Warehouse environment data acquisition method and system based on Internet of Things

Abstract

The invention relates to the technical field of data processing, in particular to a warehouse environment data acquisition method and system based on the Internet of Things. The method comprises the following steps: collecting warehouse environment data by using a laser radar, the warehouse environment data being point cloud data, and denoising the warehouse environment data by using a radius filtering algorithm to realize collection of the warehouse environment data; in the process of denoising by using a radius filtering algorithm, the method comprises the following steps: determining outlier features of data points in a target grid in a plurality of grids; determining the data density of the target grid; determining an intra-cluster aggregation degree of class clusters in the target grid; determining an inter-cluster separation degree of class clusters in the target grid; and determining the minimum neighbor number of the target grid. According to the method, the filtering parameters can be dynamically adjusted according to the local features of the point cloud data in each grid through the radius filtering of the dynamic minimum neighbor number, noise and outlier data points are more accurately removed, and more accurate warehouse environment data are collected.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a warehouse environment data collection method and system based on the Internet of Things. Background Art

[0002] Automotive supply chain warehouses store a vast number of automotive parts. These parts come in a wide variety of types and shapes, and have strict requirements for their storage environment. With the development of artificial intelligence and the Internet of Things (IoT) technologies, their application in warehouse management is gaining increasing attention and popularity. Warehouse robots collect point cloud data of the warehouse environment using devices such as lidar or 3D scanners and transmit this data via IoT technology, enabling intelligent management of the warehouse environment. However, point cloud data often contains a large amount of noise and redundant information, which affects the accuracy and reliability of the data and, in turn, the effectiveness of warehouse management. Therefore, effectively filtering point cloud data to remove noise and redundant information is key to improving data collection quality and efficiency. Radius filtering, as an effective filtering method, can be applied in the preprocessing stage of point cloud data collection in automotive supply chain warehouses. Radius filtering uses a reasonable radius range to perform a neighborhood search for each point in the point cloud data. The point is then determined to be a noise point or a redundant point based on the number or attribute characteristics of the points in the neighborhood, and the corresponding processing is performed.

[0003] The patent application document with publication number CN117422622A discloses a sonar image denoising method based on adaptive combined filtering, S1: importing and running the point cloud data measured by sonar; S2: performing radius filtering according to the point cloud spacing, setting the search radius and domain value to remove point clouds with high discreteness; S3: setting the slice thickness and adjacent slice distance to slice the point cloud data, simplifying the data set and improving the denoising effect; S4: calculating the mean and standard deviation, setting the corresponding threshold, and using the statistical filtering method to process the sonar image; S5: using combined filtering to filter the sonar data point cloud; S6: performing accuracy testing and data visualization comparison.

[0004] However, point cloud data in a warehouse environment is often non-uniform. Due to factors such as the placement, shape, size, and occlusion of objects, the density and distribution of point cloud data can vary significantly in different areas. The above patent application documents do not address the problem that when using traditional radius filtering for denoising, the minimum number of neighbors in the global neighborhood is used, which cannot adapt to the differences in the density and distribution of point cloud data in different areas, resulting in filtered data in some areas and under-filtered data in some areas, thereby reducing the accuracy of the collected warehouse environment data. Summary of the Invention

[0005] In order to solve the problem that when traditional radius filtering is used for denoising, the minimum number of neighbors in the global neighborhood is used, which cannot adapt to the differences in density and distribution of point cloud data in different areas, resulting in filtered data in some areas and under-filtered data in some areas, thereby reducing the accuracy of the collected warehouse environment data, the present invention provides a warehouse environment data collection method and system based on the Internet of Things.

[0006] In a first aspect, the present invention provides a warehouse environment data collection method based on the Internet of Things, which adopts the following technical solutions: A warehouse environment data collection method based on the Internet of Things, comprising: For a target grid among multiple grids, the outlier characteristics of the target data point are determined based on the distance between the target data point and the adjacent data points in the target grid, as well as the number of adjacent data points; the point corresponding to the mean value of the coordinates of each data point is used as a feature point, and the distance from each data point to the feature point is weightedly summed according to the outlier characteristics of each data point to obtain the point cloud aggregation degree of the target grid, and then the data density of the target grid is determined; each data point is clustered to obtain multiple clusters, and for a target cluster among the multiple clusters, the intra-cluster aggregation degree of the target cluster is determined based on the mean value of the distance from each data point in the target cluster to the cluster center; the distance between the edge point of the target cluster and the cluster closest to the target cluster is used as the inter-cluster separation degree of the target cluster; the target cluster is split and merged according to the size of the intra-cluster aggregation degree and the inter-cluster separation degree, and the minimum number of neighbors of the target grid is determined according to the number of all clusters after the split and merge, and the data density.

[0007] The beneficial effects are: by adjusting the denoising parameters according to the local density of the point cloud data, the "over-filtering" and "under-filtering" problems in traditional radius filtering are avoided, and the data processing accuracy in different areas is improved; by combining grid division, outlier detection and cluster analysis, the denoising process is optimized, which can more accurately retain the real data and reduce the impact of noise and outliers; by weighted calculation of the relationship between data points and neighbors, the accuracy of the data is enhanced, ensuring that the point cloud data in different areas are properly processed; providing higher quality point cloud data, so that warehouse automation equipment can navigate and position more accurately, improving the stability and accuracy of the system.

[0008] Furthermore, the point cloud data includes three-dimensional space coordinates, intensity information and color information.

[0009] Furthermore, the outlier feature satisfies the following relationship: Where, For the The first Outlier features of data points, For the The first The number of neighboring data points of a data point, For the The first Data points to The distance between adjacent data points.

[0010] The beneficial effects are: by calculating the average distance between each data point and its neighboring points, outliers can be effectively identified, which helps to remove noise from point cloud data and improve data accuracy; based on the calculation of outlier features, outliers can be accurately removed in areas with uneven data density or high noise, thereby optimizing data quality.

[0011] Furthermore, the data density satisfies the following relationship: Where, For the The data density of the grid, For the The number of data points in a grid, For the The degree of point cloud aggregation of a grid, For the The first Outlier features of data points, For the The first The distance from the data point to the feature point, is the normalization function.

[0012] The beneficial effects are: by combining outlier features and the distance from data points to feature points, the degree of data aggregation can be described more accurately, especially when the point cloud is unevenly distributed, it can provide more accurate local density estimation; it can adaptively adjust the density estimation according to the outlier features of the data points, avoiding errors caused by traditional methods in areas with large density changes, and helping to identify sparse and dense areas.

[0013] Furthermore, the degree of clustering within the cluster satisfies the following relationship: Where, For the The first The degree of clustering within each cluster, For the The first The number of data points in a cluster, For the The first Within the cluster data points to cluster centers distance, is the natural exponential function.

[0014] The beneficial effects are: by calculating the distance from each data point to the cluster center and applying the natural exponential function, the degree of aggregation of data points within the cluster can be accurately measured; by taking a weighted average of the distances of data points within the cluster, the degree of aggregation can be adapted to the different density of points within the cluster, and even in the case of uneven density, the overall degree of aggregation of the cluster can be judged more accurately.

[0015] Furthermore, the edge point is the data point on the straight line connecting the cluster centers of the two clusters that is farthest from the cluster center of the cluster to which it belongs; if there is no data point on the straight line, the data point in the cluster that is farthest from the cluster center and closest to the straight line is selected as the edge point.

[0016] Furthermore, the splitting and merging of the target cluster includes: in response to the degree of aggregation within the cluster being less than a preset cluster splitting threshold, splitting the target cluster; in response to the degree of separation between clusters being less than a preset cluster merging threshold, merging the target cluster and the cluster closest to the target cluster; in response to the degree of aggregation within the cluster being less than the preset cluster splitting threshold and the degree of separation between clusters being less than the preset cluster merging threshold, stopping the splitting and merging of the target cluster.

[0017] Furthermore, the minimum number of neighbors satisfies the following relationship: Where, For the The minimum number of neighbors for a grid, For the The data density of the grid, For the The number of clusters in a grid, is a hyperparameter, is the natural exponential function, The symbol for rounding up.

[0018] The beneficial effect is that by combining the minimum number of neighbors with the data density and the number of clusters, an adaptive adjustment of the number of neighbors for each grid is achieved, so that in high-density areas, the number of neighbors automatically increases, thereby capturing the local structure more accurately, while in low-density areas, the number of neighbors is reduced to avoid over-computation.

[0019] Furthermore, the method for realizing the collection of warehouse environment data includes: taking twice the minimum number of neighbors as the filter radius of the target grid, and in response to the number of neighbors within the filter radius of the target grid being less than the minimum number of neighbors of the target grid, the warehouse environment data corresponding to the target grid is eliminated; otherwise, the warehouse environment data corresponding to the target grid is denoised using radius filtering to obtain the denoised warehouse environment data corresponding to the target grid, and the denoised warehouse environment data is transmitted to the database to complete the warehouse environment data collection based on the Internet of Things.

[0020] In a second aspect, the present invention provides a warehouse environment data collection system based on the Internet of Things, which adopts the following technical solutions: A warehouse environment data collection system based on the Internet of Things includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned warehouse environment data collection method based on the Internet of Things is implemented.

[0021] By adopting the above technical solution, the above-mentioned warehouse environment data collection method based on the Internet of Things is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that a terminal device is made according to the memory and the processor for easy use.

[0022] The present invention has the following technical effects: The present invention can dynamically adjust the filtering parameters according to the local characteristics of the point cloud data in each grid through the radius filtering of the dynamic minimum number of neighbors, so as to more accurately remove noise data points and outlier data points and retain valid data points. It solves the problem that the traditional radius filtering denoising adopts the minimum number of neighbors in the global neighborhood and cannot adapt to the density and distribution differences of different regions of the point cloud data, resulting in filtering in some areas and under-filtering in some areas, thereby reducing the accuracy of the warehouse environment data. Through targeted processing, the accuracy of the collected warehouse environment data can be better guaranteed; by dividing the warehouse environment point cloud data into grids according to the preset grid width, the relevant characteristics of each grid can be obtained. For example, the outlier characteristics of each data point are first determined, and then the outlier characteristics are calculated. The degree of point cloud aggregation of the grid is calculated to clarify the data density. After clustering the data points in the grid, the degree of intra-cluster aggregation and inter-cluster separation of the clusters can be analyzed to accurately grasp the data situation in the grid from multiple aspects. At the same time, grid division divides large-scale point cloud data into multiple smaller processing units for parallel processing, making the data processing in each grid more efficient and improving the efficiency of data processing; the minimum number of neighbors of the grid is determined based on multiple factors such as the degree of intra-cluster aggregation of clusters in the grid, the degree of inter-cluster separation, the number of clusters after splitting and merging, and the data density. In this way, during the denoising process of the radius filter algorithm, the minimum number of neighbors can be more in line with the actual situation of each grid, optimize the denoising effect, and improve the quality of the overall warehouse environment data collection. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of a method for collecting warehouse environment data based on the Internet of Things in an embodiment of the present invention.

[0024] Figure 2 This is a method flow chart of step S2 in a warehouse environment data collection method based on the Internet of Things in an embodiment of the present invention.

[0025] Figure 3 This is a schematic diagram of edge points in a warehouse environment data collection method based on the Internet of Things according to an embodiment of the present invention; wherein 1 is a cluster The cluster center, 2 is the cluster The cluster center of the nearest neighbor cluster, 3 is the cluster The edge points of , 4 is the cluster The edge points of the nearest neighbor cluster. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0027] The embodiment of the present invention discloses a warehouse environment data collection method based on the Internet of Things, referring to Figure 1 , including steps S1 to S3: S1: Use laser radar to collect warehouse environment data, where the warehouse environment data is point cloud data.

[0028] Warehouse robots use lidar to scan the warehouse environment to obtain warehouse environment data.

[0029] Specifically, the point cloud data includes three-dimensional spatial coordinates, intensity information (objects with high reflectivity, such as mirrors or metal surfaces, will make the intensity of reflected light higher, while objects with low reflectivity, such as black rough cloth, will make the intensity of reflected light lower) and color information, etc.

[0030] S2: Use radius filtering algorithm to denoise warehouse environment data.

[0031] It should be noted that point cloud data from warehouse environments typically contains a large number of discrete 3D points, which reflect the shape, location, and attributes of objects within the warehouse. Processing this data requires accurately identifying the object's boundaries, features, and spatial relationships. Furthermore, because the types, sizes, and locations of objects within the warehouse may vary, the distribution and noise level of the point cloud data may also vary. This means that point cloud data in warehouse environments is often non-uniform. Radius filtering filters data based on the filter radius and the minimum number of neighbors. The global minimum number of neighbors used in traditional radius filtering cannot adapt to this uneven distribution, resulting in some areas being over-filtered and others under-filtered. Therefore, gridding the point cloud data and setting different minimum neighbor numbers for each grid for parallel (simultaneous) filtering can improve filtering efficiency while also addressing the aforementioned issues of under-filtering or over-filtering of some data points. The minimum number of neighbors corresponding to each grid can be determined based on the density characteristics of the point cloud data within that grid and the type of point cloud data (the number of clusters after clustering). This is because, on the one hand, in a warehouse environment, the distribution density of point cloud data may vary significantly due to factors such as the placement, shape, and size of objects. For example, in areas with dense shelves, point cloud data may be very dense. If the number of neighbors is too small, it may not accurately reflect the local features of the objects. In open areas or areas with fewer objects, point cloud data may be relatively sparse. If the number of neighbors is too large, it may introduce unnecessary computational complexity and noise. On the other hand, in a warehouse environment, point cloud data may contain multiple categories of objects, such as various automotive parts and shelves. Objects of different categories have different shapes, sizes, and features, so their point cloud data may also differ. If there are more categories of point cloud data in the grid, it means that there may be objects of various shapes, sizes, and features in the grid. In order to accurately describe and distinguish these objects, richer local feature information is required, that is, a larger number of neighbors is required to capture such details.

[0032] To summarize, the minimum number of neighbors corresponding to the grid is obtained here based on the distribution density of the point cloud data in the grid and the type of point cloud data (the number of clusters after clustering). The higher the distribution density of the point cloud data in the grid or the more types there are, the more neighbor points are required, and the larger the corresponding minimum number of neighbors.

[0033] Reference Figure 2 In the process of denoising using the radius filtering algorithm, the warehouse environment data is gridded according to a preset grid width (20 in this invention, which can be set by the implementer according to the specific implementation situation) to obtain multiple grids. The minimum number of neighbors for each grid used to determine the filtering radius is obtained, including steps S201 to S205, as follows: S201: For a target grid among multiple grids, determine outlier features of data points in the target grid.

[0034] It should be noted that the point cloud data density within a grid specifically refers to the relationship between the amount of data within the grid and the degree of data aggregation. The more data there is in the grid and the more aggregated it is, the higher the data density within the grid. When obtaining the degree of data aggregation within a grid, it is generally described by the distance between the point cloud data point within the grid and a certain feature point. However, due to the possible presence of noise in the point cloud data points, some data will be relatively deviated. Such data points are far away from the feature points and will have a greater impact on the degree of aggregation when participating in the calculation. They need to be eliminated to improve the accuracy of the degree of aggregation. Therefore, it is necessary to first calculate the outlier characteristics of each data point in the grid.

[0035] The outlier feature of the target data point is determined according to the distance between the target data point and the adjacent data points in the target grid, and the number of the adjacent data points, wherein the distance is the Euclidean distance.

[0036] Specifically, the outlier feature satisfies the following relationship: ; Where, For the The first Outlier features of data points, For the The first The number of neighboring data points of a data point, For the The first Data points to The distance between adjacent data points.

[0037] in, Indicates the The first The average distance from a data point to its neighboring data points. The larger the value, the better. The farther the distance between a data point and its neighboring data points, the The more obvious the outlier characteristics of the data point are.

[0038] S202: Determine the data density of the target grid.

[0039] The point corresponding to the coordinate mean of each data point is taken as the feature point. According to the outlier characteristics of each data point, the distance from each data point to the feature point is weighted summed to obtain the point cloud aggregation degree of the target grid, and then the data density of the target grid is determined.

[0040] Specifically, the data density satisfies the following relationship: ; Where, For the The data density of the grid, For the The number of data points in a grid, For the The degree of point cloud aggregation of a grid, For the The first Outlier features of data points, For the The first The distance from the data point to the feature point, is the normalization function.

[0041] in, Indicates the The first The weight of the data point in the grid The more obvious the outlier feature of a data point is, the lower the credibility of its distance from the feature points in the grid is, and the smaller the corresponding weight is. Indicates the The point cloud aggregation degree of a grid is determined by the weighted distance between the data points and the feature points in the grid. The smaller the weighted distance between the data points and the feature points in the grid, the higher the point cloud aggregation degree. The more point cloud data points in the grid and the more clustered they are, the higher the grid density.

[0042] S203: Determine the intra-cluster aggregation degree of the clusters in the target grid.

[0043] It should be noted that the simplest and most effective way to obtain the types of point cloud data within the grid is to cluster the data points within the grid. The clustering process clusters similar point cloud data points into one category, and the final number of clusters is the point cloud category within the grid. However, noise is inevitable during the point cloud data acquisition process, which leads to poor clustering effect. Therefore, the clustering results can be analyzed and some clusters can be split or merged to improve the accuracy of the final number of clusters.

[0044] Cluster each data point to obtain multiple clusters. For the target cluster among the multiple clusters, the degree of clustering within the target cluster is determined according to the mean value of the distance from each data point in the target cluster to the cluster center.

[0045] Implementers can select a clustering method based on specific implementation circumstances, for example, K-means clustering.

[0046] Specifically, the degree of clustering within the cluster satisfies the following relationship: ; Where, For the The first The degree of clustering within each cluster, For the The first The number of data points in a cluster, For the The first Within the cluster data points to cluster centers distance, is the natural exponential function.

[0047] in, Indicates the The first The average distance between the data points in a cluster and the cluster center. The smaller the value, the The higher the degree of clustering within a cluster.

[0048] S204: Determine the inter-cluster separation degree of the clusters in the target grid.

[0049] The distance between the edge points of the target cluster and the cluster closest to the target cluster is taken as the inter-cluster separation of the target cluster.

[0050] Specifically, the edge point is the data point on the straight line connecting the cluster centers of the two clusters that is farthest from the cluster center of the cluster to which it belongs; if there is no data point on the straight line, the data point in the cluster that is farthest from the cluster center and closest to the straight line is selected as the edge point, such as Figure 3 As shown, assuming that point 1 is a cluster The cluster center, point 2 is the cluster The cluster center of the nearest neighbor cluster, the cluster The edge points are the clusters on the line connecting 1 and 2. Point 3, the farthest point from the cluster center, cluster The edge point of the nearest neighbor cluster is the point 4 in the nearest neighbor cluster on the line connecting 1 and 2 that is farthest from the cluster center. If there is no point in the cluster on the line, the point farthest from the cluster center and closest to the line is selected as the edge point.

[0051] Among them, The greater the distance between the edge point in a cluster and the edge point of the nearest cluster, the The higher the separation between clusters, the higher the

[0052] S205: Determine the minimum number of neighbors of the target grid.

[0053] It should be noted that some clusters with poor clustering effects are split because they may contain subsets; some clusters with fuzzy boundaries are merged because they may be the same cluster as clusters with fuzzy boundaries.

[0054] The target clusters are split and merged according to the degree of intra-cluster aggregation and inter-cluster separation.

[0055] Specifically, the splitting and merging of the target clusters includes: In response to the degree of aggregation within the cluster being less than a preset cluster splitting threshold, splitting the target cluster; In response to the inter-cluster separation being less than a preset cluster merging threshold, merging the target cluster and the cluster closest to the target cluster; In response to the intra-cluster aggregation degree being less than a preset cluster splitting threshold, and the inter-cluster separation degree being less than a preset cluster merging threshold, the splitting and merging of the target cluster is stopped.

[0056] Among them, the cluster splitting threshold is the mean of the intra-cluster aggregation degree of all clusters in the grid. When the intra-cluster aggregation degree of a cluster in the grid is less than the set threshold, it is considered that the clustering effect of the cluster is poor and there may be a subclass structure, so it is split; the cluster merging threshold is the mean of the inter-cluster separation degree of all clusters in the grid. When the inter-cluster separation degree of a cluster in the grid is less than the set threshold, it is considered that the distance between the cluster and the nearest neighbor cluster is fuzzy, and they may belong to the same cluster, so they are merged; when the intra-cluster aggregation degree and inter-cluster separation degree of a cluster in the grid are both less than the set threshold, it is considered that the cluster may have poor clustering effect and subclass structure, and the distance with the nearest neighbor cluster is fuzzy, and they may belong to the same cluster. The situation is more complicated and difficult to handle by simple splitting or merging operations, so the cluster is not split or merged.

[0057] The minimum number of neighbors of the target grid is determined according to the number of all clusters after splitting and merging, and the data density.

[0058] Specifically, the minimum number of neighbors satisfies the following relationship: ; Where, For the The minimum number of neighbors for a grid, For the The data density of the grid, For the The number of clusters in a grid, is a hyperparameter, is the natural exponential function, The symbol for rounding up.

[0059] Implementers can set the hyperparameter to, for example, 10, based on their specific implementation.

[0060] Among them, when the data density of the grid is high or there are more types, more neighbors are needed to capture such details, and the corresponding minimum number of neighbors is also larger.

[0061] S3: To realize the collection of warehouse environment data.

[0062] Specifically, the collection of warehouse environment data includes: The twice of the minimum number of neighbors is used as the filtering radius of the target grid. In response to the number of neighbors within the filtering radius of the target grid being less than the minimum number of neighbors of the target grid, the warehouse environment data corresponding to the target grid is eliminated. Otherwise, the warehouse environment data corresponding to the target grid is denoised using radius filtering to obtain the denoised warehouse environment data corresponding to the target grid. The denoised warehouse environment data is then transmitted to a database to facilitate a more accurate analysis of the warehouse environment in the future and complete the warehouse environment data collection based on the Internet of Things.

[0063] An embodiment of the present invention also discloses a warehouse environment data collection system based on the Internet of Things, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a warehouse environment data collection method based on the Internet of Things according to the present invention is implemented.

[0064] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0065] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A warehouse environment data collection method based on the Internet of Things, characterized in that: include: Using laser radar to collect warehouse environment data, the warehouse environment data is point cloud data, and using a radius filtering algorithm to denoise the warehouse environment data to achieve the collection of warehouse environment data; In the process of denoising using the radius filter algorithm, the warehouse environment data is divided into grids to obtain multiple grids. The minimum number of neighbors for each grid used to determine the filter radius is obtained, including: For a target grid among multiple grids, the outlier characteristics of the target data point are determined based on the distance between the target data point and the adjacent data points in the target grid, as well as the number of adjacent data points; the point corresponding to the mean value of the coordinates of each data point is used as a feature point, and the distance from each data point to the feature point is weightedly summed according to the outlier characteristics of each data point to obtain the point cloud aggregation degree of the target grid, and then the data density of the target grid is determined; each data point is clustered to obtain multiple clusters, and for a target cluster among the multiple clusters, the intra-cluster aggregation degree of the target cluster is determined based on the mean value of the distance from each data point in the target cluster to the cluster center; the distance between the edge point of the target cluster and the cluster closest to the target cluster is used as the inter-cluster separation degree of the target cluster; the target cluster is split and merged according to the size of the intra-cluster aggregation degree and the inter-cluster separation degree, and the minimum number of neighbors of the target grid is determined according to the number of all clusters after the split and merge, and the data density.

2. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The point cloud data includes three-dimensional space coordinates, intensity information and color information.

3. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The outlier characteristics satisfy the following relationship: ； Where, For the The first Outlier features of data points, For the The first The number of neighboring data points of a data point, For the The first Data points to The distance between adjacent data points.

4. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The data density satisfies the following relationship: ； Where, For the The data density of the grid, For the The number of data points in a grid, For the The degree of point cloud aggregation of a grid, For the The first Outlier features of data points, For the The first The distance from the data point to the feature point, is the normalization function.

5. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The degree of clustering within the cluster satisfies the following relationship: ； Where, For the The first The degree of clustering within each cluster, For the The first The number of data points in a cluster, For the The first Within the cluster data points to cluster centers distance, is the natural exponential function.

6. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The edge point is the data point on the straight line connecting the cluster centers of the two clusters that is farthest from the cluster center of the cluster to which it belongs; if there is no data point on the straight line, the data point in the cluster that is farthest from the cluster center and closest to the straight line is selected as the edge point.

7. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The splitting and merging of the target clusters includes: In response to the degree of aggregation within the cluster being less than a preset cluster splitting threshold, splitting the target cluster; In response to the inter-cluster separation being less than a preset cluster merging threshold, merging the target cluster and the cluster closest to the target cluster; In response to the intra-cluster aggregation degree being less than a preset cluster splitting threshold, and the inter-cluster separation degree being less than a preset cluster merging threshold, the splitting and merging of the target cluster is stopped.

8. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The minimum number of neighbors satisfies the following relationship: ； Where, For the The minimum number of neighbors for a grid, For the The data density of the grid, For the The number of clusters in a grid, is a hyperparameter, is the natural exponential function, The symbol for rounding up.

9. The method for collecting warehouse environment data based on the Internet of Things according to claim 1, characterized in that: The method for collecting warehouse environment data includes: Twice the minimum number of neighbors is used as the filtering radius of the target grid. In response to the number of neighbors within the filtering radius of the target grid being less than the minimum number of neighbors of the target grid, the warehouse environment data corresponding to the target grid is eliminated. Otherwise, the warehouse environment data corresponding to the target grid is denoised using radius filtering to obtain the denoised warehouse environment data corresponding to the target grid. The denoised warehouse environment data is then transmitted to the database to complete the warehouse environment data collection based on the Internet of Things.

10. A warehouse environment data acquisition system based on the Internet of Things, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a warehouse environment data collection method based on the Internet of Things according to any one of claims 1 to 9 is implemented.

Images