A tunnel map data multi-path association optimization method, device, medium and equipment

Abstract

The application provides a tunnel map data multi-path association optimization method and device, medium and equipment, and belongs to the technical field of map making. The application mainly includes a feature extraction process, a feature association process, a check and verification process, and a consistency optimization process. The beneficial effects of the application are that, among the multiple associated point cloud pairs obtained by associating the feature data point cloud clusters in the multi-path tunnel map data obtained through multiple collection tasks, unreliable associated point cloud pairs are screened out, and the map data is optimized under reliable multi-path association constraints, the accuracy of the tunnel map data is improved, and the map building precision is improved.

Description

Technical Field

[0001] This application relates to the field of map production technology, and in particular to a method, apparatus, medium and equipment for optimizing multi-path association of tunnel map data. Background Technology

[0002] When extracting data features from objects inside tunnels during field data collection for map creation, the tunnels only have four sides (top, bottom, front, and back) and few other objects, resulting in a limited number of features that can be extracted. Furthermore, the numerous repetitive scenes can easily lead to mismatches. In addition, the poor GPS signal in tunnels causes a large discrepancy between the collected data location and the actual location. Therefore, matching these features during map creation is quite difficult, resulting in poor map accuracy. Summary of the Invention

[0003] To address the problems existing in the prior art, this application mainly provides a method, apparatus, medium, and device for multi-path association optimization of tunnel map data. This method optimizes unreliable associated point cloud pairs among multiple associated point cloud pairs obtained by associating feature data point cloud clusters from multiple tunnel map data obtained from multiple acquisition tasks, and optimizes the map data under reliable multi-path association constraints, thereby improving the accuracy of tunnel map data and thus enhancing map construction accuracy.

[0004] To achieve the above objectives, one technical solution adopted in this application is: providing a method for optimizing multi-path association of tunnel map data, comprising:

[0005] The feature extraction process extracts multiple feature data point cloud clusters from multiple tunnel map data obtained from multiple acquisition tasks. The feature association process matches and associates one feature data point cloud cluster from one tunnel map data with the feature data point cloud cluster from the other tunnel map data based on their corresponding point cloud positions to obtain associated point cloud pairs. The verification process checks and verifies multiple associated point cloud pairs according to preset rules, filtering out unreliable associated point cloud pairs to obtain at least one reliable associated point cloud pair. The consistency optimization process constructs optimized trajectory point coordinates for each trajectory point based on its absolute position coordinates in the matched tunnel map data. Based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the coordinates of adjacent optimized trajectory points of each adjacent trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair, the pose of the matched tunnel map data is adjusted to obtain optimized tunnel map data.

[0006] Another technical solution adopted in this application is: providing a tunnel map data multi-path association optimization device, which includes:

[0007] The feature extraction module extracts multiple feature data point cloud clusters from multiple tunnel map data obtained from multiple acquisition tasks, resulting in two tunnel map data sets. The feature association module matches and associates one feature data point cloud cluster from one tunnel map data set with the feature data point cloud cluster from the other tunnel map data set based on their corresponding point cloud positions to obtain associated point cloud pairs. The inspection and verification module checks and verifies multiple associated point cloud pairs according to preset rules, filtering out unreliable associated point cloud pairs to obtain at least one reliable associated point cloud pair. The consistency optimization module constructs optimized trajectory point coordinates for each trajectory point based on its absolute position coordinates in the matched tunnel map data. Based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the coordinates of the adjacent optimized trajectory points of each trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair, the matched tunnel map data is pose-adjusted to obtain optimized tunnel map data.

[0008] Another technical solution adopted in this application is to provide a computer-readable storage medium storing computer instructions that are operated to execute the tunnel map data multipath association optimization method in the above solution.

[0009] Another technical solution adopted in this application is: providing a computer device, which includes a processor and a memory, the memory storing computer instructions, which are operated to execute the tunnel map data multipath association optimization method in the above solution.

[0010] The beneficial effects that the technical solution of this application can achieve are: providing a method, device, medium and equipment for multi-path association optimization of tunnel map data. This application filters out unreliable associated point cloud pairs from multiple associated point cloud pairs obtained by associating feature data point cloud clusters in multi-path tunnel map data obtained from multiple acquisition tasks, and optimizes the map data under reliable multi-path association constraints, thereby improving the accuracy of tunnel map data and thus improving the accuracy of map construction. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a specific implementation of a tunnel map data multi-path association optimization method according to this application;

[0013] Figure 2 This is a flowchart illustrating a specific embodiment of a tunnel map data multi-path association optimization method according to this application;

[0014] Figure 3 This is a flowchart illustrating a specific implementation of a tunnel map data multipath association optimization device according to this application;

[0015] Figure 4 This is a flowchart illustrating a specific embodiment of a tunnel map data multi-path association optimization method according to this application;

[0016] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0017] The preferred embodiments of this application will now be described in detail with reference to the accompanying drawings, so that the advantages and features of this application can be more easily understood by those skilled in the art, thereby providing a clearer and more definite definition of the scope of protection of this application.

[0018] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0019] When extracting data features from objects inside tunnels during field data collection for map creation, the limited number of extractable object features—tunnels essentially consist of four sides (left, right, up, and down), with few poles like lampposts and few signs appearing every few tens to hundreds of meters—and the high repetition of signs in the surrounding environment, leads to frequent mismatches during mapping. Furthermore, poor GPS signal strength inside tunnels results in significant discrepancies between the collected data location and the actual location, making accurate tunnel map data acquisition difficult and resulting in lower map creation accuracy.

[0020] This application provides a method, apparatus, medium, and device for optimizing multi-path association of tunnel map data. This application filters out unreliable associated point cloud pairs from multiple associated point cloud pairs obtained by associating feature data point cloud clusters in multi-path tunnel map data obtained from multiple acquisition tasks, and optimizes the map data under reliable multi-path association constraints to improve the accuracy of tunnel map data, thereby improving map mapping accuracy.

[0021] The technical solutions of this application will now be described in detail with reference to specific embodiments and accompanying drawings. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0022] Figure 1 This paper illustrates a specific implementation of a multi-path association optimization method for tunnel map data according to this application.

[0023] exist Figure 1 In the specific embodiment shown, the tunnel map data multi-path association optimization method of this application includes: a feature extraction process S101, extracting multiple feature data point cloud clusters from multiple tunnel map data obtained from multiple acquisition tasks; a feature association process S102, matching and associating one feature data point cloud cluster from one tunnel map data with the feature data point cloud cluster from the other tunnel map data according to the corresponding point cloud positions to obtain associated point cloud pairs; a check and verification process S103, checking and verifying multiple associated point cloud pairs according to preset rules, and filtering out unreliable associated point cloud pairs to obtain at least one reliable associated point cloud pair; and a consistency optimization process S104, constructing optimized trajectory point coordinates for each trajectory point based on the absolute position coordinates of each trajectory point in the matched tunnel map data; and adjusting the pose of the matched tunnel map data based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the adjacent optimized trajectory point coordinates of the trajectory points adjacent to each trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair to obtain optimized tunnel map data.

[0024] This application improves the accuracy of tunnel map data by filtering out unreliable associated point cloud pairs from multiple associated point cloud pairs obtained by associating feature data point cloud clusters from multiple tunnel map data obtained from multiple acquisition tasks, and optimizing the map data under reliable multi-path association constraints.

[0025] The feature extraction process S101 represents the extraction of multiple feature data point cloud clusters from the multi-channel tunnel map data obtained from multiple acquisition tasks. This facilitates the matching of the corresponding feature data point cloud clusters of the two channels and the filtering out of unreliable matches.

[0026] In an optional embodiment of this application, the aforementioned feature data point cloud clusters include aerial planar point cloud cluster pairs, ground planar point cloud cluster pairs, and vertical pole point cloud cluster pairs.

[0027] In an optional embodiment of this application, the aforementioned aerial planar point cloud clusters are extracted. Optionally, the aerial point clouds in the two tunnel map data are subjected to planar segmentation. Specifically, planar segmentation can be performed using existing planar segmentation methods, such as using a region growing algorithm.

[0028] Optionally, the planar point cloud clusters can be clustered to filter out aerial point clouds with lower density. Specifically, existing clustering methods can be selected, such as the DPSCAN clustering algorithm.

[0029] In an optional embodiment of this application, the above-mentioned process of extracting the aerial planar point cloud cluster includes taking the centroid of the aerial planar point cloud cluster as the planar point of the aerial planar point cloud cluster, and extracting the planar point as the plane of the aerial planar point cloud cluster. Specifically, the above-mentioned descriptor is used to describe the characteristics that distinguish one point cloud plane from other point cloud planes, including information such as the normal of the corresponding plane, the position of the corresponding plane relative to the GPS track point, and the flatness of the corresponding plane.

[0030] In an optional specific embodiment of this application, the above-mentioned process of extracting the above-mentioned aerial planar point cloud cluster includes obtaining the coordinate information of all points in the above-mentioned planar point cloud cluster.

[0031] The feature association process S102 represents the process of matching and associating a feature data point cloud cluster from one tunnel map data with the feature data point cloud cluster from the other tunnel map data according to their corresponding point cloud locations to obtain associated point cloud pairs. This facilitates the deletion of unreliable associated point cloud pairs and optimizes the map data under reliable multi-path association constraints, thereby improving the accuracy of the tunnel map data.

[0032] In an optional specific embodiment of this application, the process of matching and associating a feature data point cloud cluster from one tunnel map data source with a feature data point cloud cluster from the other tunnel map data source based on their respective point cloud locations to obtain associated point cloud pairs includes: matching a feature data point cloud cluster from one tunnel map data source with a feature data point cloud cluster from the other tunnel map data source based on their respective point cloud locations to obtain multiple aerial plane matching pairs; performing matching conversion on other aerial plane matching pairs based on the matching conversion distance between each of the multiple aerial plane matching pairs and calculating the matching conversion power of the other aerial plane matching pairs; and The matching conversion distance of the highest matching power of the other aerial plane matching pair is used as the matching conversion distance between one tunnel map data and another tunnel map data; and, based on the matching conversion distance of the most accurate matching pair, the tunnel map data is matched and converted to obtain feature data point cloud clusters in the matched tunnel map data, and the feature data point cloud clusters in the matched tunnel map data and the feature data point cloud clusters in the other tunnel map data are matched and associated according to their corresponding point cloud positions to obtain associated point cloud pairs; wherein, the matching conversion power of the highest matching power of the other aerial plane matching pair is not less than a predetermined ratio threshold, and the matching conversion distance is not greater than a predetermined first distance threshold.

[0033] By selecting the most accurate aerial planar point cloud clusters from the multi-channel tunnel map data obtained from multiple data collection tasks, and then performing relatively accurate feature association matching on the multi-channel data, the workload of deleting unreliable associated point cloud pairs can be reduced.

[0034] Specifically, the aforementioned predetermined proportion threshold can be 70%, and the aforementioned first distance threshold can be 0.3-0.8m.

[0035] The inspection and verification process S103 represents the inspection and verification of multiple associated point cloud pairs according to preset rules. Unreliable associated point cloud pairs are filtered out to obtain at least one reliable associated point cloud pair. This process can filter out unreliable associated point cloud pairs and leave reliable point cloud pairs to participate in the next consistency optimization process, thereby improving the accuracy of tunnel map data, increasing mapping efficiency, and enhancing map mapping accuracy.

[0036] In an optional specific embodiment of this application, the above-mentioned inspection and verification process S103 includes deleting the corresponding associated point cloud pairs whose distances abruptly change compared to the distances of other point cloud pairs, thereby obtaining at least one reliable associated point cloud pair.

[0037] Specifically, in a normal scenario, the distance between the two point cloud clusters in a correlated point cloud pair obtained from two data collection missions should be relatively small. For example, the horizontal and vertical deviations should be -1m and 2m respectively. Therefore, the distance between the corresponding point cloud clusters in the left and right correlated point cloud pairs should be close to this value. If extreme cases occur, such as -5m horizontally and 4m vertically, the correlated point cloud pair needs to be deleted and excluded from the subsequent optimization process. This mismatch occurs in reality because there are many repetitive scenes in tunnels. For example, a sign and a wall will appear at regular intervals. This can lead to the mismatch of point clouds from signs located in different positions being incorrectly matched as a single point cloud pair.

[0038] In an optional specific embodiment of this application, the above-mentioned inspection and verification process S103 includes, based on the location of the corresponding point pairs of multiple associated point cloud pairs in the corresponding tunnel, deleting the corresponding associated point cloud pairs whose point cloud distances do not conform to the Gaussian curve distribution when the point cloud distances of the multiple associated point cloud pairs are arranged in order from the tunnel entrance to the exit, so as to obtain at least one reliable associated point cloud pair.

[0039] Specifically, in real-world scenarios, GPS signals are relatively accurate at tunnel entrances and exits, resulting in more precise absolute positioning of the tunnel map data, thus reducing the likelihood of matching errors. However, deeper into the tunnel, GPS signals weaken, and the tunnel map data becomes less accurate, increasing the probability of matching errors. Therefore, when point cloud distances are arranged in order from tunnel entrance to exit, their values ​​conform to a Gaussian curve distribution. If a point cloud distance does not conform to this distribution, it can be determined that the corresponding associated point cloud pair is mismatched, and therefore, it should be deleted.

[0040] The consistency optimization process S104 represents the process of constructing optimized trajectory point coordinates for each trajectory point based on the absolute position coordinates of each trajectory point in the matched tunnel map data; and adjusting the pose of the matched tunnel map data based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the coordinates of the adjacent optimized trajectory points of each trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair to obtain optimized tunnel map data. This process can establish constraints based on reliable associated point cloud pairs, further optimize the consistency of the tunnel map data, obtain more accurate tunnel map data, and thus improve mapping accuracy.

[0041] In an optional specific embodiment of this application, the above-mentioned consistency optimization process S104 includes adjusting the coordinates of the optimized trajectory points such that the difference between the coordinates of each optimized trajectory point and the absolute position coordinates of the corresponding trajectory point is minimized and not greater than a predetermined second distance threshold, the difference between the coordinates of each optimized trajectory point and the coordinates of adjacent optimized trajectory points is minimized and not greater than a predetermined third distance threshold, and the point cloud pair distance between the two point cloud pairs of each reliable associated point cloud pair is minimized and not greater than a preset fourth distance threshold; and the pose adjustment of the matched tunnel map data is performed according to the adjusted optimized trajectory point coordinates.

[0042] Specifically, the absolute position of the trajectory points has a certain reference value, constraining the trajectory points to be optimized; the relative position between adjacent trajectory points also constrains the position of the optimized trajectory points, and this constraint can be established using trajectory extrapolation methods; furthermore, the aforementioned associated point cloud pairs are essentially composed of point cloud clusters representing the same objects in tunnel map data obtained from other acquisition tasks and point cloud clusters in the tunnel map data of this route, thus also constraining the position of the optimized trajectory points. By constructing these three constraints and solving for the optimal value, optionally using the least squares optimization method, the position of the optimized trajectory points can be obtained. Adjusting the pose of the tunnel map data according to the position of the optimized trajectory points yields more accurate and consistent tunnel map data.

[0043] In an optional specific embodiment of this application, the aforementioned reliable associated point cloud pairs include planar point cloud cluster pairs, ground planar point cloud cluster pairs, and vertical pole point cloud cluster pairs. The aforementioned consistency optimization process S104 further includes calculating the point-to-plane distance from a point in one aerial planar point cloud cluster to the plane containing another aerial planar point cloud cluster in the aerial planar point cloud cluster pair; calculating the point-to-plane distance from a point in one ground planar point cloud cluster to the plane containing another ground planar point cloud cluster in the ground planar point cloud cluster pair; and / or calculating the point-to-line distance from one vertical pole point cloud cluster to the line containing another vertical pole point cloud cluster in the vertical pole point cloud cluster pair, thereby obtaining the point cloud pair distance between the two point cloud clusters in each reliable associated point cloud pair.

[0044] Specifically, for example, when pairing two point cloud clusters, a and b, a point in a can be used to perform a nearest neighbor search on its spatial location in b, based on its absolute coordinates. For instance, if five points are found, a plane or a straight line is constructed using these five points. Then, the distance from the point in a to this plane or straight line in b is calculated. This gives us the point-plane distance or point-line distance, which serves as the distance between the two point cloud clusters in the associated pair. Calculating point-plane or point-line distances avoids inaccurate calculations of point cloud cluster distances when occlusion causes incomplete point clouds.

[0045] In optional embodiments of this application, the tunnel map data multipath association optimization method further includes an optimization check process, such as... Figure 2 As shown in process S205, the feature extraction process and the point cloud pair distance acquisition process are repeated on the optimized tunnel map data and the corresponding matched tunnel map data to obtain the point cloud pair distance; if the difference between the values ​​of the multiple point cloud pair distances is less than a preset difference threshold, the optimized tunnel map data is used as the target to optimize the tunnel map data; if the difference between the values ​​of the point cloud pair distances is not less than the preset difference threshold, the checking, verification and position adjustment process and the consistency optimization process are continued until the point cloud pair distance is less than the difference threshold.

[0046] In practice, the initially obtained reliable associated point cloud pairs may only include associated point cloud pairs composed of feature data point cloud clusters with relatively accurate GPS positioning at the tunnel entrances and exits. After each optimization, the positions of point cloud data in deeper parts of the tunnel can be optimized. Through multiple iterations, the optimization can be continuously pushed into the depths of the tunnel until the positions of all data point clouds in the tunnel map data are optimized.

[0047] Figure 3 This paper illustrates a specific embodiment of a tunnel map data multipath association optimization device according to this application.

[0048] exist Figure 3 In the specific embodiment shown, the tunnel map data multi-path association optimization device of this application includes: a feature extraction module 301, used to extract multiple feature data point cloud clusters of two tunnel map data from multiple tunnel map data obtained from multiple acquisition tasks; a feature association module 302, used to match and associate one feature data point cloud cluster of one tunnel map data with the feature data point cloud cluster of the other tunnel map data according to the corresponding point cloud positions to obtain associated point cloud pairs; an inspection and verification module 303, used to inspect and verify the multiple associated point cloud pairs according to preset rules, and filter out unreliable associated point cloud pairs to obtain at least one reliable associated point cloud pair; and a consistency optimization module 304, used to construct optimized trajectory point coordinates for each trajectory point based on the absolute position coordinates of each trajectory point in the matched tunnel map data; and to adjust the pose of the matched tunnel map data based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the adjacent optimized trajectory point coordinates of the trajectory points adjacent to each trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair to obtain optimized tunnel map data.

[0049] This application improves the accuracy of tunnel map data by filtering out unreliable associated point cloud pairs from multiple associated point cloud pairs obtained by associating feature data point cloud clusters from multiple tunnel map data obtained from multiple acquisition tasks, and optimizing the map data under reliable multi-path association constraints.

[0050] The feature extraction module 301 is used to extract multiple feature data point cloud clusters of two tunnel map data from multiple tunnel map data obtained from multiple acquisition tasks. It can facilitate the matching of the corresponding feature data point cloud clusters of the two data and filter out unreliable matches.

[0051] Feature association module 302 is used to match and associate one feature data point cloud cluster of one tunnel map data with the feature data point cloud cluster of the other tunnel map data according to the corresponding point cloud positions to obtain associated point cloud pairs. This facilitates the deletion of unreliable associated point cloud pairs and optimizes the map data under reliable multi-path association constraints, thereby improving the accuracy of tunnel map data.

[0052] The inspection and verification module 303 is used to check and verify multiple associated point cloud pairs according to preset rules, and to filter out unreliable associated point cloud pairs to obtain at least one reliable associated point cloud pair. It can filter out unreliable associated point cloud pairs and leave reliable point cloud pairs to participate in the next consistency optimization process, which can improve the accuracy of tunnel map data, improve mapping efficiency, and improve map mapping accuracy.

[0053] The consistency optimization module 304 is used to construct optimized trajectory point coordinates for each trajectory point based on its absolute position coordinates in the matched tunnel map data. Based on these optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the coordinates of adjacent optimized trajectory points of each adjacent trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair, the module adjusts the pose of the matched tunnel map data to obtain optimized tunnel map data. This module can establish constraints based on reliable associated point cloud pairs to further optimize the consistency of the tunnel map data, resulting in more accurate tunnel map data and improving mapping accuracy. Specifically, the absolute position of the trajectory point has a certain reference value, constraining the trajectory points to be optimized; the relative position between adjacent trajectory points also constrains the position of the optimized trajectory point, which can be established using trajectory extrapolation methods; furthermore, the aforementioned associated point cloud pairs are essentially composed of point cloud clusters representing the same objects in tunnel map data obtained from other acquisition tasks and point cloud clusters in the current tunnel map data, thus also constraining the position of the optimized trajectory points. By constructing these three constraints and solving for the optimal value, the position of the optimized trajectory point can be obtained by using the least squares optimization method. The tunnel map data can then be adjusted in pose according to the position of the optimized trajectory point to obtain more accurate and consistent tunnel map data.

[0054] In an optional embodiment of this application, the tunnel map data multipath association optimization device further includes an optimization checking device, such as... Figure 4 As shown in module 405, it is used to re-perform the feature extraction process and the point cloud pair distance acquisition process on the optimized tunnel map data and the corresponding matched tunnel map data to obtain the point cloud pair distance; if the difference between the values ​​of the multiple point cloud pair distances is less than a preset difference threshold, then the optimized tunnel map data is used as the target to optimize the tunnel map data; if the difference between the values ​​of the point cloud pair distances is not less than the preset difference threshold, then the checking and verification and position adjustment process and the consistency optimization process are continued until the point cloud pair distance is less than the difference threshold.

[0055] In practice, the initially obtained reliable associated point cloud pairs may only include associated point cloud pairs composed of feature data point cloud clusters with relatively accurate GPS positioning at the tunnel entrances and exits. After each optimization, the positions of point cloud data in deeper parts of the tunnel can be optimized. Through multiple iterations, the optimization can be continuously pushed into the depths of the tunnel until the positions of all data point clouds in the tunnel map data are optimized.

[0056] The extended tunnel map data multi-path association optimization device provided in this application can be used to execute the extended tunnel map data multi-path association optimization method described in any of the above embodiments. Its implementation principle and technical effect are similar, and will not be repeated here.

[0057] In one specific embodiment of this application, the functional modules in the tunnel map data multipath association optimization device of this application can be directly in hardware, in software modules executed by a processor, or in a combination of both.

[0058] Software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in this art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium.

[0059] The processor can 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, discrete hardware components, or any combination thereof. A general-purpose processor can be a microprocessor, but alternatively, it can be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors incorporating a DSP core, or any other such configuration. Alternatively, the storage medium can be integrated with the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in the user terminal. Alternatively, the processor and storage medium can reside as discrete components in the user terminal.

[0060] In another specific embodiment of this application, a computer-readable storage medium stores computer instructions that are operated to perform the extended tunnel map data multipath association optimization method described above.

[0061] In another specific embodiment of this application, a computer device includes a processor and a memory, the memory storing computer instructions that are operated to execute the tunnel map data multipath association optimization method in the above scheme.

[0062] Another technical solution adopted in this application is: to provide a map production system, which includes a tunnel map data multipath association optimization device, which is used in the tunnel map data multipath association optimization method in the above solution.

[0063] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0064] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0065] The above are merely embodiments of this application and do not limit the scope of this patent application. Any equivalent structural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of this application.

Claims

1. A method for optimizing multi-path association of tunnel map data, characterized in that, include, The feature extraction process extracts multiple feature data point clusters from two tunnel map data sources obtained from multiple collection tasks. The feature association process involves matching and associating a feature data point cloud cluster from one tunnel map data source with a feature data point cloud cluster from the other tunnel map data source based on their corresponding point cloud locations to obtain associated point cloud pairs. The multiple feature data point cloud clusters include aerial planar point cloud clusters representing aerial objects. The feature association process includes... The aerial plane point cloud clusters of one tunnel map data and the aerial plane point cloud clusters of the other tunnel map data are matched according to their corresponding point cloud positions to obtain multiple aerial plane matching pairs. Based on the matching transformation distance between each of the plurality of aerial plane matching pairs, matching transformation is performed on the other aerial plane matching pairs among the plurality of aerial plane matching pairs, and the matching transformation power of the other aerial plane matching pairs is calculated; and the matching transformation distance of the aerial plane matching pair corresponding to the highest matching transformation power of the other aerial plane matching pair is taken as the matching transformation distance between one tunnel map data and another tunnel map data; and... Based on the matching transformation distance of the most accurate matching pair, the tunnel map data of one channel is matched and transformed to obtain the feature data point cloud cluster in the matched tunnel map data. The feature data point cloud cluster in the matched tunnel map data is then matched and associated with the feature data point cloud cluster in the other tunnel map data according to the corresponding point cloud positions to obtain associated point cloud pairs. Wherein, the matching conversion power of the highest other aerial plane matching pair is not less than a predetermined proportional threshold, and the matching conversion distance is not greater than a predetermined first distance threshold; The verification process involves checking and verifying the multiple associated point cloud pairs according to preset rules, and filtering out unreliable associated point cloud pairs to obtain at least one reliable associated point cloud pair. The consistency optimization process involves constructing optimized trajectory point coordinates for each trajectory point based on its absolute position coordinates in the matched tunnel map data; and then adjusting the pose of the matched tunnel map data based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the adjacent optimized trajectory point coordinates of the trajectory points adjacent to each trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair to obtain optimized tunnel map data.

2. The tunnel map data multi-path association optimization method according to claim 1, characterized in that, The inspection and verification process includes, Among the multiple associated point cloud pairs, the associated point cloud pairs whose distances abruptly change compared to the distances of other point cloud pairs are deleted to obtain at least one reliable associated point cloud pair.

3. The tunnel map data multi-path association optimization method according to claim 1, characterized in that, The inspection and verification process includes, Based on the location of the corresponding point pairs in the corresponding tunnels, when the point cloud distances of the multiple associated point cloud pairs are arranged in order from the tunnel entrance to the exit, the corresponding associated point cloud pairs whose point cloud distances do not conform to the Gaussian curve distribution are deleted, thereby obtaining the at least one reliable associated point cloud pair.

4. The tunnel map data multi-path association optimization method according to claim 1, characterized in that, It also includes, The inspection process is optimized by re-performing the feature extraction process and the point cloud pair distance acquisition process on the optimized tunnel map data and the corresponding matched tunnel map data to obtain the point cloud pair distance. If the difference between the distance values ​​of the multiple point cloud pairs is less than a preset difference threshold, then the optimized tunnel map data is used as the location target to optimize the tunnel map data; if the difference between the distance values ​​of the point cloud pairs is not less than the preset difference threshold, then the checking, verification, and position adjustment process, as well as the consistency optimization process, continue until the distance between the point cloud pairs is less than the difference threshold.

5. The tunnel map data multi-path association optimization method according to claim 1, characterized in that, The process of adjusting the pose of the matched tunnel map data based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the coordinates of the adjacent optimized trajectory points of each trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair includes: The coordinates of the optimized trajectory points are adjusted such that the difference between the coordinates of each optimized trajectory point and the absolute position coordinates of the corresponding trajectory point is minimized and not greater than a predetermined second distance threshold, the difference between the coordinates of each optimized trajectory point and the coordinates of adjacent optimized trajectory points is minimized and not greater than a predetermined third distance threshold, and the point cloud pair distance between the two point cloud pairs of each reliable associated point cloud pair is minimized and not greater than a preset fourth distance threshold; the pose of the matched tunnel map data is adjusted according to the adjusted coordinates of the optimized trajectory points.

6. The tunnel map data multi-path association optimization method according to claim 5, characterized in that, The feature data point cloud clusters include aerial planar point cloud clusters, ground planar point cloud clusters, and vertical pole point clouds; the reliable associated point cloud pairs include aerial planar point cloud cluster pairs, ground planar point cloud cluster pairs, and vertical pole point cloud cluster pairs. The process of adjusting the pose of the matched tunnel map data based on the distance between the reliable associated point cloud pairs further includes... Calculate the point-to-plane distance from a point in one aerial plane point cloud cluster to the plane containing another aerial plane point cloud cluster in the aerial plane point cloud cluster pair; calculate the point-to-plane distance from a point in one ground plane point cloud cluster to the plane containing another ground plane point cloud cluster in the ground plane point cloud cluster pair; and / or calculate the point-to-line distance from one vertical pole point cloud cluster to the line containing another vertical pole point cloud cluster in the vertical pole point cloud cluster pair, to obtain the point cloud pair distance between the two point cloud clusters in each of the reliable associated point cloud pairs.

7. A tunnel map data multi-path correlation optimization device, characterized in that, include, The feature extraction module is used to extract multiple feature data point clusters from the multi-channel tunnel map data obtained from multiple acquisition tasks to obtain two tunnel map data. The feature association module is used to match and associate a feature data point cloud cluster from one tunnel map data source with a feature data point cloud cluster from the other tunnel map data source based on their respective point cloud locations to obtain associated point cloud pairs. The multiple feature data point cloud clusters include aerial planar point cloud clusters representing aerial objects. The feature association module includes matching an aerial planar point cloud cluster from one tunnel map data source with an aerial planar point cloud cluster from the other tunnel map data source based on their respective point cloud locations to obtain multiple aerial planar matching pairs; and performing matching transformation on other aerial planar matching pairs based on the matching transformation distance between each aerial planar matching pair and calculating the other aerial planar matching pairs. The matching conversion power of the mid-plane matching pair is calculated; and the matching conversion distance of the aerial plane matching pair corresponding to the highest matching conversion power is used as the matching conversion distance between one tunnel map data and another tunnel map data; and, based on the matching conversion distance of the most accurate matching pair, the tunnel map data is matched and converted to obtain feature data point cloud clusters in the matched tunnel map data, and the feature data point cloud clusters in the matched tunnel map data and the feature data point cloud clusters in the other tunnel map data are matched and associated according to their corresponding point cloud positions to obtain associated point cloud pairs; wherein, the matching conversion power of the highest aerial plane matching pair is not less than a predetermined proportion threshold, and the matching conversion distance is not greater than a predetermined first distance threshold; The inspection and verification module is used to inspect and verify the multiple associated point cloud pairs according to preset rules, and to filter out unreliable associated point cloud pairs to obtain at least one reliable associated point cloud pair. The consistency optimization module is used to construct optimized trajectory point coordinates for each trajectory point based on the absolute position coordinates of each trajectory point in the matched tunnel map data; and to perform pose adjustment on the matched tunnel map data based on the optimized trajectory point coordinates, the absolute position coordinates of the trajectory points, the adjacent optimized trajectory point coordinates of the trajectory points adjacent to each trajectory point, and the point cloud pair distance between the two point cloud clusters of each reliable associated point cloud pair to obtain optimized tunnel map data.

8. A computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are operated to perform the tunnel map data multipath association optimization method according to any one of claims 1-6.

9. A computer device comprising a processor and a memory storing computer instructions, wherein the processor operates the computer instructions to perform the tunnel map data multipath association optimization method according to any one of claims 1-6.

Images