Track offset detection method and device, electronic equipment and medium

Abstract

The invention provides a track offset detection method and device, electronic equipment and a medium, and the method comprises the steps: obtaining the multi-protocol original data of a target route, and generating a to-be-analyzed track under a unified coordinate system according to the multi-protocol original data; according to a standard route corresponding to the to-be-analyzed track and the target route, obtaining a track image with the same geographic range and coordinate mapping relation with the standard route, and inputting the track image into a recognition model for comparison to obtain an offset area and an offset direction of the to-be-analyzed track relative to the standard route; and according to the offset area and the offset direction, quantitatively calculating the offset of the to-be-analyzed track relative to the standard route, so as to generate an offset analysis result of the target route according to the offset. According to the invention, high-precision track offset detection and analysis of multi-source data unified processing and image comparison are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of trajectory analysis, in particular to a trajectory deviation detection method and device, electronic equipment and medium. BACKGROUND

[0002] Current positioning and trajectory analysis technology generally relies on application layer post-processing and direct latitude and longitude sequence comparison, and has the prominent problem of low automation and intelligence. Especially in complex environments (such as urban canyons, tunnels, viaducts), positioning data is prone to drift and jump, and traditional methods are difficult to achieve efficient and accurate anomaly identification and rule analysis. The specific performance is as follows: lacking automated analysis means, relying on manual comparison in the face of massive data, low efficiency and easy to miss; unable to accurately quantify the deviation distance between the trajectory and the standard route, and also difficult to output statistical results; unable to introduce image recognition technology for automatic comparison of trajectory images, limiting intuitive and batch drift pattern analysis. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a trajectory deviation detection method, device, electronic equipment and medium, which aims to overcome at least one of the above-mentioned defects.

[0004] In a first aspect, the present application provides a trajectory deviation detection method, which comprises: obtaining multi-protocol raw data of a target route, and generating a to-be-analyzed trajectory in a unified coordinate system according to the multi-protocol raw data; According to the to-be-analyzed trajectory and the standard route corresponding to the target route, a trajectory image with the same geographical range and coordinate mapping relationship as the standard route is obtained, and the trajectory image is input into a recognition model for comparison to obtain a deviation area and a deviation direction of the to-be-analyzed trajectory relative to the standard route; According to the deviation area and the deviation direction, the deviation amount of the to-be-analyzed trajectory relative to the standard route is quantitatively calculated, and the deviation analysis result of the target route is generated according to the deviation amount.

[0005] In a possible implementation, the to-be-analyzed trajectory is generated by the following method: identifying and analyzing the protocol type of the multi-protocol raw data to extract a position coordinate sequence; Batch converting the position coordinate sequence from its original coordinate system to a predetermined unified geographical coordinate system to generate the to-be-analyzed trajectory.

[0006] In a possible implementation, the deviation area and the deviation direction are obtained by the following method: The to-be-analyzed trajectory and the standard route are rasterized respectively to generate a first trajectory image and a second trajectory image; performing pixel-level difference operation on the first trajectory image and the second trajectory image to obtain the offset region and the offset direction.

[0007] In a possible implementation, the offset amount is obtained by: based on the offset direction, calculating distances from points in the trajectory to be analyzed to the standard route in the offset region by using a Hausdorff distance algorithm based on a sliding window; based on the distances, generating statistical indicators including mean, variance and confidence interval, to determine the statistical indicators as the offset amount.

[0008] In a possible implementation, the method further includes: associating the offset analysis result with geographic scene information in which the offset occurs, the geographic scene information being determined based on a preset road type label library and an obstacle label library; performing cluster analysis on the associated data to generate analysis charts representing positioning offset rules in different scenes.

[0009] In a possible implementation, the method further includes: generating a structured report including abnormal statistics, trend analysis and improvement suggestions according to the offset analysis result and the analysis charts.

[0010] In a second aspect, the present application provides a trajectory offset detection device, the device comprising: an acquisition module configured to acquire multi-protocol raw data and generate a trajectory to be analyzed in a unified coordinate system according to the multi-protocol raw data; a trajectory generation module configured to obtain a trajectory image of the same geographic range and coordinate mapping relationship as a standard route according to the trajectory to be analyzed and the standard route corresponding to the trajectory to be analyzed, and input the trajectory image into a recognition model for comparison to obtain an offset region and an offset direction of the trajectory to be analyzed relative to the standard route; an analysis module configured to quantitatively calculate an offset amount of the trajectory to be analyzed relative to the standard route according to the offset region and the offset direction, and generate an offset analysis result according to the offset amount.

[0011] In a possible implementation, the acquisition module is further configured to: perform protocol type identification and analysis on the multi-protocol raw data to extract a position coordinate sequence; batch convert the position coordinate sequence from its original coordinate system to a predetermined unified geographic coordinate system to generate the trajectory to be analyzed.

[0012] In a third aspect, the present application also provides an electronic device, comprising a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor communicates with the memory through the bus, and the machine readable instructions are executed by the processor to perform the steps of the above method.

[0013] In a fourth aspect, the present application also provides a computer readable storage medium, the computer readable storage medium stores a computer program, when the computer program is executed by a processor, the steps of the above method are performed.

[0014] The present application provides a trajectory deviation detection method and device, an electronic device and a medium, wherein the method comprises: obtaining multi-protocol raw data, and generating a to-be-analyzed trajectory in a unified coordinate system according to the multi-protocol raw data; obtaining a trajectory image with the same geographical range and coordinate mapping relationship as a standard route according to the to-be-analyzed trajectory and the standard route corresponding to the to-be-analyzed trajectory, and inputting the trajectory image into an identification model for comparison to obtain a deviation area and a deviation direction of the to-be-analyzed trajectory relative to the standard route; and quantitatively calculating a deviation amount of the to-be-analyzed trajectory relative to the standard route according to the deviation area and the deviation direction, to generate a deviation analysis result according to the deviation amount. Through the present application, high-precision trajectory deviation detection and analysis of multi-source data unified processing and image comparison are realized.

[0015] In order to make the above objectives, features and advantages of the present application more apparent and understandable, the following will describe a preferred embodiment in detail, and the accompanying drawings will be described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments, and it should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0017] Figure 1 A flowchart of a trajectory deviation detection method provided by the embodiments of the present application; Figure 2 A structural schematic diagram of a trajectory deviation detection device provided by the embodiments of the present application; Figure 3 A structural schematic diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the following will be combined with the accompanying drawings of the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. The components of the embodiments of the present application generally described and shown in the accompanying drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, every other embodiment obtained by a person skilled in the art without creative work belongs to the scope protected by the present application.

[0019] Firstly, the application scenarios applicable to the present application are introduced. The present application can be applied to the technical field of trajectory analysis.

[0020] It is found through research that at present, positioning and trajectory analysis technology is crucial in many fields such as automobiles, logistics, agriculture, engineering surveying and mapping, and emergency rescue. However, the existing technology is generally limited to data post-processing at the application layer, and its core still relies on direct latitude and longitude sequence comparison, which faces significant bottlenecks in complex environments. For example, in signal-shielded areas such as urban canyons, tunnels, and viaducts, positioning data is prone to drift, jump, and path deviation, and traditional methods are difficult to automatically and accurately identify these abnormalities and their occurrence patterns. Existing solutions mainly rely on manual or semi-automatic methods for trajectory comparison and deviation analysis, which is inefficient and prone to omissions; they cannot accurately quantify the deviation distance between the actual trajectory and the standard route, and lack statistical analysis results; especially they fail to introduce image recognition technology to automatically compare trajectories and route images, thus failing to intuitively and efficiently analyze drift patterns. In addition, cross-industry, multi-protocol positioning data formats are not unified, and different coordinate systems (such as GCJ-02, BD-09, WGS-84, etc.) used by different map vendors are complex, making data alignment difficult, and the "standard route" itself may have systematic deviations, further affecting the accuracy of deviation analysis. In summary, the existing technology lacks a full-process intelligent solution from multi-source data access, automatic comparison, deviation quantification to scenario-based root cause analysis, and cannot effectively establish the causal relationship between the environment, module, and drift, limiting its application in large-scale, high-precision positioning data analysis.

[0021] Based on this, the embodiment of the present application provides a trajectory deviation detection method, device, electronic equipment and medium, which realizes automatic and high-precision analysis of multi-source heterogeneous positioning data, and achieves significant comprehensive effect: the processing efficiency is improved by orders of magnitude, meeting the demand of big data scene; the deviation quantization precision reaches meter level, and comprehensive statistical indicators are automatically generated, improving the credibility and direct applicability of the results; the drift root cause can be batch summarized, and regularity analysis is automatically output through scene clustering, helping product optimization and calibration, making the analysis results intuitive and visual, and greatly reducing the technical communication cost.

[0022] Please refer to Figure 1 , Figure 1 The embodiment of the present application provides a flow chart of a trajectory deviation detection method. As shown in Figure 1 The trajectory deviation detection method provided by the embodiment of the present application comprises the following steps: S101, obtaining multi-protocol original data of a target route, and generating a to-be-analyzed trajectory in a unified coordinate system according to the multi-protocol original data.

[0023] In the embodiment of the present application, the to-be-analyzed trajectory is generated in the following manner: The multi-protocol original data is identified and analyzed by protocol type to extract a position coordinate sequence; and the position coordinate sequence is batch converted from its original coordinate system to a predetermined unified geographic coordinate system to generate the to-be-analyzed trajectory.

[0024] Specifically, the original data stream of the target route collected from various heterogeneous data sources such as automobile TBOX, agricultural machinery positioning module, unmanned aerial vehicle positioning system, surveying and mapping receiver, logistics vehicle terminal, etc. is obtained, that is, multi-protocol original data. These data may comply with NMEA-0183, UBX, RTCM, CAN matrix, J1939, OBD-II or manufacturer's private protocol, etc. An AI protocol identification model can automatically identify and analyze the input data stream. The model extracts n-gram features from the data stream, and combines a support vector machine (SVM) classifier to establish a mapping relationship between the data stream syntax features and known protocol type labels, so as to automatically identify the structure of unknown or private protocols and generate an analysis template under the support of a small amount of samples. After analysis, the original position coordinate sequence is extracted. Subsequently, for different coordinate systems (such as WGS-84, GCJ-02, BD-09, UTM, etc.) that different map suppliers or data sources may use, the system performs batch coordinate conversion to unify all position coordinates to a predetermined unified geographic coordinate system (for example, WGS-84), aligns the trajectory data on the spatial reference, and forms the to-be-analyzed trajectory.

[0025] S102, obtain a track image with the same geographical range and coordinate mapping relationship as the standard route according to the standard route corresponding to the track to be analyzed, input the track image into the recognition model for comparison, to obtain the offset area and offset direction of the track to be analyzed relative to the standard route.

[0026] In the embodiment of the present application, the offset area and offset direction are obtained in the following way: The track to be analyzed and the standard route are rasterized respectively to generate a first track image and a second track image; pixel-level difference operation is performed on the first track image and the second track image to obtain the offset area and the offset direction.

[0027] Specifically, the track to be analyzed with the unified coordinate system generated in step S101 and the corresponding preset standard route are rasterized respectively. Specifically, the system converts the vector track data into a binary image (for example, the pixel points passed by the track are 1, and the rest are 0) according to the preset geographical range and resolution, generates a first track image representing the actual driving track and a second track image representing the standard route. The two images have the same geographical range, resolution and coordinate-to-pixel mapping relationship, ensuring the consistency of the basis for comparison. Then, the two images are input into a pre-trained deep learning model (for example, a convolutional neural network dedicated to image difference segmentation). The model can automatically and accurately identify the difference area of the first track image relative to the second track image, that is, the offset area, and further determine the direction of the offset (for example, left or right relative to the standard route) through pixel-level difference operation and feature learning.

[0028] S103, according to the offset area and the offset direction, quantitatively calculate the offset amount of the track to be analyzed relative to the standard route, to generate the offset analysis result of the target route according to the offset amount.

[0029] In the embodiment of the present application, the offset amount is obtained in the following way: The Hausdorff distance algorithm based on sliding window is used to calculate the distance from the points in the track to be analyzed to the standard route; based on the distance, statistical indicators including mean, variance and confidence interval are generated to determine the statistical indicators as the offset amount.

[0030] Specifically, after obtaining the offset region and direction, a precise quantitative calculation is performed. For the trajectory points in the identified offset region, a sliding window Hausdorff distance algorithm is used to calculate the maximum and minimum distances from the actual trajectory point set in the window to the standard route center line point set, as a measure of the offset degree of the window. By sliding the window through the entire trajectory, a series of offset data can be obtained. Based on these original offset data, the system further calculates and generates a plurality of statistical indicators including the mean, variance, 95% confidence interval, and process capability index (such as CP / CPK). These indicators collectively constitute a comprehensive quantitative description of the offset from the central tendency, dispersion, to the statistical reliability, forming the preliminary offset analysis results. This process realizes the transition from image-based qualitative differences to precise numerical measurements.

[0031] In the embodiments of the present application, further comprising: The offset analysis results are associated with the geographical scene information where the offset occurs, and the geographical scene information is determined based on a preset road type label library and an obstacle label library; and the associated data is subjected to cluster analysis to generate analysis charts representing the positioning offset regularity in different scenes.

[0032] Specifically, a plurality of scene label libraries can be established, including road type labels (such as expressway, tunnel, bridge, mountainous area, farmland, etc.) and obstacle environment labels (such as high-voltage line, tree shade, steel structure building, etc.). The offset analysis results (including offset amount and position) generated in step S103 are matched with the geographical position information where the offset occurs, and the corresponding road type and obstacle scene are automatically associated according to the label library. Then, an AI clustering model (such as K-means, DBSCAN, etc.) is used to perform cluster analysis on the offset data of these associated scene labels, and the typical patterns and influence laws of positioning offset in different scene combinations are summarized. Finally, the system automatically generates visual analysis charts, such as “drift heat map”, “scene-error correlation matrix”, or “industry drift trend report”, to intuitively display the error distribution in high-risk road sections, specific environments, etc., and realize the sedimentation of systematic knowledge from isolated problem points.

[0033] In the embodiments of the present application, further comprising: According to the offset analysis results and the analysis charts, a structured report including abnormal statistics, trend analysis, and improvement suggestions is generated.

[0034] Specifically, all the key information generated by the preceding steps, including the quantified offset statistical indicators, the offset area visualization results generated by image comparison, and the regular charts obtained from scene clustering analysis, are taken as inputs. Using integrated large language models (LLMs) and natural language generation techniques, the system automatically organizes, summarizes, and polishes these structured and unstructured data according to pre-set or customized report templates (such as templates conforming to ISO GPS accuracy standards, agricultural A-B line control requirements, or specific vehicle manufacturer road test acceptance formats) to generate complete technical reports. The report content covers abnormal point position statistical lists, offset amount trend analysis over time or space, risk area rankings, and targeted improvement suggestions based on historical data and scene regularity. The report can be output in multiple formats such as Word, PPT, PDF, and Excel, greatly improving the efficiency and standardization of data analysis to conclusion output.

[0035] Compared with the prior art, the present application realizes automatic and high-precision analysis of multi-source heterogeneous positioning data, achieving significant comprehensive effects: the processing efficiency is improved by orders of magnitude, meeting the needs of big data scenarios; the offset quantization precision reaches the meter level, and comprehensive statistical indicators are automatically generated, improving the reliability and direct applicability of the results; drift root causes can be summarized in batches, and regularity analysis is automatically output through scene clustering, helping product optimization and calibration, making the analysis results intuitive and visual, and greatly reducing technical communication costs.

[0036] Based on the same inventive concept, the present application also provides a trajectory offset detection device corresponding to the trajectory offset detection method. Since the principle of solving problems in the device of the present application is similar to the above-mentioned trajectory offset detection method of the present application, the implementation of the device can be referred to the implementation of the method, and the repeated parts will not be described again.

[0037] Please refer to Figure 2 , Figure 2 The structure diagram of a trajectory offset detection device provided by the present application embodiment, the trajectory offset detection device 200 comprises: The acquisition module 201 is configured to acquire multi-protocol raw data and generate a to-be-analyzed trajectory in a unified coordinate system according to the multi-protocol raw data.

[0038] The trajectory generation module 202 is configured to obtain a trajectory image with the same geographical range and coordinate mapping relationship as the standard route according to the to-be-analyzed trajectory and the standard route corresponding to the to-be-analyzed trajectory, and input the trajectory image into the recognition model for comparison to obtain an offset area and an offset direction of the to-be-analyzed trajectory relative to the standard route.

[0039] The analysis module 203 is configured to quantitatively calculate the offset amount of the to-be-analyzed trajectory relative to the standard route according to the offset area and the offset direction, and generate an offset analysis result according to the offset amount.

[0040] In the embodiment of the present application, the acquisition module 201 is further configured to: identify and analyze the multi-protocol raw data by protocol type to extract a sequence of location coordinates; and convert the sequence of location coordinates from its original coordinate system to a predetermined unified geographic coordinate system in batches to generate a trajectory to be analyzed.

[0041] Please refer to Figure 3 , Figure 3 A structural schematic diagram of an electronic device provided by an embodiment of the present application. As shown in Figure 3 The electronic device 300 includes a processor 310, a memory 320 and a bus 330.

[0042] The memory 320 stores machine-readable instructions executable by the processor 310. When the electronic device 300 is running, the processor 310 and the memory 320 communicate through the bus 330. The machine-readable instructions, when executed by the processor 310, can perform the steps of the above method. For details, refer to the method embodiments, which will not be repeated here.

[0043] The embodiment of the present application further provides a computer-readable storage medium, which stores a computer program. When the computer program is run by a processor, the steps of the above method can be executed. For details, refer to the method embodiments, which will not be repeated here.

[0044] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here.

[0045] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented by other ways. The device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some communication interface, device or unit, which can be electrical, mechanical or other forms.

[0046] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e. may be located in one place, or may be distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0047] In addition, the functional units in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0048] If the functions are realized in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer readable storage medium executable by a processor. Based on this understanding, the technical solutions of the present application or the part of the present application that essentially contributes to the prior art or the part of the technical solutions can be embodied in the form of a software product, which is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0049] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present application, used to illustrate the technical solutions of the present application, and not to limit them, the protection scope of the present application is not limited thereto, although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand: any skilled person in the art within the technical scope disclosed by the present application, they can still modify or easily think of changes to the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part of the technical features; and these modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and all should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A trajectory offset detection method, characterized in that, The method includes: Acquire multi-protocol raw data of the target route, and generate the trajectory to be analyzed in a unified coordinate system based on the multi-protocol raw data; Based on the standard route corresponding to the trajectory to be analyzed and the target route, a trajectory image with the same geographical range and coordinate mapping relationship as the standard route is obtained, and the trajectory image is input into the recognition model for comparison to obtain the offset area and offset direction of the trajectory to be analyzed relative to the standard route; Based on the offset region and the offset direction, the offset of the trajectory to be analyzed relative to the standard route is quantitatively calculated, so as to generate the offset analysis result of the target route based on the offset.

2. The method according to claim 1, characterized in that, The trajectory to be analyzed is generated in the following manner: The protocol type of the original multi-protocol data is identified and parsed to extract the location coordinate sequence; The location coordinate sequence is converted from the original coordinate system to a predetermined unified geographic coordinate system in batches to generate the trajectory to be analyzed.

3. The method according to claim 1, characterized in that, The offset region and the offset direction are obtained in the following way: The trajectory to be analyzed and the standard route are rasterized respectively to generate a first trajectory image and a second trajectory image; A pixel-level difference operation is performed on the first trajectory image and the second trajectory image to obtain the offset region and the offset direction.

4. The method according to claim 1, characterized in that, The offset is obtained in the following way: Based on the offset direction, the Hausdorff distance algorithm based on a sliding window is used within the offset region to calculate the distance from the point in the trajectory to be analyzed to the standard route. Based on the distance, a statistical indicator including mean, variance, and confidence interval is generated to determine the statistical indicator as the offset.

5. The method according to claim 1, characterized in that, Also includes: The offset analysis results are correlated with the geographic scene information where the offset occurred, and the geographic scene information is determined based on a preset road type label library and obstacle label library; Cluster analysis is performed on the correlated data to generate analytical charts representing the positioning offset patterns under different scenarios.

6. The method according to claim 5, characterized in that, Also includes: Based on the offset analysis results and the analysis charts, a structured report is generated that includes anomaly statistics, trend analysis, and improvement suggestions.

7. A trajectory offset detection device, characterized in that, The device includes: The acquisition module is used to acquire multi-protocol raw data of the target route and generate the trajectory to be analyzed in a unified coordinate system based on the multi-protocol raw data; The trajectory generation module is used to obtain a trajectory image with the same geographical range and coordinate mapping relationship as the standard route corresponding to the trajectory to be analyzed and the target route, and input the trajectory image into the recognition model for comparison to obtain the offset area and offset direction of the trajectory to be analyzed relative to the standard route; The analysis module is used to quantitatively calculate the offset of the trajectory to be analyzed relative to the standard route based on the offset region and the offset direction, so as to generate the offset analysis result of the target route based on the offset.

8. The apparatus according to claim 7, characterized in that, The acquisition module is also used for: The protocol type of the original multi-protocol data is identified and parsed to extract the location coordinate sequence; The location coordinate sequence is converted in batches from its original coordinate system to a predetermined unified geographic coordinate system to generate the trajectory to be analyzed.

9. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of the method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, performs the steps of the method as described in any one of claims 1 to 6.

Images