Sensor data conflict detection method and system for train dynamic seat display system

Abstract

This application provides a sensor data conflict detection method and system for a train dynamic seat display system. The method includes: comparing the occupancy status of seats belonging to the same seat identifier within a multi-source status signal set to identify conflicting seat sets with inconsistent statuses; generating an arbitration weight allocation table based on the historical accuracy of the detection sources corresponding to the conflicting seat sets; calculating the instantaneous jump frequency of the state switching process within a unit time window for the multi-source status signal set, and generating a signal fluctuation feature vector based on the instantaneous jump frequency; cumulatively calculating the multi-source data conflict events occurring in the conflicting seat set during a continuous monitoring period to generate a conflict accumulation index to characterize the conflict persistence trend; and performing multi-source data consistency fusion adjudication on the conflicting seat set to obtain the true occupancy status of each conflicting seat. This application can improve the instantaneous false alarm filtering capability, enhance the accuracy of continuous conflict assessment, and improve the stability of seat status adjudication.

Description

Technical Field

[0001] This application relates to the technical field of train dynamic seat display systems, and in particular to a sensor data conflict detection method and system for train dynamic seat display systems. Background Technology

[0002] With the improvement of the intelligence level of rail transit, dynamic seat display systems are increasingly being used for displaying seat status in carriages, guiding passengers, and managing operational services. To improve the real-time performance and accuracy of seat occupancy status perception, existing systems typically combine multi-source information such as ticketing data, pressure detection data, and human body sensing data to make a comprehensive judgment on whether a seat is occupied, and display the judgment result on the carriage display terminal or passenger service platform.

[0003] Existing methods for sensing train seat status mostly employ single-sensor determination, fixed-priority determination, or simple multi-source voting. When the outputs from all data sources are consistent, the seat status can be determined relatively directly. However, during train operation, the seat detection environment is complex. Factors such as vehicle vibration, short-term passenger stops, luggage placement, obstruction interference, and passenger seat changes can all lead to inconsistent outputs from different detection sources.

[0004] Existing technologies still have shortcomings in handling the aforementioned conflicts. On the one hand, a single detection source is susceptible to environmental interference, resulting in momentary false alarms and frequent changes in seat occupancy status within a short period. Existing methods typically rely solely on the current state for judgment, making it difficult to quantify the instantaneous changes in the state signal, which can easily lead to repeated changes in seat display results. On the other hand, when multiple source data continuously conflict during a continuous monitoring period, existing systems lack a quantitative assessment mechanism for the cumulative effect and persistent trend of conflict events, making it difficult to distinguish between occasional and long-term conflicts. This results in insufficient stability of the final decision, affecting the accuracy and reliability of the train's dynamic seat display. Summary of the Invention

[0005] In view of this, embodiments of this application provide a sensor data conflict detection method and system for a train dynamic seating display system to solve the problems of difficulty in identifying instantaneous false alarms, difficulty in assessing continuous conflicts, and insufficient decision stability in the prior art.

[0006] A first aspect of this application provides a sensor data conflict detection method for a train dynamic seat display system, comprising: acquiring status data reported by multiple seat status detection sources in the train dynamic seat display system according to seat identifiers, aligning the status data based on a unified time reference to generate a multi-source status signal set; performing a consistency comparison of the occupancy statuses belonging to the same seat identifier in the multi-source status signal set to determine a conflicting set of seats with inconsistent statuses; generating an arbitration weight allocation table to characterize the reliability of different detection sources based on the historical accuracy of the detection sources corresponding to the conflicting set; calculating the instantaneous jump frequency for the state switching process of the multi-source status signal set within a unit time window, and generating a signal fluctuation feature vector to characterize environmental interference-type signal fluctuations based on the instantaneous jump frequency; accumulating and calculating the multi-source data conflict events occurring in the conflicting set during a continuous monitoring period to generate a conflict accumulation index to characterize the conflict persistence trend; performing multi-source data consistency fusion adjudication on the conflicting set based on the arbitration weight allocation table, the signal fluctuation feature vector, and the conflict accumulation index to obtain the true occupancy status of each conflicting seat; updating the true occupancy status to the train dynamic seat display system, and generating a conflict adjudication log corresponding to the true occupancy status.

[0007] A second aspect of this application provides a sensor data conflict detection system for a train dynamic seat display system, comprising: an acquisition module, configured to acquire status data reported by multiple seat status detection sources in the train dynamic seat display system according to seat identifiers, and align the status data based on a unified time reference to generate a multi-source status signal set; a comparison module, configured to perform consistency comparison on the occupancy status of seats belonging to the same seat identifier in the multi-source status signal set to determine a conflicting set of seats with inconsistent statuses; a first generation module, configured to generate an arbitration weight allocation table characterizing the reliability of different detection sources based on the historical accuracy of the detection sources corresponding to the conflicting seat set; and a second generation module, configured to target the multi-source status signal set with the sensor data conflict detection data. The system calculates the instantaneous jump frequency of the seat set during the state switching process within a unit time window, and generates a signal fluctuation feature vector based on the instantaneous jump frequency to characterize environmental interference-type signal fluctuations; the accounting module is used to accumulate and calculate the multi-source data conflict events that occur in the conflict seat set during continuous monitoring periods, and generate a conflict accumulation index to characterize the conflict persistence trend; the adjudication module is used to perform multi-source data consistency fusion adjudication on the conflict seat set based on the arbitration weight allocation table, signal fluctuation feature vector, and conflict accumulation index to obtain the true occupancy status of each conflict seat; the third generation module is used to update the true occupancy status to the train dynamic seat display system and generate a conflict adjudication log corresponding to the true occupancy status.

[0008] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: This application acquires status data reported by multiple seat status detection sources according to seat identifiers in a train dynamic seat display system, aligns the status data based on a unified time reference, and generates a multi-source status signal set. It then performs a consistency comparison of the occupancy statuses belonging to the same seat identifier within the multi-source status signal set to identify conflicting seat sets with inconsistent statuses. Based on the historical accuracy of the detection sources corresponding to the conflicting seat sets, it generates an arbitration weight allocation table to characterize the reliability of different detection sources. The application calculates the instantaneous jump frequency of the multi-source status signal set's status switching process within a unit time window and generates a signal fluctuation feature vector to characterize environmental interference-type signal fluctuations based on the instantaneous jump frequency. It accumulates and calculates multi-source data conflict events occurring in the conflicting seat sets during continuous monitoring periods, generating a conflict accumulation index to characterize the continuous trend of conflict. Based on the arbitration weight allocation table, signal fluctuation feature vector, and conflict accumulation index, it performs multi-source data consistency fusion adjudication on the conflicting seat sets to obtain the true occupancy status of each conflicting seat. Finally, it updates the true occupancy status to the train dynamic seat display system and generates a conflict adjudication log corresponding to the true occupancy status. This application improves the instantaneous false alarm filtering capability, enhances the accuracy of continuous conflict assessment, and improves the stability of seat status adjudication. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating the sensor data conflict detection method for the train dynamic seat display system provided in this application embodiment; Figure 2 This is a schematic diagram of the sensor data conflict detection system of the train dynamic seat display system provided in this application embodiment; Figure 3 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0011] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0012] Existing train dynamic seating display systems typically determine seat occupancy status using multi-source information such as ticketing data, pressure detection data, and human body sensing data, and then use the results for carriage seating display and passenger guidance. In practical applications, multi-source detection data can improve the completeness of seat status perception to some extent, but existing solutions mostly use single-source detection, fixed-priority detection, or simple voting fusion methods, resulting in a rather crude handling of conflicts between different detection sources.

[0013] However, the train operating environment is complex. Factors such as vehicle vibration, short passenger stops, luggage placement, obstruction, and passenger seat changes can all cause abnormal reporting by seat status detection sources. When a single detection source is affected by environmental interference, seat occupancy status may change frequently within a short period of time. Existing methods struggle to quantify and identify such transient changes, easily misinterpreting interference fluctuations as genuine status changes. Furthermore, when multiple detection sources continuously output inconsistent results during a continuous monitoring period, existing systems lack mechanisms to assess the cumulative degree and persistent trend of conflict events, making it difficult to establish stable and reliable status decisions.

[0014] To address the aforementioned issues, this application proposes a multi-sensor data conflict detection method for a train dynamic seat display system. The method involves acquiring status data reported by multiple seat status detection sources according to seat identifiers and generating a multi-source status signal set based on a unified time reference. Consistency comparisons are performed on the occupancy status corresponding to the same seat identifier to determine a conflicting seat set. An arbitration weight allocation table is generated based on the historical accuracy of the detection sources corresponding to the conflicting seat set. The instantaneous jump frequency is calculated for the state switching process of the multi-source status signal set within a unit time window, generating a signal fluctuation feature vector. Multi-source data conflict events within a continuous monitoring period are cumulatively calculated to generate a conflict accumulation index. Finally, combining the arbitration weight allocation table, the signal fluctuation feature vector, and the conflict accumulation index, a multi-source data consistency fusion adjudication is performed on the conflicting seats to obtain the true occupancy status and update it to the train dynamic seat display system.

[0015] Through the above technical solutions, this application can use instantaneous frequency jumps to identify short-term signal fluctuations caused by environmental interference, reducing the impact of instantaneous false alarms on seat display results; it can use the conflict accumulation index to reflect the continuous trend of multi-source data conflicts during continuous monitoring periods, improving the quantitative assessment capability of long-term conflict states; at the same time, it combines the historical accuracy of the detection source to form an arbitration weight, so that the final adjudication process simultaneously considers the credibility of the detection source, the current signal fluctuation characteristics, and the conflict persistence trend, thereby improving the accuracy, stability, and traceability of train seat status display.

[0016] The technical solution of this application will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0017] Figure 1 This is a flowchart illustrating the sensor data conflict detection method for the train dynamic seating display system provided in this application embodiment. Figure 1 As shown, the method may specifically include: S101: Obtain status data reported by multiple seat status detection sources in the train dynamic seat display system according to seat identification, and align the status data based on a unified time reference to generate a multi-source status signal set. S102, perform a consistency comparison of the occupancy status of the multi-source status signals belonging to the same seat identifier, and determine the set of conflicting seats with inconsistent status. S103, Generate an arbitration weight allocation table to characterize the credibility of different detection sources based on the historical accuracy of the detection sources corresponding to the conflict seat set; S104, calculates the instantaneous jump frequency for the state switching process of a multi-source state signal set within a unit time window, and generates a signal fluctuation feature vector based on the instantaneous jump frequency to characterize environmental interference signal fluctuations. S105, cumulatively calculates the multi-source data conflict events that occur in the conflict seat set during the continuous monitoring period, and generates a conflict accumulation index to characterize the continuous trend of conflict. S106, based on the arbitration weight allocation table, signal fluctuation feature vector and conflict accumulation index, performs multi-source data consistency fusion adjudication on the conflict seat set to obtain the true occupancy status of each conflict seat; S107 updates the actual occupancy status to the train dynamic seat display system and generates a conflict resolution log corresponding to the actual occupancy status.

[0018] In some embodiments, status data reported by multiple seat status detection sources according to seat identifiers in a train dynamic seat display system are acquired, and the status data is aligned based on a unified time reference to generate a multi-source status signal set, including: Receive status data reported by each seat status detection source, perform source verification, seat identifier parsing and status format normalization on the status data, and generate a standardized status record carrying the detection source identifier, seat identifier, reporting time and occupancy status; Using the unified time reference of the train dynamic seat display system as the alignment constraint, time window merging and timing compensation are performed on the standardized status records corresponding to different detection sources, and the standardized status records that are in the same alignment window and correspond to the same seat identifier are associated. Based on the correlation results, a multi-source state signal set is generated, which includes the multi-source occupancy status of each seat at the alignment time and the corresponding detection source identifier.

[0019] Specifically, after the train dynamic seat display system is activated, the onboard seat status processing unit uses the unified clock provided by the carriage controller as the time reference and receives status data reported by the ticketing interface, pressure detection terminal, and infrared detection terminal. Taking the third carriage of a certain EMU train as an example, the ticketing interface periodically pushes seat ticketing and boarding status according to train number, carriage number, seat number, and ticket occupancy status; the pressure detection terminal reports seat cushion pressure changes and pressure occupancy judgment results according to the pressure sampling cycle; and the infrared detection terminal reports the human presence judgment results of the corresponding seat area according to the human body sensing cycle.

[0020] After various status data enter the seat status processing unit, source verification is performed first. The verification includes checking whether the source registration identifier is valid, whether the reporting link matches the preset carriage area, whether the status data carries a valid reporting time, and whether there are any missing or duplicate data frames. Status data that fails the source verification is marked as invalid status data and written into the abnormal reception record, and will not participate in the subsequent multi-source status signal set generation.

[0021] After the source verification is successful, the seat status processing unit parses the seat identifiers from the status data. Since the seat identifier formats may differ between different seat status detection sources—for example, the ticketing interface uses a code of carriage number plus seat number, the pressure detection terminal uses a code of sensor channel number mapped to installation location, and the infrared detection terminal uses a code of detection area number mapped to seat area—the system pre-establishes a seat identifier mapping table to convert the original reported identifiers from each detection source into a unified seat identifier. For seat 12A in carriage 3, the 3rd carriage 12A in the ticketing interface, the P03-12A channel in the pressure detection terminal, and the IR03-12A area in the infrared detection terminal all resolve to the same unified seat identifier. If the same original reported identifier cannot be matched with a valid seat, or if one detection channel matches multiple seats simultaneously, an identifier parsing anomaly record is generated, and the alignment process is temporarily suspended.

[0022] After parsing the seat identifiers, the seat status processing unit normalizes the status format. Specifically, it converts ticket statuses such as sold, checked, unsold, and refunded into occupancy or availability candidate statuses; pressure detection results such as pressure above the threshold, pressure below the threshold, and abnormal pressure changes into occupancy, availability, or low-confidence statuses; and infrared detection results such as human presence, human absence, and abnormal occupancy into occupancy, availability, or low-confidence statuses. During normalization, the original status code, detection source identifier, detection source confidence level, and reporting time are also retained to generate standardized status records. Each standardized status record carries at least the detection source identifier, seat identifier, reporting time, and occupancy status, enabling data from different sources to participate in subsequent processing under a unified data structure.

[0023] Furthermore, the seat status processing unit uses the unified time reference of the train dynamic seat display system as an alignment constraint to merge standardized status records into time windows. The unified time reference can be provided by the onboard controller's time synchronization module, and the alignment window can be set according to the seat display refresh cycle, for example, every 2 seconds as an alignment window. Within a certain alignment window, the system merges the standardized status records of the ticketing interface, pressure detection terminal, and infrared detection terminal to the same alignment time according to the reporting time.

[0024] For detection sources with long reporting cycles or network transmission delays, the system performs timing compensation based on the most recent valid status record, the detection source sampling cycle, and the allowable delay range. For example, ticketing data is updated every 10 seconds, pressure detection data every 1 second, and infrared detection data every 2 seconds. When no new ticketing status is received within the current alignment window, the most recent ticketing status can be used as long as the ticketing status has not exceeded the valid retention time, and a compensation flag will be written in the standardized status record. When the pressure or infrared status exceeds the allowable delay range, no compensation will be provided, and it will be marked as a missing detection status.

[0025] Furthermore, after completing time window merging and timing compensation, the seat status processing unit establishes a standardized association between status records based on the same seat identifier. For seat 12A in carriage 3, within the alignment window corresponding to 10:15:20, if the ticketing interface outputs a candidate occupancy status, the pressure detection terminal outputs an idle status, and the infrared detection terminal outputs a human presence status, the system associates these three types of status records as a multi-source occupancy status combination for the same seat at the same alignment time, and retains the detection source identifier, reporting time, compensation identifier, and validity identifier corresponding to each status. For cases where the same detection source reports multiple times within the same alignment window, a representative status record can be determined based on the principle of reporting time closest to the alignment time, highest data validity, or longer status duration, avoiding duplicate reporting that could affect subsequent conflict identification.

[0026] Based on the above correlation results, the system generates a multi-source state signal set. This multi-source state signal set uses seat identifier and alignment time as indexes to record the multi-source occupancy status of each seat at the corresponding alignment time and the corresponding detection source identifier, while also recording the state source, timing compensation status, and validity information.

[0027] Through the above processing method, heterogeneous reported data from different seat status detection sources such as ticketing, pressure, and infrared are converted into unified, comparable, and traceable multi-source status signals. This provides a stable data foundation for subsequent identification of conflicting seat sets, calculation of instantaneous jump frequency, calculation of conflict accumulation index, and execution of consistency fusion adjudication. This reduces misjudgments caused by format differences, timing deviations, and abnormal sources, and improves the data processing reliability of the train dynamic seat display system.

[0028] In some embodiments, a consistency comparison is performed on the occupancy status of seats belonging to the same seat identifier in a multi-source status signal set to determine a set of conflicting seats with inconsistent statuses, including: Based on the seat identifier, extract the multi-source occupancy status at the same alignment time from the multi-source status signal set to generate a seat status comparison unit; Based on the preset state consistency rules, the occupancy status corresponding to different seat state detection sources in the seat state comparison unit is judged to obtain a state consistency identifier. When the state consistency identifier indicates that the occupancy status of at least two seat state detection sources is inconsistent, the corresponding seat identifier is determined as the conflict seat identifier, and the conflict seat identifier and the corresponding conflict source relationship are written into the conflict seat set.

[0029] Specifically, after generating the multi-source status signal set, the seat status processing unit establishes a comparison index according to the seat identifier and alignment time, and extracts the multi-source occupancy status of the same seat within the same alignment window from the multi-source status signal set to generate a seat status comparison unit. Taking seat 12A in carriage 3 as an example, at the alignment time corresponding to 10:15:20, the multi-source status signal set records the ticketing interface output occupancy candidate status, the pressure detection terminal output idle status, and the infrared detection terminal output human presence status. The system encapsulates the above statuses together with the corresponding detection source identifier, reporting time, validity identifier, and compensation identifier into a seat status comparison unit. If seat 12B in carriage 3 only contains ticketing status and pressure status at the same alignment time, and the infrared detection terminal is marked as missing status due to occupancy exceeding the effective holding time, the seat status comparison unit also records the source and reason for the missing status, avoiding the use of the missing status as occupancy or idle status in the consistency comparison.

[0030] Furthermore, after generating the seat status comparison unit, the seat status processing unit invokes preset status consistency rules to determine consistency. These preset status consistency rules may include status category consistency rules, low-confidence status exclusion rules, compensation status validity rules, and abnormal status isolation rules. For the status category consistency rule, if all valid detection sources in the same seat status comparison unit point to either an occupied state or an idle state, it is determined to be consistent; if both occupied and idle states coexist among the valid detection sources, it is determined to be inconsistent.

[0031] For the low-confidence state exclusion rule, if a detection source is marked as low-confidence due to signal abnormalities, obstruction, communication delays, etc., it can be initially excluded from primary comparison and instead recorded as auxiliary conflict information. For the compensation state validity rule, if the compensation state is still within its valid retention period, it is allowed to participate in the comparison; if the compensation state has exceeded its valid retention period, it is treated as a missing detection. Through these rules, invalid, expired, or abnormal data can be avoided from being mistakenly included in conflict judgment.

[0032] For example, in the seat status comparison unit of seat 12A in carriage 3, the ticketing interface points to the occupancy candidate state, the pressure detection terminal points to the idle state, and the infrared detection terminal points to the presence state. The system classifies the ticketing interface and the infrared detection terminal as the occupancy side, and the pressure detection terminal as the idle side. Since there are at least two different occupancy state directions at the same alignment time, the state consistency rule outputs inconsistent results and generates a state consistency identifier. This state consistency identifier can be represented as conflict, non-conflict, or pending observation.

[0033] For example, if the ticketing interface, pressure detection terminal, and infrared detection terminal of seat 13C in carriage 3 all output an idle status, a non-conflicting status consistency identifier is generated and not written into the conflict seat set. If seat 14D in carriage 3 only has a ticketing occupancy status and other detection sources are missing, an observation-type status consistency identifier is generated, and the system does not temporarily identify this seat as a conflict seat, but waits for subsequent alignment windows to continue supplementing the judgment.

[0034] When a conflict is identified by a consistent status identifier, the seat status processing unit determines the corresponding seat identifier as the conflicting seat identifier and further generates a conflict source relationship. The conflict source relationship records which seat status detection sources the conflict originates from and the corresponding status directions of different detection sources. For example, the conflict source relationship for seat 12A in carriage 3 can be recorded as the ticketing interface and infrared detection terminal supporting occupancy, and the stress detection terminal supporting idleness, carrying the corresponding reporting time, status validity, whether compensation is required, and status confidence flags. If a seat has an idle ticketing interface, an occupied stress detection terminal, and an occupied infrared detection terminal, the conflict source relationship is recorded as the stress detection terminal and infrared detection terminal supporting occupancy, and the ticketing interface supporting idleness. If the outputs of different detection sources are inconsistent, but one detection source is marked as low confidence, the system can still write the conflict source relationship into the set and reduce the impact of that detection source in subsequent arbitration weight allocation and signal fluctuation analysis.

[0035] Furthermore, the conflicting seat set can be indexed according to train number, carriage, seat identifier, and alignment time. Each record includes the conflicting seat identifier, seat status comparison unit, status consistency identifier, and conflict source relationship. The system can also mark conflicting seats that appear repeatedly in multiple consecutive alignment windows, providing basic data for subsequent conflict accumulation index calculation.

[0036] Through the above processing method, the system can accurately filter out seats with inconsistent states from the unified and aligned multi-source state signal set, and clarify the source and direction of the conflict. This provides structured input for subsequent generation of arbitration weights based on historical accuracy, identification of instantaneous signal fluctuations, and execution of multi-source data consistency fusion adjudication, thereby improving the accuracy and traceability of the conflict detection process.

[0037] In some embodiments, an arbitration weight allocation table is generated to characterize the credibility of different detection sources based on the historical accuracy of the detection sources corresponding to the conflicting seat set, including: Based on the conflict seat identifiers and conflict source relationships in the conflict seat set, query the status judgment results and review results of the corresponding seat status detection source in the historical adjudication records, and generate the historical accuracy of the detection source. Based on the historical accuracy of the detection sources, the credibility of the detection sources for different seat states under the same conflicting seat identifier is normalized to obtain the corresponding arbitration weight. The conflict seat identifier, seat status detection source identifier, and arbitration weight are linked to generate an arbitration weight allocation table.

[0038] Specifically, after determining the set of conflicting seats, the seat status processing unit uses the conflicting seat identifier and the relationship between the conflict sources as search criteria to extract the historical adjudication records of the corresponding seat status detection sources from the historical adjudication database. These historical adjudication records can be generated by the train dynamic seat display system during each conflict adjudication, manual review, crew terminal confirmation, and subsequent status retrospective processes. Each record includes the seat identifier, detection source identifier, alignment time, detection source output status, system adjudication status, review confirmation status, operating scenario identifier, and record validity identifier.

[0039] For example, taking seat 12A in carriage 3 as an example, the current conflict source relationship shows that the ticketing interface and infrared detection terminal support occupancy, while the stress detection terminal supports idleness. The system will query the historical judgment results of the ticketing interface, stress detection terminal, and infrared detection terminal in the same carriage area, similar operating time, and similar conflict scenarios, and prioritize the selection of valid records that have been manually reviewed or confirmed by subsequent status backtracking.

[0040] When generating historical accuracy rates for detection sources, the system does not directly use the global average accuracy rate of the detection sources. Instead, it performs stratified statistics by combining conflict seat identifiers, conflict source relationships, and operational scenarios. For example, pressure detection terminals may experience more false alarms during peak hours due to passenger luggage placement, infrared detection terminals may experience more short-term false alarms in window areas due to changes in lighting, and ticketing interfaces may not be entirely consistent with the actual seating status in scenarios where passengers frequently change seats. The system calculates the effective hit count and effective participation count of each detection source in the current conflict scenario based on the consistency between the detection source output status and the verification confirmation status in the historical adjudication records. It then combines the record validity, time decay coefficient, and sample quantity reliability to generate the historical accuracy rate of the detection sources. If the number of effective samples for a certain detection source is insufficient, the system can use the regional accuracy rate of the same type of detection source in the same carriage as a supplement, and add a sample insufficiency flag to avoid abnormally high or low weights due to small samples.

[0041] For example, in the historical adjudication records of the past 30 days, the pressure detection terminal in carriage 3 participated in 120 valid adjudications in scenarios where seats were briefly covered by luggage, with 90 of them matching the verified status, corresponding to a basic accuracy rate of 75%; the infrared detection terminal participated in 100 valid adjudications in scenarios with minimal human obstruction, with 88 matching, corresponding to a basic accuracy rate of 88%; and the ticketing interface participated in 150 valid adjudications in scenarios where passengers with tickets had been checked were seated normally, with 120 matching, corresponding to a basic accuracy rate of 80%. The system further performs scenario adaptation processing on the above basic accuracy rates based on the current conflict source relationship. If the current carriage vibration state is normal, the infrared obstruction anomaly has not been triggered, and the pressure fluctuation anomaly has been triggered, the applicable confidence level of the accuracy rate corresponding to the pressure detection terminal can be reduced, while the higher applicable confidence level of the ticketing interface and the infrared detection terminal can be maintained.

[0042] Furthermore, after obtaining the historical accuracy of each seat status detection source, the seat status processing unit normalizes the credibility of multiple detection sources under the same conflicting seat identifier. Specifically, the historical accuracy of the detection sources can be first converted into a non-negative credibility score, then combined with the sample quantity credibility and scenario adaptation coefficient to form a comprehensive credibility score, and then normalized according to the sum of the comprehensive credibility scores of each detection source within the same conflicting seat to obtain the corresponding arbitration weight.

[0043] If a detection source is flagged as having communication anomalies, a low-confidence state, or a severe shortage of samples, the system can set a weight cap or adopt a default minimum weight to ensure that the detection source is not completely excluded due to missing historical data, nor does it dominate the decision-making process under abnormal conditions. Taking seat 12A in carriage 3 as an example, after normalization, the arbitration weight for the ticketing interface can be 0.32, the arbitration weight for the pressure detection terminal can be 0.24, and the arbitration weight for the infrared detection terminal can be 0.44. Subsequent fusion decisions can then be based on a comprehensive judgment combining the above weights with signal fluctuation characteristics and conflict accumulation indicators.

[0044] Finally, the system establishes a correlation between the conflicting seat identifier, the seat status detection source identifier, and the arbitration weight, generating an arbitration weight allocation table. The arbitration weight allocation table can be indexed by train number, carriage, seat identifier, and alignment time. Each row records the weight value of the corresponding detection source for a conflicting seat within the current adjudication period, historical accuracy, sample quantity reliability, scene adaptation identifier, and weight generation time.

[0045] Through the above processing method, the system can transform the historical credibility of different detection sources in the current conflict scenario into arbitration weights that can be used for calculation, so that the multi-source data consistency fusion adjudication no longer depends on fixed priorities or simple voting, thereby improving the accuracy, stability and traceability of determining the true occupancy status of conflict seats.

[0046] In some embodiments, the instantaneous jump frequency is calculated for the state switching process of a multi-source state signal set within a unit time window, and a signal fluctuation feature vector is generated based on the instantaneous jump frequency to characterize environmental interference signal fluctuations, including: Using a unit time window as the statistical range, a continuous occupancy state sequence is extracted from the multi-source state signal set according to the seat identifier and the seat status detection source identifier; The switching events of occupancy states between adjacent alignment times in a continuous occupancy state sequence are identified, and the instantaneous jump frequency of the corresponding seat state detection source is calculated based on the number of switching events and the unit time window length. Based on the matching results between the instantaneous jump frequency and the preset fluctuation judgment conditions, the fluctuation intensity characterization of the corresponding continuous occupancy state sequence is determined. The seat identifier, seat status detection source identifier, and fluctuation intensity characterization are associated and encoded to generate a signal fluctuation feature vector.

[0047] Specifically, after generating the multi-source state signal set, the seat state processing unit uses a unit time window as the statistical range to extract and analyze the changes in occupancy status of each seat at consecutive alignment moments. The unit time window can be determined based on the refresh cycle of the train dynamic seat display system and the sensor reporting cycle, for example, using 10 seconds or 30 seconds as a statistical window, with the alignment moment as the sampling node. The system extracts the continuous occupancy status sequence of the same seat and the same detection source within the unit time window from the multi-source state signal set according to the seat identifier and the seat state detection source identifier.

[0048] For example, taking seat 12A in carriage 3 as an example, from 10:15:00 to 10:15:30, the corresponding state sequence of the pressure detection terminal could be occupied, idle, occupied, idle, occupied, and the corresponding state sequence of the infrared detection terminal could be occupied, occupied, occupied, occupied, occupied, occupied, and the corresponding state sequence of the ticketing interface could remain in the candidate occupied state. The system establishes a continuous occupancy state sequence at the detection source level for the above state sequences and records the alignment time, validity identifier, and compensation identifier corresponding to each state.

[0049] When identifying handover events, the seat status processing unit traverses the continuous occupancy state sequence according to the alignment time order and determines the differences between occupancy states between adjacent alignment times. When two adjacent valid states change from occupied to idle, or from idle to occupied, the system marks this change process as a handover event. If there are missing states, compensation states that exceed the valid holding time, or low-confidence abnormal states among the adjacent states, these adjacent states are not directly counted as handover events, but are written into the auxiliary fluctuation record to avoid communication delays or missing states being mistaken for real transitions.

[0050] For example, the pressure detection terminal at seat 12A in carriage 3 showed four valid switches between occupied and idle statuses within a 30-second window, while the infrared detection terminal did not show any valid switches, and the ticketing interface also remained stable without any switches. The system calculates the instantaneous transition frequency of the corresponding seat status detection source based on the proportional relationship between the number of switching events and the effective statistical length of the unit time window. If the pressure detection terminal shows four valid switches within 30 seconds, the corresponding instantaneous transition frequency can be represented as four times every 30 seconds, or converted to a unit time frequency.

[0051] Furthermore, the seat status processing unit matches the instantaneous jump frequency with preset fluctuation judgment conditions to determine the fluctuation intensity characterization of the continuous occupancy state sequence. The preset fluctuation judgment conditions can be set according to the sampling characteristics of different detection sources, train operation status, and historical false alarm patterns. For example, for a pressure detection terminal, if frequent alternations between occupancy and vacancy occur within a short period, and each state duration is relatively short, it can be judged as a high-intensity fluctuation caused by vibration, short-term pressure on luggage, or brief contact by passengers; for an infrared detection terminal, if the presence and absence of a human body alternate multiple times between adjacent alignment moments, it can be judged as a medium-to-high-intensity fluctuation caused by obstruction, personnel passing by, or light interference; for the ticketing interface, due to the low data update frequency, if the state remains stable within the effective statistical window, the fluctuation intensity remains low. The system can classify fluctuation intensity characterization into stable, slight fluctuation, significant fluctuation, and strong fluctuation, or it can use a continuous scoring method to represent the degree of fluctuation.

[0052] When generating the signal fluctuation feature vector, the system correlates and encodes the seat identifier, seat status detection source identifier, and fluctuation intensity characterization. It can also add instantaneous jump frequency, number of effective switching events, effective statistical length, and anomaly auxiliary identifiers. For example, taking seat 12A in carriage 3 as an example, the pressure detection terminal is encoded as having strong fluctuations due to frequent switching within 30 seconds, the infrared detection terminal is encoded as stable, and the ticketing interface is encoded as stable, thus forming the signal fluctuation feature vector for this seat within the current statistical window. This signal fluctuation feature vector is then correlated with the arbitration weight allocation table and conflict accumulation index to suppress the credibility of detection sources exhibiting environmental interference-type signal fluctuations during the fusion adjudication phase.

[0053] Through the above processing method, the system can transform sensor state transitions that are difficult to judge directly in a short period of time into quantifiable instantaneous transition frequency and fluctuation intensity characteristics. This effectively identifies environmental interference signal fluctuations caused by factors such as vehicle vibration, obstruction interference, and temporary luggage pressing on the seat, reduces the impact of instantaneous false alarms on seat occupancy status determination, and improves the stability and accuracy of subsequent multi-source data consistency fusion determination.

[0054] In some embodiments, the switching events of occupancy states between adjacent alignment times in a continuous occupancy state sequence are identified, and the instantaneous jump frequency of the corresponding seat state detection source is calculated based on the number of switching events and the unit time window length, including: Traverse the sequence of consecutive occupied states according to the alignment time order, and determine the differences between the occupied states corresponding to adjacent alignment times. When the occupancy status corresponding to an adjacent alignment moment changes and the corresponding status data meets the validity condition, the corresponding change process is marked as a switching event; The number of switching events is counted within a unit time window, and the instantaneous switching frequency of the corresponding seat status detection source is generated based on the ratio between the number of switching events and the effective statistical length of the unit time window.

[0055] Specifically, after obtaining the continuous occupancy state sequence, the seat state processing unit traverses the state sequence in ascending order of alignment time, and uses two adjacent alignment times as the smallest judgment unit to determine whether the occupancy state output by the same seat state detection source has changed. Taking the pressure detection terminal of seat 12A in carriage 3 as an example, within a unit time window from 10:15:00 to 10:15:30, the system sequentially reads the pressure occupancy states corresponding to alignment times such as 10:15:00, 10:15:02, and 10:15:04. If 10:15:00 corresponds to an occupied state and 10:15:02 corresponds to an idle state, the system identifies this adjacent state pair as having a state difference; if 10:15:02 is an idle state and 10:15:04 is also an idle state, then this adjacent state pair does not constitute a state difference.

[0056] When determining differences, the system does not only compare the status values ​​themselves, but also simultaneously checks whether the corresponding status data meets the validity conditions. Validity conditions may include: the detection source identifier passes verification, the seat identifier is correctly parsed, the reporting time is within the current unit time window, the status data does not exceed the valid retention time, and the status record is not marked as a communication anomaly or a missing test anomaly. For status records obtained through timing compensation, if the compensation time is still within the allowable retention range, it can participate in the difference determination; if the compensation time exceeds the allowable retention range, the status record is not considered a valid status for the switching event statistics. For example, if the ticketing interface uses the previous valid ticketing status within the current unit time window due to its long reporting cycle, and the ticketing status is within 10 seconds of the current alignment time, then this status can still be considered a stable status for determination; if the stress detection terminal misses a data packet at a certain moment, the system skips the direct switching statistics between the missing status and the preceding and following statuses to avoid mistaking communication packet loss for a sudden change in seat status.

[0057] When the occupancy status changes at adjacent alignment times and the corresponding status data meets the validity conditions, the seat status processing unit marks the change process as a switching event and records the seat identifier, seat status detection source identifier, start alignment time, end alignment time, change direction, and event validity identifier corresponding to the switching event. The change direction can include changing from occupied to idle or from idle to occupied.

[0058] For example, taking seat 12A in carriage 3 as an example, the pressure detection terminal forms a continuous occupancy state sequence of occupied, idle, occupied, idle, occupied within a 30-second window. The system identifies four valid state transitions and records them as four transition events. If the infrared detection terminal continuously outputs a human presence status within the same window, then no valid transition event occurs. If the ticketing interface continuously outputs occupancy candidate statuses, no transition event is generated. For cases where low-confidence states are interspersed in the sequence, the system can first mark the low-confidence states as auxiliary states, without directly generating transition events, but they can be used as anomaly indication parameters in subsequent fluctuation intensity characterization.

[0059] After the unit time window ends, the system counts the number of switching events corresponding to each seat status detection source and calculates the effective statistical length of the unit time window. The effective statistical length is not simply the total window length, but the statistical time after deducting invalid segments such as missing detections, communication anomalies, and timeout compensation. For example, if a pressure detection terminal has 4 seconds of invalid communication segments in a 30-second unit time window, the effective statistical length is 26 seconds. The system generates an instantaneous jump frequency based on the ratio between the number of switching events and the effective statistical length. If the pressure detection terminal in seat 12A of carriage 3 experiences 4 switching events within the 26-second effective statistical length, the system generates a high instantaneous jump frequency; if the infrared detection terminal does not experience any switching events within the 30-second effective statistical length, the instantaneous jump frequency is 0 or close to the stable threshold; if the ticketing interface does not generate any switching events due to stable business status, the corresponding instantaneous jump frequency remains low.

[0060] Furthermore, the seat status processing unit associates the instantaneous frequency jump with the seat identifier and the seat status detection source identifier, and writes this information into the signal fluctuation calculation cache. If the instantaneous frequency jump is significantly higher than the normal fluctuation range of the detection source, the system can mark the detection source as a suspected source of environmental interference within the current window. For example, if the pressure detection terminal frequently switches between occupied and idle states, and both the infrared detection terminal and the ticketing interface maintain the occupied direction, it can be inferred that the pressure state change is more likely caused by train vibration, passenger posture adjustments, or short-term luggage compression.

[0061] Through the above processing method, the system can accurately distinguish between valid state switching and invalid data disturbance, and transform short-term occupancy state changes into quantifiable instantaneous jump frequencies, providing a reliable basis for subsequent generation of signal fluctuation feature vectors and suppression of environmental interference signals, thereby improving the stability of multi-source data conflict detection and seat state fusion adjudication.

[0062] In some embodiments, the multi-source data conflict events occurring in the conflict seat set during a continuous monitoring period are cumulatively calculated to generate a conflict accumulation index to characterize the persistence trend of the conflict, including: Using the continuous monitoring period as the statistical range, the corresponding conflict event records are extracted from the conflict seat set according to the seat identifier; Based on the time of conflict occurrence and the relationship between conflict sources in the conflict event records, time-series aggregation is performed on multi-source data conflict events under the same seat identifier to generate a continuous conflict record. Based on the continuous conflict record, the intensity of conflict occurrence at corresponding conflict seats during the continuous monitoring period is calculated, and a conflict accumulation index is generated to characterize the trend of conflict persistence.

[0063] Specifically, after the set of conflicting seats is formed, the seat status processing unit accumulates and calculates the conflict events according to the continuous monitoring period set by the train dynamic seat display system. The continuous monitoring period can be set according to the seat display refresh stability requirements, such as 5 minutes, 10 minutes, or a section of operation as the statistical range. The system extracts the corresponding conflict event records from the set of conflicting seats using train number, carriage, seat identifier, and alignment time as indexes.

[0064] In some examples, each conflict event record includes the conflicting seat identifier, the time of the conflict, the identifier of the seat status detection source involved in the conflict, the occupancy direction corresponding to different detection sources, the status consistency identifier, and the conflict source relationship. Taking seat 12A in carriage 3 as an example, if the ticketing interface and the infrared detection terminal both support occupancy within the alignment windows corresponding to 10:15:20, 10:15:24, and 10:15:28, while the pressure detection terminal supports idleness, then the system will extract the above records as same-direction conflict events for the same seat within the continuous monitoring period.

[0065] After extracting the conflict event records, the system performs time-series aggregation based on the time of conflict occurrence and the relationship between conflict sources. For adjacent conflict events under the same seat identifier, if the conflict source relationship is the same or the conflict direction is consistent, and the interval between the occurrence times of adjacent conflicts does not exceed a preset continuous interval threshold, then multiple conflict events are merged into a single conflict duration segment. For example, if seat 12A in carriage 3 experiences consecutive conflicts between 10:15:20 and 10:15:40 regarding the ticketing interface being occupied and the pressure detection terminal being idle, the system aggregates this conflict segment into a single duration segment and records the segment's start time, end time, number of conflicts, and dominant conflict source relationship. If the conflict direction changes for the same seat in subsequent time, for example, from the pressure detection terminal being idle to the infrared detection terminal being idle, the system generates a new conflict duration segment to avoid mixing different conflict patterns in the statistics.

[0066] When generating conflict persistence records, the system can also distinguish between occasional conflicts and persistent conflicts. If a conflicting seat only shows inconsistency in a single alignment window and subsequently returns to consistency in multiple alignment windows, the conflict event is marked as an occasional conflict; if the same seat repeatedly shows the same or similar conflict source relationship in multiple consecutive alignment windows, it is marked as a persistent conflict.

[0067] For conflict events caused by missing detection sources, communication delays, or low-reliability anomalies, the system can write them into an auxiliary field of the conflict persistence record, but will not directly include them in the main conflict count to prevent invalid data from amplifying the cumulative conflict result. Taking seat 12A in carriage 3 as an example, if the pressure detection terminal's alignment window is inconsistent with the ticketing interface and the infrared detection terminal for 6 consecutive times, and the pressure detection terminal does not experience communication anomalies, the corresponding conflict persistence record will be marked as a high-continuity conflict; if seat 14D in carriage 3 only experiences a conflict due to a single occlusion anomaly of the infrared detection terminal, the corresponding record will be marked as an occasional conflict.

[0068] Furthermore, the seat status processing unit calculates the conflict intensity based on continuous conflict recording. Conflict intensity can be determined by the number of conflict events within a continuous monitoring period, the length of conflict duration segments, the degree of conflict repetition, and the effective monitoring length. For example, if seat 12A in carriage 3 experiences 18 effective conflicts within a 5-minute continuous monitoring period, with the longest continuous conflict segment lasting 60 seconds and the main conflict source relationship being inconsistency between the pressure detection terminal and other detection sources, then the system generates a high conflict intensity. Conversely, if seat 13C in carriage 3 only experiences two conflicts with relatively long intervals, then the conflict intensity is low. The system can also normalize the number of conflicts by incorporating the effective monitoring length to avoid underestimating or overestimating the conflict intensity due to missing data in a particular period.

[0069] Finally, the system generates a conflict accumulation index to characterize the persistence trend of conflicts based on the intensity of the conflicts. The conflict accumulation index can be indexed by seat identification and carries the cumulative number of conflicts, the duration of continuous conflicts, the main conflict source relationship, the conflict stability level, and the statistical period identifier. For seat 12A in carriage 3, if the conflict duration is long and the conflict source relationship is stable, the conflict accumulation index can characterize the presence of a persistent conflict trend for that seat; for seats with only short-term changes, the conflict accumulation index characterizes low-persistence conflicts.

[0070] Through the above processing method, the system can transform scattered conflict events within a continuous monitoring period into quantifiable trend indicators, enabling subsequent fusion decisions to make stable judgments not only based on the state differences at the current alignment time, but also in combination with long-term conflict accumulation. This improves the train dynamic seat display system's ability to identify continuous conflict states and the reliability of the final seat status decision.

[0071] In some embodiments, based on the arbitration weight allocation table, signal fluctuation feature vector, and conflict accumulation index, a multi-source data consistency fusion adjudication is performed on the conflict seat set to obtain the true occupancy status of each conflict seat, including: The conflicting seat set is associated with the arbitration weight allocation table, signal fluctuation feature vector and conflict accumulation index according to the seat identifier to generate the conflict adjudication input representation; The reliability of seat status detection sources with environmental interference-type signal fluctuations is suppressed based on the signal fluctuation feature vector, and the continuous conflict correction parameters of the corresponding conflict seats are determined based on the conflict accumulation index. Based on the conflict adjudication input representation and the continuous conflict correction parameters, the fusion confidence of different occupancy states is calculated, and the occupancy states that meet the preset adjudication conditions are determined as the true occupancy states.

[0072] Specifically, after generating the arbitration weight allocation table, signal fluctuation feature vector, and conflict accumulation index, the seat status processing unit associates relevant data according to seat identifiers to form a conflict adjudication input representation for a single conflicting seat. Taking seat 12A in carriage 3 as an example, at the current alignment time, the conflicting seat set records the ticketing interface outputting the candidate occupancy status, the pressure detection terminal outputting the idle status, and the infrared detection terminal outputting the presence status of a human body; the arbitration weight allocation table records that the ticketing interface weight is 0.32, the pressure detection terminal weight is 0.24, and the infrared detection terminal weight is 0.44; the signal fluctuation feature vector records that the pressure detection terminal experiences strong fluctuations within a 30-second unit time window, while the ticketing interface and the infrared detection terminal remain stable; the conflict accumulation index records that this seat experiences multiple instances of continuous conflict where the pressure detection terminal is inconsistent with other detection sources within a 5-minute continuous monitoring period. The system associates the above information according to the same seat identifier, the same alignment time, and the same adjudication cycle to form a conflict adjudication input representation that includes state direction, detection source weight, fluctuation intensity, conflict persistence trend, and data validity.

[0073] After generating the conflict resolution input representation, the seat status processing unit suppresses the credibility of seat status detection sources with environmental interference signal fluctuations based on the signal fluctuation feature vector. Specifically, the suppression level can be determined based on the fluctuation intensity representation. When a detection source frequently switches between occupied and idle states within a unit time window, and the switching event meets the environmental interference characteristics, the system reduces the effective credibility of that detection source in the current resolution cycle.

[0074] For example, taking seat 12A in carriage 3 as an example, if the pressure detection terminal repeatedly changes from occupied to idle and vice versa within a short period of time, while the ticketing interface and infrared detection terminal remain stable, the system will identify the pressure detection terminal as a potential source of interference from vehicle vibration, short-term pressure from luggage, or passenger posture adjustments. The system will then adjust the pressure detection terminal's original arbitration weight of 0.24 to a lower, fluctuation-corrected reliability. Similarly, if the infrared detection terminal experiences frequent switching of human presence status due to obstruction, the system can reduce the effective reliability of the infrared detection terminal to prevent momentary false alarms from dominating the final decision.

[0075] Meanwhile, the system determines the continuous conflict correction parameters for the corresponding conflicting seats based on the conflict accumulation index. The conflict accumulation index characterizes the cumulative number of conflict events, the duration of continuous conflicts, and the main conflict source relationships for a particular seat within a continuous monitoring period. When conflicts occur only in a single or a few alignment windows, the system sets the conflict stability level to low, and the continuous conflict correction parameters are mainly used to suppress the impact of occasional conflicts on the actual occupancy status. When the same conflict source relationship occurs repeatedly in multiple consecutive alignment windows, the system sets the conflict stability level to medium-high and adjusts the fusion confidence of different occupancy statuses according to the direction of continuous conflicts. For example, if seat 12A in carriage 3 experiences continuous conflicts within 5 minutes, indicating that the pressure detection terminal supports idleness, the ticketing interface, and the infrared detection terminal supports occupancy, the continuous conflict correction parameters can enhance the fusion confidence of the stable and consistent side and reduce the impact of the continuous abnormal side on the adjudication results.

[0076] Subsequently, the seat status processing unit calculates the fusion confidence of different occupancy states based on the conflict resolution input representation and persistent conflict correction parameters. The system can aggregate detection sources supporting the occupancy state using fluctuation correction confidence to obtain the occupancy fusion confidence; aggregate detection sources supporting the idle state using fluctuation correction confidence to obtain the idle fusion confidence; and then compensate or suppress the confidence in both directions based on the persistent conflict correction parameters.

[0077] For example, taking seat 12A in carriage 3 as an example, both the ticketing interface and the infrared detection terminal support occupancy, and their fluctuation intensity is low. The system aggregates the confidence levels of both after correction to form an occupancy fusion confidence level. The stress detection terminal supports idleness, but it exhibits strong fluctuations and is inconsistent with other detection sources multiple times during continuous monitoring periods. The system suppresses the confidence level of the stress detection terminal to form an idle fusion confidence level. If the occupancy fusion confidence level exceeds a preset adjudication threshold, and the difference between it and the idle fusion confidence level meets the stable adjudication condition, the system determines the occupancy status as the true occupancy status of the conflicting seat.

[0078] For seats whose confidence difference does not meet the stable decision criteria, the system can make a conservative decision by combining the actual occupancy status of the previous decision period, the current conflict persistence trend, and the detection source validity indicator. For example, when the occupancy fusion confidence and the idle fusion confidence are close, and the conflict accumulation index shows that the conflict is still in the short-term observation stage, the system can temporarily maintain the previous actual occupancy status and mark the seat as pending continuous observation; when the same state direction continuously obtains a high fusion confidence in multiple subsequent alignment windows, the actual occupancy status is then updated. Through this processing method, the system can avoid frequent changes in seat display results due to a single conflict or instantaneous false alarms.

[0079] Through the above implementation methods, the system incorporates historical reliability, short-term fluctuation characteristics, and long-term conflict trends into the same adjudication process. This ensures that the determination of the true occupancy status not only relies on the comparison of multiple sources at the current moment but also combines the reliability of the detection sources and the conflict evolution characteristics for stable judgment. This improves the accuracy, anti-interference ability, and display stability of seat status adjudication in environmental interference scenarios.

[0080] In some embodiments, confidence suppression is performed on seat status detection sources with environmental interference signal fluctuations based on signal fluctuation feature vectors, and persistent conflict correction parameters for corresponding conflict seats are determined based on conflict accumulation indices, including: Based on the fluctuation intensity characterization of the corresponding seat state detection source in the signal fluctuation feature vector, the interference level of environmental interference signal fluctuation is determined. The arbitration weights of the corresponding seat status detection sources in the arbitration weight allocation table are suppressed according to the interference level to generate the confidence level of the detection sources after fluctuation correction. The conflict stability level of the corresponding conflict seat is determined based on the conflict accumulation index, which characterizes the intensity and duration of the conflict. Based on the conflict stability level, continuous conflict correction parameters are generated to adjust the fusion confidence of different occupancy states.

[0081] Specifically, before executing the multi-source data consistency fusion decision, the seat status processing unit first reads the signal fluctuation feature vector corresponding to the currently conflicting seat and extracts the fluctuation intensity characterization according to the seat status detection source identifier. The fluctuation intensity characterization can be formed by combining the instantaneous jump frequency, the number of effective switching events, the effective statistical length, and the anomaly auxiliary identifier.

[0082] For example, taking seat 12A in carriage 3 as an example, within a 30-second time window, the pressure detection terminal's output state effectively switches between occupied and idle four times, while the ticketing interface and infrared detection terminal maintain a stable occupied direction within the same window. Therefore, the system identifies the pressure detection terminal's fluctuation intensity as strong fluctuation, and the ticketing interface and infrared detection terminal as stable. If the infrared detection terminal frequently alternates between the presence and absence of a person within a short period due to people passing by or obstruction, the infrared detection terminal can also be identified as exhibiting significant or strong fluctuation.

[0083] After determining the fluctuation intensity characterization, the system determines the interference level of environmental interference signal fluctuations according to a preset interference level mapping rule. This interference level can include low interference, medium interference, and high interference, or it can be represented by a continuous interference score. For pressure detection terminals, if the instantaneous jump frequency is significantly higher than the baseline fluctuation range of similar seats under normal operating conditions, and each state duration is short, the system determines the interference level to be high interference, indicating that the current pressure state is more likely to be affected by vehicle vibration, passenger posture adjustments, or short-term luggage contact. For infrared detection terminals, if the jump frequency increases and is accompanied by an abnormal occlusion indicator, the system determines the interference level to be medium-high interference. For ticketing interfaces, since state changes are usually triggered by ticketing business events, the interference level remains low unless there are short-term, frequent changes.

[0084] Subsequently, the system suppresses the arbitration weights of the corresponding seat status detection sources in the arbitration weight allocation table according to the interference level, generating a fluctuation-corrected confidence level for the detection sources. Specifically, suppression coefficients can be configured for different interference levels, and the original arbitration weights of the detection sources are combined with the corresponding suppression coefficients for calculation. For example, in seat 12A of carriage 3, the original arbitration weight of the pressure detection terminal is 0.24, but it is identified as having high interference. The system corrects the confidence level of the pressure detection terminal to a lower value according to the suppression coefficient corresponding to high interference. The original arbitration weight of the infrared detection terminal is 0.44, and its fluctuation intensity is stable, so the confidence level of the infrared detection terminal remains basically unchanged. The original arbitration weight of the ticketing interface is 0.32, also maintaining a stable confidence level. In this way, detection sources with short-term frequent fluctuations will not excessively affect the judgment of the actual occupancy status in the current adjudication cycle due to their historically high accuracy.

[0085] Furthermore, the seat status processing unit reads the cumulative conflict index corresponding to the currently conflicting seat and determines the conflict stability level based on the conflict intensity and duration. The conflict intensity can be determined by the number of valid conflicts within a continuous monitoring period, the longest consecutive conflict duration, and the degree of repetition of the main conflict source relationships; the conflict duration can indicate whether the conflict is sporadic, intermittent, or stable. Taking seat 12A in carriage 3 as an example, if 18 valid conflicts occur within a 5-minute continuous monitoring period, and the main manifestations are that the pressure detection terminal supports idleness, while the ticketing interface and infrared detection terminal support occupancy, then the system determines the conflict stability level to be high, indicating that the conflict is not a sporadic phenomenon at a single alignment moment, but rather has a stable conflict direction. If a seat only experiences inconsistency within one alignment window and subsequently quickly recovers consistency, then the conflict stability level is determined to be low.

[0086] Furthermore, after determining the conflict stability level, the system generates persistent conflict correction parameters to adjust the fusion confidence of different occupancy states. For high-stability conflicts, if the conflict source relationship remains consistent throughout the continuous monitoring period, the system can increase the confidence compensation for the long-term consistent side and reduce the confidence impact of the continuously abnormal side. For example, in seat 12A of carriage 3, where the ticketing interface and infrared detection terminal consistently support occupancy, while the pressure detection terminal consistently supports idleness with strong fluctuations, the persistent conflict correction parameters are used to enhance the fusion confidence of the occupancy direction while suppressing the fusion confidence of the idle direction. For low-stability conflicts, the system can reduce the correction magnitude to avoid over-correction caused by occasional conflicts. For seats with frequently changing conflict directions, the system can set the persistent conflict correction parameters to a conservative correction mode and continue observation in conjunction with subsequent alignment windows.

[0087] Through the above processing methods, the system can dynamically correct the confidence of the detection source from two aspects before fusion decision: short-term signal fluctuations and long-term conflict trends. This ensures that environmental interference signals do not directly dominate the state decision, and that continuous conflict trends can effectively participate in confidence adjustment. This improves the anti-interference capability of judging the actual occupancy status of conflicting seats, reduces frequent changes in seat display, and enhances the stability and reliability of the train dynamic seat display system in complex operating environments.

[0088] In some embodiments, the actual occupancy status is updated to the train dynamic seat display system, and a conflict resolution log corresponding to the actual occupancy status is generated, including: Based on the seat identifiers of each conflicting seat, the actual occupancy status is matched with the seat display records in the train dynamic seat display system to generate a status update command. When the status update command meets the preset update conditions, the corresponding seat display record will be updated to show the actual occupied status, and the status update time will be recorded. The actual occupancy status, the corresponding multi-source status signal set, the arbitration weight allocation table, the signal fluctuation feature vector, the conflict accumulation index, and the status update time are correlated to generate a conflict adjudication log.

[0089] Specifically, after the multi-source data consistency fusion decision determines the actual occupancy status of each conflicting seat, the seat status processing unit matches the conflicting seat identifier with the seat display record in the train dynamic seat display system. The seat display record may include train number, carriage number, seat number, current display status, last update time, display terminal identifier, and status source identifier.

[0090] For example, taking seat 12A in carriage 3 as an example, if the system determines that the actual occupancy status is occupied within the current adjudication cycle, it will retrieve the carriage seat display records and passenger service platform seat records based on the seat identifier 12A in carriage 3 to confirm whether the current displayed status is still vacant, pending confirmation, or occupied in the previous cycle. If the current displayed status is inconsistent with the actual occupancy status, the system generates a status update instruction; if the current displayed status is already consistent with the actual occupancy status, the system can generate a status hold instruction and only update the adjudication log without triggering a display refresh.

[0091] The status update command can carry seat identifier, target display status, decision generation time, target display terminal, update priority, and update reason identifier. To avoid frequent changes in seat display results within a short period, the system also needs to determine whether the status update command meets preset update conditions before executing the status update. Preset update conditions may include: the fusion confidence level of the actual occupancy status reaching the display update threshold; the decision result not being in an observation state; the interval between the previous display update time and the current update time meeting the minimum refresh interval; the target display terminal having normal communication status; and the seat record not being manually locked.

[0092] For example, taking seat 12A in carriage 3 as an example, if the occupancy confidence level reaches 0.82, which is higher than the preset threshold of 0.70, and more than 5 seconds have passed since the last display update, the system allows the update to be performed; if the occupancy confidence level is insufficient or the conflict is still in the short-term observation stage, the system will not change the display status for the time being, but will record the current decision result in the background.

[0093] When the status update command meets the preset update conditions, the train dynamic seat display system updates the corresponding seat display record to the actual occupancy status and simultaneously records the status update time. The status update time can be an aligned time under a unified time base or the actual time the system writes the data. For multiple display objects such as carriage displays, seat indicator lights, and passenger service platforms, the system can use a unified status update sequence: first update the onboard master status table, and then send refresh messages to each display terminal to ensure consistent display results across different terminals.

[0094] For example, when seat 12A in carriage 3 changes from vacant to occupied, the color or label of that seat on the carriage display screen changes accordingly, and the seat status list in the crew terminal is also updated simultaneously. If a display terminal is temporarily offline, the system retains the status update command to be issued and reissues the update based on the latest actual occupancy status after the terminal resumes communication, thus avoiding long-term inconsistencies in display across different terminals.

[0095] Furthermore, after completing the status update or status maintenance process, the seat status processing unit generates a conflict resolution log. The conflict resolution log does not only record the final result, but also stores the actual occupancy status, the corresponding multi-source status signal set, the arbitration weight allocation table, the signal fluctuation feature vector, the conflict accumulation index, and the status update time in a correlated manner. Taking seat 12A in carriage 3 as an example, the log records the ticketing interface output occupancy candidate status, the pressure detection terminal output idle status, and the infrared detection terminal output human presence status at the current alignment time. It also records the arbitration weights of the ticketing interface, pressure detection terminal, and infrared detection terminal, the strong fluctuation characteristic corresponding to the pressure detection terminal, the conflict accumulation index for continuous conflicts between the pressure detection terminal and other detection sources within 5 minutes, and the final actual occupancy status as occupied and the status update time. Through the above correlation, subsequent maintenance personnel can trace back the arbitration basis from the log to determine whether the status update was caused by an anomaly, short-term fluctuation, or long-term conflict trend of a certain detection source.

[0096] Furthermore, the conflict resolution log can be indexed by train number, carriage, seat identifier, resolution period, and log sequence number, and a log validity period and review identifier can be set. For seats with low fusion confidence but which have still been updated, the system can write a review prompt in the log for spot checks by the crew terminal or the back-end maintenance system. For seats with the same conflict source relationship appearing multiple times, the log can also serve as a basis for subsequently updating the historical accuracy of the detection source and optimizing the arbitration weight. For example, if the crew later confirms that seat 12A in carriage 3 is indeed occupied, the resolution log of this instance can, in turn, enhance the historical accuracy of the ticketing interface and infrared detection terminal in similar scenarios, and reduce the credible contribution of the stress detection terminal in scenarios with strong fluctuations.

[0097] Through the above processing method, the system can stably update the actual occupancy status obtained from the fusion adjudication to the train dynamic seat display system, and completely retain the data source, weight basis, fluctuation characteristics, conflict trend and update time through the conflict adjudication log. This ensures that the seat display results can not only reflect the actual occupancy status in a timely manner, but also support subsequent traceability, verification and parameter optimization, thereby improving the display consistency, operational reliability and maintenance interpretability of the train dynamic seat display system.

[0098] The following are system embodiments of this application, which can be used to execute the method embodiments of this application. For details not disclosed in the system embodiments of this application, please refer to the method embodiments of this application.

[0099] Figure 2 This is a schematic diagram of the sensor data conflict detection system of the train dynamic seat display system provided in this application embodiment. Figure 2 As shown, the system includes: The acquisition module 201 is used to acquire the status data reported by multiple seat status detection sources in the train dynamic seat display system according to the seat identifier, and to align the status data based on a unified time reference to generate a multi-source status signal set. The comparison module 202 is used to perform a consistency comparison of the occupancy status of the multi-source status signals belonging to the same seat identifier, and to determine the set of conflicting seats with inconsistent status. The first generation module 203 is used to generate an arbitration weight allocation table to characterize the credibility of different detection sources based on the historical accuracy of the detection sources corresponding to the conflict seat set. The second generation module 204 is used to calculate the instantaneous jump frequency for the state switching process of the multi-source state signal set within a unit time window, and generate a signal fluctuation feature vector based on the instantaneous jump frequency to characterize the fluctuation of environmental interference signal. The accounting module 205 is used to perform cumulative accounting on the multi-source data conflict events that occur in the conflict seat set during a continuous monitoring period, and generate a conflict accumulation index to characterize the continuous trend of conflict. The adjudication module 206 is used to perform multi-source data consistency fusion adjudication on the set of conflicting seats based on the arbitration weight allocation table, signal fluctuation feature vector and conflict accumulation index, so as to obtain the true occupancy status of each conflicting seat. The third generation module 207 is used to update the actual occupancy status to the train dynamic seat display system and generate a conflict resolution log corresponding to the actual occupancy status.

[0100] In some embodiments, Figure 2 The acquisition module 201 receives status data reported by each seat status detection source, performs source verification, seat identifier parsing, and status format normalization on the status data, and generates a standardized status record carrying the detection source identifier, seat identifier, reporting time, and occupancy status. Using the unified time reference of the train dynamic seat display system as an alignment constraint, the standardized status records corresponding to different detection sources are merged into time windows and time-series compensated, and standardized status records that are in the same alignment window and correspond to the same seat identifier are associated. Based on the association result, a multi-source status signal set containing the multi-source occupancy status of each seat at the alignment time and the corresponding detection source identifier is generated.

[0101] In some embodiments, Figure 2 The comparison module 202 extracts the multi-source occupancy status at the same alignment time from the multi-source status signal set according to the seat identifier, and generates a seat status comparison unit; based on the preset status consistency rule, it performs consistency judgment on the occupancy status corresponding to different seat status detection sources in the seat status comparison unit to obtain a status consistency identifier; when the status consistency identifier indicates that the occupancy status corresponding to at least two seat status detection sources is inconsistent, the corresponding seat identifier is determined as a conflict seat identifier, and the conflict seat identifier and the corresponding conflict source relationship are written into the conflict seat set.

[0102] In some embodiments, Figure 2 The first generation module 203 queries the status judgment results and review results of the corresponding seat status detection source in the historical arbitration record based on the conflict seat identifier and conflict source relationship in the conflict seat set, and generates the historical accuracy of the detection source; based on the historical accuracy of the detection source, the credibility of different seat status detection sources under the same conflict seat identifier is normalized to obtain the corresponding arbitration weight; the conflict seat identifier, seat status detection source identifier and arbitration weight are associated to generate an arbitration weight allocation table.

[0103] In some embodiments, Figure 2 The second generation module 204 uses a unit time window as the statistical range and extracts a continuous occupancy state sequence from the multi-source state signal set according to the seat identifier and the seat status detection source identifier; it identifies the switching events of the occupancy state between adjacent alignment times in the continuous occupancy state sequence, and calculates the instantaneous jump frequency of the corresponding seat status detection source according to the number of switching events and the unit time window length; based on the matching result of the instantaneous jump frequency and the preset fluctuation judgment condition, it determines the fluctuation intensity characterization of the corresponding continuous occupancy state sequence; and it associates and encodes the seat identifier, the seat status detection source identifier, and the fluctuation intensity characterization to generate a signal fluctuation feature vector.

[0104] In some embodiments, Figure 2 The second generation module 204 traverses the continuous occupancy state sequence according to the alignment time order and makes a difference judgment on the occupancy state corresponding to adjacent alignment times. When the occupancy state corresponding to adjacent alignment times changes and the corresponding state data meets the validity condition, the corresponding change process is marked as a switching event. The number of switching events is counted within a unit time window, and the instantaneous jump frequency of the corresponding seat state detection source is generated according to the proportional relationship between the number of switching events and the effective statistical length of the unit time window.

[0105] In some embodiments, Figure 2The accounting module 205 takes the continuous monitoring period as the statistical range, extracts the corresponding conflict event records from the conflict seat set according to the seat identifier; based on the conflict occurrence time and conflict source relationship in the conflict event records, it performs time-series aggregation of multi-source data conflict events under the same seat identifier to generate a conflict persistence record; based on the conflict persistence record, it calculates the conflict occurrence intensity of the corresponding conflict seat in the continuous monitoring period and generates a conflict accumulation index to characterize the conflict persistence trend.

[0106] In some embodiments, Figure 2 The adjudication module 206 associates the conflicting seat set with the arbitration weight allocation table, signal fluctuation feature vector, and conflict accumulation index according to the seat identifier, generating a conflict adjudication input representation; based on the signal fluctuation feature vector, it performs confidence suppression on the seat status detection source with environmental interference signal fluctuation, and determines the continuous conflict correction parameter of the corresponding conflicting seat based on the conflict accumulation index; according to the conflict adjudication input representation and the continuous conflict correction parameter, it calculates the fusion confidence of different occupancy states, and determines the occupancy state that meets the preset adjudication conditions as the true occupancy state.

[0107] In some embodiments, Figure 2 The third generation module 207 matches the actual occupancy status with the seat display record in the train dynamic seat display system based on the seat identifier of each conflicting seat, and generates a status update command; when the status update command meets the preset update conditions, the corresponding seat display record is updated to the actual occupancy status, and the status update time is recorded; the actual occupancy status, the corresponding multi-source status signal set, the arbitration weight allocation table, the signal fluctuation feature vector, the conflict accumulation index, and the status update time are associated to generate a conflict adjudication log.

[0108] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0109] Figure 3 This is a schematic diagram of the electronic device 3 provided in an embodiment of this application. Figure 3 As shown, the electronic device 3 of this embodiment includes: a processor 301, a memory 302, and a computer program 303 stored in the memory 302 and executable on the processor 301. When the processor 301 executes the computer program 303, it implements the steps in the various method embodiments described above. Alternatively, when the processor 301 executes the computer program 303, it implements the functions of each module / unit in the various system embodiments described above.

[0110] Electronic device 3 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 3 may include, but is not limited to, processor 301 and memory 302. Those skilled in the art will understand that... Figure 3 This is merely an example of electronic device 3 and does not constitute a limitation on electronic device 3. It may include more or fewer components than shown, or different components.

[0111] The processor 301 may be a central processing unit (CPU), or 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.

[0112] The memory 302 can be an internal storage unit of the electronic device 3, such as a hard disk or memory of the electronic device 3. The memory 302 can also be an external storage device of the electronic device 3, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 3. The memory 302 can also include both internal and external storage units of the electronic device 3. The memory 302 is used to store computer programs and other programs and data required by the electronic device.

[0113] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the system can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0114] If integrated modules / units are implemented as software functional units and sold or used as independent products, they can be stored in a readable storage medium (e.g., a computer-readable storage medium). Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program may include computer program code, which may be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable storage medium may include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0115] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for detecting sensor data conflict in a train dynamic seating display system, characterized in that, include: The status data reported by multiple seat status detection sources in the train dynamic seat display system according to the seat identification are acquired, and the status data is aligned based on a unified time reference to generate a multi-source status signal set. A consistency comparison is performed on the occupancy status of seats belonging to the same seat identifier in the multi-source status signal set to determine the set of conflicting seats with inconsistent status. Based on the historical accuracy of the detection sources corresponding to the conflict seat set, an arbitration weight allocation table is generated to characterize the credibility of different detection sources; The instantaneous jump frequency is calculated for the state switching process of the multi-source state signal set within a unit time window, and a signal fluctuation feature vector is generated based on the instantaneous jump frequency to characterize the fluctuation of environmental interference signals. The conflict seat set is cumulatively calculated for the multi-source data conflict events that occur within a continuous monitoring period to generate a conflict accumulation index to characterize the trend of conflict persistence. Based on the arbitration weight allocation table, signal fluctuation feature vector and conflict accumulation index, the multi-source data consistency fusion adjudication is performed on the conflict seat set to obtain the true occupancy status of each conflict seat. The actual occupancy status is updated to the train dynamic seat display system, and a conflict resolution log corresponding to the actual occupancy status is generated.

2. The method according to claim 1, characterized in that, The process involves acquiring status data reported by multiple seat status detection sources according to seat identifiers in the train dynamic seat display system, aligning the status data based on a unified time reference, and generating a multi-source status signal set, including: Receive status data reported by each seat status detection source, perform source verification, seat identifier parsing and status format normalization on the status data, and generate a standardized status record carrying the detection source identifier, seat identifier, reporting time and occupancy status. Using the unified time reference of the train dynamic seat display system as the alignment constraint, time window merging and timing compensation are performed on the standardized status records corresponding to different detection sources, and the standardized status records that are in the same alignment window and correspond to the same seat identifier are associated. Based on the correlation results, a multi-source state signal set is generated, which includes the multi-source occupancy status of each seat at the alignment time and the corresponding detection source identifier.

3. The method according to claim 1, characterized in that, The step of performing a consistency comparison of the occupancy statuses belonging to the same seat identifier in the multi-source status signal set to determine the conflicting set of seats with inconsistent statuses includes: According to the seat identifier, extract the multi-source occupancy status at the same alignment time from the multi-source status signal set to generate a seat status comparison unit; Based on the preset state consistency rule, the occupancy status corresponding to different seat state detection sources in the seat state comparison unit is judged to obtain a state consistency identifier. When the status consistency identifier indicates that the occupancy status of at least two seat status detection sources is inconsistent, the corresponding seat identifier is determined as a conflict seat identifier, and the conflict seat identifier and the corresponding conflict source relationship are written into the conflict seat set.

4. The method according to claim 1, characterized in that, The step of generating an arbitration weight allocation table to characterize the credibility of different detection sources based on the historical accuracy of the detection sources corresponding to the conflicting seat set includes: Based on the conflict seat identifiers and conflict source relationships in the conflict seat set, query the status judgment results and review results of the corresponding seat status detection source in the historical adjudication records, and generate the historical accuracy of the detection source. Based on the historical accuracy of the detection sources, the credibility of different seat status detection sources under the same conflicting seat identifier is normalized to obtain the corresponding arbitration weight. The conflict seat identifier, seat status detection source identifier, and arbitration weight are associated to generate the arbitration weight allocation table.

5. The method according to claim 1, characterized in that, The step of calculating the instantaneous jump frequency for the state switching process of the multi-source state signal set within a unit time window, and generating a signal fluctuation feature vector based on the instantaneous jump frequency to characterize environmental interference signal fluctuations, includes: Using the unit time window as the statistical range, a continuous occupancy state sequence is extracted from the multi-source state signal set according to the seat identifier and the seat status detection source identifier; The switching events of occupancy states between adjacent alignment times in the continuous occupancy state sequence are identified, and the instantaneous jump frequency of the corresponding seat state detection source is calculated based on the number of switching events and the unit time window length. Based on the matching result between the instantaneous jump frequency and the preset fluctuation judgment condition, the fluctuation intensity characterization of the corresponding continuous occupancy state sequence is determined. The seat identifier, seat status detection source identifier, and fluctuation intensity characterization are associated and encoded to generate the signal fluctuation feature vector.

6. The method according to claim 5, characterized in that, The step of identifying the switching events of occupancy states between adjacent alignment times in the continuous occupancy state sequence, and calculating the instantaneous jump frequency of the corresponding seat state detection source based on the number of switching events and the unit time window length, includes: The sequence of consecutive occupancy states is traversed according to the alignment time order, and the differences between the occupancy states corresponding to adjacent alignment times are determined. When the occupancy status corresponding to an adjacent alignment moment changes and the corresponding status data meets the validity condition, the corresponding change process is marked as a switching event. The number of switching events is counted within the unit time window, and the instantaneous switching frequency of the corresponding seat status detection source is generated based on the ratio between the number of switching events and the effective statistical length of the unit time window.

7. The method according to claim 1, characterized in that, The step involves cumulatively calculating and analyzing multi-source data conflict events occurring within the conflict seat set during a continuous monitoring period to generate a conflict accumulation index characterizing the persistence trend of the conflict, including: Using the continuous monitoring period as the statistical range, the corresponding conflict event records are extracted from the conflict seat set according to the seat identifier; Based on the conflict occurrence time and conflict source relationship in the conflict event record, time-series aggregation is performed on multi-source data conflict events under the same seat identifier to generate a conflict persistence record; Based on the continuous conflict records, the intensity of conflict occurrence at the corresponding conflict seats during the continuous monitoring period is calculated, and the cumulative conflict index is generated to characterize the trend of conflict persistence.

8. The method according to claim 1, characterized in that, The process of performing multi-source data consistency fusion adjudication on the conflicting seat set based on the arbitration weight allocation table, signal fluctuation feature vector, and conflict accumulation index to obtain the true occupancy status of each conflicting seat includes: The conflicting seat set is associated with the arbitration weight allocation table, the signal fluctuation feature vector, and the conflict accumulation index according to the seat identifier to generate a conflict adjudication input representation; Based on the signal fluctuation feature vector, the reliability of seat status detection sources with environmental interference signal fluctuations is suppressed, and the continuous conflict correction parameters of the corresponding conflict seats are determined based on the conflict accumulation index. Based on the conflict resolution input representation and the persistent conflict correction parameters, the fusion confidence of different occupancy states is calculated, and the occupancy state that meets the preset resolution conditions is determined as the true occupancy state.

9. The method according to claim 1, characterized in that, The step of updating the actual occupancy status to the train dynamic seat display system and generating a conflict resolution log corresponding to the actual occupancy status includes: Based on the seat identifiers of each conflicting seat, the actual occupancy status is matched with the seat display records in the train dynamic seat display system to generate a status update command; When the status update instruction meets the preset update conditions, the corresponding seat display record is updated to the actual occupancy status, and the status update time is recorded; The actual occupancy status, the corresponding multi-source status signal set, the arbitration weight allocation table, the signal fluctuation feature vector, the conflict accumulation index, and the status update time are correlated to generate the conflict arbitration log.

10. A sensor data conflict detection system for a train dynamic seating display system, characterized in that, include: The acquisition module is used to acquire status data reported by multiple seat status detection sources in the train dynamic seat display system according to seat identification, and to align the status data based on a unified time reference to generate a multi-source status signal set. The comparison module is used to perform a consistency comparison of the occupancy status of the multi-source status signals belonging to the same seat identifier, and to determine the set of conflicting seats with inconsistent status. The first generation module is used to generate an arbitration weight allocation table to characterize the credibility of different detection sources based on the historical accuracy of the detection sources corresponding to the conflict seat set. The second generation module is used to calculate the instantaneous jump frequency for the state switching process of the multi-source state signal set within a unit time window, and generate a signal fluctuation feature vector based on the instantaneous jump frequency to characterize the fluctuation of environmental interference-type signals. The accounting module is used to perform cumulative accounting on the multi-source data conflict events that occur in the conflict seat set during a continuous monitoring period, and generate a conflict accumulation index to characterize the trend of conflict persistence. The adjudication module is used to perform multi-source data consistency fusion adjudication on the set of conflicting seats based on the arbitration weight allocation table, signal fluctuation feature vector and conflict accumulation index, so as to obtain the true occupancy status of each conflicting seat. The third generation module is used to update the actual occupancy status to the train dynamic seat display system and generate a conflict resolution log corresponding to the actual occupancy status.

Images