A multi-source vehicle-mounted data fusion method, medium and computer device

Abstract

The application discloses a kind of multi-source vehicle data fusion method, medium and computer equipment, wherein the method comprises: extracting multi-source vehicle data from the vehicle data storage device of the vehicle of traffic accident;Different data processing methods are provided for single-source time period data and single-source continuous data obtained from a single vehicle data storage device, efficient integration and accurate alignment of single-source data are realized to ensure the consistency of time series and the reliability of data source in subsequent analysis;Merge multiple single-source time period data and multiple single-source continuous data onto the same time axis.The application can automatically and systematically extract data from different electronic control units of the vehicle and merge them onto the same time reference data axis with high precision and high reliability to support high-quality vehicle accident data forensics analysis.

Description

Technical Field

[0001] This invention relates to the field of vehicle safety and accident analysis technology, and more specifically to a multi-source vehicle data fusion method, medium, and computer equipment for vehicle accident data forensic analysis. Background Technology

[0002] Modern vehicles are typically equipped with multiple data recording systems, such as Event Data Recording (EDR), Automated Streaming Data Recording (DSSAD), and Telematics Processors (T-Boxes). As important components of the vehicle's electronic architecture, these systems independently record relevant vehicle dynamic parameters and environmental information based on preset trigger conditions during an accident, forming multi-source, heterogeneous data records. However, in actual forensic analysis, due to the independent operation of each system, the lack of effective synchronization mechanisms for internal clocks, differences in data sampling frequencies, and variations in storage formats and protocols, the recorded data is scattered over time. While it possesses some complementarity in terms of content, it is difficult to directly and accurately correlate and comprehensively utilize it.

[0003] Existing technical methods are typically limited to independent analysis of a single data source, or rely heavily on manual data comparison and time alignment by technicians when performing multi-source analysis. This approach is not only inefficient and susceptible to errors due to subjective factors, but more importantly, it fails to guarantee the consistency of multi-source data at both the temporal and physical-logical levels. Ultimately, it cannot provide a complete, reliable, and unified time-series data foundation for accurate reconstruction of the accident process.

[0004] Therefore, the current technological status quo urgently requires a technical method that can automatically and systematically fuse multi-source vehicle data into a unified timeline with high precision and high reliability to support high-quality vehicle accident data forensic analysis. Summary of the Invention

[0005] The present invention provides a method, medium, and computer equipment for automatically and systematically fusing multi-source vehicle data into a unified timeline with high precision and high reliability, which can at least solve one of the above-mentioned technical problems.

[0006] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: A method for fusing multi-source vehicle data includes the following steps: S1. Extract multi-source vehicle data from the vehicle's on-board data storage device. The multi-source vehicle data includes at least relevant data from the Event Data Recording System (EDR), Vehicle Driving Recorder (VDR), Telematics Processor (T-Box), and Autonomous Driving Data Recording System (DSSAD). S2. Different data processing methods are provided for single-source time-period data and single-source continuous data obtained from a single vehicle-mounted data storage device, so as to achieve efficient integration and accurate alignment of single-source data, and ensure the consistency of time series and the reliability of data source in subsequent analysis. S3. Merge multi-source data by combining multiple single-source time-period data and multiple single-source continuous data onto the same time axis.

[0007] Furthermore, in S1, a single vehicle has multiple on-board data storage devices, each storing single-source on-board data. All single-source on-board data are aggregated to form multi-source on-board data, which is divided into two categories: One type is time-period data, which has a trigger point for a data record, and the recorded data is the data within a certain period of time around that trigger point; Another type is continuous data. In continuous data, one part has a trigger point and an end point, which is used to record the data of the entire time period from the trigger point to the end point, and the other part is used to record data of a fixed duration.

[0008] Furthermore, in step S2, the data processing procedure for single-source time-period data obtained from a single in-vehicle data storage device further includes: S2.a1. Screen for invalid data. If invalid data is found, the data will no longer participate in the subsequent merging process according to the time axis, and will be marked as non-time-related data and transferred to the auxiliary information database for subsequent manual judgment. S2.a2 After removing invalid data, screen all remaining data to see if they have a strong correlation with time. If a strong correlation with time is found, the data is retained. If no strong correlation with time is found, the data will no longer participate in the subsequent merging process according to the time axis, and will be marked as non-time-related data and transferred to the auxiliary information database for subsequent manual judgment. S2.a3. After removing data without time association, determine whether the time data associated with all remaining data is relative or absolute time. S2.a4. Determine if absolute time data exists within each time period: If absolute time data exists, all relative time data are aligned with that absolute time point as the reference, the relative time offset is converted into the corresponding absolute timestamp and a simple verification is performed to verify the continuity of time and the logical rationality. If absolute time data does not exist, the data within that time period needs to be merged according to relative time, and the data within that time period should be marked as a relative time data group, and then merged after comparison with other data. S2.a5. For cases where a single vehicle-mounted data storage device contains data for multiple time periods, based on the time period data with absolute time within the target time period, align the remaining multiple time period data according to the absolute time of that time period data. For relative time data sets, determine whether their storage order is adjacent to data that has already undergone alignment: If other time periods adjacent to it are involved in time alignment, then the data in this relative time data group will be merged after being compared with other data. If other time periods adjacent to it are not included in the time alignment, the data in that relative time data group is unlikely to participate and will not be compared or merged with other data in the future.

[0009] Furthermore, in step S2, the data processing procedure for single-source continuous data acquired from a single vehicle-mounted data storage device further includes: S2.b1. Due to the long storage time and large amount of data, single-source continuous data is stored in multiple files or folders. The continuous data in these multiple files or folders needs to be reconstructed into a time series using the following two methods: (1) For continuous data in a folder or file where each piece of data has an absolute timestamp, sort them directly in ascending order according to the timestamp; (2) For continuous data containing a mixture of absolute and relative times in folders or files, firstly extract the absolute time points in each file as anchor points, and then interpolate and align adjacent relative time data segments based on the nearest predecessor or successor absolute timestamp to form a unified time series. Secondly, after completing the time series reconstruction, perform integrity verification and redundancy filtering on the data to remove duplicate sampling points and outliers whose time deviates significantly from the normal range. S2.b2 Further data deduplication: First, identify data from the same type of bus and under the same ID. If there are multiple sets of data with the same ID, select the best to retain them based on the integrity of the recorded signal and the continuity of sampling of each bus. Prioritize retaining data segments with complete signals and high sampling continuity, and eliminate redundant records with frame loss or time discontinuity. Secondly, for multiple sets of data with the same ID and similar signal integrity, only one set needs to be retained.

[0010] Furthermore, the merging process of multi-source data in S3 further includes: S3.1 Select a single-source continuous data set as the reference time axis, prioritizing data sets with wide time coverage, high sampling frequency, and high timestamp accuracy; S3.2. Time-align all remaining single-source data based on absolute time. S3.3 After merging the unit data containing absolute time, for the relative time data group lacking absolute timestamps, compare each consecutive data value in the data group without relative time with the same field in the merged data segment. The specific steps are as follows: S3.3.1 Starting from the entire merged time starting point, traverse the entire baseline time axis, search for each continuous data in the relative time data group one by one, and screen whether there is a time period that can be matched with the data segment that is precisely aligned with the merged absolute time series. S3.3.2. Compare all continuous data in the relative time data group to screen for cases where all data items have completely identical time periods: If it does not exist, the relative time data group will not participate in the merging; If only one set of perfectly matching time periods exists, then the relative time data set is mapped to the corresponding absolute time interval. If there are multiple candidate time periods, each candidate time period will be merged as an option. S3.3.3 Manually determine and select a candidate time period to ensure accurate mapping of relative time data on the global time axis.

[0011] Furthermore, in the process of merging multi-source data, since the acquisition accuracy of different data sources is different, after the merging work is completed, some low-precision continuous data need to be fine-tuned: connect the high-precision data, find the corresponding values ​​of the low-precision data on each connection line, and make left and right fine adjustments on the reference time axis so that the low-precision data points are aligned with the high-precision data as completely as possible.

[0012] A computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the multi-source vehicle data fusion method described above.

[0013] A computer device includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the multi-source vehicle data fusion method described above.

[0014] The beneficial effects of this invention are reflected in: This invention achieves efficient fusion and precise alignment of multi-source heterogeneous vehicle data under a unified time reference data axis. The fused data enables users to analyze the continuous operating status and behavioral characteristics of vehicles more intuitively and accurately, providing a highly reliable data foundation for accident tracing, driving behavior modeling, and intelligent diagnosis. Attached Figure Description

[0015] The accompanying drawings, which are provided to further illustrate this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application.

[0016] Figure 1 This is a schematic diagram of the overall process of the multi-source vehicle data fusion method according to an embodiment of the present invention.

[0017] Figure 2 This is a detailed flowchart illustrating the multi-source vehicle data fusion method according to an embodiment of the present invention.

[0018] Figure 3 This is a structural block diagram of a computer device according to an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Unless otherwise specified, the embodiments and features in the embodiments of this application can be combined with each other. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] It should be noted that the meaning of "and / or" throughout the text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, solution B, or a solution that simultaneously satisfies A and B. Furthermore, "multiple" refers to two or more. Additionally, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by this invention.

[0021] See Figures 1-2 This invention provides a multi-source vehicle data fusion method. This method is mainly used to merge data from multiple different data sources (data storage devices) of a single vehicle in a traffic accident into a single timeline according to chronological order. It should be noted that the merged data is within a certain time range, which is determined manually and covers a period before and after the accident. Only data within a specific timeframe before and after the accident is considered; therefore, data fusion only needs to consider data within this event segment. Specifically, this method includes the following implementation steps: S1. Extract multi-source vehicle data from the vehicle's on-board data storage device. The multi-source vehicle data includes at least relevant data from the Event Data Recording System (EDR), Vehicle Driving Recorder (VDR), Telematics Processor (T-Box), and Autonomous Driving Data Recording System (DSSAD).

[0022] In S1, a single vehicle has multiple on-board data storage devices. Each on-board data storage device stores single-source on-board data. All single-source on-board data are aggregated to form multi-source on-board data, which is divided into two categories: One type is time-period data, which has a trigger point for a data record, and the recorded data is the data within a certain period of time around that trigger point; Another type is continuous data. In continuous data, one part has a trigger point and an end point, which is used to record the data for the entire time period from the trigger point to the end point (for example, the DSSAD Type II system records the data during the period when the assisted driving is activated), and the other part is used to record data for a fixed duration (for example, T-BOX records the data up to 7*24 hours before the current time).

[0023] Since the two major categories of data differ in their recording and storage methods, different data processing methods will be provided for these two categories of data in subsequent steps.

[0024] S2. Different data processing methods are provided for single-source time-period data and single-source continuous data obtained from a single vehicle-mounted data storage device, so as to achieve efficient integration and accurate alignment of single-source data, and ensure the consistency of time series and the reliability of data source in subsequent analysis.

[0025] In step S2, the data processing procedure for single-source time-period data obtained from a single in-vehicle data storage device further includes: S2.a1. Screen for invalid data. If invalid data is found, the data will no longer participate in the subsequent merging process according to the time axis, and will be marked as non-time-related data and transferred to the auxiliary information database for subsequent manual judgment. Among them, the determination of invalid values ​​includes: (1) "invalid value" and "unobtainable value" that are directly recorded (the data will directly record the value as invalid); (2) values ​​that cannot be parsed, some of which are defined by the enterprise and have no clear meaning (for example, ACC status: some enterprises will define various conditions or various states of the functions that are enabled); (3) values ​​that are recorded beyond the range of valid values ​​(for example, brake pedal opening exceeds 100%, steering wheel angle exceeds 1080°, etc.).

[0026] S2.a2 After removing invalid data, screen all remaining data to see if they have a strong correlation with time. If a strong correlation with time is found, the data is retained. If no strong correlation with time is found (e.g., the hardware number of the device, the manufacturing date of the device, etc., do not have a strong correlation with time), the data will no longer participate in the subsequent merging process according to the time axis, and will be marked as non-time-related data and transferred to the auxiliary information database for subsequent manual judgment. S2.a3. After removing data without time association, determine whether the time data associated with all remaining data is relative or absolute (for example, the first 5 seconds of data stored in EDR is relative time data, and there is no absolute time; the time collision point stored in EDR is absolute time). S2.a4. Determine if absolute time data exists within each time period: If absolute time data exists, all relative time data are aligned with that absolute time point as the reference. The relative time offset is converted into the corresponding absolute timestamp and a simple verification is performed to check the continuity and logical rationality of the time (e.g., abnormal cases such as relative time reversal or jump). If absolute time data does not exist, the data within that time period needs to be merged according to relative time, and the data within that time period should be marked as a relative time data group, and then merged after comparison with other data. S2.a5. For cases where a single vehicle-mounted data storage device contains data for multiple time periods, based on the time period data with absolute time within the target time period, align the remaining multiple time period data according to the absolute time of that time period data. For relative time data sets, determine whether their storage order is adjacent to data that has already undergone alignment: Since the data in the vehicle's storage device is stored sequentially, if data from other time periods that are immediately adjacent to it participates in time alignment, the data from that relative time period will be merged after being compared with other data later. If other time periods adjacent to it are not included in the time alignment, the data in that relative time data group is unlikely to participate and will not be compared or merged with other data in the future.

[0027] In addition, if there is no data segment containing absolute time, the data in this vehicle data storage device is considered as a relative time data group and will only be processed in the subsequent multi-source data cross-validation stage.

[0028] In S2, the data processing procedure for single-source continuous data acquired from a single in-vehicle data storage device further includes: S2.b1. Due to the long storage time and large amount of data, single-source continuous data is stored in multiple files or folders. The continuous data in these multiple files or folders needs to be reconstructed into a time series using the following two methods: (1) For continuous data in a folder or file where each piece of data has an absolute timestamp, sort them directly in ascending order according to the timestamp; (2) For continuous data containing a mixture of absolute and relative times in folders or files, firstly extract the absolute time points in each file as anchor points, and then interpolate and align adjacent relative time data segments based on the nearest predecessor or successor absolute timestamp to form a unified time series. Secondly, after completing the time series reconstruction, perform integrity verification and redundancy filtering on the data to remove duplicate sampling points and outliers whose time deviates significantly from the normal range. S2.b2. Since the stored data comes from the vehicle's onboard bus, and there are multiple onboard buses, with data in each bus potentially originating from the same data source. Furthermore, if a data conflict occurs on the bus, a priority level will prevent lower-priority data from being collected. Therefore, after completing the above data merging and deduplication operations, further deduplication processing is required. The specific process is as follows: First, identify data originating from the same type of bus and under the same ID (e.g., CAN bus, CAN ID is the same, data under the same CAN ID is collected on multiple CAN buses). If there are multiple sets of data with the same ID, select the best to retain based on the integrity of the recorded signal and the continuity of sampling for each bus. Prioritize retaining data segments with complete signals and high sampling continuity, and eliminate redundant records with frame loss or time discontinuity. Secondly, for multiple sets of data with the same ID and similar signal integrity, only one set needs to be retained.

[0029] S3. Merge multi-source data by combining multiple single-source time-period data and multiple single-source continuous data onto the same time axis.

[0030] The merging process of multi-source data in S3 further includes: S3.1 Select a single-source continuous data set as the reference time axis, prioritizing data sets with wide time coverage, high sampling frequency, and high timestamp accuracy; S3.2. Time-align all remaining single-source data based on absolute time. S3.3 After merging the unit data containing absolute time, for the relative time data group lacking absolute timestamps, compare each consecutive data value in the data group without relative time with the same field in the merged data segment. The specific steps are as follows: S3.3.1 Starting from the entire merged time starting point, traverse the entire baseline time axis, search for each continuous data in the relative time data group one by one, and screen whether there is a time period that can be matched with the data segment that is precisely aligned with the merged absolute time series. S3.3.2. Compare all continuous data in the relative time data group to screen for cases where all data items have completely identical time periods: If it does not exist, the relative time data group will not participate in the merging; If only one set of perfectly matching time periods exists, then the relative time data set is mapped to the corresponding absolute time interval. If there are multiple candidate time periods, each candidate time period will be merged as an option. S3.3.3 Manually determine and select a candidate time period to ensure accurate mapping of relative time data on the global time axis.

[0031] In addition, during the process of merging multi-source data in S3, since the acquisition accuracy of different data sources is different (some are accurate to 1ms, and some are accurate to 1s), after the merging work is completed, some low-precision continuous data need to be fine-tuned: connect the high-precision data, find the corresponding values ​​of the low-precision data on each connection line, and make left and right fine adjustments on the reference time axis to make the low-precision data points as perfectly aligned with the high-precision data as possible.

[0032] It should also be noted that multiple data points collected from multi-source data may be the same data point, but multiple copies should still be retained because they will be used as evidence in the end. However, for multiple identical data points collected from single-source data, only one copy needs to be retained, because one copy is representative and used as evidence, while the others are completely duplicates.

[0033] It should also be noted that the main purpose of this method is to efficiently fuse and accurately align multi-source heterogeneous vehicle data on the same time axis. The fusion result can be used for subsequent accident data analysis models to analyze the continuous operating status and behavioral characteristics of vehicles, while this method only provides a high-reliability data foundation.

[0034] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the steps of the multi-source vehicle data fusion method described above.

[0035] See Figure 3 The present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the multi-source vehicle data fusion method described above.

[0036] This invention also provides a computer program product containing instructions that, when run on a computer, causes the computer to perform the steps of the multi-source vehicle data fusion method described above.

[0037] It is understood that the systems, devices and storage media provided in the embodiments of the present invention correspond to the methods provided in the embodiments of the present invention, and the explanations, examples and beneficial effects of the relevant content can be referred to the corresponding parts of the above-mentioned multi-source vehicle data fusion method.

[0038] It should be noted that those skilled in the art will understand that all or part of the steps implemented in the embodiments of the present invention can be implemented entirely or partially by software, hardware, firmware, or any combination thereof. When implemented in hardware, it can be implemented entirely or partially by purchasing standard parts or modifications. When implemented in software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid state disks (SSDs)).

[0039] In summary, this invention addresses the problem of accurately reconstructing the accident process during evidence collection and analysis of existing accident vehicles by providing a multi-source vehicle-mounted data fusion method for vehicle-mounted accident data forensic analysis. This method extracts multi-source vehicle-mounted data from the vehicle's onboard data storage device. Different data processing methods are then provided for single-source time-segment data and single-source continuous data obtained from individual onboard data storage devices. This achieves efficient integration and precise alignment of single-source data, ensuring consistency of time series and reliability of data sources in subsequent analysis. Finally, multiple single-source time-segment data and multiple single-source continuous data are merged onto the same timeline. Thus, multi-source vehicle-mounted data can be automatically and systematically fused onto a unified timeline with high precision and reliability, supporting high-quality vehicle-mounted accident data forensic analysis.

[0040] It should be understood that the examples and embodiments described herein are for illustrative purposes only and are not intended to limit the invention. Those skilled in the art can make various modifications or changes based on them. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the invention should be included within the protection scope of the invention.

Claims

1. A method for fusing multi-source vehicle-mounted data, characterized in that, Includes the following steps: S1. Extract multi-source vehicle data from the vehicle's on-board data storage device. The multi-source vehicle data includes at least relevant data from the Event Data Recording System (EDR), Vehicle Driving Recorder (VDR), Telematics Processor (T-Box), and Autonomous Driving Data Recording System (DSSAD). S2. Different data processing methods are provided for single-source time-period data and single-source continuous data obtained from a single vehicle-mounted data storage device, so as to achieve efficient integration and accurate alignment of single-source data, and ensure the consistency of time series and the reliability of data source in subsequent analysis. S3. Merge multi-source data by combining multiple single-source time-period data and multiple single-source continuous data onto the same time axis.

2. The multi-source vehicle data fusion method as described in claim 1, characterized in that, In S1, a single vehicle has multiple on-board data storage devices. Each on-board data storage device stores single-source on-board data. All single-source on-board data are aggregated to form multi-source on-board data, which is divided into two categories: One type is time-period data, which has a trigger point for a data record, and the recorded data is the data within a certain period of time around that trigger point; Another type is continuous data. In continuous data, one part has a trigger point and an end point, which is used to record the data of the entire time period from the trigger point to the end point, and the other part is used to record data of a fixed duration.

3. The multi-source vehicle data fusion method as described in claim 1, characterized in that, In step S2, the data processing procedure for single-source time-period data obtained from a single in-vehicle data storage device further includes: S2.a1. Screen for invalid data. If invalid data is found, the data will no longer participate in the subsequent merging process according to the time axis, and will be marked as non-time-related data and transferred to the auxiliary information database for subsequent manual judgment. S2.a2 After removing invalid data, screen all remaining data to see if they have a strong correlation with time. If a strong correlation with time is found, the data is retained. If no strong correlation with time is found, the data will no longer participate in the subsequent merging process according to the time axis, and will be marked as non-time-related data and transferred to the auxiliary information database for subsequent manual judgment. S2.a3. After removing data without time association, determine whether the time data associated with all remaining data is relative or absolute time. S2.a4. Determine if absolute time data exists within each time period: If absolute time data exists, all relative time data are aligned with that absolute time point as the reference, the relative time offset is converted into the corresponding absolute timestamp and a simple verification is performed to verify the continuity of time and the logical rationality. If absolute time data does not exist, the data within that time period needs to be merged according to relative time, and the data within that time period should be marked as a relative time data group, and then merged after comparison with other data. S2.a5. For cases where a single vehicle-mounted data storage device contains data for multiple time periods, based on the time period data with absolute time within the target time period, align the remaining multiple time period data according to the absolute time of that time period data. For relative time data sets, determine whether their storage order is adjacent to data that has already undergone alignment: If other time periods adjacent to it are involved in time alignment, then the data in this relative time data group will be merged after being compared with other data. If other time periods adjacent to it are not included in the time alignment, the data in that relative time data group is unlikely to participate and will not be compared or merged with other data in the future.

4. The multi-source vehicle data fusion method as described in claim 1, characterized in that, In step S2, the data processing procedure for single-source continuous data obtained from a single vehicle-mounted data storage device further includes: S2.b1. Due to the long storage time and large amount of data, single-source continuous data is stored in multiple files or folders. The continuous data in these multiple files or folders needs to be reconstructed into a time series using the following two methods: (1) For continuous data in a folder or file where each piece of data has an absolute timestamp, sort them directly in ascending order according to the timestamp; (2) For continuous data containing a mixture of absolute and relative times in folders or files, firstly extract the absolute time points in each file as anchor points, and then interpolate and align adjacent relative time data segments based on the nearest predecessor or successor absolute timestamp to form a unified time series. Secondly, after completing the time series reconstruction, perform integrity verification and redundancy filtering on the data to remove duplicate sampling points and outliers whose time deviates significantly from the normal range. S2.b2 Further data deduplication: First, identify data from the same type of bus and under the same ID. If there are multiple sets of data with the same ID, select the best to retain them based on the integrity of the recorded signal and the continuity of sampling of each bus. Prioritize retaining data segments with complete signals and high sampling continuity, and eliminate redundant records with frame loss or time discontinuity. Secondly, for multiple sets of data with the same ID and similar signal integrity, only one set needs to be retained.

5. The multi-source vehicle data fusion method as described in claim 1, characterized in that, The merging process of multi-source data in S3 further includes: S3.1 Select a single-source continuous data set as the reference time axis, prioritizing data sets with wide time coverage, high sampling frequency, and high timestamp accuracy; S3.

2. Time-align all remaining single-source data based on absolute time. S3.3 After merging the unit data containing absolute time, for the relative time data group lacking absolute timestamps, compare each consecutive data value in the data group without relative time with the same field in the merged data segment. The specific steps are as follows: S3.3.1 Starting from the entire merged time starting point, traverse the entire baseline time axis, search for each continuous data in the relative time data group one by one, and screen whether there is a time period that can be matched with the data segment that is precisely aligned with the merged absolute time series. S3.3.

2. Compare all continuous data in the relative time data group to screen for cases where all data items have completely identical time periods: If it does not exist, the relative time data group will not participate in the merging; If only one set of perfectly matching time periods exists, then the relative time data set is mapped to the corresponding absolute time interval. If there are multiple candidate time periods, each candidate time period will be merged as an option. S3.3.3 Manually determine and select a candidate time period to ensure accurate mapping of relative time data on the global time axis.

6. The multi-source vehicle data fusion method as described in claim 5, characterized in that, In the process of merging multi-source data, since the acquisition accuracy of different data sources is different, after the merging work is completed, some low-precision continuous data need to be fine-tuned: connect the high-precision data, find the corresponding values ​​of the low-precision data on each connection line, and make left and right fine adjustments on the reference time axis to make the low-precision data points as perfectly aligned with the high-precision data as possible.

7. A computer-readable storage medium, characterized in that, The system contains a computer program that, when executed by a processor, causes the processor to perform the steps of the multi-source vehicle data fusion method as described in any one of claims 1-6.

8. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the multi-source vehicle data fusion method as described in any one of claims 1-6.

Images