repair unit 304, and an update unit 305.

[0150] A data acquisition unit 301 is configured to acquire a ground loop signal, a DSRC communication status, and transaction flow data to obtain multimodal data; a feature analysis unit 302 is configured to perform feature analysis on the multimodal data to obtain an analysis result; an anomaly classification unit 303 is configured to input the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal and classify it according to different anomaly types to obtain corresponding anomaly types; a repair unit 304 is configured to execute corresponding repair strategies according to the anomaly types to obtain a repair result; an update unit 305 is configured to confirm the repair result through a self-checking method, ensure that vehicles are allowed to pass after the anomaly is resolved, and update the health status library.

[0151] In one embodiment, as Figure 6 shown, the feature analysis unit 302 includes an extraction subunit 3021, an error rate calculation subunit 3022, and a verification subunit 3023.

[0152] The extraction subunit 3021 is configured to perform fast Fourier transform analysis on the ground loop signal to extract frequency and amplitude features; the error rate calculation subunit 3022 is configured to calculate the error rate of the DSRC communication link corresponding to the DSRC communication status; the verification subunit 3023 is configured to verify the timestamp and amount in the transaction flow data to obtain a verification result.

[0153] In one embodiment, as Figure 7 shown, the anomaly classification unit 303 includes an anomaly judgment subunit 3031 and a classification subunit 3032.

[0154] The anomaly judgment subunit 3031 is configured to input the analysis result into a decision tree model, and the decision tree model combines a status code table to determine whether the current state of the lane controller is abnormal; the classification subunit 3032 is configured to classify according to different anomaly types when the current state of the lane controller is abnormal to obtain corresponding anomaly types.

[0155] In one embodiment, the repair unit 304 is configured to determine a corresponding repair strategy from a preset security mechanism table according to the anomaly type and execute the corresponding repair strategy to obtain a repair result.

[0156] In one embodiment, the update unit 305 is configured to confirm the repair result through a self-checking method, ensure that vehicles are allowed to pass after the anomaly is resolved, and update the health status library. When there is still an anomaly in the lane controller, an audible and visual alarm is triggered.

[0157] In one embodiment, the update unit 305 is configured to confirm the repair result through a self-checking method, ensure that vehicles are allowed to pass after the anomaly is resolved, and update the current state and repair record to the health status library.

[0158] It should be noted that those skilled in the art can clearly understand that the specific implementation processes of the above-mentioned real-time lane controller anomaly detection system 300 based on multi-modal data fusion and each unit can refer to the corresponding descriptions in the foregoing method embodiments. For the sake of convenience and brevity of description, they will not be elaborated herein.

[0159] The above-mentioned real-time lane controller anomaly detection system 300 based on multi-modal data fusion can be implemented in the form of a computer program, and this computer program can run on a computer device as shown in Figure 8 the following.

[0160] Please refer to Figure 8 , Figure 8 which is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device 500 can be a server. Among them, the server can be an independent server or a server cluster composed of multiple servers.

[0161] Referring to Figure 8 , the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501. Among them, the memory can include a non-volatile storage medium 503 and an internal memory 504.

[0162] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, and when these program instructions are executed, the processor 502 can be made to execute a real-time lane controller anomaly detection method based on multi-modal data fusion.

[0163] The processor 502 is used to provide computing and control capabilities to support the operation of the entire computer device 500.

[0164] The internal memory 504 provides an environment for the operation of the computer program 5032 in the non-volatile storage medium 503. When the computer program 5032 is executed by the processor 502, the processor 502 can be made to execute a real-time lane controller anomaly detection method based on multi-modal data fusion.

[0165] The network interface 505 is used for network communication with other devices. Those skilled in the art can understand that Figure 8 the structure shown in

[0166] Among them, the processor 502 is used to run the computer program 5032 stored in the memory to implement the following steps:

[0167] Obtain the ground loop signal, DSRC communication status, and transaction flow data to obtain multi-modal data; perform feature analysis on the multi-modal data to obtain an analysis result; input the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classify according to different abnormal types to obtain the corresponding abnormal type; execute the corresponding repair strategy according to the abnormal type to obtain a repair result; confirm the repair result through a self-check method, ensure that vehicles are allowed to pass after the abnormality is resolved, and update the health status library.

[0168] In one embodiment, when the processor 502 implements the step of performing feature analysis on the multi-modal data to obtain an analysis result, the specific implementation is as follows:

[0169] Perform fast Fourier transform analysis on the ground loop signal to extract frequency and amplitude features; calculate the bit error rate of the DSRC communication link corresponding to the DSRC communication status; verify the timestamp and amount in the transaction flow data to obtain a verification result;

[0170] Among them, the analysis result includes frequency and amplitude features, bit error rate, and verification result.

[0171] In one embodiment, when the processor 502 implements the step of inputting the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classifying according to different abnormal types to obtain the corresponding abnormal type, the specific implementation is as follows:

[0172] Input the analysis result into a decision tree model, and the decision tree model combines the status code table to determine whether the current state of the lane controller is abnormal; when the current state of the lane controller is abnormal, classify according to different abnormal types to obtain the corresponding abnormal type.

[0173] Among them, the abnormal type includes at least one of sensor failure, communication timeout, and data abnormality.

[0174] In one embodiment, when the processor 502 implements the step of executing the corresponding repair strategy according to the abnormal type to obtain a repair result, the specific implementation is as follows:

[0175] Determine the corresponding repair strategy from a preset security mechanism table according to the abnormal type, and execute the corresponding repair strategy to obtain a repair result.

[0176] Among them, the repair strategy includes:

[0177] When a sensor failure is detected, immediately switch to the backup inductive loop to ensure uninterrupted signal acquisition;

[0178] In the case of communication timeout, dynamically adjust the transmission power of the DSRC module and enable the 4G backup channel to ensure smooth communication;

[0179] When abnormal transaction data is detected, call the compensation mechanism, store the abnormal transaction record and generate audit evidence.

[0180] In one embodiment, when the processor 502 implements the step of confirming the repair result by self-checking, ensuring that passing vehicles are allowed after the abnormality is resolved, and updating the health status library, the following steps are specifically implemented:

[0181] Confirm the repair result by self-checking, ensure that passing vehicles are allowed after the abnormality is resolved, and update the health status library. When there is still an abnormality in the lane controller, trigger an audible and visual alarm.

[0182] In one embodiment, when the processor 502 implements the step of confirming the repair result by self-checking, ensuring that passing vehicles are allowed after the abnormality is resolved, and updating the health status library, the following steps are specifically implemented:

[0183] Confirm the repair result by self-checking, ensure that passing vehicles are allowed after the abnormality is resolved, and update the current status and repair record to the health status library.

[0184] It should be understood that in the embodiments of the present application, the processor 502 may be a central processing unit (CPU), and the processor 502 may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0185] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program includes program instructions, and the computer program can be stored in a storage medium, and the storage medium is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the embodiments of the above methods.

[0186] Therefore, the present invention also provides a storage medium. The storage medium may be a computer-readable storage medium. The storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following steps:

[0187] Obtain a ground loop signal, a DSRC communication status, and transaction flow data to obtain multimodal data; perform feature analysis on the multimodal data to obtain an analysis result; input the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classify according to different abnormal types to obtain corresponding abnormal types; execute corresponding repair strategies according to the abnormal types to obtain repair results; confirm the repair results through a self-check method, ensure that vehicles are allowed to pass after the abnormality is resolved, and update the health status library.

[0188] In one embodiment, when the processor executes the computer program to implement the step of performing feature analysis on the multimodal data to obtain an analysis result, the following steps are specifically implemented:

[0189] Perform fast Fourier transform analysis on the ground loop signal to extract frequency and amplitude characteristics; calculate the bit error rate of the DSRC communication link corresponding to the DSRC communication status; verify the timestamp and amount in the transaction flow data to obtain a verification result; wherein, the analysis result includes frequency and amplitude characteristics, bit error rate, and verification result.

[0190] In one embodiment, when the processor executes the computer program to implement the step of inputting the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classifying according to different abnormal types to obtain corresponding abnormal types, the following steps are specifically implemented:

[0191] Input the analysis result into a decision tree model, and the decision tree model combines with a status code table to determine whether the current state of the lane controller is abnormal; when the current state of the lane controller is abnormal, classify according to different abnormal types to obtain corresponding abnormal types.

[0192] Wherein, the abnormal types include at least one of sensor failure, communication timeout, and data abnormality.

[0193] In one embodiment, when the processor executes the computer program to implement the step of executing corresponding repair strategies according to the abnormal types to obtain repair results, the following steps are specifically implemented:

[0194] Determine corresponding repair strategies from a preset security mechanism table according to the abnormal types, and execute the corresponding repair strategies to obtain repair results.

[0195] Among them, the repair strategy includes:

[0196] When a sensor failure is detected, immediately switch to the backup inductive loop to ensure uninterrupted signal acquisition;

[0197] In the case of communication timeout, dynamically adjust the transmission power of the DSRC module and enable the 4G backup channel to ensure smooth communication;

[0198] When abnormal transaction data is detected, call the compensation mechanism, store the abnormal transaction record and generate audit evidence.

[0199] In one embodiment, when the processor executes the computer program to implement the step of confirming the repair result by self-checking, ensuring that passing vehicles are allowed after the abnormality is resolved, and updating the health status library, the following steps are specifically implemented:

[0200] Confirm the repair result by self-checking, ensure that passing vehicles are allowed after the abnormality is resolved, and update the health status library. When there is still an abnormality in the lane controller, trigger an audible and visual alarm.

[0201] In one embodiment, when the processor executes the computer program to implement the step of confirming the repair result by self-checking, ensuring that passing vehicles are allowed after the abnormality is resolved, and updating the health status library, the following steps are specifically implemented:

[0202] Confirm the repair result by self-checking, ensure that passing vehicles are allowed after the abnormality is resolved, and update the current status and repair record to the health status library.

[0203] The storage medium can be various computer-readable storage media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a magnetic disk, or an optical disc that can store program codes.

[0204] Those of ordinary skill in the art can realize that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been generally described according to their functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0205] In several embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of each unit is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0206] The steps in the method embodiments of the present invention can be adjusted, combined, and deleted according to actual needs. The units in the system embodiments of the present invention can be combined, divided, and deleted according to actual needs. In addition, the functional units in each embodiment of the present invention can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0207] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention.

[0208] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of various equivalent modifications or substitutions, and these modifications or substitutions should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A real-time lane controller anomaly detection method based on multimodal data fusion, characterized in that: include: Obtain ground sensor coil signals, DSRC communication status, and transaction flow data to obtain multi-modal data; Performing feature analysis on the multimodal data to obtain analysis results; Input the analysis result into the decision tree model to determine whether the current state of the lane controller is abnormal, and classify it according to different abnormal types to obtain the corresponding abnormal type; Execute a corresponding repair strategy according to the abnormality type to obtain a repair result; The repair result is confirmed by self-checking to ensure that vehicles are allowed to pass after the abnormality is resolved and the health status library is updated.

2. The lane controller abnormality real-time detection method based on multimodal data fusion according to claim 1 is characterized in that: The performing feature analysis on the multimodal data to obtain analysis results includes: Performing fast Fourier transform analysis on the ground sensing coil signal to extract frequency and amplitude characteristics; Calculating a bit error rate of a DSRC communication link corresponding to the DSRC communication state; Verify the timestamp and amount in the transaction flow data to obtain a verification result; The analysis results include frequency and amplitude characteristics, bit error rate and verification results.

3. The lane controller abnormality real-time detection method based on multimodal data fusion according to claim 1 is characterized in that: The analysis results are input into the decision tree model to determine whether the current state of the lane controller is abnormal, and classified according to different abnormal types to obtain corresponding abnormal types, including: The analysis result is input into a decision tree model, and the decision tree model is combined with a state code table to determine whether the current state of the lane controller is abnormal; When the current state of the lane controller is abnormal, it is classified according to different abnormal types to obtain the corresponding abnormal type.

4. The lane controller abnormality real-time detection method based on multimodal data fusion according to claim 3 is characterized in that: The abnormality type includes at least one of sensor failure, communication timeout, and data abnormality.

5. The lane controller abnormality real-time detection method based on multimodal data fusion according to claim 1 is characterized in that: The executing a corresponding repair strategy according to the abnormality type to obtain a repair result includes: A corresponding repair strategy is determined from a preset safety mechanism table according to the abnormality type, and the corresponding repair strategy is executed to obtain a repair result.

6. The lane controller abnormality real-time detection method based on multimodal data fusion according to claim 5 is characterized in that: The repair strategy includes: When a sensor failure is detected, it will immediately switch to the backup ground sensing coil to ensure uninterrupted signal acquisition; In case of communication timeout, the transmission power of the DSRC module is dynamically adjusted and the 4G backup channel is enabled to ensure smooth communication; When transaction data anomalies are detected, the compensation mechanism is called to store abnormal transaction records and generate audit evidence.

7. The lane controller abnormality real-time detection method based on multimodal data fusion according to claim 1 is characterized in that: The self-checking method is used to confirm the repair result, ensure that the vehicle is allowed to pass after the abnormality is resolved, and update the health status library, including: The repair result is confirmed by self-checking to ensure that vehicles are allowed to pass after the abnormality is resolved, and the health status library is updated. When the lane controller still has an abnormality, an audible and visual alarm is triggered.

8. The method for real-time detection of lane controller anomalies based on multimodal data fusion according to claim 1, characterized in that: The self-checking method is used to confirm the repair result, ensure that the vehicle is allowed to pass after the abnormality is resolved, and update the health status library, including: The repair result is confirmed by self-checking to ensure that vehicles are allowed to pass after the abnormality is resolved, and the current status and repair records are updated in the health status library.

9. The lane controller abnormality real-time detection system based on multimodal data fusion is characterized by: include: A data acquisition unit, used to acquire ground sensor coil signals, DSRC communication status, and transaction flow data to obtain multi-modal data; A feature analysis unit, used to perform feature analysis on the multimodal data to obtain an analysis result; An abnormality classification unit, used for inputting the analysis result into a decision tree model to determine whether the current state of the lane controller is abnormal, and classifying it according to different abnormality types to obtain a corresponding abnormality type; A repair unit, configured to execute a corresponding repair strategy according to the abnormality type to obtain a repair result; The updating unit is used to confirm the repair result by self-checking, ensure that the vehicle is allowed to pass after the abnormality is resolved, and update the health status library.

10. The lane controller abnormality real-time detection system based on multimodal data fusion according to claim 9 is characterized in that: The feature analysis unit comprises: An extraction subunit, used for performing fast Fourier transform analysis on the ground sensing coil signal to extract frequency and amplitude characteristics; A bit error rate calculation subunit, used to calculate the bit error rate of the DSRC communication link corresponding to the DSRC communication state; The verification subunit is used to verify the timestamp and amount in the transaction flow data to obtain a verification result.

Images