Method, device, storage medium and computer equipment for constructing geomagnetic navigation reference map

Abstract

The present application relates to the field of geomagnetic navigation technology, and discloses a method, device, storage medium, and computer equipment for constructing a geomagnetic navigation reference map. The method comprises: obtaining a geomagnetic sequence of at least one vehicle when it is traveling in the area to be mapped; matching the geomagnetic sequence with the atomic path in the area to be mapped based on the position data in the geomagnetic sequence to determine the path sequence of the atomic path; clustering the path sequence based on the number of targets to determine the target sequence of the atomic path; performing position alignment on the target sequence to determine the calibration sequence of the atomic path; performing position point-by-position averaging on the calibration sequence to determine the fusion sequence of the atomic path; and constructing a geomagnetic navigation reference map of the area to be mapped based on the fusion sequence. The present application aims to reduce the cost of constructing a geomagnetic navigation reference map. On the basis of ensuring the accuracy of mapping, it realizes a geomagnetic navigation reference map construction process that rapidly acquires geomagnetic data and refines the fusion of multi-source geomagnetic data.

Description

Technical Field

[0001] The present application relates to the field of geomagnetic navigation technology, and in particular to a method, device, storage medium and computer equipment for constructing a geomagnetic navigation reference map. Background Art

[0002] Geomagnetic navigation technology involves measuring magnetic field information, constructing a geomagnetic reference map, and designing geomagnetic positioning methods. The construction of a high-precision geomagnetic reference map is fundamental and crucial, requiring extensive and precise geomagnetic measurements. However, these extensive and precise measurements require significant human and material resources, resulting in high map construction costs and severely restricting the implementation of geomagnetic navigation technology.

[0003] In related technologies, crowdsourcing solutions introduced to reduce costs have improved data collection efficiency, but have also led to serious data heterogeneity issues, making it difficult to ensure data quality. Furthermore, there are issues with low data discernibility due to heterogeneous devices, user data fluctuations, and massive amounts of crowdsourced data in large-scale scenarios. Summary of the Invention

[0004] In view of this, the present application provides a method, device, storage medium and computer equipment for constructing a geomagnetic navigation reference map, with the goal of reducing the cost of constructing a geomagnetic navigation reference map. On the basis of ensuring the accuracy of mapping, it realizes the geomagnetic navigation reference map construction process of rapid acquisition of geomagnetic data, preprocessing of multi-source geomagnetic data, and fusion and refinement of multi-source geomagnetic data.

[0005] According to one aspect of the present application, a method for constructing a geomagnetic navigation reference map is provided, comprising:

[0006] Acquire a geomagnetic sequence of at least one vehicle when it is traveling in the area to be mapped, the geomagnetic sequence including geomagnetic data and its corresponding position data;

[0007] According to the position data in the geomagnetic sequence, the geomagnetic sequence is matched with the atomic paths in the area to be mapped to determine a path sequence of the atomic paths, where the atomic paths are determined according to the roads in the area to be mapped;

[0008] performing clustering processing on the path sequence according to the number of targets to determine the target sequence of the atomic path;

[0009] Performing position alignment on the target sequence to determine a calibration sequence of the atomic path;

[0010] performing position-by-position averaging processing on the calibration sequence to determine a fusion sequence of the atomic path;

[0011] A geomagnetic navigation reference map of the area to be mapped is constructed according to the fusion sequence.

[0012] According to another aspect of the present application, a device for constructing a geomagnetic navigation reference map is provided, comprising:

[0013] an acquisition module, configured to acquire a geomagnetic sequence of at least one vehicle when it is traveling in the area to be mapped, the geomagnetic sequence including geomagnetic data and its corresponding position data;

[0014] a processing module, configured to match the geomagnetic sequence with the atomic paths in the area to be mapped based on the position data in the geomagnetic sequence to determine a path sequence of the atomic paths, where the atomic paths are determined based on the roads in the area to be mapped; and to cluster the path sequences based on the number of targets to determine a target sequence of the atomic paths; and to positionally align the target sequences to determine a calibration sequence of the atomic paths; and to average the calibration sequences point by point to determine a fusion sequence of the atomic paths;

[0015] A construction module is used to construct a geomagnetic navigation reference map of the area to be mapped according to the fusion sequence.

[0016] According to another aspect of the present application, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the above-mentioned geomagnetic navigation reference map construction method are implemented.

[0017] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor. When the processor executes the program, the steps of the above-mentioned method for constructing a geomagnetic navigation reference map are implemented.

[0018] By means of the above technical solution, the present application provides a method, device, storage medium and computer equipment for constructing a geomagnetic navigation reference map, which adopts a crowdsourcing acquisition method to perform low-cost perception of the geomagnetic features of the mapping area, obtain the geomagnetic sequence of the mapping area, simplify the acquisition process and reduce the difficulty of acquisition. In addition, the roads in the mapping area are subdivided to obtain atomic paths, so as to attribute the geomagnetic sequences to different atomic paths, realize the conversion of crowdsourced disordered data to ordered data, and provide a complete data foundation for subsequent refined mapping. Then, the multi-source, massive and heterogeneous path sequences of the atomic paths are clustered to select representative and consistent target sequences from the numerous path sequences, exclude data with large deviations and that do not conform to the overall characteristics, and improve the data quality of the mapping. Then, the different target sequences of the atomic paths are aligned on the time axis to achieve sequence calibration, eliminate the sequence inconsistency caused by various factors in the crowdsourcing acquisition process, and obtain a calibration sequence. Further, the calibration sequence of the atomic path is averaged point by point to obtain a fused sequence by fusing the calibration sequence, so that the fused sequence is closer to the true geomagnetic value and improves the accuracy of the fused sequence. Therefore, a geomagnetic navigation reference map of the area to be mapped is constructed according to the fusion sequence of different atomic paths, thereby improving the accuracy and reliability of the geomagnetic navigation reference map.

[0019] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0021] Figure 1 A schematic diagram showing a flow chart of a method for constructing a geomagnetic navigation reference map provided in an embodiment of the present application is shown;

[0022] Figure 2 A distribution diagram of a segment of a scenario provided by an embodiment of the present application is shown;

[0023] Figure 3 The following shows a road section distribution diagram for scenario 2 provided in an embodiment of the present application;

[0024] Figure 4 A schematic diagram of the path boundary of the atomic path in scenario 2 provided in an embodiment of the present application is shown;

[0025] Figure 5A comparison diagram of geomagnetic data collected by a mobile phone in a geomagnetic sequence provided in an embodiment of the present application is shown;

[0026] Figure 6 A comparison chart of the accuracy of using the average value of the target sequence as the fusion sequence provided in the embodiment of the present application is shown;

[0027] Figure 7 A comparison chart of the accuracy of using the average value of the calibration sequence as the fusion sequence provided in the embodiment of the present application is shown;

[0028] Figure 8 The structural block diagram of the geomagnetic navigation reference map construction device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0029] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0030] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and are not to be construed as limiting the present application.

[0031] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of this application refers to the presence of the described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, or there may be intermediate elements. In addition, "connected" or "connected" as used herein may include wireless connection or wireless fusion. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0032] Now, exemplary embodiments according to the present application will be described in more detail with reference to the accompanying drawings. However, these exemplary embodiments may be implemented in a variety of different forms and should not be construed as being limited to the embodiments set forth herein. It should be understood that these embodiments are provided to make the disclosure of this application thorough and complete and to fully convey the concepts of these exemplary embodiments to those of ordinary skill in the art.

[0033] With the rapid development of society and the accelerating pace of informatization, accurate and timely location-based services have become a vital aid to social production and convenient living. The full completion of the deployment of my country's Beidou global satellite navigation system constellation will provide high-precision, all-weather continuous positioning coverage worldwide, empowering various industries with improved and enhanced capabilities. Despite this, current satellite navigation technology still faces numerous drawbacks, including signal interference, signal spoofing, signal suppression, and even attacks on positioning infrastructure, making it difficult to meet the robust positioning requirements in extreme environments, such as indoors and in enclosed areas. Geomagnetic navigation, on the other hand, offers advantages such as strong interference and damage resistance, a wide range of usable areas, no cumulative error, and is passive and highly concealed. Furthermore, it requires no pre-installed signal sources, resulting in extremely low hardware investment costs. Therefore, geomagnetic navigation technology is expected to provide emergency and reliable positioning support in extreme environments, such as unknown areas, highly damaged areas, and abnormal weather.

[0034] Geomagnetic navigation technology includes three aspects: magnetic field information measurement, geomagnetic reference map construction, and geomagnetic positioning method design. Among them, the construction of a high-precision geomagnetic reference map is the foundation and core, which usually requires a large amount of accurate geomagnetic measurements. However, a large number of accurate measurements require a lot of manpower and material costs, resulting in high map construction costs, which seriously restricts the implementation of geomagnetic navigation technology. In related technologies, although crowdsourcing collection avoids the difficulties of large-scale in-depth collection and effectively improves collection efficiency, many uncontrollable factors in the collection method make it difficult to guarantee data quality. In addition, there are still problems with low data recognition caused by heterogeneous devices, user data fluctuations, and massive crowdsourcing data in large-scale scenarios.

[0035] In order to solve the above problems, a method for constructing a geomagnetic navigation reference map is provided in this embodiment. Figure 1 As shown, the method includes:

[0036] Step 101: Obtain a geomagnetic sequence of at least one vehicle when it is traveling in an area to be mapped.

[0037] The geomagnetic sequence includes geomagnetic data and its corresponding position data.

[0038] It should be noted that the principle of geomagnetic navigation is based on the variations in the Earth's magnetic field at different locations and directions. To build a high-precision geomagnetic navigation reference map, it is necessary to measure geomagnetic data and corresponding position data at countless locations to obtain detailed geomagnetic fingerprints.

[0039] In this embodiment, a crowdsourcing collection method is adopted, with vehicles as carriers, to achieve dynamic collection of geomagnetic fingerprints based on a large number of users and general mobile terminals.

[0040] Specifically, during the data collection process, the user controls the vehicle within the area to be mapped. Simultaneously, the user uses their mobile device to record the location data of different collection points along the route, along with the corresponding geomagnetic data, in real time. This allows the recorded location data and corresponding geomagnetic data for each trip to be arranged into a geomagnetic sequence, chronologically.

[0041] Here, a crowdsourcing approach is used to collect location and geomagnetic data from multiple vehicles to rapidly construct a geomagnetic navigation reference map for a large area. This data is collected using mobile devices with positioning and geomagnetic measurement capabilities, such as cell phones, resulting in low-cost and rapid data collection.

[0042] For example, the location data may be longitude and latitude, and the geomagnetic data may be the geomagnetic field strength.

[0043] It is worth mentioning that during the collection process, there is no need to pre-specify the user's collection section and driving direction, and there is no need to specify the user's collection device type and collection posture, which greatly reduces the collection difficulty and the requirements for users.

[0044] Considering the negative impact of complex road conditions on geomagnetic data during vehicle-based collection, congested roads should be avoided and collection should be conducted at night whenever possible to prevent prolonged vehicle parking and magnetic field interference from nearby vehicles, which can lead to extremely low accuracy and render the data unusable for geomagnetic map construction. Furthermore, the vehicle speed should be kept below or equal to 20 km / h to avoid unstable offsets in the geomagnetic sequence caused by excessive speed, and increased collection time and costs caused by excessive speed.

[0045] Understandably, during the crowdsourcing process, multiple users collected data at different road sections and distances, resulting in spatial misalignment of the geomagnetic data. Furthermore, differences in vehicle speed, collection time, collection posture, and equipment further contribute to inconsistencies in the geomagnetic sequences.

[0046] Step 102 : Match the geomagnetic sequence with the atomic paths in the area to be mapped based on the position data in the geomagnetic sequence to determine the path sequence of the atomic paths.

[0047] Among them, the atomic path is determined according to the roads in the area to be mapped.

[0048] In this embodiment, based on the road network distribution of the area to be mapped, different roads in the area are segmented at their intersections, further refining the roads in the area to be mapped into atomic paths. This allows the massive amount of geomagnetic sequences collected to be matched with the atomic paths, transforming crowdsourced disordered data into stable and ordered data. This meets the requirements of fused mapping generated by different mobile terminals and provides a comprehensive data foundation for subsequent refined mapping.

[0049] Step 103: cluster the path sequence according to the number of targets to determine the target sequence of the atomic path.

[0050] The number of targets is determined according to the number of path sequences and the geomagnetic data in the path sequences.

[0051] It should be noted that when there are no abnormal behaviors such as vehicle movement or lane changes during crowdsourcing data collection, the path sequences collected by all mobile terminals should have basically the same trend. However, in actual application scenarios, it is impossible to guarantee that there will be no abnormalities.

[0052] In this embodiment, the atomic path is used as the smallest unit to cluster and filter the path sequence of the atomic path, eliminate abnormal path sequences whose trends obviously deviate from the group in the crowdsourced data, and select high-quality path sequences with consistent trends to form the optimal cluster as the target sequence of the atomic path, providing a reliable, complete and expressive data foundation for subsequent steps, and effectively improving the accuracy and reliability of fusion mapping.

[0053] Step 104: perform position alignment on the target sequence to determine the calibration sequence of the atomic path.

[0054] It should be noted that within the optimal cluster selected after atomic path selection, the overall trend of the target sequence should theoretically be consistent, and the average target sequence within the optimal cluster can be directly selected as the final fusion result for that atomic path. However, in practical application scenarios, the presence of vehicles passing each other during crowdsourcing collection, congestion on certain road sections, and inconsistent sensor update frequencies within mobile terminals can all cause lags in geomagnetic key points such as peaks and troughs, which can lead to significant errors when directly fusion is performed on the target sequence.

[0055] In this embodiment, based on the geomagnetic data in the target sequence, the target sequence after atomic path clustering is aligned on the time axis to adapt to non-equal length sequences, achieve calibration, obtain a calibrated sequence, improve data accuracy, solve the deviation problem of actual collected data, and provide a more reliable basis for sequence fusion in subsequent steps.

[0056] Step 105 : Perform position-by-position averaging processing on the calibration sequence to determine the fusion sequence of the atomic path.

[0057] In this embodiment, the calibration sequence of the atomic path is averaged point by point to achieve fusion of the calibration sequences, thereby obtaining a fused sequence that can accurately reflect the actual geomagnetic conditions of the atomic path, thereby improving the accuracy of the fused sequence.

[0058] Step 106: construct a geomagnetic navigation reference map of the area to be mapped based on the fusion sequence.

[0059] In this embodiment, a geomagnetic navigation reference map of the area to be mapped is accurately constructed based on the fusion sequence of different atomic paths.

[0060] Furthermore, as a refinement and expansion of the specific implementation methods of the above-mentioned embodiment, in order to fully illustrate the specific implementation process of this embodiment, the geomagnetic sequence is matched with the atomic path in the area to be mapped according to the position data in the geomagnetic sequence, and the path sequence of the atomic path is determined, specifically including: matching the position data in the geomagnetic sequence with the coordinate data of the atomic path to generate a matching degree between the geomagnetic sequence and the atomic path; using the geomagnetic sequence corresponding to the target matching degree as the path sequence of the atomic path corresponding to the target matching degree, and the target matching degree is a matching degree greater than a first preset threshold.

[0061] In this embodiment, different geomagnetic sequences collected by multi-source terminals are matched on each atomic path, thereby achieving the initial orderly organization of massive crowd-sourced disordered data and providing a complete data foundation for subsequent refined mapping.

[0062] Specifically, a path boundary is set for the atomic path based on its shape. For example, if the atomic path is a straight road, a rectangular boundary is set along the outer perimeter of the atomic path. Next, the complete coordinate data of the atomic path within the path boundary is obtained using a positioning device such as a GPS (Global Positioning System). It can be understood that the coordinate data of the atomic path is continuous and orderly distributed.

[0063] Then, by comparing the position data in the geomagnetic sequence with the coordinate data of the atomic path, it is determined whether the position data in the geomagnetic sequence is within the path boundary of the atomic path, and the degree of overlap between the vehicle's driving trajectory and the atomic path when the geomagnetic sequence is collected is quantified, thereby generating a matching degree between the geomagnetic sequence and the atomic path. For example, for any geomagnetic sequence and any atomic path, if the position data in the geomagnetic sequence contains the complete coordinate data on the atomic path, the matching degree of the geomagnetic sequence and the atomic path can be set to 1. If the position data in the geomagnetic sequence contains most of the coordinate data on the atomic path, and the position data in the geomagnetic sequence is highly consistent with the coordinate data of the atomic path in terms of sequence and spatial distribution, without obvious jumps or confusion, the matching degree of the geomagnetic sequence and the atomic path can be set to a higher value, such as 0.85. On the contrary, if the position data in the geomagnetic sequence only contains a small amount of coordinate data on the atomic path, and these overlapping data are relatively scattered and do not show obvious rules or order, the matching degree of the geomagnetic sequence and the atomic path can be set to a lower value, such as 0.2.

[0064] Furthermore, a match degree greater than a first preset threshold is used as a target match degree. The geomagnetic sequence corresponding to the target match degree is attributed to the atomic path corresponding to the target match degree, that is, the geomagnetic sequence corresponding to the target match degree is used as the path sequence of the atomic path corresponding to the target match degree. In this way, the path sequence of each atomic path in the area to be mapped can be obtained.

[0065] Here, the first preset threshold is specifically set according to the accuracy required for data collection and mapping in actual application scenarios, and this embodiment does not impose any specific restrictions. For example, the first preset threshold is set to 0.8.

[0066] Furthermore, as a refinement and extension of the specific implementation methods of the above-mentioned embodiments, in order to fully illustrate the specific implementation process of this embodiment, the geomagnetic navigation reference map construction method also includes: removing the first target position data of the coordinate data that does not belong to the atomic path in the path sequence and the geomagnetic data corresponding to the first target position data to update the path sequence; obtaining the number of available satellites at the collection point corresponding to the position data in the path sequence; removing the second target position data in the path sequence whose corresponding number of available satellites is less than a second preset threshold and the geomagnetic data corresponding to the second target position data to update the path sequence; and preprocessing the path sequence using the average difference method to update the path sequence.

[0067] In this embodiment, the path sequence of the matched atomic path is optimized to improve the quality and usability of the path sequence.

[0068] Specifically, for any path sequence of any atomic path: in the process of crowdsourcing collection, some noise data or data irrelevant to the current atomic path may be introduced. In this embodiment, the position data of the coordinate data that does not belong to the atomic path in the path sequence is used as the first target position data, and the first target position data and its corresponding geomagnetic data are removed from the path sequence to update the path sequence and reduce the interference of irrelevant data. In addition, the number of available satellites at the collection point corresponding to the position data in the path sequence is obtained from the positioning software of the mobile terminal used to collect the data, and the position data corresponding to the available satellite data in the path sequence is less than the second preset threshold as the second target position data. For the unreliable second target position data with low available satellite data in the path sequence, it is directly eliminated to update the path sequence and ensure positioning accuracy. At the same time, the missing jump points in the path sequence are supplemented by the average interpolation method to update the path sequence, reduce the noise and fluctuation in the path sequence, make the path sequence more stable and continuous, and further improve the data quality of the path sequence.

[0069] Here, the second preset threshold is specifically set according to the positioning software used, and this embodiment does not impose any limitation thereto.

[0070] Furthermore, as a refinement and extension of the specific implementation methods of the above-mentioned embodiment, in order to fully illustrate the specific implementation process of this embodiment, the path sequence is clustered according to the target number to determine the target sequence of the atomic path, specifically including: randomly selecting the target number of path sequences as cluster centers; calculating the Euclidean distances from the path sequence to different cluster centers based on the geomagnetic data in the path sequence; sorting the Euclidean distances to obtain a distance order, and adding the path sequence to the cluster to which the cluster center that ranks first in the distance order belongs; updating the cluster centers within the cluster according to the average value of the path sequences within the cluster until the cluster reaches the preset convergence condition; sorting the clusters that reach the preset convergence condition according to the number of path sequences within the cluster that reach the preset convergence condition to determine the cluster order; and taking the path sequence within the cluster that ranks first in the cluster order as the target sequence.

[0071] In this embodiment, for a large number of path sequences of atomic paths, a K-means clustering algorithm with low computational complexity is used to screen out abnormal path sequences with trends that obviously deviate from group behavior, thereby quickly screening out high-quality path sequences, i.e., target sequences.

[0072] For example, for any atomic path, the path sequence set defining the atomic path is ,in, is the number of path sequences in the path sequence set of the atomic path, They are the first path sequence, the second path sequence, ..., the The dimension of geomagnetic data in the path sequence is m, and here m is set to 4, which means that the geomagnetic data is in x 、 y 、 z The components in the direction and the size of the overall geomagnetic data. Based on the distribution differences of the path sequences, this embodiment needs to divide the path sequence set of the atomic path into k '(i.e. the target number) clusters, expressed as: ,in, They are the first cluster, the second cluster, ..., the k For each cluster, the average of the path sequences within the cluster can be used as the centroid to measure the similarity of the path sequences within the cluster and the differences between clusters.

[0073] It is worth mentioning that in this embodiment, K-means clustering is performed based on minimizing the square error of the path sequence within the cluster, which can be expressed as:

[0074] ,

[0075] in, is the sum of the squares of the Euclidean distances from each path sequence to the cluster center within the cluster to which it belongs, Indicates the i Clusters S i Internal path sequence With the i Clusters S i Cluster Center The square of the Euclidean distance between them.

[0076] The goal of K-means clustering in this embodiment is to find The smallest cluster center. Specifically, it includes the following steps:

[0077] Step 1: Use random initialization, that is, randomly select from the path sequence set of atomic paths k 'Path sequences are used as initial cluster centers;

[0078] Step 2: Based on the minimum distance principle, for each path sequence in the path sequence set of atomic paths, calculate the Euclidean distance from the path sequence to each cluster center, select the cluster center closest to the path sequence, and use the path sequence as a member of the cluster to which the closest cluster center belongs. That is, sort the Euclidean distances to obtain the distance order, and add the path sequence to the cluster to which the cluster center that ranks first in the distance order belongs.

[0079] Step 3: Calculate the average value of the path sequence in each cluster respectively, and use the average value of the path sequence in each cluster as the new cluster center of the cluster. S i Cluster Center The update is expressed as: ;

[0080] Step 4: Repeat steps 2 and 3 until the clusters no longer change significantly. The clustering process is considered to have converged, thereby avoiding unnecessary iterations and effectively saving computing resources. For example, in this embodiment, the iteration is stopped when the change in cluster centers between two consecutive iterations is less than a threshold.

[0081] Therefore, the cluster containing the largest number of path sequences after the iteration is selected as the optimal cluster, and the path sequence within the optimal cluster is used as the target sequence of the atomic path (that is, according to the number of path sequences within the cluster that meet the preset convergence conditions, the clusters that meet the preset convergence conditions are sorted, the cluster order is determined, and the path sequence within the cluster that ranks first in the cluster order is used as the target sequence), and the remaining clusters are eliminated as abnormal data.

[0082] It should be noted that the calculations involved in this embodiment, such as the Euclidean distance and the average value of the path sequence, are all calculated based on the geomagnetic data in the path sequence.

[0083] Furthermore, as a refinement and extension of the specific implementation methods of the above-mentioned embodiment, in order to fully illustrate the specific implementation process of this embodiment, the geomagnetic navigation reference map construction method also includes: clustering the path sequence to obtain a preset number of comparison clusters; determining the sequence cohesion of the path sequence based on the average Euclidean distance from the path sequence to other path sequences in the comparison cluster to which it belongs; determining the sequence separation of the path sequence based on the average Euclidean distance from the path sequence to the path sequences in its neighboring clusters, the neighboring cluster of the path sequence being the comparison cluster with the smallest Euclidean distance to the path sequence among other comparison clusters excluding the comparison cluster to which the path sequence belongs; determining the silhouette coefficient of the path sequence based on the sequence cohesion and sequence separation of the path sequence; determining the average silhouette coefficient of the sequence under the preset number of clusters based on the average value of the silhouette coefficients of the path sequence; and screening out the target number from the preset number based on the average silhouette coefficient of the sequence.

[0084] It's important to note that for the K-means clustering algorithm, the number of clusters affects not only the algorithm's execution efficiency but also the accuracy of path sequence classification. For crowdsourced data collection, different atomic paths vary in length and road conditions, leading to significant variations in the overall distribution of path sequences. This makes uniformly setting a value for all atomic paths highly likely to result in poor fusion results for certain sections.

[0085] In this embodiment, before performing K-means clustering processing, the number of clusters for each atomic path cluster is adaptively set based on the silhouette coefficient method according to the uniqueness of the path sequence set of each atomic path, thereby improving the reliability of the clustering processing.

[0086] Specifically, for any set of atomic path sequences: First, set the K-means clustering by enumeration. k The range of values, for example, k The value ranges from 2 to 10 (i.e. the preset number). For each k value, according to the k Perform K-means clustering on the path sequence set of the atomic path and get k For each path sequence in the set of atomic path sequences, the average Euclidean distance from that path sequence to all other path sequences in its comparison cluster is calculated and used as the sequence cohesion of that path sequence. Furthermore, the average Euclidean distance from that path sequence to all path sequences in its closest cluster (i.e., neighboring cluster) is calculated and used as the sequence separation of that path sequence. The effectiveness of clustering is evaluated by combining sequence cohesion and sequence separation.

[0087] For example, the path sequence set of atomic paths Any path sequence in ( j =1, .., n) Expressed as:

[0088] ,

[0089] in, for The contrast cluster to which it belongs, for Remove Other path sequences than for The number of path sequences in .

[0090] For example, the path sequence set of atomic paths Middle path sequence Sequence separation Expressed as:

[0091] ,

[0092] in, For the exception Belonging In addition, the other comparison clusters The nearest contrast cluster is the neighboring cluster. Here, we calculate The Euclidean distances between the cluster centers of other contrasting clusters are sorted from small to large to obtain a contrast sorting, and the other contrasting clusters corresponding to the Euclidean distances that are ranked first in the contrast sorting are used as neighboring clusters. for Internal path sequence. for The number of inner path sequences.

[0093] Therefore, the silhouette coefficient of the path sequence is calculated according to the sequence cohesion and sequence separation of the path sequence, so as to reflect the clustering quality of the path sequence through the silhouette coefficient.

[0094] For example, the path sequence set of atomic paths Middle path sequence Silhouette coefficient Expressed as:

[0095] ,

[0096] in, express and The maximum value.

[0097] Furthermore, the calculation is based on k When the values ​​are clustered, the average value of the silhouette coefficients of all path sequences is the sequence average silhouette coefficient.

[0098] For example, according to k When clustering is performed on the values, the average silhouette coefficient of the sequence Expressed as:

[0099] .

[0100] It should be noted that for different k The average silhouette coefficient of the sequences under the value clustering, the value with the largest average silhouette coefficient has better clustering quality.

[0101] In this embodiment, the sequence with the largest average silhouette coefficient is k The target number of atomic paths to cluster is used to make the clusters have high cohesion and good separation.

[0102] It should be noted that the Euclidean distance involved in this embodiment is calculated based on the geomagnetic data in the path sequence.

[0103] Furthermore, as a refinement and expansion of the specific implementation methods of the above-mentioned embodiment, in order to fully illustrate the specific implementation process of this embodiment, the target sequence is positionally aligned and the calibration sequence of the atomic path is determined, specifically including: based on the geomagnetic data in the target sequence, using Euclidean distance to calculate the cumulative distance matrix between the target sequence and other target sequences; based on the cumulative distance matrix, determining the matching path between the target sequence and other target sequences, the matching path including the correspondence between the geomagnetic data in the target sequence and the geomagnetic data in the other target sequences; taking the average value of the geomagnetic data corresponding to the geomagnetic data in the target sequence in other target sequences as the calibration result of the geomagnetic data in the target sequence; based on the calibration result of the geomagnetic data in the target sequence, updating the target sequence to obtain the calibration sequence.

[0104] In this embodiment, the DTW (Dynamic Time Warping) algorithm is used to align the target sequences of the atomic paths, thereby achieving calibration of the target sequences.

[0105] It should be noted that the DTW algorithm is primarily used to evaluate the similarity between two sequences. The DTW distance can better characterize the temporal similarity of sequences of unequal length. Furthermore, the DTW algorithm can achieve nonlinear "distortion" by stretching or shrinking the time axis of a sequence.

[0106] For example, for any target sequence of any atomic path :When the target sequence When calibrating, first, the DTW algorithm is used to construct the target sequence Other target sequences with this atomic path ( , is the cumulative distance matrix between the target sequence set of atomic paths. It should be noted that the cumulative distance matrix is ​​calculated based on the geomagnetic data in the target sequence using Euclidean distance.

[0107] Next, continue to use the DTW algorithm to find the target sequence on the cumulative distance matrix With other target sequences The shortest path between them, and take the shortest path as the target sequence With other target sequences Matching paths between It should be noted that the matching path is used to represent the corresponding relationship between geomagnetic data in different target sequences.

[0108] The length is Target sequence and For example, the matching path can be expressed as: . Where K is the length of the matching path, (k=1,2,...,K) is the matching path The kth element in Represents the target sequence Middle m Geomagnetic data With other target sequences Middle n' Geomagnetic data Corresponding.

[0109] It is worth mentioning that there may be a one-to-many correspondence between the geomagnetic data in the target sequence.

[0110] Furthermore, when all other sequences are obtained To the target sequence After matching the path, from the target sequence The first geomagnetic data Start by matching the path Determine all other sequences Neutralization of target sequence The first geomagnetic data The corresponding geomagnetic data, and all other sequences Neutralization of target sequence The first geomagnetic data The average value of the corresponding geomagnetic data is used as the target sequence The first geomagnetic data The calibration results, and so on, determine the target sequence The calibration results of all geomagnetic data in the target sequence are completed. The calibrated target sequence is obtained , that is, the calibration sequence . Then all calibration sequences of atomic paths can be obtained.

[0111] Furthermore, as a refinement and extension of the specific implementation methods of the above-mentioned embodiment, in order to fully illustrate the specific implementation process of this embodiment, the calibration sequence is averaged point by point to determine the fusion sequence of the atomic path, specifically including: using the position data in the calibration sequence as the calibration position point; averaging the geomagnetic data corresponding to the calibration position point in different calibration sequences to obtain the fused geomagnetic data corresponding to the calibration position point; and determining the fusion sequence of the atomic path based on the fused geomagnetic data corresponding to the calibration position point.

[0112] It is understandable that each geomagnetic data in the calibration sequence corresponds to unique position data when it is collected.

[0113] In this embodiment, the geomagnetic data corresponding to the same position in all calibration sequences of the atomic path are averaged to achieve point-by-point averaging of the calibration sequence, reduce the measurement deviation of the geomagnetic data, and thus fuse all calibration sequences of the atomic path into a fused sequence, thereby improving the accuracy of the fused sequence.

[0114] Specifically, for the calibration sequence of any atomic path: all position data involved in the calibration sequence of the atomic path are extracted and used as calibration position points. For any calibration position point, the geomagnetic data corresponding to the calibration position point in different calibration sequences are averaged to obtain the fused geomagnetic data corresponding to the calibration position point. Then, the calibration position points can be sorted along the length direction of the atomic path, thereby forming a fused sequence based on the sorted calibration position points and their corresponding fused geomagnetic data, so that the fused sequence can more accurately reflect the geomagnetic conditions of the atomic path.

[0115] For example, the position-by-position averaging process of the calibration sequence can be expressed as:

[0116] ,

[0117] in, is the fusion sequence, p is the number of calibration sequences.

[0118] This application provides a method, device, storage medium and computer equipment for constructing a geomagnetic navigation reference map. Based on the road network cutting method, it realizes the conversion of disordered crowdsourcing data into a path sequence of atomic paths, providing high-quality data support for map construction. In addition, for the path sequence on each atomic path, the K-means algorithm and k Value selection enables filtering of atomic path sequences with low computational complexity. Furthermore, the DTW calibration mechanism further reduces differences in geomagnetic acquisition methods and achieves feature alignment of the atomic path target sequence, effectively improving the accuracy of the final fused sequence.

[0119] In one embodiment, the geomagnetic navigation reference map construction method described in this application is respectively carried out in two experimental scenes with large differences. Considering that the road sections in the area to be mapped are usually relatively dense, the geomagnetic navigation reference map construction method described in this application must not only ensure the mapping accuracy but also ensure the operating efficiency. Therefore, the mapping accuracy test of the geomagnetic navigation reference map construction method described in this application is carried out for scene 1. The distribution of the road sections in the scene is as follows Figure 2 As shown, it includes a target street, a first target road and a second target road. The roads are unobstructed, the environment is open and the length is long. It is mainly used to verify the accuracy and efficiency of the geomagnetic navigation reference map construction method described in this application. Among them, the target street is a north-south road with a total length of about 6.5 kilometers and 12 atomic paths; the first target road is an east-west road with a total length of about 4 kilometers and 8 atomic paths; the second target road is an east-west road with a total length of 4.5 kilometers and 10 atomic paths. In the actual positioning scenario, due to the complex environment, the collected geomagnetic sequence is easily disturbed by the surrounding environment and produces obvious fluctuations. This requires that the matching between the geomagnetic sequence and the atomic path can more accurately identify the changes in the shape of the geomagnetic sequence. Therefore, the test of the matching between the geomagnetic sequence and the atomic path is carried out in Scene 2 where the surrounding environment is more complex. Scene 2 is as shown Figure 3 As shown in the figure, the atomic paths are divided by the turning points of the paths, including 6 atomic paths with a length range of [106 m, 622 m], and there are many buildings around them. Figure 4 The red dots represent the four vertices on the path boundary of the atomic path in Scenario 2, and the blue rectangle represents the path restriction area within the path boundary of the atomic path. Before the positioning test, the geomagnetic navigation reference map construction method described in this application was used to complete the construction of the geomagnetic reference map in this scenario.

[0120] Crowdsourcing data collection primarily uses commercially available mobile phones with built-in, highly sensitive magnetometers, specifically brands A, B, C, and D. Based on research and data analysis, the magnetometer information for different phone types is shown in Table 1. In scenario one, during crowdsourcing data collection, to ensure the required data volume, vehicle speeds were kept to a minimum, generally controlled at 20 km / h. Furthermore, to fully leverage the crowdsourcing advantage of "compensating for precision with quantity," each phone collected data at least twice on each road section. Therefore, each atomic path contains at least eight geomagnetic sequences. However, the length of geomagnetic data sequences collected by different atomic paths varies depending on factors such as path length, speed bumps encountered during the collection process, lane changes and encounters, and sensor inconsistencies between mobile phones. Therefore, the overall length of geomagnetic sequences corresponding to atomic paths ranges from 200 to 1000. In scenario two, during crowdsourcing data collection, the length of a single acquisition for all atomic paths was approximately 7100 sequences. To verify the accuracy of the geomagnetic navigation reference map construction method described in this application, a mag690-fl100 three-axis fluxgate probe was used to collect geomagnetic reference values. The acquisition frequency was 10-60 kHz, with an accuracy of <0.5% of the reading ±10 nT (nanotesla). The magnetometer was supported on a tripod with a level in the vehicle's central armrest. A computer onboard the vehicle was connected to the magnetometer via a wired connection, and dedicated control software was used to capture and store real-time magnetometer readings.

[0121] Table 1

[0122]

[0123] First, verify that the K-means algorithm k The clustering performance under the selected value is compared in Table 2. The data in the table shows the percentage of target sequence accuracy of some atomic paths less than 200 nT. As can be seen from the table, this application can achieve good results as a whole, but the optimal value of different atomic paths is not fixed, which shows that k Necessity of value selection.

[0124] Table 2

[0125]

[0126] Furthermore, to verify the accuracy of the fused sequence for the atomic path, Table 3 compares the accuracy of individual geomagnetic sequences collected on a single atomic path in two batches using four different devices, as well as the accuracy of the fused sequence after fusing all geomagnetic sequences. The data in Table 3 demonstrates that the fused sequence obtained after applying the K-means algorithm is more consistent with the geomagnetic reference value, thus validating the effectiveness of the geomagnetic navigation reference map construction method described in this application.

[0127] Table 3

[0128]

[0129] To verify the effectiveness of the DTW calibration mechanism, this example compares the accuracy of using the target sequence average as the fusion sequence and the calibration sequence average as the fusion sequence. Figure 5 As shown, each type of mobile phone is collected twice on this atomic path. Figure 5 In this embodiment, the geomagnetic sequence is first differentially processed, and then clustered and calibrated to obtain the target sequence and the calibration sequence. The accuracy comparison results of the target sequence average value as the fusion sequence are shown in Figure 2. Figure 6 As shown in the figure, the accuracy comparison results of using the average value of the calibration sequence as the fusion sequence are shown in the figure. Figure 7 As shown. It can be seen that Figure 6 The magnetic field intensity of the fusion sequence shows large fluctuations in the entire observation interval, and there is a significant offset from the reference sequence at many peaks and troughs, which directly leads to poor accuracy. Figure 7 After DTW calibration, the fusion sequence has obvious improvements in the amplitude and change trend of magnetic field intensity.

[0130] It should be noted that the accuracy of the above sequence is determined by the ratio of the geomagnetic data in the sequence to its corresponding geomagnetic reference value.

[0131] It should be noted that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0132] Furthermore, if Figure 8 As shown, as a specific implementation of the above-mentioned geomagnetic navigation reference map construction method, an embodiment of the present application provides a geomagnetic navigation reference map construction device 800, and the geomagnetic navigation reference map construction device 200 includes: an acquisition module 801, a processing module 802 and a construction module 803.

[0133] The acquisition module 801 is used to acquire a geomagnetic sequence of at least one vehicle when it is traveling in the area to be mapped. The geomagnetic sequence includes geomagnetic data and its corresponding position data.

[0134] Processing module 802 is configured to match the geomagnetic sequence with atomic paths in the area to be mapped based on the position data in the geomagnetic sequence to determine a path sequence of the atomic paths, where the atomic paths are determined based on the roads in the area to be mapped; cluster the path sequences based on the number of targets to determine a target sequence of the atomic paths; align the target sequences to determine a calibration sequence of the atomic paths; and average the calibration sequences point by point to determine a fused sequence of the atomic paths.

[0135] The construction module 803 is used to construct a geomagnetic navigation reference map of the area to be mapped according to the fusion sequence.

[0136] Optionally, the processing module 802 is specifically used to match the position data in the geomagnetic sequence with the coordinate data of the atomic path to generate a matching degree between the geomagnetic sequence and the atomic path; the geomagnetic sequence corresponding to the target matching degree is used as the path sequence of the atomic path corresponding to the target matching degree, and the target matching degree is a matching degree greater than a first preset threshold.

[0137] Optionally, the processing module 802 is specifically used to remove the first target position data of the coordinate data that does not belong to the atomic path in the path sequence and the geomagnetic data corresponding to the first target position data to update the path sequence; obtain the number of available satellites at the collection point corresponding to the position data in the path sequence; remove the second target position data in the path sequence whose corresponding number of available satellites is less than a second preset threshold and the geomagnetic data corresponding to the second target position data to update the path sequence; and pre-process the path sequence using the average difference method to update the path sequence.

[0138] Optionally, the processing module 802 is specifically used to randomly select a target number of path sequences as cluster centers; calculate the Euclidean distances from the path sequences to different cluster centers based on the geomagnetic data in the path sequences; sort the Euclidean distances to obtain a distance order, and add the path sequences to the cluster to which the cluster center that ranks first in the distance order belongs; update the cluster centers within the cluster based on the average value of the path sequences within the cluster until the cluster reaches a preset convergence condition; sort the clusters that reach the preset convergence condition based on the number of path sequences within the cluster that reach the preset convergence condition to determine the cluster order; and use the path sequence within the cluster that ranks first in the cluster order as the target sequence.

[0139] Optionally, the processing module 802 is specifically configured to perform clustering processing on the path sequence to obtain a preset number of comparison clusters; determine the sequence cohesion of the path sequence based on the average Euclidean distance between the path sequence and other path sequences in the comparison cluster to which it belongs; determine the sequence separation of the path sequence based on the average Euclidean distance between the path sequence and path sequences in its neighboring clusters, where the neighboring clusters of the path sequence are the comparison clusters with the smallest Euclidean distance to the path sequence among other comparison clusters excluding the comparison cluster to which the path sequence belongs; determine the silhouette coefficient of the path sequence based on the sequence cohesion and sequence separation of the path sequence; determine the average silhouette coefficient of the sequence under the preset number of clusters based on the average value of the silhouette coefficients of the path sequence; and screen out the target number from the preset number based on the average silhouette coefficient of the sequence.

[0140] Optionally, the processing module 802 is specifically used to calculate the cumulative distance matrix between the target sequence and other target sequences based on the geomagnetic data in the target sequence using Euclidean distance; determine the matching path between the target sequence and other target sequences based on the cumulative distance matrix, and the matching path includes the correspondence between the geomagnetic data in the target sequence and the geomagnetic data in other target sequences; take the average value of the geomagnetic data corresponding to the geomagnetic data in the target sequence in other target sequences as the calibration result of the geomagnetic data in the target sequence; update the target sequence based on the calibration result of the geomagnetic data in the target sequence to obtain a calibration sequence.

[0141] Optionally, the processing module 802 is specifically used to use the position data in the calibration sequence as the calibration position point; average the geomagnetic data corresponding to the calibration position point in different calibration sequences to obtain the fused geomagnetic data corresponding to the calibration position point; and determine the fused sequence of the atomic path based on the fused geomagnetic data corresponding to the calibration position point.

[0142] For the specific definition of the geomagnetic navigation reference map construction device, please refer to the definition of the geomagnetic navigation reference map construction method above, and will not be repeated here. The various modules in the above-mentioned geomagnetic navigation reference map construction device can be implemented in whole or in part by software, hardware, or a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0143] Based on the above Figure 1 The method shown in FIG. 1 is a method for performing the above-mentioned operation. Accordingly, the embodiment of the present application further provides a readable storage medium having a computer program stored thereon. When the computer program is executed by the processor, the computer program is executed as shown in FIG. Figure 1 The method for constructing the geomagnetic navigation reference map is shown.

[0144] Based on this understanding, the technical solution of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, USB flash drive, mobile hard disk, etc.), and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each implementation scenario of the present application.

[0145] Based on the above Figure 1 The method shown, and Figure 8 In order to achieve the above-mentioned purpose, the embodiment of the present application further provides a computer device, which can be a personal computer, a server, a network device, etc. The computer device includes a storage medium and a processor; the storage medium is used to store a computer program; the processor is used to execute the computer program to achieve the above-mentioned Figure 1 The method for constructing the geomagnetic navigation reference map is shown.

[0146] Optionally, the computer device may also include a user interface, a network interface, a camera, a radio frequency (RF) circuit, a sensor, an audio circuit, a Wi-Fi module, etc. The user interface may include a display, an input unit such as a keyboard, etc. Optional user interfaces may also include a USB interface, a card reader interface, etc. The network interface may optionally include a standard wired interface, a wireless interface (such as a Bluetooth interface, a Wi-Fi interface), etc.

[0147] Those skilled in the art will understand that the computer device structure provided in this embodiment does not constitute a limitation on the computer device, and may include more or fewer components, or a combination of certain components, or different component arrangements.

[0148] The storage medium may also include an operating system and a network communication module. An operating system is a program that manages and stores the hardware and software resources of a computer device, supporting the execution of information processing programs and other software and / or programs. The network communication module facilitates communication between components within the storage medium, as well as with other hardware and software within the physical device.

[0149] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present application can be implemented with the help of software plus the necessary general hardware platform, and can also implement the embodiments of the present application through hardware.

[0150] Those skilled in the art will understand that the accompanying drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the accompanying drawings are not necessarily required to implement the present application. Those skilled in the art will understand that the modules in the devices in the implementation scenario can be distributed in the devices of the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more devices different from the implementation scenario. The modules of the above-mentioned implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0151] The serial numbers of the above application are for descriptive purposes only and do not represent the advantages or disadvantages of the implementation scenarios. The above disclosure only discloses several specific implementation scenarios of the present application, but the present application is not limited thereto. Any changes that can be conceived by those skilled in the art should fall within the scope of protection of the present application.

Claims

1. A method for constructing a geomagnetic navigation reference map, characterized in that: The method comprises: Acquire a geomagnetic sequence of at least one vehicle when it is traveling in the area to be mapped, the geomagnetic sequence including geomagnetic data and its corresponding position data; According to the position data in the geomagnetic sequence, the geomagnetic sequence is matched with the atomic paths in the area to be mapped to determine a path sequence of the atomic paths, where the atomic paths are determined according to the roads in the area to be mapped; performing clustering processing on the path sequence according to the number of targets to determine the target sequence of the atomic path, wherein the number of targets is determined according to the number of the path sequences and the geomagnetic data in the path sequence; Performing position alignment on the target sequence to determine a calibration sequence of the atomic path; performing position-by-position averaging processing on the calibration sequence to determine a fusion sequence of the atomic path; A geomagnetic navigation reference map of the area to be mapped is constructed according to the fusion sequence.

2. The method for constructing a geomagnetic navigation reference map according to claim 1, wherein: The matching of the geomagnetic sequence with the atomic paths in the area to be mapped based on the position data in the geomagnetic sequence to determine the path sequence of the atomic paths specifically includes: Matching the position data in the geomagnetic sequence with the coordinate data of the atomic path to generate a matching degree between the geomagnetic sequence and the atomic path; The geomagnetic sequence corresponding to the target matching degree is used as the path sequence of the atomic path corresponding to the target matching degree, and the target matching degree is the matching degree greater than a first preset threshold.

3. The method for constructing a geomagnetic navigation reference map according to claim 1, wherein: The method further comprises: removing the first target position data of the coordinate data that does not belong to the atomic path in the path sequence and the geomagnetic data corresponding to the first target position data to update the path sequence; Obtaining the number of available satellites at a collection point corresponding to the position data in the path sequence; Removing the second target position data corresponding to the path sequence and the geomagnetic data corresponding to the second target position data, for which the number of available satellites is less than a second preset threshold, to update the path sequence; The path sequence is preprocessed by using an average difference method to update the path sequence.

4. The method for constructing a geomagnetic navigation reference map according to claim 1, wherein: The clustering process of the path sequence according to the number of targets to determine the target sequence of the atomic path specifically includes: Randomly selecting the target number of path sequences as cluster centers; Calculating the Euclidean distance between the path sequence and different cluster centers according to the geomagnetic data in the path sequence; Sorting the Euclidean distances to obtain a distance order, and adding the path sequence to the cluster to which the cluster center that is first in the distance order belongs; updating the cluster center within the cluster according to the average value of the path sequence within the cluster until the cluster reaches a preset convergence condition; Sorting the clusters that meet the preset convergence condition according to the number of the path sequences in the clusters that meet the preset convergence condition to determine a cluster order; The path sequence in the cluster that is first in the cluster order is used as the target sequence.

5. The method for constructing a geomagnetic navigation reference map according to claim 1, wherein: The method further comprises: Performing clustering processing on the path sequence to obtain a preset number of comparison clusters; determining the sequence cohesion of the path sequence according to the average Euclidean distance between the path sequence and other path sequences in the comparison cluster to which the path sequence belongs; Determine the sequence separation of the pathway sequence based on the average Euclidean distance between the pathway sequence and the pathway sequences in its neighboring clusters, wherein the neighboring cluster of the pathway sequence is the comparison cluster with the smallest Euclidean distance to the pathway sequence among the comparison clusters other than the comparison cluster to which the pathway sequence belongs; determining a silhouette coefficient of the path sequence according to the sequence cohesion and sequence separation of the path sequence; Determining an average silhouette coefficient of the sequence under the preset number of clusters according to the average silhouette coefficient of the path sequence; The target number is selected from the preset number according to the sequence average silhouette coefficient.

6. The method for constructing a geomagnetic navigation reference map according to claim 1, wherein: The position alignment of the target sequence to determine the calibration sequence of the atomic path specifically includes: Calculating the cumulative distance matrix between the target sequence and other target sequences using Euclidean distance according to the geomagnetic data in the target sequence; determining, according to the cumulative distance matrix, a matching path between the target sequence and the other target sequences, wherein the matching path includes a corresponding relationship between the geomagnetic data in the target sequence and the geomagnetic data in the other target sequences; taking the average value of the geomagnetic data in the target sequence and the corresponding geomagnetic data in other target sequences as the calibration result of the geomagnetic data in the target sequence; The target sequence is updated according to the calibration result of the geomagnetic data in the target sequence to obtain the calibration sequence.

7. The method for constructing a geomagnetic navigation reference map according to claim 1, wherein: The performing position-by-position averaging processing on the calibration sequence to determine the fusion sequence of the atomic path specifically includes: Using the position data in the calibration sequence as calibration position points; Averaging the geomagnetic data corresponding to the calibration position point in different calibration sequences to obtain fused geomagnetic data corresponding to the calibration position point; The fusion sequence of the atomic path is determined according to the fused geomagnetic data corresponding to the calibration position point.

8. A device for constructing a geomagnetic navigation reference map, characterized in that: The device comprises: an acquisition module, configured to acquire a geomagnetic sequence of at least one vehicle when it is traveling in the area to be mapped, the geomagnetic sequence including geomagnetic data and its corresponding position data; a processing module, configured to match the geomagnetic sequence with the atomic paths in the area to be mapped based on the position data in the geomagnetic sequence to determine a path sequence of the atomic paths, where the atomic paths are determined based on the roads in the area to be mapped; and to cluster the path sequences based on the number of targets to determine a target sequence of the atomic paths; and to positionally align the target sequences to determine a calibration sequence of the atomic paths; and to average the calibration sequences point by point to determine a fusion sequence of the atomic paths; A construction module is used to construct a geomagnetic navigation reference map of the area to be mapped according to the fusion sequence.

9. A readable storage medium having a program or instruction stored thereon, characterized in that: When the program or instruction is executed by a processor, the steps of the method for constructing a geomagnetic navigation reference map according to any one of claims 1 to 7 are implemented.

10. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein: When the processor executes the program, the method for constructing a geomagnetic navigation reference map according to any one of claims 1 to 7 is implemented.

Images