Workstation identification method in robot mapping

Abstract

The invention relates to a workstation identification method in robot mapping. The workstation identification method in robot mapping comprises the following steps: in a mapping process, carrying out base station feature identification on a clustering set of point cloud data of a current frame collected by a host; if the clustering set of the point cloud data of the current frame identifies suspected base station features, performing multi-frame splicing on multi-frame point cloud data acquired by the host in situ and the point cloud data of the current frame to obtain spliced point cloud data; if the cluster set of the spliced point cloud data identifies base station features, performing position score calculation on the cluster set of the spliced point cloud data to obtain a base station position and a base station score corresponding to the passive base station; and controlling the host to return to the passive base station for self-cleaning based on the base station position and the base station score. According to the scheme of the invention, the passive base station identification success rate can be improved.

Description

Technical Field

[0001] The present disclosure relates to the technical field of cleaning equipment, and in particular to a method for identifying workstations in robot mapping. Background Art

[0002] With the continuous improvement of living standards and the development of automation technology, the application of sweeping robots is becoming more and more extensive. As one of the important functions of sweeping robots, the technology of "recharging" has also been continuously studied and developed. When cleaning, sweeping robots need to constantly go back and forth to the base station for self-cleaning and charging. At present, sweeping robots mostly rely on the infrared guidance signal of the charging pile for recharging (returning to the charging pile), but once the charging pile is powered off, the robot will not be able to recharge, so we need to ensure that the robot can recharge normally in the event of a power outage. At present, similar patents have been found, robots and methods, systems, storage media, and workstations for identifying workstations. Patent publication number CN115587603A: Point cloud data is clustered based on coordinate information and reflection intensity information, and point cloud data sets are matched based on the regional information of the identification image to identify the workstation. This method needs to control the robot to move to the corresponding predetermined area for identification based on the known base station location, and the degree of intelligence is not high.

[0003] In the related art, a sweeping robot is usually equipped with only one base station. When building a map, the sweeping robot usually starts from this base station, so the position of the base station in the map is known. However, for the design of the sweeping robot with two base stations, one charging base station and one passive sewage discharge base station, if the sweeping robot gets on and off the charging base station for rapid mapping, the position of the passive base station cannot be determined. Since the sweeping robot needs to constantly go to the passive sewage discharge base station for self-cleaning during the cleaning task, it is necessary to ensure that the passive base station is found in the rapid mapping to ensure that the task can be performed normally during the cleaning task. Since the rapid mapping requires the sweeping robot to build the overall environmental map in the shortest possible time, the robot's walking path needs to be as short as possible during the rapid mapping process, and it cannot traverse the entire environmental area. Moreover, due to the obstruction of the radar cover column, the single-frame point cloud scanned by the robot based on the rapid mapping path may not fully meet the characteristics of the base station, and thus the passive base station cannot be identified. Summary of the invention

[0004] In order to solve the above technical problems, the present disclosure provides a method for identifying workstations in robot mapping.

[0005] In a first aspect, the present disclosure provides a method for identifying workstations in robot mapping, which is applied to a robot mapping workstation identification system. The system includes a host and at least one passive base station, including:

[0006] During the mapping process, base station feature recognition is performed on the clustered set of current frame point cloud data collected by the host;

[0007] If the cluster set of the current frame point cloud data identifies a suspected base station feature, performing multi-frame splicing of the multi-frame point cloud data collected on-site by the host and the current frame point cloud data to obtain spliced ​​point cloud data;

[0008] If the cluster set of the spliced ​​point cloud data recognizes a base station feature, a position score calculation is performed on the cluster set of the spliced ​​point cloud data to obtain a base station position and a base station score corresponding to the passive base station;

[0009] Based on the base station position and the base station score, the host is controlled to return to the passive base station for self-cleaning.

[0010] In some embodiments, the passive base station is provided with a reflective marker, and the host is provided with a laser radar;

[0011] Wherein, before performing base station feature recognition on the cluster set of the current frame point cloud data collected by the host, the method further includes:

[0012] Acquire the current frame point cloud data corresponding to the reflective mark in the surrounding environment by scanning with the laser radar, wherein the current frame point cloud data includes position information and intensity information;

[0013] The current frame point cloud data is filtered.

[0014] In some embodiments, after filtering the current frame point cloud data, the method further includes:

[0015] By determining the relationship between the position information of the two point clouds and a first preset distance threshold, and determining the relationship between the intensity information and a preset intensity threshold, Euclidean clustering is performed on the current frame point cloud data to obtain a cluster set of the current frame point cloud data, wherein the cluster set includes a strong cluster set and a weak cluster set.

[0016] In some embodiments, the performing base station feature recognition on the clustered set of current frame point cloud data collected by the host includes:

[0017] Performing set size filtering on the cluster set of the current frame point cloud data based on a preset size;

[0018] It is determined whether the cluster set of the current frame point cloud data is identified as a base station feature, and whether the distance between the strong cluster set and the weak cluster set meets a second preset distance threshold.

[0019] In some embodiments, the performing position score calculation on the cluster set of the spliced ​​point cloud data to obtain the base station position and base station score corresponding to the passive base station includes:

[0020] Respectively calculating the set score of each set in the cluster set of the spliced ​​point cloud data, and summing the set scores of each set to obtain the base station score;

[0021] Performing least square straight line fitting on the spliced ​​point cloud data to obtain a corresponding slope;

[0022] The base station position is calculated based on the slope and the center value of the spliced ​​point cloud data.

[0023] In some embodiments, after performing base station feature recognition on the clustered set of current frame point cloud data collected by the host, the method further includes:

[0024] If the cluster set of the current frame point cloud data recognizes the base station feature, a position score calculation is performed on the cluster set of the current frame point cloud data to obtain the base station position and the base station score corresponding to the passive base station.

[0025] In some embodiments, the multi-frame point cloud data collected on-site by the host and the current frame point cloud data are stitched together to obtain stitched point cloud data, including:

[0026] Each frame of point cloud data in the multiple frames of point cloud data is sequentially spliced ​​with the current frame of point cloud data until the obtained spliced ​​point cloud data with no more than a preset number of frames spliced ​​together can identify the base station feature.

[0027] In some embodiments, the multi-frame point cloud data collected on-site by the host and the current frame point cloud data are stitched together to obtain stitched point cloud data, including:

[0028] Determining in turn whether the base station feature is recognized for each frame of point cloud data in the multiple frames of point cloud data;

[0029] Each frame of point cloud data that identifies the base station feature is spliced ​​with the current frame of point cloud data for multiple frames until no more than a preset number of frames of spliced ​​point cloud data are obtained to identify the base station feature.

[0030] In some embodiments, if the cluster set of the current frame point cloud data identifies a suspected base station feature, the multi-frame point cloud data collected on-site by the host and the current frame point cloud data are spliced ​​to obtain spliced ​​point cloud data, including:

[0031] If the cluster set of the current frame point cloud data identifies a suspected base station feature, the host rotates left and right at a first preset angle to collect first multi-frame point cloud data, and performs multi-frame splicing of the first multi-frame point cloud data and the current frame point cloud data to obtain first spliced ​​point cloud data;

[0032] If the cluster set of the first spliced ​​point cloud data does not identify the base station feature but identifies the enhanced suspected base station feature, the host rotates left and right at a second preset angle to collect second multi-frame point cloud data, and splices the second multi-frame point cloud data with the current frame point cloud data to obtain second spliced ​​point cloud data;

[0033] If the cluster set of the second stitched point cloud data does not identify the base station feature but identifies the enhanced suspected base station feature, the host rotates left and right at a third preset angle to collect third multi-frame point cloud data, and stitches the third multi-frame point cloud data with the current frame point cloud data to obtain third stitched point cloud data.

[0034] In some embodiments, if the cluster set of the current frame point cloud data identifies a suspected base station feature, the multi-frame point cloud data collected on-site by the host and the current frame point cloud data are spliced ​​to obtain spliced ​​point cloud data, including:

[0035] If the cluster set of the current frame point cloud data identifies a suspected base station feature, determining a target rotation angle according to the suspected base station feature and the position information of the current frame point cloud data;

[0036] The host rotates left and right at the target rotation angle in situ to collect multiple frames of point cloud data, and performs multi-frame splicing of the multiple frames of point cloud data with the current frame of point cloud data to obtain the spliced ​​point cloud data.

[0037] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages:

[0038] The workstation identification method in robot mapping of the disclosed embodiment can perform base station feature identification on the clustered set of current frame point cloud data collected by the host during the mapping process. If the clustered set of the current frame point cloud data identifies a suspected base station feature, multi-frame point cloud data collected on-site by the host and the current frame point cloud data are spliced ​​to obtain spliced ​​point cloud data. If the clustered set of the spliced ​​point cloud data identifies a base station feature, the position score is calculated for the clustered set of the spliced ​​point cloud data to obtain the base station position and base station score corresponding to the passive base station. Then, based on the base station position and the base station score, the host is controlled to return to the passive base station for self-cleaning. In this way, the spliced ​​point cloud data is obtained by splicing the multi-frame point cloud data with the current frame point cloud data, and the base station position and base station score corresponding to the passive base station are obtained by performing position score calculation based on the spliced ​​point cloud data that identifies the base station feature, thereby improving the success rate of passive base station identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and the originals and elements are not necessarily drawn to scale.

[0040] Figure 1 A schematic diagram of a process flow of a method for identifying workstations in robot mapping provided by an embodiment of the present disclosure;

[0041] Figure 2 A schematic diagram of the structure of a reflective logo provided by an embodiment of the present disclosure;

[0042] Figure 3 A schematic diagram of the structure of another reflective logo provided by an embodiment of the present disclosure;

[0043] Figure 4 A schematic diagram of a process of Euclidean clustering calculation provided in an embodiment of the present disclosure;

[0044] Figure 5 A schematic diagram of a flow chart of another method for identifying workstations in robot mapping provided in an embodiment of the present disclosure;

[0045] Figure 6 A schematic diagram of the structure of a workstation identification device in robot mapping provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0046] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments described herein, which are instead provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0047] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0048] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0049] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0050] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0051] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0052] In order to solve the above problems, the present disclosure provides a method for identifying workstations in robot mapping. Figures 1 to 5 The workstation identification method in robot mapping provided by an embodiment of the present disclosure is described in detail.

[0053] Figure 1 A schematic flow chart of a method for identifying workstations in robot mapping provided by an embodiment of the present disclosure is shown.

[0054] In the embodiment of the present disclosure, the workstation identification method in robot mapping can be executed by an electronic device, wherein the electronic device can be a device including a cleaning robot, a workstation, various devices, etc.

[0055] Among them, the robot mapping workstation identification method can be applied to the robot mapping workstation identification system, which includes a host and at least one passive base station. The passive base station can be a base station that does not need to be connected to a power supply. For example, the passive base station can be at least one of a water addition base station, a sewage discharge base station, and a water addition and sewage discharge base station.

[0056] like Figure 1 As shown, the method for identifying workstations in robot mapping may include the following steps.

[0057] S110 . During the mapping process, base station feature recognition is performed on the clustered set of current frame point cloud data collected by the host.

[0058] In an embodiment of the present disclosure, during the mapping process, the electronic device may perform base station feature recognition on a clustered set of current frame point cloud data collected by the host.

[0059] Optionally, the point cloud data of the current frame can be obtained by the host in real time scanning to obtain point cloud data in three-dimensional coordinates of the current frame.

[0060] Optionally, the cluster set of the current frame point cloud data may be a set obtained after clustering the current frame point cloud data. For a specific implementation, see below.

[0061] Optionally, the base station feature recognition may be used to identify whether the point cloud data in the set has a preset base station feature.

[0062] Optionally, the base station feature may be a feature used to characterize the data strength of the base station. For example, the base station feature may be a "strong weak strong weak strong" feature.

[0063] Specifically, before cleaning, the cleaning robot needs to quickly build an overall environmental map and determine the location of passive base stations. For example, for a multi-base station design, the cleaning robot starts from one of the base stations and needs to determine the corresponding environmental map and the locations of other base stations. Therefore, during the mapping process, the electronic device can perform base station feature recognition on the clustered set of current frame point cloud data collected by the host.

[0064] S120: If the cluster set of the current frame point cloud data identifies a suspected base station feature, multi-frame point cloud data collected on-site by the host and the current frame point cloud data are spliced ​​to obtain spliced ​​point cloud data.

[0065] In an embodiment of the present disclosure, if the clustered set of the current frame point cloud data identifies a suspected base station feature, the electronic device can perform multi-frame splicing of the multi-frame point cloud data collected on-site by the host and the current frame point cloud data to obtain spliced ​​point cloud data.

[0066] Optionally, the suspected base station feature may be a data strength feature used to characterize data similar to the base station feature. For example, the suspected base station feature may be a "weak strong weak", "strong weak", "weak strong" feature, etc.

[0067] Optionally, the multi-frame point cloud data may be point cloud data in three-dimensional coordinates of multiple frames collected in situ by the host.

[0068] Optionally, the stitched point cloud data may be data obtained by stitching point cloud data of multiple frames.

[0069] Specifically, after the electronic device performs base station feature recognition on the clustered set of current frame point cloud data, if the clustered set of current frame point cloud data recognizes suspected base station features, such as "weak strong weak", "strong weak", or "weak strong" features, the host collects multiple frames of point cloud data on the spot, and performs multi-frame splicing of the multi-frame point cloud data with the current frame point cloud data, for example, through the ICP (Iterative Closest Point) matching method to perform multi-frame splicing, thereby obtaining spliced ​​point cloud data.

[0070] S130: If the cluster set of the spliced ​​point cloud data recognizes a base station feature, a position score calculation is performed on the cluster set of the spliced ​​point cloud data to obtain a base station position and a base station score corresponding to the passive base station.

[0071] In an embodiment of the present disclosure, if the clustered set of stitched point cloud data recognizes a base station feature, the electronic device may calculate a position score for the clustered set of stitched point cloud data to obtain a base station position and a base station score corresponding to the passive base station.

[0072] Optionally, the position score calculation may be used to calculate the base station position and base station score corresponding to the point cloud data.

[0073] Optionally, the base station position may be a corresponding position of the base station in three-dimensional coordinates.

[0074] Optionally, the base station score may be used to characterize the credibility of the base station.

[0075] Specifically, after the electronic device performs base station feature recognition on the spliced ​​point cloud data, if the cluster set of the spliced ​​point cloud data recognizes the base station feature, it means that the cluster set of the spliced ​​point cloud data may correspond to a passive base station. The electronic device can calculate the position score of the cluster set of the spliced ​​point cloud data to obtain the base station position and base station score corresponding to the passive base station.

[0076] S140: Based on the base station position and the base station score, control the host to return to the passive base station for self-cleaning.

[0077] In the embodiment of the present disclosure, the electronic device may control the host to return to the passive base station for self-cleaning based on the base station position and the base station score.

[0078] Specifically, the electronic device can control the host to return to the passive base station for self-cleaning according to the base station position and the base station score. For example, the electronic device first determines a passive base station with a higher base station score, and controls the host to return to the passive base station according to the corresponding base station position. If the base station position is not a passive base station, the electronic device continues to determine the passive base station with the second highest base station score, and controls the host to return to the passive base station according to the corresponding base station position, until the correct passive base station is found.

[0079] Therefore, in the disclosed embodiment, during the mapping process, base station feature recognition can be performed on the clustered set of current frame point cloud data collected by the host. If the clustered set of the current frame point cloud data recognizes a suspected base station feature, multi-frame point cloud data collected on-site by the host and the current frame point cloud data are spliced ​​to obtain spliced ​​point cloud data. If the clustered set of the spliced ​​point cloud data recognizes a base station feature, a position score calculation is performed on the clustered set of the spliced ​​point cloud data to obtain the base station position and base station score corresponding to the passive base station. Then, based on the base station position and the base station score, the host is controlled to return to the passive base station for self-cleaning. Thus, the spliced ​​point cloud data is obtained by splicing the multi-frame point cloud data with the current frame point cloud data. The base station position and base station score corresponding to the passive base station are obtained by performing position score calculation based on the spliced ​​point cloud data that recognizes the base station feature, thereby improving the success rate of passive base station recognition.

[0080] In some embodiments, the passive base station is provided with a reflective marker, and the host is equipped with a laser radar. The host can release a laser signal to the environment through the laser radar, and the reflective marker can be a marker made of a material with a significantly higher reflectivity than most objects in the home scene, and the light spot intensity information presented by it has a higher intensity. According to the intensity information of the laser point information scanned by the laser radar, the high-reflective material and the general object material in the home environment can be distinguished.

[0081] Optionally, before S110, the robot mapping workstation identification method may also include: acquiring current frame point cloud data corresponding to the reflective mark in the surrounding environment through the laser radar scanning, the current frame point cloud data including position information and intensity information; and filtering the current frame point cloud data.

[0082] In an embodiment of the present disclosure, the electronic device can obtain the current frame point cloud data corresponding to the reflective mark in the surrounding environment through laser radar scanning.

[0083] Optionally, the current frame point cloud data includes position information and intensity information.

[0084] Specifically, during the mapping process of the cleaning robot, the electronic device can emit a laser signal to the environment through the laser radar on the host, and receive the laser point cloud generated by the reflection of the laser signal by different object materials in the environment, thereby obtaining the corresponding current frame point cloud data. The current frame point cloud data may include position information (in the Cartesian coordinate system) and intensity information (0-255), wherein when the intensity information is 255, it indicates that the point cloud data is a high-reflection point cloud; when the intensity information is less than 255, it indicates that the point cloud data is not a high-reflection point cloud.

[0085] Figure 2 A schematic structural diagram of a reflective logo provided by an embodiment of the present disclosure is shown.

[0086] like Figure 2 As shown, a plurality of reflective markers (markers) with a spacing (such as 5 cm) may be attached to the base station, such as 2, 3, etc., which are not limited here. Figure 2 The three reflective marks are on the same plane. The host can transmit laser signals to the environment through the laser radar, and receive the laser point cloud generated by the reflection of laser signals by different objects in the environment, such as the high-reflection point cloud reflected by the reflective mark and the low-reflection point cloud reflected by other materials. Assuming that the recognition position of the host and the base station is 1.5m, there are 3 reflective marks on the base station, and the laser angle resolution is 0.72 degrees, according to the preset reflective mark size calculation formula:

[0087] W=1.5*2*tan((3-1)*0.72 degrees / 2).

[0088] The calculated size W of the reflective mark is 5 cm.

[0089] Furthermore, the electronic device may filter the point cloud data of the current frame.

[0090] Specifically, the current frame point cloud data may contain useless data, so the electronic device can filter the current frame point cloud data. For example, it can filter out the point cloud data blocked by the top cover pillar of the laser radar, filter the point cloud data within the range of the host body, identify point cloud data with a long distance (such as more than 1.5m), filter point cloud data with intensity information of 0, etc.

[0091] Optionally, after filtering the current frame point cloud data, the robot mapping workstation identification method may also include: performing Euclidean clustering on the current frame point cloud data by judging the relationship between the position information of the two point clouds and a first preset distance threshold, and judging the relationship between the intensity information and a preset intensity threshold, to obtain a cluster set of the current frame point cloud data, wherein the cluster set includes a strong cluster set and a weak cluster set.

[0092] In an embodiment of the present disclosure, the electronic device can perform Euclidean clustering on the current frame point cloud data by determining the relationship between the position information of the two point clouds and a first preset distance threshold, and determining the relationship between the intensity information and a preset intensity threshold, to obtain a cluster set of the current frame point cloud data.

[0093] Optionally, the first preset distance threshold may be a clustered Euclidean distance threshold determined according to the distance between the host and the reflective mark.

[0094] Figure 3 A schematic structural diagram of another reflective marking provided in an embodiment of the present disclosure is shown.

[0095] like Figure 3 As shown in the figure, when the laser radar scans a high-reflectivity object (reflective mark), due to the strong reflection and scattering effect of the high-reflectivity object on the laser, the point cloud echo signal will be stronger than when scanning a low-reflectivity object, resulting in diffusion of the point cloud data when it is generated, as shown in the figure. Figure 3 When the laser radar scans a low-reflectivity object or a high-reflectivity object at close range, the point cloud data will appear in a straight line shape and will not jump dramatically; when the laser radar scans a high-reflectivity object, the point cloud data jump will increase with the distance to the object. For example, the maximum recognition distance of a laser radar point cloud is 10m, and low-reflectivity objects will be well recognized within 10m; but for high-reflectivity objects, point cloud data may not be generated if the distance is greater than 2m, and the point cloud data within 2m will also jump to varying degrees, and the closer the distance to the object, the smaller the jump will be. Therefore, it is defined that when the distance between the cleaning robot and the object in the environment exceeds 1.5m, no clustering processing is performed, and the distance between the cleaning robot and the object is greater than 0.8m, the first preset distance threshold is 0.04m, and the distance between the cleaning robot and the object is less than 0.8m, the first preset distance threshold is 0.015m.

[0096] Optionally, the preset intensity threshold may be a preset intensity value, for example, the preset intensity threshold may be 255.

[0097] Optionally, the cluster set includes a strong cluster set and a weak cluster set, wherein the strong cluster set is used to store point cloud data with intensity information of 255, and the weak cluster set is used to store point cloud data with intensity information less than 255.

[0098] Specifically, the electronic device can perform Euclidean clustering on the current frame point cloud data by judging the relationship between the position information of the two point clouds and the first preset distance threshold, and judging the relationship between the intensity information and the preset intensity threshold, that is, clustering multiple current frame point cloud data to obtain a cluster set including a strong cluster set and a weak cluster set.

[0099] Figure 4 A schematic diagram of a process of Euclidean clustering calculation provided by an embodiment of the present disclosure is shown.

[0100] like Figure 4 As shown, after filtering the point cloud data of the current frame, the electronic device determines whether it is the first point cloud data. If it is the first point cloud data, the point cloud data is placed in a strong clustering set A1 or a weak clustering set B1 according to the strength information and marked; if it is not the first point cloud data, whether the distance is less than a first preset distance threshold is calculated based on the Euclidean method, that is, whether the distance of the point cloud data is less than the first preset distance threshold based on the Euclidean method. If it is not less than the first preset distance threshold, the point cloud data is placed in a new strong clustering set A2 or a new weak clustering set B2 according to the strength information and marked; if it is less than the first preset distance threshold, the point cloud data is combined with the previous point cloud data. Perform strength comparison. If the previous point cloud data is 255 and the current point cloud data is 255, add the current point cloud data to the same strong clustering set as the previous point cloud data; if the previous point cloud data is less than 255 and the current point cloud data is 255, add the current point cloud data to a new strong clustering set; if the previous point cloud data is less than 255 and the current point cloud data is less than 255, add the current point cloud data to the same weak clustering set as the previous point cloud data; if the previous point cloud data is 255 and the current point cloud data is less than 255, add the current point cloud data to a new weak clustering set. Finally, a clustering set of the current frame point cloud data including a strong clustering set and a weak clustering set is obtained.

[0101] Optionally, S110 may specifically include: performing set size filtering on the cluster set of the current frame point cloud data based on a preset size; determining whether the cluster set of the current frame point cloud data is identified as a base station feature, and whether the distance between the strong cluster set and the weak cluster set meets a second preset distance threshold.

[0102] In an embodiment of the present disclosure, the electronic device may perform set size filtering on the cluster set of the current frame point cloud data based on a preset size.

[0103] Optionally, the preset size may be a pre-set filtering size. For example, the preset size may be 10, 20, etc., which is used as a limitation.

[0104] Specifically, the electronic device can perform set size filtering on the cluster set of the current frame point cloud data based on a preset size. For example, for a strong cluster set or a weak cluster set of the cluster set, there must be at least two point cloud data and the maximum size does not exceed the preset size of 10.

[0105] Further, the electronic device determines whether the cluster set of the current frame point cloud data is identified as a base station feature, and whether the distance between the strong cluster set and the weak cluster set meets a second preset distance threshold.

[0106] Optionally, the second preset distance threshold may be a distance threshold between pre-set sets. For example, the second preset distance threshold may be 0.03-0.06 m, based on the set center.

[0107] Specifically, the electronic device can determine whether the cluster set of the current frame point cloud data is identified as a base station feature, for example, whether the cluster set of the current frame point cloud data is a "strong-weak-strong-weak-strong" feature, and whether the distance between the strong cluster set and the weak cluster set meets the second preset distance threshold, for example, whether the distance between the set center of the strong cluster set and the set center of the weak cluster set meets the second preset distance threshold (0.03-0.06m).

[0108] Therefore, after the set size filtering is performed, even if the cluster set meets the base station characteristics, the distance between the sets will not meet the second preset distance threshold, thereby reducing the recognition error.

[0109] Optionally, S130 may specifically include: respectively calculating the set score of each set in the cluster set of the spliced ​​point cloud data, and summing the set scores of each set to obtain the base station score; performing least squares straight line fitting on the spliced ​​point cloud data to obtain the corresponding slope; and calculating the base station position based on the slope and the center value of the spliced ​​point cloud data.

[0110] In the embodiment of the present disclosure, the set score of each set in the cluster set of the spliced ​​point cloud data is calculated respectively, and the set scores of each set are summed to obtain the base station score.

[0111] Specifically, the electronic device can respectively calculate the set score of each set in the cluster set of the spliced ​​point cloud data, for example, the set score of each set is calculated by the preset formula score (Score) = weight (Weight) * point cloud size (Size) of the cluster set, wherein for the weight (Weight), when the cleaning robot is more than 0.8m away from the center of the set, the weight is 0.3; when the cleaning robot is not more than 0.8m away from the center of the set, the weight is 1-0.3, thereby calculating the set scores corresponding to each strong cluster set and weak cluster set in the cluster set, and then summing the set scores of each set to obtain the score of the cluster set of the spliced ​​point cloud data, that is, the base station score.

[0112] Furthermore, the electronic device may perform least square straight line fitting on the spliced ​​point cloud data to obtain a corresponding slope, and calculate the base station position according to the slope and a center value of the spliced ​​point cloud data.

[0113] Specifically, the electronic device can perform least squares straight line fitting on the spliced ​​point cloud data, use computer vision (CV) methods to obtain the slope of the fitting line, and obtain the posture of the cleaning robot relative to the base station based on the slope and the center value of the spliced ​​point cloud data, that is, the base station position.

[0114] Figure 5 A schematic flow chart of another method for identifying workstations in robot mapping provided by an embodiment of the present disclosure is shown.

[0115] like Figure 5 As shown, the electronic device can traverse the cluster set to determine whether the cluster set has the characteristics of a base station and whether the distance between each set meets the second preset distance threshold. If not, continue to traverse the cluster set; if satisfied, calculate the set score of each set in the cluster set respectively, and sum the set scores of each set to obtain the base station score, then perform least squares straight line fitting on the spliced ​​point cloud data to obtain the corresponding slope, and calculate the base station position based on the slope and the center value of the spliced ​​point cloud data.

[0116] Optionally, after S110, the robot mapping workstation identification method may further include: if the clustered set of the current frame point cloud data recognizes the base station feature, performing a position score calculation on the clustered set of the current frame point cloud data to obtain the base station position and the base station score corresponding to the passive base station.

[0117] In an embodiment of the present disclosure, if the clustering set of the current frame point cloud data recognizes the base station feature, such as recognizing the "strong-weak-strong-weak-strong" feature, the electronic device can perform a position score calculation on the clustering set of the current frame point cloud data to obtain the base station position and the base station score corresponding to the passive base station. For example, the specific real-time method refers to the above-mentioned method for calculating the position score of the clustering set of the spliced ​​point cloud data, first traverse the clustering set of the current frame point cloud data, determine whether the clustering set has the base station feature and the distance between each set meets the second preset distance threshold, if not, continue to traverse the clustering set; if satisfied, respectively calculate the set score of each set in the clustering set of the current frame point cloud data, and sum the set scores of each set to obtain the base station score, then perform least squares straight line fitting on the spliced ​​point cloud data to obtain the corresponding slope, and calculate the base station position based on the slope and the center value of the spliced ​​point cloud data.

[0118] Optionally, S120 may specifically include: performing multi-frame splicing of each frame of point cloud data in the multi-frame point cloud data with the current frame of point cloud data in sequence, until the obtained spliced ​​point cloud data with no more than a preset number of frames spliced ​​together can identify the base station feature.

[0119] In an embodiment of the present disclosure, the electronic device may perform multi-frame splicing of each frame of point cloud data in the multi-frame point cloud data with the current frame of point cloud data in sequence until the obtained spliced ​​point cloud data with no more than a preset number of frames spliced ​​together can identify the base station feature.

[0120] For example, when the current frame point cloud data identifies a suspected base station feature, the electronic device can control the host to continue to collect multiple frames of point cloud data in situ, and directly splice the second frame of point cloud data in the multi-frame point cloud data with the current frame point cloud data (the first frame of point cloud data) for multiple frames. The solution method of the spliced ​​rotation matrix adopts the existing ICP (Iterative Closest Point) matching method is used, and base station feature recognition is performed on the stitched point cloud data after stitching. If the stitched point cloud data recognizes the base station feature, the stitching is terminated; if the stitched point cloud data does not recognize the base station feature, the electronic device continues to perform multi-frame stitching of the third frame of point cloud data in the multi-frame point cloud data with the stitched point cloud data, and performs base station feature recognition. If the new stitched point cloud data recognizes the base station feature, the stitching is terminated; if the new stitched point cloud data does not recognize the base station feature, the electronic device will discard the current frame point cloud data (the first frame point cloud data), and continue to perform multi-frame stitching of the fourth frame of point cloud data in the multi-frame point cloud data with the stitched point cloud data, until the stitched point cloud data obtained by splicing no more than a preset number of frames recognizes the base station feature, such as the preset number of frames can be 3, 4, etc., that is, the point cloud data used for multi-frame stitching does not exceed 3 frames.

[0121] Optionally, S120 may specifically include: determining in turn for each frame of point cloud data in the multiple frames of point cloud data whether the base station feature is recognized; and performing multi-frame splicing of each frame of point cloud data that recognizes the base station feature with the current frame of point cloud data until no more than a preset number of frames of spliced ​​point cloud data recognize the base station feature.

[0122] In an embodiment of the present disclosure, when the current frame of point cloud data identifies a suspected base station feature, the electronic device can determine in turn whether each frame of point cloud data in the multiple frames of point cloud data collected by the host in situ identifies the base station feature. For example, the electronic device determines whether the second frame of point cloud data in the multiple frames of point cloud data identifies the base station feature.

[0123] Furthermore, the electronic device may perform multi-frame splicing of each frame of point cloud data that identifies the base station feature with the current frame of point cloud data until no more than a preset number of frames of spliced ​​point cloud data are obtained to identify the base station feature.

[0124] For example, when the current frame point cloud data recognizes a suspected base station feature, the electronic device can determine whether the second frame point cloud data in the multi-frame point cloud data recognizes the base station feature. If the second frame point cloud data does not recognize the base station feature, and recognizes a suspected base station feature (such as "weak strong weak", "strong weak strong weak", etc.), the second frame point cloud data and the current frame point cloud data (first frame point cloud data) are spliced ​​into multiple frames, and the base station feature is identified for the spliced ​​point cloud data. If the spliced ​​point cloud data recognizes the base station feature, the splicing is terminated; if the spliced ​​point cloud data does not recognize the base station feature, the electronic device continues to judge whether the third frame point cloud data in the multi-frame point cloud data is similar to the second frame point cloud data, which is not repeated here, until the spliced ​​point cloud data spliced ​​with no more than a preset number of frames recognizes the base station feature.

[0125] Optionally, S120 may specifically include: if the clustering set of the current frame point cloud data identifies a suspected base station feature, the host rotates left and right at a first preset angle on the spot to collect first multi-frame point cloud data, and performs multi-frame splicing of the first multi-frame point cloud data and the current frame point cloud data to obtain first spliced ​​point cloud data; if the clustering set of the first spliced ​​point cloud data does not identify the base station feature but identifies the enhanced suspected base station feature, the host rotates left and right at a second preset angle on the spot to collect second multi-frame point cloud data, and performs multi-frame splicing of the second multi-frame point cloud data and the current frame point cloud data to obtain second spliced ​​point cloud data; if the clustering set of the second spliced ​​point cloud data does not identify the base station feature but identifies the enhanced suspected base station feature, the host rotates left and right at a third preset angle on the spot to collect third multi-frame point cloud data, and performs multi-frame splicing of the third multi-frame point cloud data and the current frame point cloud data to obtain third spliced ​​point cloud data.

[0126] In an embodiment of the present disclosure, if the clustered set of the current frame point cloud data identifies a suspected base station feature, the host rotates left and right at a first preset angle in situ to collect first multi-frame point cloud data, and performs multi-frame splicing of the first multi-frame point cloud data and the current frame point cloud data to obtain first spliced ​​point cloud data.

[0127] Optionally, the first preset angle may be a preset angle. For example, the first preset angle may be 45°.

[0128] Optionally, the first multi-frame point cloud data may be multi-frame point cloud data collected when the host is rotated left or right by a first preset angle.

[0129] Specifically, if the clustering set of the current frame point cloud data identifies suspected base station features, such as "weak strong weak", "strong weak", and "weak strong" features, the host of the cleaning robot can rotate left and right at a first preset angle in situ to collect first multi-frame point cloud data, such as rotating left and right 45° in situ to collect first multi-frame point cloud data, and perform multi-frame splicing of the first multi-frame point cloud data and the current frame point cloud data, such as multiple three-frame splicing, to obtain first spliced ​​point cloud data.

[0130] Furthermore, if the cluster set of the first spliced ​​point cloud data does not identify the base station feature but identifies the enhanced suspected base station feature, the host rotates left and right at a second preset angle in situ to collect second multi-frame point cloud data, and performs multi-frame splicing of the second multi-frame point cloud data with the current frame point cloud data to obtain second spliced ​​point cloud data.

[0131] For example, the enhanced suspected base station characteristics may be "strong-weak-strong" or "weak-strong-weak-strong" characteristics.

[0132] Optionally, the second preset angle may be a pre-set angle. For example, the first preset angle may be 90°.

[0133] Optionally, the second multi-frame point cloud data may be multi-frame point cloud data collected when the host is rotated left and right by a second preset angle.

[0134] Specifically, when the host rotates left and right by a first preset angle (45°) in situ to collect a first plurality of frames of point cloud data and stitches the first stitched point cloud data, when the base station feature is not identified but the enhanced suspected base station feature is identified, such as changing from "weak strong weak", "strong weak", and "weak strong" features to "strong weak strong", "weak strong weak strong" features, the host can rotate left and right by a second preset angle (90°) in situ to collect a second plurality of frames of point cloud data, and perform multi-frame stitching of the second plurality of frames of point cloud data with the current frame of point cloud data, such as multiple three-frame stitching, to obtain a second stitched point cloud data.

[0135] Furthermore, if the cluster set of the second stitched point cloud data does not identify the base station features but identifies the enhanced suspected base station features, the host rotates left and right at a third preset angle in situ to collect third multi-frame point cloud data, and stitches the third multi-frame point cloud data with the current frame point cloud data to obtain third stitched point cloud data.

[0136] For example, the enhanced suspected base station characteristics may be "strong-weak-strong-weak" or "weak-strong-weak-strong-weak" characteristics.

[0137] Optionally, the third preset angle may be a pre-set angle. For example, the first preset angle may be 180°.

[0138] Optionally, the third multi-frame point cloud data may be multi-frame point cloud data collected when the host is rotated left or right by a third preset angle.

[0139] Specifically, when the host rotates left and right at a second preset angle (90°) in situ to collect a second multi-frame point cloud data and stitches the second stitched point cloud data, when the base station features are not identified but the enhanced suspected base station features are identified, such as from "strong weak strong", "weak strong weak strong" features to "strong weak strong weak", "weak strong weak strong weak" features, the host can rotate left and right at a third preset angle (180°) in situ to collect a third multi-frame point cloud data, and perform multi-frame stitching of the third multi-frame point cloud data with the current frame point cloud data, such as multiple three-frame stitching, to obtain the third stitched point cloud data, until the stitched point cloud data obtained identifies the base station features.

[0140] Optionally, S120 may specifically include: if the cluster set of the current frame point cloud data identifies a suspected base station feature, determining the target rotation angle according to the suspected base station feature and the position information of the current frame point cloud data; the host rotates left and right in situ by the target rotation angle to collect multi-frame point cloud data, and performs multi-frame splicing of the multi-frame point cloud data with the current frame point cloud data to obtain the spliced ​​point cloud data.

[0141] In an embodiment of the present disclosure, if the clustered set of the current frame point cloud data identifies a suspected base station feature, the electronic device determines the target rotation angle based on the suspected base station feature and the position information of the current frame point cloud data.

[0142] Optionally, the target rotation angle may be an angle at which the host rotates left and right in situ.

[0143] Specifically, if the clustering set of the current frame point cloud data identifies a suspected base station feature, the electronic device can determine the target rotation angle of the host's left-right rotation in situ based on the suspected base station feature and the location information of the current frame point cloud data. For example, if the suspected base station feature is a "strong-weak-strong-weak" feature, the electronic device can accurately calculate the host's left-right rotation target angle of 30° based on the location information to obtain the base station feature; if the suspected base station feature is a "strong-weak" feature, the electronic device can accurately calculate the host's left-right rotation target angle of 160° based on the location information to obtain the base station feature, etc., which is not limited here.

[0144] In the disclosed embodiment, during the mapping process, the passive base station may be provided with 1, 2, 3, etc. reflective marks, the size of the reflective marks may be 5cm*5cm, and the spacing between the reflective marks is 5cm. When the passive base station is provided with 1 reflective mark, the cleaning robot rotates 45° left and right near the reflective mark to collect multi-frame point cloud data, and performs multi-frame splicing of the multi-frame point cloud data with the current frame point cloud data to obtain spliced ​​point cloud data; when the passive base station is provided with 2 reflective marks, the cleaning robot first rotates 45° left and right near the reflective mark to collect multi-frame point cloud data, and performs multi-frame splicing of the multi-frame point cloud data with the current frame point cloud data to obtain spliced ​​point cloud data, and then rotates 90° and 180° left and right near the reflective mark to collect multi-frame point cloud data, and performs multi-frame splicing of the multi-frame point cloud data with the current frame point cloud data to obtain spliced ​​point cloud data, etc., which are not limited here.

[0145] Figure 6 A schematic structural diagram of a workstation identification device in robot mapping provided by an embodiment of the present disclosure is shown.

[0146] In some embodiments of the present disclosure, Figure 6The workstation identification equipment in the robot mapping shown may include equipment such as cleaning robots, workstations, and various devices.

[0147] like Figure 6 As shown, the workstation identification device in the robot mapping may include a processor 601 and a memory 602 storing computer program instructions.

[0148] Specifically, the processor 601 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0149] The memory 602 may include a large capacity memory for information or instructions. By way of example and not limitation, the memory 602 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive or a combination of two or more of these. Where appropriate, the memory 602 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 602 may be inside or outside the integrated gateway device. In a particular embodiment, the memory 602 is a non-volatile solid-state memory. In a particular embodiment, the memory 602 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (Electrically Erasable Programmable ROM, EEPROM), an electrically rewritable ROM (EAROM) or a flash memory, or a combination of two or more of these.

[0150] The processor 601 reads and executes the computer program instructions stored in the memory 602 to perform the steps of the workstation identification method in the robot mapping provided by the embodiment of the present disclosure.

[0151] In one example, the robot mapping workstation identification device may also include a transceiver 603 and a bus 604. Figure 6 As shown, the processor 601, the memory 602 and the transceiver 603 are connected via a bus 604 and communicate with each other.

[0152] The bus 604 includes hardware, software, or both. For example, but not limitation, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, the bus 604 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0153] The embodiments of the present disclosure also provide a computer-readable storage medium, which may store a computer program. When the computer program is executed by a processor, the processor implements the method for identifying workstations in robot mapping provided by the embodiments of the present disclosure.

[0154] The above-mentioned storage medium may, for example, include a memory 602 of computer program instructions, and the above-mentioned instructions may be executed by a processor 601 of a robot mapping workstation identification device to complete the robot mapping workstation identification method provided in the embodiment of the present disclosure. Optionally, the storage medium may be a non-temporary computer-readable storage medium, for example, a non-temporary computer-readable storage medium may be a ROM, a random access memory (Random Access Memory, RAM), a compact disc read-only memory (Compact Disc ROM, CD-ROM), a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0155] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0156] The above description is only a specific embodiment of the present disclosure, so that those skilled in the art can understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to the embodiments described herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for identifying workstations in robot mapping, characterized in that: Applied to a workstation identification system for robot mapping, the system includes a host and at least one passive base station, and the method includes: During the mapping process, base station feature recognition is performed on the clustered set of current frame point cloud data collected by the host; If the cluster set of the current frame point cloud data identifies a suspected base station feature, performing multi-frame splicing of the multi-frame point cloud data collected on-site by the host and the current frame point cloud data to obtain spliced ​​point cloud data; If the cluster set of the spliced ​​point cloud data recognizes a base station feature, a position score calculation is performed on the cluster set of the spliced ​​point cloud data to obtain a base station position and a base station score corresponding to the passive base station; Based on the base station position and the base station score, the host is controlled to return to the passive base station for self-cleaning.

2. The method according to claim 1, characterized in that The passive base station is provided with a reflective mark, and the host is equipped with a laser radar; Wherein, before performing base station feature recognition on the cluster set of the current frame point cloud data collected by the host, the method further includes: Acquire the current frame point cloud data corresponding to the reflective mark in the surrounding environment by scanning with the laser radar, wherein the current frame point cloud data includes position information and intensity information; The current frame point cloud data is filtered.

3. The method according to claim 2, characterized in that After filtering the current frame point cloud data, the method further includes: By determining the relationship between the position information of the two point clouds and a first preset distance threshold, and determining the relationship between the intensity information and a preset intensity threshold, Euclidean clustering is performed on the current frame point cloud data to obtain a cluster set of the current frame point cloud data, wherein the cluster set includes a strong cluster set and a weak cluster set.

4. The method according to claim 3, characterized in that The performing base station feature recognition on the clustered set of current frame point cloud data collected by the host includes: Performing set size filtering on the cluster set of the current frame point cloud data based on a preset size; It is determined whether the cluster set of the current frame point cloud data is identified as a base station feature, and whether the distance between the strong cluster set and the weak cluster set meets a second preset distance threshold.

5. The method according to claim 1, characterized in that The step of calculating the position score of the clustered set of the spliced ​​point cloud data to obtain the base station position and base station score corresponding to the passive base station includes: Respectively calculating the set score of each set in the cluster set of the spliced ​​point cloud data, and summing the set scores of each set to obtain the base station score; Performing least square straight line fitting on the spliced ​​point cloud data to obtain a corresponding slope; The base station position is calculated based on the slope and the center value of the spliced ​​point cloud data.

6. The method according to claim 1, characterized in that After performing base station feature recognition on the clustered set of current frame point cloud data collected by the host, the method further includes: If the cluster set of the current frame point cloud data recognizes the base station feature, a position score calculation is performed on the cluster set of the current frame point cloud data to obtain the base station position and the base station score corresponding to the passive base station.

7. The method according to claim 1, characterized in that The step of performing multi-frame splicing on the multi-frame point cloud data collected on-site by the host and the current frame point cloud data to obtain spliced ​​point cloud data includes: Each frame of point cloud data in the multiple frames of point cloud data is sequentially spliced ​​with the current frame of point cloud data until the obtained spliced ​​point cloud data with no more than a preset number of frames spliced ​​together can identify the base station feature.

8. The method according to claim 1, characterized in that The step of performing multi-frame splicing on the multi-frame point cloud data collected on-site by the host and the current frame point cloud data to obtain spliced ​​point cloud data includes: Determining in turn whether the base station feature is recognized for each frame of point cloud data in the multiple frames of point cloud data; Each frame of point cloud data that identifies the base station feature is spliced ​​with the current frame of point cloud data for multiple frames until no more than a preset number of frames of spliced ​​point cloud data are obtained to identify the base station feature.

9. The method according to claim 1, characterized in that: If the cluster set of the current frame point cloud data identifies a suspected base station feature, the multi-frame point cloud data collected on-site by the host and the current frame point cloud data are spliced ​​to obtain spliced ​​point cloud data, including: If the cluster set of the current frame point cloud data identifies a suspected base station feature, the host rotates left and right at a first preset angle to collect first multi-frame point cloud data, and performs multi-frame splicing of the first multi-frame point cloud data and the current frame point cloud data to obtain first spliced ​​point cloud data; If the cluster set of the first spliced ​​point cloud data does not identify the base station feature but identifies the enhanced suspected base station feature, the host rotates left and right at a second preset angle to collect second multi-frame point cloud data, and splices the second multi-frame point cloud data with the current frame point cloud data to obtain second spliced ​​point cloud data; If the cluster set of the second stitched point cloud data does not identify the base station feature but identifies the enhanced suspected base station feature, the host rotates left and right at a third preset angle to collect third multi-frame point cloud data, and stitches the third multi-frame point cloud data with the current frame point cloud data to obtain third stitched point cloud data.

10. The method according to claim 1, characterized in that If the cluster set of the current frame point cloud data identifies a suspected base station feature, the multi-frame point cloud data collected on-site by the host and the current frame point cloud data are spliced ​​to obtain spliced ​​point cloud data, including: If the cluster set of the current frame point cloud data identifies a suspected base station feature, determining a target rotation angle according to the suspected base station feature and the position information of the current frame point cloud data; The host rotates left and right at the target rotation angle in situ to collect multiple frames of point cloud data, and performs multi-frame splicing of the multiple frames of point cloud data with the current frame of point cloud data to obtain the spliced ​​point cloud data.

Images