Dynamic Environment Relocation Method, Device, Equipment and Storage Medium for Floor Sweeping Machine

Abstract

The present application provides a dynamic environment relocalization method, device, equipment and storage medium for a floor sweeping robot. The method includes: obtaining current three-dimensional map information, and obtaining current object point cloud information in the current three-dimensional map information, and simultaneously parsing and obtaining first strong attribute point cloud information in the current object point cloud information; wherein, the strong attribute point cloud information is point cloud information of immovable objects; obtaining historical three-dimensional map information, and obtaining historical object point cloud information in the historical three-dimensional map information, and simultaneously parsing and obtaining second strong attribute point cloud information in the historical object point cloud information; if the first strong attribute point cloud information matches the second strong attribute point cloud information, it is determined that the floor sweeping robot has completed relocalization. Since the present application only compares the strong attribute point cloud information, the amount of calculation is small and the relocalization speed is faster; and because the strong attribute point cloud information is point cloud information of immovable objects, the accuracy of the floor sweeping robot relocalization is improved.

Description

Technical Field

[0001] The present invention relates to the field of floor sweeping robots, and particularly to a dynamic environment relocalization method, device, equipment and storage medium for a floor sweeping robot. Background Art

[0002] Currently, floor sweeping robots based on lidar for positioning are gradually developing in the market. Because of their advantages in environmental mapping and navigation, they have prominent advantages in the mid- to high-end fields. However, the floor sweeping robots using lidar for positioning mainly solve the problems of positioning and mapping in static scenarios. In dynamic scenarios, due to environmental changes, such as moving stools and ornaments, etc., the success rate of positioning will be affected, resulting in a decrease in the success rate when the floor sweeping robot executes the relocalization instruction. Summary of the Invention

[0003] The purpose of this application is to provide a dynamic environment relocalization method, device, equipment and storage medium for a floor sweeping robot, aiming to solve the problem that the success rate of relocalization of the floor sweeping robot decreases in dynamic scenarios in the prior art.

[0004] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] This application provides a dynamic environment relocalization method for a floor sweeping robot, and the method includes:

[0006] Obtain the current three-dimensional map information, obtain the current object point cloud information in the current three-dimensional map information, and at the same time parse and obtain the first strong attribute point cloud information in the current object point cloud information; wherein, the strong attribute point cloud information is the point cloud information of immovable objects;

[0007] Obtain the historical three-dimensional map information, obtain the historical object point cloud information in the historical three-dimensional map information, and at the same time parse and obtain the second strong attribute point cloud information in the historical object point cloud information;

[0008] If the first strong attribute point cloud information matches the second strong attribute point cloud information, it is determined that the floor sweeping robot has completed relocalization.

[0009] Further, the step of obtaining the current three-dimensional map information, obtaining the current object point cloud information in the current three-dimensional map information, and at the same time parsing and obtaining the first strong attribute point cloud information in the current object point cloud information includes:

[0010] Obtain the current three-dimensional map information, and obtain the current object point cloud information in the current three-dimensional map information;

[0011] Parse and obtain the first weak attribute point cloud information in the current object point cloud information; wherein, the weak attribute point cloud information is the point cloud information of movable objects;

[0012] Delete the first weak attribute point cloud information from the current 3D map information;

[0013] After deleting the first weak attribute point cloud information, parse and obtain the first strong attribute point cloud information in the current object point cloud information.

[0014] Further, the step of obtaining the historical 3D map information, obtaining the historical object point cloud information in the historical 3D map information, and simultaneously parsing and obtaining the second strong attribute point cloud information in the historical object point cloud information includes:

[0015] Obtain the historical 3D map information and obtain the historical object point cloud information in the historical 3D map information;

[0016] Parse and obtain the second weak attribute point cloud information in the historical object point cloud information;

[0017] Delete the second weak attribute point cloud information from the historical 3D map information;

[0018] After deleting the second weak attribute point cloud information, parse and obtain the second strong attribute point cloud information in the historical object point cloud information.

[0019] Further, before the step of determining that the sweeping robot has completed repositioning if the first strong attribute point cloud information matches the second strong attribute point cloud information, it includes:

[0020] If the proportion of the first strong attribute point cloud information in the current object point cloud information is higher than a preset threshold, and / or the proportion of the second strong attribute point cloud information in the historical object point cloud information is higher than the preset threshold, then determine whether the first strong attribute point cloud information matches the second strong attribute point cloud information.

[0021] Further, after the step of obtaining the historical 3D map information, obtaining the historical object point cloud information in the historical 3D map information, and simultaneously parsing and obtaining the second strong attribute point cloud information in the historical object point cloud information, it includes:

[0022] If the proportion of the first strong attribute point cloud information in the current object point cloud information is lower than the preset threshold, and the proportion of the second strong attribute point cloud information in the historical object point cloud information is lower than the preset threshold, then parse and obtain the first medium attribute point cloud information in the current object point cloud information, and parse and obtain the second medium attribute point cloud information in the historical object point cloud information; wherein, the movement probability of the object corresponding to the medium attribute point cloud information is less than that of the moving object and greater than that of the immovable object;

[0023] If the first type of attribute point cloud information matches the second type of attribute point cloud information, it is determined that the floor cleaning machine has completed repositioning.

[0024] Further, the obtaining of the current object point cloud information in the current three-dimensional map information includes:

[0025] Clustering the current three-dimensional map information to obtain the current object point cloud information.

[0026] Further, after the steps of obtaining historical three-dimensional map information, obtaining historical object point cloud information in the historical three-dimensional map information, and simultaneously parsing and obtaining second strong attribute point cloud information in the historical object point cloud information, it includes:

[0027] If the first strong attribute point cloud information does not match the second strong attribute point cloud information, then first weak attribute point cloud information is parsed from the current object point cloud information, and second weak attribute point cloud information is parsed from the historical object point cloud information, where the weak attribute point cloud information is point cloud information of a moving object

[0028] Deleting the first weak attribute point cloud from the current object point cloud information to form a first matching point cloud, and deleting the second weak attribute point cloud from the historical object point cloud information to form a second matching point cloud;

[0029] Determine whether the first matching point cloud is consistent with the second matching point cloud. If so, it is determined that the floor cleaning machine has completed repositioning.

[0030] The present application also provides a dynamic environment repositioning device for a floor cleaning machine, including:

[0031] A first obtaining and parsing unit, configured to obtain current three-dimensional map information, obtain current object point cloud information in the current three-dimensional map information, and simultaneously parse and obtain first strong attribute point cloud information in the current object point cloud information; where the strong attribute point cloud information is point cloud information of an immovable object;

[0032] A second obtaining and parsing unit, configured to obtain historical three-dimensional map information, obtain historical object point cloud information in the historical three-dimensional map information, and simultaneously parse and obtain second strong attribute point cloud information in the historical object point cloud information;

[0033] A determination unit, configured to determine that the floor cleaning machine has completed repositioning if the first strong attribute point cloud information matches the second strong attribute point cloud information.

[0034] The present application also provides a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, it implements the dynamic environment repositioning method of the floor cleaning machine as described in any one of the above.

[0035] The present application further provides a computer device, which includes a processor, a memory, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the dynamic environment relocalization method of the floor sweeper as described above is implemented.

[0036] Advantages of the present invention: For the dynamic environment relocalization method, device, equipment, and storage medium of the floor sweeper of the present application, first, the first strong attribute point cloud information in the current object point cloud information of the current cleaning environment is obtained, and the second strong attribute point cloud information in the historical object point cloud information of the current cleaning environment is obtained. Secondly, when the first strong attribute point cloud information matches the second strong attribute point cloud information, it is determined that the floor sweeper has completed relocalization. In this way, the present application only needs to match the first strong attribute point cloud information and the second strong attribute point cloud information to determine whether relocalization is completed. Because only the strong attribute point cloud information is compared, the comparison calculation amount is small, so the relocalization speed is faster; and because the strong attribute point cloud information is the point cloud information of immovable objects, the accuracy of the floor sweeper relocalization is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a flowchart of the method for the dynamic environment relocalization method of the floor sweeper of the present application;

[0038] Figure 2 is a schematic structural diagram of the dynamic environment relocalization device of the floor sweeper of the present application;

[0039] Figure 3 is a structural block diagram of an embodiment of the storage medium of the present application;

[0040] Figure 4 is a structural block diagram of an embodiment of the computer device of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0042] Those skilled in the art of the present technology can understand that unless specifically stated otherwise, the singular forms "a", "an", "the", and "said" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present invention means 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 their groups. The term "and / or" used herein includes all or any unit and all combinations of one or more of the associated listed items.

[0043] Refer to Figure 1, this application provides a dynamic environment relocalization method for a floor cleaning robot, including:

[0044] S1. Obtain the current 3D map information, obtain the current object point cloud information in the current 3D map information, and at the same time parse and obtain the first strong attribute point cloud information in the current object point cloud information; wherein, the strong attribute point cloud information is the point cloud information of immovable objects;

[0045] S2. Obtain the historical 3D map information, obtain the historical object point cloud information in the historical 3D map information, and at the same time parse and obtain the second strong attribute point cloud information in the historical object point cloud information;

[0046] S3. If the first strong attribute point cloud information matches the second strong attribute point cloud information, it is determined that the floor cleaning robot has completed relocalization.

[0047] As described in step S1 above, the floor cleaning robot obtains the graphic information around the area to be cleaned and converts the image information into the current 3D map information; then obtains the current object point cloud information in the current 3D map information, and the current object point cloud information is used to identify each item in the current 3D map information, parses the current object point cloud information, and obtains the point cloud information corresponding to immovable objects, that is, the first strong attribute point cloud information. The above immovable objects are generally fixed items such as walls.

[0048] As described in step S2 above, the floor cleaning robot obtains the historical 3D map information, which is the 3D map information recorded by the floor cleaning robot in history, and then obtains the historical object point cloud information in the historical 3D map information, and parses and obtains the second strong attribute point cloud information in the historical object point cloud information.

[0049] As described in step S3 above, since both the first strong attribute point cloud information and the second strong attribute point cloud information are the point cloud information of corresponding immovable items, when the first strong attribute point cloud information and the second strong attribute point cloud information match, it can be considered that the floor cleaning robot has completed relocalization. The matching of the above first strong attribute point cloud information and the second strong attribute point cloud information can be that the similarity between the first strong attribute point cloud information and the second strong attribute point cloud information is greater than a specified similarity threshold, etc.

[0050] In the embodiments of this application, only by matching the first strong attribute point cloud information and the second strong attribute point cloud information, it can be determined whether relocalization is completed. Because only the strong attribute point cloud information is compared, the comparison calculation amount is small, and the relocalization speed is faster; and because the strong attribute point cloud information is the point cloud information of immovable objects, the accuracy of the floor cleaning robot relocalization is improved.

[0051] In one embodiment, step S1 of obtaining the current three-dimensional map information, obtaining the current object point cloud information in the current three-dimensional map information, and simultaneously parsing and obtaining the first strong attribute point cloud information in the current object point cloud information includes:

[0052] Obtain the current three-dimensional map information, and obtain the current object point cloud information in the current three-dimensional map information;

[0053] Parse and obtain the first weak attribute point cloud information in the current object point cloud information; wherein, the weak attribute point cloud information is the point cloud information of a moving object;

[0054] Delete the first weak attribute point cloud information from the current three-dimensional map information;

[0055] After deleting the first weak attribute point cloud information, parse and obtain the first strong attribute point cloud information in the current object point cloud information.

[0056] As described above, first, the sweeping robot obtains the current three-dimensional map information of the current environment through an image acquisition device or the like, so as to obtain the current object point cloud information in the current three-dimensional map information; secondly, identify the point cloud information of each moving item in the current object point cloud information, that is, the first weak attribute point cloud information; then, delete the first weak attribute point cloud information from the current object point cloud information, and parse the first strong attribute point cloud information from the remaining current object point cloud information. The above-mentioned moving objects are items with relatively high mobility, so this part of the items is not helpful for repositioning and affects the accuracy of repositioning. Therefore, first delete the point cloud information corresponding to this part of the moving items, leaving the point cloud information corresponding to immovable items and items that are not easily movable usually (the moving probability of the object is less than that of the moving object and greater than that of the immovable object, such as cabinets, beds, sofas, etc.), and then further parse this part of the remaining point cloud information to obtain the first strong attribute point cloud information. During the process of obtaining the first strong attribute point cloud information, first deleting the interference factors can improve the accuracy of the first strong attribute point cloud information.

[0057] In one embodiment, step S2 of obtaining the historical three-dimensional map information, obtaining the historical object point cloud information in the historical three-dimensional map information, and simultaneously parsing and obtaining the second strong attribute point cloud information in the historical object point cloud information includes:

[0058] Obtain the historical three-dimensional map information, and obtain the historical object point cloud information in the historical three-dimensional map information;

[0059] Parse and obtain the second weak attribute point cloud information in the historical object point cloud information;

[0060] Delete the second weakest attribute point cloud information from the historical three-dimensional map information;

[0061] After deleting the second weakest attribute point cloud information, parse and obtain the second strongest attribute point cloud information in the historical object point cloud information.

[0062] As described above, first, the sweeping robot obtains the historical three-dimensional map information, so as to obtain the historical object point cloud information in the historical three-dimensional map information; second, identify the point cloud information of each moving item in the historical object point cloud information, that is, the second weakest attribute point cloud information; then, delete the second weakest attribute point cloud information from the historical object point cloud information, and parse out the second strongest attribute point cloud information from the remaining historical object point cloud information. The above-mentioned moving items are items with relatively high mobility, so this part of the items is not helpful for relocalization and affects the accuracy of relocalization. Therefore, first delete the point cloud information corresponding to this part of the moving items, leaving the point cloud information corresponding to immovable items and items that are not convenient to move usually, and then further parse this part of the remaining point cloud information to obtain the second strongest attribute point cloud information. In the process of obtaining the second strongest attribute point cloud information, first deleting the interference factors can improve the accuracy of the second strongest attribute point cloud information.

[0063] In one embodiment, before the step S2 of determining that the sweeping robot has completed relocalization if the first strongest attribute point cloud information matches the second strongest attribute point cloud information, it includes:

[0064] If the proportion of the first strongest attribute point cloud information in the current object point cloud information is higher than a preset threshold, and / or the proportion of the second strongest attribute point cloud information in the historical object point cloud information is higher than the preset threshold, then determine whether the first strongest attribute point cloud information matches the second strongest attribute point cloud information.

[0065] As described above, before performing the matching calculation on the first strong attribute point cloud information and the second strong attribute point cloud information, it is also necessary to first determine whether the proportion of the first strong attribute point cloud information in the current object point cloud information is higher than a preset threshold, and / or determine whether the proportion of the second strong attribute point cloud information in the historical object point cloud information is higher than a preset threshold. When the proportion of the first strong attribute point cloud information in the current object point cloud information is not higher than the preset threshold, and / or when the proportion of the second strong attribute point cloud information in the historical object point cloud information is not higher than the preset threshold, it means that there is no need to perform the matching calculation between the first strong attribute point cloud information and the second strong attribute point cloud information. Specifically, when the proportion of the first strong attribute point cloud information in the current object point cloud information is not higher than the preset threshold, it means that the proportion of the immovable object corresponding to the first strong attribute point cloud information in the current 3D map is relatively small and is not sufficient to be used as a reference for accurate repositioning; when the proportion of the second strong attribute point cloud information in the historical object point cloud information is not higher than the preset threshold, it means that the proportion of the immovable object corresponding to the second strong attribute point cloud information in the historical 3D map is relatively small and is also not sufficient to be used as a reference for comparison with the current 3D map, etc. Further, when the proportion of the first strong attribute point cloud information in the current object point cloud information is not higher than the preset threshold and the proportion of the second strong attribute point cloud information in the historical object point cloud information is also not higher than the preset threshold, there is no need to perform the matching calculation on the first strong attribute point cloud information and the second strong attribute point cloud information. In summary, only when the proportion of the first strong attribute point cloud information in the current object point cloud information is higher than the preset threshold, and / or when the proportion of the second strong attribute point cloud information in the historical object point cloud information is higher than the preset threshold, is it necessary to further determine whether the first strong attribute point cloud information and the second strong attribute point cloud information match. In this way, the matching calculation is only performed on the first strong attribute point cloud information and the second strong attribute point cloud information when the matching calculation is required, saving the calculation overhead.

[0066] In one embodiment, after step S2 of obtaining the historical 3D map information, obtaining the historical object point cloud information in the historical 3D map information, and simultaneously parsing and obtaining the second strong attribute point cloud information in the historical object point cloud information, it includes:

[0067] If the proportion of the first strong attribute point cloud information in the current object point cloud information is lower than the preset threshold, and the proportion of the second strong attribute point cloud information in the historical object point cloud information is lower than the preset threshold, then parse and obtain the first medium attribute point cloud information in the current object point cloud information, and parse and obtain the second medium attribute point cloud information in the historical object point cloud information; wherein, the movement probability of the object corresponding to the medium attribute point cloud information is less than that of the movable object and greater than that of the immovable object.

[0068] If the first intermediate attribute point cloud information matches the second intermediate attribute point cloud information, it is determined that the floor sweeper has completed repositioning.

[0069] As described above, when the proportion of the first strong attribute point cloud information in the current object point cloud information is lower than the preset threshold, and the proportion of the second strong attribute point cloud information in the historical object point cloud information is lower than the preset threshold, it indicates that the proportion of the immovable object corresponding to the first strong attribute point cloud information in the current three-dimensional map information is relatively small, and the proportion of the immovable object corresponding to the second strong attribute point cloud information in the historical three-dimensional map information is relatively small. Comparing the two cannot accurately perform repositioning. Therefore, when the proportion of the first strong attribute point cloud information in the current object point cloud information is lower than the preset threshold, and the proportion of the second strong attribute point cloud information in the historical object point cloud information is lower than the preset threshold, the intermediate attribute point cloud information is introduced. The movement probability of the object corresponding to the intermediate attribute point cloud information is less than that of the movable object and greater than that of the immovable object, such as items that are not convenient to move usually, such as cabinets, beds, sofas, etc. Then, using the matching result of the first intermediate attribute point cloud information and the second intermediate attribute point cloud information to determine whether the floor sweeper has completed repositioning. If the first intermediate attribute point cloud information and the second intermediate attribute point cloud information match, it is determined that the floor sweeper has completed repositioning; otherwise, it is determined that the floor sweeper has not completed repositioning.

[0070] In one embodiment, the above-mentioned obtaining the current object point cloud information in the current three-dimensional map information includes:

[0071] Clustering the current three-dimensional map information to obtain the current object point cloud information.

[0072] As described above, it is to cluster the object information with the same attribute, and then parse the clustered current three-dimensional map information to obtain the corresponding current object point cloud information. Further, continue to cluster the point cloud, identify the shape of the point cloud, and then infer its attributes, such as stools / tables / decorations / walls, etc. Then, assign the strong and weak attributes of the corresponding point cloud according to the attributes. For example, define immovable objects such as walls / large objects as strong attributes, and movable objects such as tables / stools that are easy to move and often change positions as weak attributes, etc. By eliminating the point clouds with weak attributes in the point cloud matching, multi-level matching is performed to improve the accuracy of repositioning.

[0073] It should be further noted that clustering is to divide a data set into different clusters according to a specific standard, so that the data within the same cluster is as similar as possible, and the data not in the same cluster is as different as possible. Among them, clustering algorithms include but are not limited to partition-based clustering, hierarchical clustering, density-based clustering, grid-based clustering, model-based clustering, and fuzzy-based clustering. In another embodiment, the k-means clustering algorithm is used. K-means is a typical partition-based clustering algorithm. For a scattered point set, the number of clusters is selected and the center points are randomly initialized, and the distances between the points within the class are shortened and the distances between the points between classes are increased through iterative calculations.

[0074] In one embodiment, after step S2 of obtaining the historical three-dimensional map information, obtaining the historical object point cloud information in the historical three-dimensional map information, and simultaneously parsing and obtaining the second strong attribute point cloud information in the historical object point cloud information, it includes:

[0075] If the first strong attribute point cloud information does not match the second strong attribute point cloud information, the first weak attribute point cloud information is parsed from the current object point cloud information, and the second weak attribute point cloud information is parsed from the historical object point cloud information, where the weak attribute point cloud information is the point cloud information of a moving object

[0076] The first weak attribute point cloud is deleted from the current object point cloud information to form a first matching point cloud, and the second weak attribute point cloud is deleted from the historical object point cloud information to form a second matching point cloud;

[0077] It is judged whether the first matching point cloud is consistent with the second matching point cloud. If so, it is determined that the sweeping robot has completed repositioning.

[0078] As described above, that is, when the first strong attribute point cloud information does not match the second strong attribute point cloud information, the current object point cloud information (first matching point cloud) after deleting the first weak attribute point cloud information is used to match the historical object point cloud information (second matching point cloud) after deleting the second weak attribute point cloud information. If the matching is successful, it is also determined that the sweeping robot has completed repositioning.

[0079] Reference Figure 2 , the present application also provides a dynamic environment repositioning device for a sweeping robot, including:

[0080] The first acquisition and parsing unit 1 is used to acquire the current three-dimensional map information, acquire the current object point cloud information in the current three-dimensional map information, and simultaneously parse and obtain the first strong attribute point cloud information in the current object point cloud information; where the strong attribute point cloud information is the point cloud information of an immovable object;

[0081] The second acquisition and parsing unit 2 is configured to acquire historical three-dimensional map information, acquire historical object point cloud information in the historical three-dimensional map information, and parse and obtain second strongest attribute point cloud information in the historical object point cloud information;

[0082] The determination unit 3 is configured to determine that the floor sweeper has completed repositioning if the first strongest attribute point cloud information matches the second strongest attribute point cloud information.

[0083] The above units are for implementing the dynamic environment repositioning device of the floor sweeper as described above, and will not be introduced one by one here.

[0084] Reference Figure 3 , this application also provides a storage medium 100, in which a computer program 200 is stored. When it runs on a computer, it causes the computer to execute the dynamic environment repositioning method of the floor sweeper described in the above embodiments.

[0085] Reference Figure 4 , this application also provides a computer device 300 containing instructions. When it runs on the computer device 300, it causes the computer device 300 to execute the above-mentioned dynamic environment repositioning method of the floor sweeper through a processor 400 provided inside it.

[0086] Based on the above embodiments, it can be seen that the greatest beneficial effect of this application is that by converting the items in the cleaning range into various types of point cloud information and identifying the point cloud information, when the cleaning range or the positions of the items in the cleaning range change, the floor sweeper can still determine whether the current cleaning range has changed by identifying the point cloud information of other items, and update the positioning information in a timely manner, improving the accuracy of the positioning information obtained by the floor sweeper.

[0087] Those skilled in the art can understand the operation method of the intelligent device described in the present invention and the devices involved above for performing one or more of the methods described in the present application. These devices can be specially designed and manufactured for the required purposes, or they can also include known devices in general-purpose computers. These devices have computer programs or application programs stored therein, and these computer programs are selectively activated or reconstructed. Such computer programs can be stored in a device (e.g., a computer) readable medium or in any type of medium suitable for storing electronic instructions and coupled to the bus respectively. The computer readable medium includes, but is not limited to, any type of disk (including floppy disks, hard disks, optical disks, CD-ROMs, and magneto-optical disks), ROM (Read-Only Memory), RAM (Random Access Memory), EPROM (Erasable Programmable Read-Only Memory), EEPROM (Electrically Erasable Programmable Read-Only Memory), flash memory, magnetic cards or optical cards. That is, the readable medium includes any medium that stores or transmits information in a form readable by a device (e.g., a computer).

[0088] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall be included in the patent protection scope of the present invention by the same token.

Claims

1. A dynamic environment repositioning method for a floor sweeper, characterized in that, The method includes: Obtain the current 3D map information, obtain the current object point cloud information in the current 3D map information, and at the same time parse and obtain the first strong attribute point cloud information in the current object point cloud information; wherein, the strong attribute point cloud information is the point cloud information of immovable objects; Obtain the historical 3D map information, obtain the historical object point cloud information in the historical 3D map information, and at the same time parse and obtain the second strong attribute point cloud information in the historical object point cloud information; If the proportion of the first strong attribute point cloud information in the current object point cloud information is lower than a preset threshold, and the proportion of the second strong attribute point cloud information in the historical object point cloud information is lower than the preset threshold, then parse and obtain the first medium attribute point cloud information in the current object point cloud information, and parse and obtain the second medium attribute point cloud information in the historical object point cloud information; wherein, the movement probability of the object corresponding to the medium attribute point cloud information is less than that of the movable object and greater than that of the immovable object; If the first medium attribute point cloud information matches the second medium attribute point cloud information, it is determined that the sweeping robot has completed repositioning.

2. The dynamic environment relocalization method of the floor sweeper according to claim 1, characterized in that The step of obtaining the current 3D map information, obtaining the current object point cloud information in the current 3D map information, and at the same time parsing and obtaining the first strong attribute point cloud information in the current object point cloud information includes: Obtain the current 3D map information, and obtain the current object point cloud information in the current 3D map information; Parse and obtain the first weak attribute point cloud information in the current object point cloud information; wherein, the weak attribute point cloud information is the point cloud information of movable objects; Delete the first weak attribute point cloud information from the current 3D map information; After deleting the first weak attribute point cloud information, parse and obtain the first strong attribute point cloud information in the current object point cloud information.

3. The dynamic environment relocalization method of the floor sweeper according to claim 2, wherein, The step of obtaining the historical 3D map information, obtaining the historical object point cloud information in the historical 3D map information, and at the same time parsing and obtaining the second strong attribute point cloud information in the historical object point cloud information includes: Obtain the historical 3D map information, and obtain the historical object point cloud information in the historical 3D map information; Parse and obtain the second weak attribute point cloud information in the historical object point cloud information; Delete the second weak attribute point cloud information from the historical 3D map information; After deleting the second weak attribute point cloud information, parse and obtain the second strong attribute point cloud information in the historical object point cloud information.

4. The dynamic environment relocalization method of the floor sweeper according to claim 1, characterized in that The obtaining of the current object point cloud information in the current 3D map information includes: Cluster the current 3D map information to obtain the current object point cloud information.

5. A dynamic environment relocalization device for a floor sweeper, characterized in that Includes: The first acquisition and parsing unit is used to obtain the current 3D map information, obtain the current object point cloud information in the current 3D map information, and at the same time parse and obtain the first strong attribute point cloud information in the current object point cloud information; wherein, the strong attribute point cloud information is the point cloud information of immovable objects; A second acquisition and parsing unit, configured to acquire historical three-dimensional map information, acquire historical object point cloud information in the historical three-dimensional map information, and simultaneously parse and obtain second strong attribute point cloud information in the historical object point cloud information; A determination unit, configured to, if the proportion of the first strong attribute point cloud information in the current object point cloud information is lower than a preset threshold, and the proportion of the second strong attribute point cloud information in the historical object point cloud information is lower than the preset threshold, then parse and obtain first medium attribute point cloud information in the current object point cloud information, and parse and obtain second medium attribute point cloud information in the historical object point cloud information; wherein, the movement probability of the object corresponding to the medium attribute point cloud information is less than that of the moving object and greater than that of the immovable object; if the first medium attribute point cloud information matches the second medium attribute point cloud information, it is determined that the sweeping robot has completed repositioning.

6. A storage medium, characterized in that, It is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, it implements the dynamic environment repositioning method of the sweeping robot according to any one of claims 1 to 4.

7. A computer device, characterized in that, It includes a processor, a memory, and a computer program stored on the memory and executable on the processor, and when the processor executes the computer program, it implements the dynamic environment repositioning method of the sweeping robot according to any one of claims 1 to 4.

Images