A method for a cleaning robot to return to a base station

Abstract

This application provides a method for a cleaning robot to return to a base station. The cleaning robot is equipped with a lidar, and the base station has an opening with a positioning component at the opening, forming a cavity for accommodating the cleaning robot. The method includes: identifying the base station based on a set of laser point clouds obtained by lidar scanning; if identification is successful, identifying geometric features from the laser point cloud set and determining the position information of the positioning component based on the geometric features; determining the central axis of the cavity based on the position information and generating a movement path based on the central axis; and controlling the cleaning robot to move along the movement path back to the base station. This solution eliminates the influence of laser point cloud measurement errors on the positioning accuracy of the cavity's central axis, improves the positioning accuracy of the cavity's central axis, and thus ensures the accuracy of the movement path generation, enabling the cleaning robot to accurately return to the base station.

Description

Technical Field

[0001] This application relates to the field of robotics, and in particular to a method for a cleaning robot to return to a base station. Background Technology

[0002] Currently, most household cleaning robots on the market are equipped with base stations. The cleaning robots can autonomously return to the base station to recharge or perform operations such as washing mops and refilling water. This requires the cleaning robots to be able to automatically identify and locate the position and direction of the base station.

[0003] Regarding base station identification and positioning, the laser signature-based positioning solution is widely used in the market. This solution utilizes the built-in LiDAR on the cleaning robot, eliminating the need for additional sensors. It boasts lower hardware costs, doesn't occupy structural space, and doesn't require a power supply to the base station, offering significant advantages. The laser signature-based positioning solution scans the contours of surrounding objects using LiDAR, then identifies and positions the base station based on the resulting laser point cloud data. Specifically, to further improve positioning performance and effectiveness, the industry often uses highly reflective film as the reflective material, affixed to specific parts of the base station. This allows the LiDAR on the cleaning robot to scan the base station from a greater distance and at a wider angle. Simultaneously, by differentiating the reflective characteristics of different areas on the reflective film, alternating high-reflectivity and low-reflectivity regions are created. The width of each region is then adjusted to ensure that the laser point cloud scanned by the reflective film area has unique shape and size characteristics.

[0004] The base station includes a cavity for housing the cleaning robot. When using a laser signature positioning scheme, the base station is first identified based on laser point cloud data obtained from LiDAR scanning. Once identified, the central axis of the cavity is determined based on the point cloud data. Then, a movement path is generated based on the central axis of the cavity, allowing the cleaning robot to travel along this path back to the base station.

[0005] However, in existing technologies, the area where reflective film is placed on the base station is generally small, or the overall size of the reflective film is generally small due to cost considerations. This results in a limited number of laser points formed on the reflective film. On the other hand, because the cost of the LiDAR used in home robots is relatively low, the number of laser points detected and their positions can fluctuate. These two factors combined lead to a large measurement error in the laser point cloud. In this situation, determining the central axis of the cavity solely based on the laser point cloud data has a large error. Furthermore, the movement path generated based on the central axis of the cavity also has a large error. This causes the cleaning robot to deviate from the path while returning to the base station, ultimately preventing it from returning. Even worse, it can lead to a collision between the cleaning robot and the base station. Summary of the Invention

[0006] The purpose of this application is to provide a method for a cleaning robot to return to a base station, which eliminates the influence of laser point cloud measurement errors on the positioning accuracy of the central axis of the cavity, improves the positioning accuracy of the central axis of the cavity, and thus ensures the accuracy of the movement path generation, enabling the cleaning robot to accurately return to the base station.

[0007] This application provides a method for a cleaning robot to return to a base station. The cleaning robot is equipped with a lidar, and the base station has an opening with a positioning component at the opening. The positioning component forms a receiving cavity for accommodating the cleaning robot.

[0008] The methods for cleaning robots to return to the base station include:

[0009] Base stations are identified based on the laser point cloud set obtained from lidar scanning;

[0010] If the identification is successful, the geometric features are extracted from the laser point cloud set, and the position information of the positioning component is determined based on the geometric features.

[0011] The position information of the central axis of the receiving cavity is determined based on the position information of the positioning component, and a movement path is generated based on the position information of the central axis of the receiving cavity;

[0012] Control the cleaning robot to travel along the movement path and return to the base station.

[0013] In one embodiment, the method for the cleaning robot to return to the base station before identifying the base station based on the set of laser point clouds obtained from the lidar scan further includes:

[0014] Control the cleaning robot to move to the first target location at a preset distance from the base station.

[0015] In one embodiment, the positioning component is provided with a positioning tag;

[0016] Based on the laser point cloud set obtained from lidar scanning, base stations are identified, including:

[0017] Determine whether a target point cloud set corresponding to a positioning tag exists in the laser point cloud set;

[0018] If a target point cloud set exists, the base station has been successfully identified;

[0019] If the target point cloud set does not exist, the base station was not successfully identified.

[0020] In one embodiment, the location tag includes a first location tag and a second location tag;

[0021] A first positioning part and a second positioning part are provided at intervals on the surface of the positioning component, a first positioning label is provided on the first positioning part, and a second positioning label is provided on the second positioning part;

[0022] Determining whether a target point cloud set corresponding to a positioning tag exists in the laser point cloud set includes:

[0023] Determine whether a first point cloud set and a second point cloud set that meet preset conditions exist in the laser point cloud set;

[0024] Among them, the first point cloud set and the second point cloud set are the point cloud sets corresponding to the first positioning label and the second positioning label.

[0025] In one embodiment, the method for the cleaning robot to return to the base station before extracting geometric features from the laser point cloud set further includes:

[0026] The feature point cloud set corresponding to the base station is selected from the laser point cloud set in order to extract geometric features from the feature point cloud set.

[0027] In one embodiment, a first positioning part, a feature part, and a second positioning part are sequentially provided on the surface of the positioning component. The first positioning part and the feature part intersect at a first straight line, and the second positioning part and the feature part intersect at a second straight line.

[0028] Geometric features are extracted from the laser point cloud dataset, and the position information of the positioning component is determined based on these geometric features, including:

[0029] Geometric features are extracted from the feature point cloud set, and the position information of the first and second straight lines is determined based on the geometric features.

[0030] In one embodiment, the first positioning part and the second positioning part are planar structures and are located on the same plane;

[0031] Geometric features are extracted from the feature point cloud set, and the position information of the first and second straight lines is determined based on the geometric features, including:

[0032] Extract the straight line features corresponding to the first and second localization parts from the feature point cloud set;

[0033] Extract the circular arc features corresponding to the feature portion from the feature point cloud set;

[0034] Determine the first and second intersection points of the arc and straight line features, and determine the position information of the first and second straight lines based on the first and second intersection points.

[0035] In one embodiment, determining the position information of the central axis of the receiving cavity based on the position information of the positioning component includes:

[0036] Based on the first intersection point, the second intersection point, and the straight line characteristics, determine the position information of the central axis of the receiving cavity.

[0037] In one embodiment, the method for the cleaning robot to return to the base station further includes:

[0038] If the base station is not successfully identified, control the cleaning robot to perform rotation and / or translation trigger actions within the preset range;

[0039] During or after the triggering action is executed, the laser point cloud set scanned by the LiDAR is reacquired, and the base station is identified based on the reacquired laser point cloud set.

[0040] In one embodiment, a cleaning component is provided at the opening, and the cleaning component and the positioning component form a receiving cavity; the cleaning robot is provided with a cleaning component for wiping the surface to be cleaned, and the cleaning component is provided with a cleaning component for cleaning the cleaning component.

[0041] Controlling the cleaning robot to travel along the designated path back to the base station includes:

[0042] The cleaning robot is controlled to move to the second target position according to the movement path; wherein, at the second target position, the central axis of the cleaning robot is basically coincident with the central axis of the receiving cavity;

[0043] The cleaning robot is driven to rotate at a preset angle threshold so that the cleaning parts are placed on the cleaning components when the rotating cleaning robot returns to the base station along the movement path.

[0044] In this application, when the cleaning robot needs to return to the base station, the base station is first identified based on the laser point cloud set obtained from the LiDAR scan. After successful identification, geometric features are extracted from the laser point cloud data, and then the position information of the positioning component on the base station is determined based on these geometric features. After identifying the position information of the positioning component on the base station, the central axis of the receiving cavity is determined based on the position information of the positioning component, and then a movement path is generated based on the central axis of the receiving cavity, so that the cleaning robot can travel along the movement path to return to the base station.

[0045] As can be seen from the above, because the position information of the positioning component is determined based on the extracted contour information in this application, changes in the position and number of laser points will not interfere with the extraction result of the contour information, resulting in high accuracy of the obtained position information of the positioning component. Based on this, the central axis of the receiving cavity determined based on the position information of the positioning component will also have high accuracy; furthermore, it ensures the accuracy of the generated movement path, fully guaranteeing that the cleaning robot will not deviate when returning to the base station according to the movement path, and can ensure that the cleaning robot accurately returns to the base station. In addition, ensuring that the cleaning robot can accurately return to the base station also avoids collisions between the cleaning robot and the base station, thus protecting the cleaning robot to a certain extent.

[0046] In summary, this application eliminates the influence of laser point cloud measurement errors on the positioning accuracy of the central axis of the cavity, improves the positioning accuracy of the central axis of the cavity, and thus ensures the accuracy of the movement path generation, enabling the cleaning robot to accurately return to the base station. Attached Figure Description

[0047] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly described below.

[0048] Figure 1 This is a schematic diagram of the structure of a base station provided in the first embodiment of this application;

[0049] Figure 2 This is a schematic diagram of the structure of a cleaning robot provided in one embodiment of this application;

[0050] Figure 3 This is a schematic diagram of the structure of a base station provided in the second embodiment of this application;

[0051] Figure 4 Provided for an embodiment of this application Figure 3 A magnified view of a portion of point A in the middle;

[0052] Figure 5 This is a schematic diagram of the structure of a positioning component provided in an embodiment of this application;

[0053] Figure 6 This is a schematic diagram of the structure of a control module provided in one embodiment of this application;

[0054] Figure 7 A flowchart illustrating a method for a cleaning robot to return to a base station according to an embodiment of this application;

[0055] Figure 8A schematic diagram of a laser radar scanning a base station according to an embodiment of this application;

[0056] Figure 9 A schematic diagram of the outline of the first side as viewed from a top angle, provided for an embodiment of this application;

[0057] Figure 10 This is a schematic diagram of the laser point cloud set formed behind the first side of the lidar scanning and positioning component provided in an embodiment of this application;

[0058] Figure 11 Provided for an embodiment of this application Figure 10 A schematic diagram of feature extraction after feature point cloud set in the image;

[0059] Figure 12 This is a schematic diagram illustrating the determination of the position information of the central axis of the receiving cavity according to an embodiment of this application;

[0060] Figure 13 This is a schematic diagram showing the position of a cleaning robot provided in one embodiment of this application;

[0061] Figure 14 A schematic diagram of an optical theoretical model provided in an embodiment of this application;

[0062] Figure 15 This is a schematic diagram of the movement of the cleaning robot provided in the first embodiment of this application;

[0063] Figure 16 This is a schematic diagram of the movement of the cleaning robot provided in the second embodiment of this application;

[0064] Figure 17 This is a schematic diagram of the movement of the cleaning robot provided in the third embodiment of this application;

[0065] Figure 18 This is a schematic diagram of the movement of the cleaning robot provided in the fourth embodiment of this application;

[0066] Figure 19 This is a schematic diagram of the movement of the cleaning robot provided in the fifth embodiment of this application.

[0067] Figure label:

[0068] 1-Base station; 10-Positioning component; 110-First side; 111-First positioning part; 112-Feature part; 113-Second positioning part; 114-First straight line; 115-Second straight line; 120-Second side; 30-Opening; 40-Receiving cavity; 50-Cleaning assembly; 51-Cleaning component; 2-Cleaning robot; 21-Protective cover; 22-LiDAR; 23-Control module; 231-Memory; 232-Bus; 233-Processor; 24-Cleaning component. Detailed Implementation

[0069] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0070] Similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0071] Please refer to Figure 1 This is a schematic diagram of the structure of base station 1 provided in the first embodiment of this application. Please refer to... Figure 2 This is a schematic diagram of the structure of a cleaning robot 2 provided in an embodiment of this application. This application provides a cleaning system, such as... Figure 1 and Figure 2 As shown, the cleaning system in this application includes a base station 1 and a cleaning robot 2. The cleaning system is used to clean surfaces in a user's home that are awaiting cleaning. Figure 1 As shown, the base station 1 in this application has an opening 30, and a positioning component 10 is provided at the opening 30. The positioning component 10 forms a receiving cavity 40 for accommodating the cleaning robot 2. The positioning component 10 is used to fix the cleaning robot 2 in the receiving cavity 40. When the cleaning robot 2 is located in the receiving cavity 40, the base station 1 can perform operations such as cleaning and charging on the cleaning robot 2.

[0072] like Figure 2 As shown, the cleaning robot 2 in this application is equipped with a lidar 22. The lidar 22 can rotate 360° and scan objects around the cleaning robot 2, forming multiple laser points after scanning. The cleaning robot 2 uses these multiple laser points to locate the position of the base station 1. The lidar 22 scans objects by rotating and emitting and receiving infrared laser beams. Specifically, as... Figure 2 As shown, a protective cover 21 for protecting the lidar 22 can be provided on the top surface of the cleaning robot 2. The lidar 22 is located inside the protective cover 21, as shown. Figure 2 As shown, the bottom surface of the cleaning robot 2 is provided with a cleaning component 24 for wiping the surface to be cleaned.

[0073] In one embodiment, please refer to Figure 3 This is a schematic diagram of the structure of base station 1 provided in the second embodiment of this application. Please refer to... Figure 4 This is provided as an embodiment of the present application. Figure 3 A magnified view of a portion at point A. Please refer to... Figure 5This is a schematic diagram of the structure of the positioning component 10 provided in an embodiment of this application. Figure 3 , Figure 4 and Figure 5 As shown, the positioning component 10 in this application includes a first side surface 110 and a second side surface 120; the second side surface 120 is fixed inside the base station 1, and the first side surface 110 is exposed at the opening 30, so that the lidar 22 can scan the first side surface 110 when scanning the positioning component 10. The first side surface 110 is provided with a first positioning part 111, a feature part 112, and a second positioning part 113; the first positioning part 111 and the second positioning part 113 are spaced apart; the feature part 112 is located between the first positioning part 111 and the second positioning part 113; the first positioning part 111 and the feature part 112 intersect at a first straight line 114, and the second positioning part 113 and the feature part 112 intersect at a second straight line 115; the first positioning part 111 and the second positioning part 113 are planar structures and are located on the same plane; the feature part 112 is an arc-shaped groove structure.

[0074] In one embodiment, such as Figure 3 As shown, base station 1 also includes a cleaning assembly 50, which is located at the opening 30. The cleaning assembly and positioning component 10 form a receiving cavity 40. The cleaning assembly 50 is provided with a cleaning component 51 for cleaning the cleaning component 24.

[0075] Please refer to Figure 6 This is a schematic diagram of the structure of the control module 23 provided in one embodiment of this application. The cleaning robot 2 also includes a control module 23, such as... Figure 6 As shown, the control module 23 includes: at least one processor 233 and a memory 231. Figure 6 Taking a processor 233 as an example, the processor 233 and the memory 231 are connected via a bus 232. The memory 231 stores instructions that can be executed by the processor 233. The instructions are executed by the processor 233 to enable the control module 23 to execute all or part of the process of the method in the following embodiments. In this application, the control module 23 is connected to the lidar 22. The control module 23 is used to control the working state of the cleaning robot 2 when it is performing a cleaning task. When the cleaning robot 2 is not within the base station 1, the control module 23 is used to receive multiple laser points obtained by the lidar 22 scanning, locate the position of the base station 1 based on the multiple laser points, and then control the cleaning robot 2 to return to the base station 1 based on the positioning result.

[0076] The memory 231 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable red-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0077] This application also provides a computer-readable storage medium storing a computer program that can be executed by a processor 233 to perform the method provided in this application for the cleaning robot 2 to return to the base station 1.

[0078] Please refer to Figure 7 This is a flowchart illustrating a method for a cleaning robot 2 to return to base station 1 according to an embodiment of this application. Figure 7 As shown, the method includes the following steps S210-S240.

[0079] Step S210: Identify base station 1 based on the laser point cloud set obtained by the LiDAR 22 scanning.

[0080] In this step, when the user needs to clean the surface awaiting cleaning, the control module 23 controls the cleaning robot 2 to begin the cleaning task. During or after the cleaning task, the cleaning robot 2 needs to return to the base station 1, where the base station 1 will charge and clean the robot. When the cleaning robot 2 needs to return to the base station 1, the control module 23 drives the lidar 22 to scan the surrounding environment in real time. After scanning, the lidar 22 transmits the scanned multiple laser points to the control module 23, which then identifies the base station 1 based on the laser point cloud set composed of these laser points. If the control module 23 finds multiple laser points corresponding to the base station 1 in the laser point cloud set, it indicates that the approximate location of the base station 1 has been identified, completing the identification of the base station 1. The multiple laser points corresponding to the base station 1 are generated by the lidar 22 scanning the base station 1; the location of these laser points is the location of the base station 1.

[0081] Step S220: If the recognition is successful, extract the geometric features from the laser point cloud set, and determine the position information of the positioning component 10 based on the geometric features.

[0082] Among them, the geometric features are the contour information of the positioning component 10 extracted by the cleaning robot 2 from multiple laser points corresponding to the base station 1 through feature extraction.

[0083] In this step, after successfully identifying base station 1, control module 23 can extract features from multiple laser points corresponding to base station 1. By sequentially connecting these laser points corresponding to positioning component 10 through feature extraction, geometric features reflecting the contour information of positioning component 10 can be obtained. Then, control module 23 can determine the position information of positioning component 10 based on these geometric features.

[0084] Step S230: Determine the position information of the central axis of the receiving cavity 40 based on the position information of the positioning component 10, and generate a movement path based on the position information of the central axis of the receiving cavity 40.

[0085] In this step, since the receiving cavity 40 is formed at the positioning component 10, the control module 23 can determine the position information of the central axis of the receiving cavity 40 based on the position information of the positioning component 10. After the position information of the central axis of the receiving cavity 40 is determined, the control module 23 can plan the movement path of the cleaning robot 2 when returning to the base station 1 based on the position information of the central axis of the receiving cavity 40. The method of generating the movement path is the same as that in the prior art, and will not be described in detail here.

[0086] Step S240: Control the cleaning robot 2 to travel back to base station 1 along the movement path.

[0087] In this step, after the movement path is generated, the control module 23 can control the cleaning robot 2 to travel back to the base station 1 according to the above movement path.

[0088] As can be seen from the above, since the position information of the positioning component 10 is determined based on the extracted contour information in this application, changes in the position and number of laser points will not interfere with the extraction result of the contour information, resulting in high accuracy of the obtained position information of the positioning component 10. Based on this, the central axis of the receiving cavity 40 determined based on the position information of the positioning component 10 will also have high accuracy; furthermore, it ensures the accuracy of the generated movement path, fully guaranteeing that the cleaning robot 2 will not deviate when returning to the base station 1 according to the movement path, and ensuring that the cleaning robot 2 accurately returns to the base station 1. In addition, ensuring that the cleaning robot 2 can accurately return to the base station 1 also avoids collisions between the cleaning robot 2 and the base station 1, thus protecting the cleaning robot 2 to a certain extent.

[0089] In summary, this application eliminates the influence of laser point cloud measurement errors on the positioning accuracy of the central axis of the cavity 40, improves the positioning accuracy of the central axis of the cavity 40, and thus ensures the accuracy of the movement path generation, enabling the cleaning robot 2 to accurately return to the base station 1.

[0090] In one embodiment, the positioning component 10 is provided with a positioning tag. At this time, the control module 23 can identify the base station 1 by the following steps: determining whether there is a target point cloud set corresponding to the positioning tag in the laser point cloud set obtained by the laser radar 22 scanning.

[0091] In one embodiment, the positioning tag includes a first positioning tag and a second positioning tag. The first positioning tag is disposed on the first positioning part 111, and the second positioning tag is disposed on the second positioning part 113. In this case, the control module 23 can also determine whether there is a target point cloud set corresponding to the positioning tag in the laser point cloud set through the following step S310.

[0092] Step S310: Determine whether there is a first point cloud set and a second point cloud set that meet preset conditions in the laser point cloud set obtained by the LiDAR 22. The first point cloud set and the second point cloud set are the point cloud sets corresponding to the first positioning tag and the second positioning tag, respectively; the target point cloud set includes the first point cloud set and the second point cloud set.

[0093] In this step, after receiving the laser point cloud set sent by the lidar 22, the control module 23 can determine whether there are a first point cloud set and a second point cloud set that meet the preset conditions in the laser point cloud set, and then identify the base station 1 based on the judgment result. If the first point cloud set and the second point cloud set are found in the set of point clouds, the base station 1 is identified; the location of the first point cloud set and the second point cloud set is the location of the base station 1.

[0094] like Figure 8 As shown, the lidar 22 rotates and emits infrared laser beams, all of which are located in plane S1, parallel to the ground S2. When the lidar 22 follows... Figure 8 During the scanning posture shown, the first side 110 of the positioning component 10 is scanned, forming multiple laser points. The contour information formed by connecting these multiple laser points should essentially match the contour information of the first side 110 of the positioning component 10 as observed from a top-down angle. Furthermore, the multiple laser points include a first point cloud set formed after scanning the first positioning tag, and a second point cloud set formed after scanning the second positioning tag. For example... Figure 1 As shown, the top-down angle refers to the direction from the top surface of base station 1 to the bottom surface of base station 1.

[0095] According to the above description, the preset conditions should be that the first point cloud set and the second point cloud set are on a straight line, the length of the first point cloud set is equal to the length of the first positioning tag observed from a top-down angle, and / or, the length of the second point cloud set is equal to the length of the second positioning tag observed from a top-down angle, and / or, the distance between the first point cloud set and the second point cloud set is equal to a first threshold, and / or, the brightness value of each laser point in the first point cloud set and the second point cloud set is greater than the brightness threshold. The first threshold and the brightness threshold are stored in the control module 23. Because the positioning tag has a stronger reflective effect, the brightness value of each laser point in the first point cloud set and the second point cloud set is greater than the brightness threshold.

[0096] like Figure 9 As shown, it is a schematic outline of the first side 110 as viewed from above. Figure 9 As shown, from a top-down view, the length of the first positioning tag is 'a', the length of the second positioning tag is 'c', and the distance between the first and second positioning tags is 'b'. Figure 10 As shown, this is a schematic diagram of the laser point cloud set formed after the LiDAR 22 scans the first side 110 of the positioning component 10. For example, the length a of the first positioning tag and the length c of the second positioning tag can be 40-50 mm; the distance between the first positioning tag and the second positioning tag can be 300-320 mm.

[0097] The following is based on Figure 9 and Figure 10 The method for identifying base station 1 in the above embodiments will be explained in detail using an example:

[0098] The LiDAR 22 scans the environment around the cleaning robot 2 to generate a laser point cloud set, and sends the generated laser point cloud set to the control module 23. The control module 23 then identifies the base station 1 based on the received laser point cloud set. Specifically, during identification, the control module 23 first determines whether there are multiple laser points in the laser point cloud set with brightness values ​​greater than a brightness threshold. If so, these multiple laser points are filtered out. After filtering, it determines whether there is a point cloud set with a length equal to the length of the first positioning tag and a point cloud set with a length equal to the length of the second positioning tag. After successful determination, it further determines whether the two point cloud sets are on a straight line and whether the distance between the two point cloud sets is equal to the first threshold. If the distance between the two point cloud sets is equal to the first threshold and they are on a straight line, then the two filtered point cloud sets are the first point cloud set and the second point cloud set. Specifically, as shown... Figure 9 As shown, after judgment, the control module 23 finds that the length of point cloud set D1 among multiple laser points is equal to the length a of the first positioning tag observed from the top angle, the length of point cloud set D2 in the target point cloud set is equal to the length c of the second positioning tag observed from the top angle, the distance D3 between point cloud set D1 and point cloud set D2 is equal to the first threshold b, and point cloud set D1 and point cloud set D2 are on a straight line. Therefore, point cloud set D1 can be determined as the first point cloud set, and point cloud set D2 as the second point cloud set. After identifying the first and second point cloud sets, the identification of base station 1 is completed, and the locations of the first and second point cloud sets are the locations of base station 1.

[0099] By taking the above measures, base station 1 is identified by determining whether there is a laser point cloud set with a specific width and length corresponding to the positioning tag in the laser point cloud set. The identification method is simple and has a small amount of computation, which improves the identification speed of base station 1 while ensuring the identification accuracy.

[0100] In one embodiment, the positioning tag can be a national standard Class V reflective film. This type of reflective film has the characteristics of being resistant to temperature, humidity and corrosion, has a long service life, has a good reflective effect, and has weak directional sensitivity and excellent wide-angle performance. It can obtain strong reflected signals from different angles, which is beneficial to the identification of base station 1.

[0101] In one embodiment, after identifying base station 1, control module 23 may further perform the following steps: filter out the feature point cloud set corresponding to base station 1 from the laser point cloud set, so as to identify geometric features from the feature point cloud set. The geometric features are the contour information of the positioning component 10 extracted from the feature point cloud set by the cleaning robot 2 through feature extraction.

[0102] In this embodiment, because the lidar 22 performs a 360° rotating scan, the laser point cloud set transmitted by the lidar 22 contains not only the point cloud data corresponding to the base station 1, but also the point cloud data of other objects around the cleaning robot 2. Therefore, to reduce noise interference, the cleaning robot 2 can first filter out the feature point cloud set corresponding to the base station 1 from the laser point cloud set before extracting the contour information of the positioning component 10. For example, Figure 8 As shown, when the lidar 22 scans the base station 1, it can only scan the positioning component 10. Therefore, the feature point cloud set is the point cloud set corresponding to the positioning component 10. After filtering out the feature point cloud set corresponding to the positioning component 10, the control module 23 can extract the contour information of the positioning component 10 from the feature point cloud set.

[0103] For example, such as Figure 10 As shown, after the control module 23 identifies the first point cloud set D1 and the second point cloud set D2 from the laser point cloud set, it can filter the feature point cloud set in the following way: extract the first point cloud set D1, the second point cloud set D2 and the point cloud set located between the first point cloud set D1 and the second point cloud set D2 from the laser point cloud set.

[0104] Through the above measures, the feature point cloud set corresponding to the positioning component 10 is selected from the laser point cloud set. The contour information of the positioning component 10 is extracted only from the feature point cloud set, eliminating interference factors and ensuring the accuracy of the contour information extraction of the positioning component 10. At the same time, feature extraction is not performed on all laser point cloud sets, but only the contour information of the positioning component 10 is extracted from the feature point cloud set, which reduces the amount of computation and improves the positioning speed of base station 1.

[0105] In one embodiment, after the control module 23 filters out the feature point cloud set from the laser point cloud set, it can first extract geometric features from the feature point cloud set, and then determine the position information of the first straight line 114 and the second straight line 115 based on the geometric features. The geometric features are the contour information of the positioning component 10 extracted from the feature point cloud set by the cleaning robot 2 through feature extraction; the first positioning part 111 and the feature part 112 in the positioning component 10 intersect at the first straight line 114, and the second positioning part 113 and the feature part 112 in the positioning component 10 intersect at the second straight line 115.

[0106] In this embodiment, the control module 23 can first extract features from the feature point cloud set to extract the contour information of the positioning component 10; after feature extraction, the position information of the first straight line 114 and the second straight line 115 is determined based on the contour information of the positioning component 10, and then the position information of the positioning component 10 is determined based on the position information of the first straight line 114 and the second straight line 115. Specifically, when the control module 23 extracts features from the feature point cloud set, it can first extract the straight line features corresponding to the first point part and the second positioning part 113 from the feature point cloud set, and then extract the arc features corresponding to the feature part 112 from the feature point cloud set; then, the position information of the first straight line 114 and the second straight line 115 is determined based on the intersection of the arc features and the straight line features.

[0107] The following is based on Figure 11 The above process will be explained using an example:

[0108] like Figure 11 As shown, in this embodiment, the control module 23 performs feature extraction on the feature point cloud set. During extraction, the Hough line transform is first used to extract the contour information corresponding to the first positioning part 111 and the second positioning part 113 from the feature point cloud set. Specifically, since the first point cloud set D1 and the second point cloud set D2 are the point cloud sets corresponding to the first positioning tag and the second positioning tag, and the first and second positioning tags are relatively thin, it can be approximately considered that the first point cloud set D1 is the point cloud set corresponding to the first positioning part 111, and the second point cloud set D2 is the point cloud set corresponding to the second positioning part 113. Based on this, the Hough line transform is used to sequentially connect the laser points in the first point cloud set D1 and the second point cloud set D2. The resulting straight line H2 is the contour information of the first positioning part 111 and the second positioning part 113 observed from the top view angle. After extracting the straight line H2, the control module 23 then uses the Hough circle transform to extract the contour information corresponding to the feature part 112 from the feature point cloud set. Specifically, the laser points located between the first point cloud set D1 and the second point cloud set D2 are sequentially connected using the Hough circle transform. The resulting arc H1 is the contour information of the feature part 112. Since the first straight line 114 and the second straight line 115 are the intersection lines of the first positioning part 111 and the second positioning part 113 with the feature part 112, the intersection point of the two contour information can reflect the position of the first straight line 114 and the second straight line 115. Specifically, as shown... Figure 10As shown, straight line H2 intersects arc H1 at a first intersection point P1 and a second intersection point P2. The position of the first intersection point P1 reflects the position of the first straight line 114, and the position of the second intersection point P2 reflects the position of the second straight line 115. Then, the control module 23 determines the position of the positioning component 10 based on the positions of the first straight line 114 and the second straight line 115. The position of the positioning component 10 includes the distance between the positioning component 10 and the cleaning robot 2, and the angle of the positioning component 10 relative to the cleaning robot 2.

[0109] like Figure 11 As shown, in the above method, the Hough transform can filter the laser points, and laser points far from the straight line H2 and the arc H1 can be filtered out, overcoming certain noise interference. Even if the laser point cloud is floated due to the use of a low-cost and low-precision lidar 22, it will not interfere with the extraction result of the contour information. At the same time, this method allows for the absence of local laser points. Even if the positioning component 10 is equipped with pipes for the flow of cleaning water and charging electrodes for charging the cleaning robot 2, resulting in the absence of local laser point cloud, it will not interfere with the extraction result of the contour information. Therefore, it can be seen that the floating and absence of laser point cloud will not interfere with the extraction result of contour information, fully ensuring the extraction accuracy of contour information, and further ensuring the positioning accuracy of the positioning component 10. In addition, the control module 23 only extracts features from the laser points in the feature point cloud set, eliminating interference factors and ensuring the extraction accuracy of straight line features and arc features. This makes the intersection position determined based on the straight line features and arc features more stable and accurate, further ensuring the positioning accuracy of the positioning component 10.

[0110] In one embodiment, such as Figure 11 As shown, after determining the positions of the first straight line 114 and the second straight line 115, i.e., after determining the positions of the first intersection point P1 and the second intersection point P2, the control module 23 can determine the position information of the central axis of the receiving cavity 40 based on the first intersection point P1, the second intersection point P2, and the straight line feature H2. After determining the position information of the central axis of the receiving cavity 40, the control module 23 can generate a movement path based on the position information of the central axis of the receiving cavity 40. After the movement path is generated, the control module can control the cleaning robot 2 to move back to the base station 1 according to the movement path.

[0111] Specifically, such as Figure 11 As shown, the straight line feature H2 is divided into three segments by the first intersection point P1 and the second intersection point P2. These three segments are the straight line feature corresponding to the first point cloud set D1, the straight line feature corresponding to the second point cloud set D2, and the straight line feature P1P2 located between the first intersection point P1 and the second intersection point P2. Then, the straight line feature perpendicular to and bisects P1P2 is calculated; this straight line feature is the central axis of the receiving cavity 40. Specifically, as shown... Figure 12 As shown in the figure, the dashed line P3P4 perpendicularly bisects the straight line feature P1P2, and therefore, the dashed line P3P4 is the central axis of the receiving cavity 40. Simultaneously, the positional information of the dashed line P3P4 is the positional information of the central axis of the receiving cavity 40.

[0112] In one embodiment, after the control module 23 extracts the line H2 and the arc H1 using the Hough linear transform, it can further verify the identification result of the base station 1. The verification method involves calculating whether the radius of the formed arc H1 is the same as the radius of the arc of the feature portion 112. If they are the same, it indicates that the arc H1 is indeed the contour information corresponding to the feature portion 112, verifying the previous identification results of the first point cloud set D1 and the second point cloud set D2, and ensuring the identification result of the base station 1. For example, the radius of the arc of the feature portion 112 can be 160–170 mm.

[0113] Through the above measures, this application fully guarantees the accuracy of the identification of base station 1 by verifying the identification result of base station 1.

[0114] In one embodiment, if the control module 23 fails to identify base station 1 based on the laser point cloud set sent by the lidar 22, it may further perform the following steps: control the cleaning robot 2 to perform rotation and / or translation trigger actions within a preset range; during or after the trigger actions are performed, reacquire the laser point cloud set scanned by the lidar 22, and identify base station 1 based on the reacquired laser point cloud set. The preset range can be all floor areas in the user's home, or it can be a partial floor area within the user's home. For example, the floor areas in the user's home may include sub-areas such as the study floor, bathroom floor, living room floor, bedroom floor, and kitchen floor; in this case, the preset range can be a combination of one, more, or all of the aforementioned sub-areas.

[0115] Because the cleaning robot 2 is far from the base station 1, and / or, the cleaning robot 2 and the base station 1 are obstructed by obstacles such as walls, and / or the cleaning robot 2 is trapped by obstacles, the lidar 22 cannot scan the base station 1, thus causing the control module 23 to be unable to identify the base station 1 based on the laser point cloud set scanned by the lidar 22. To solve this problem, in the above embodiment, the cleaning robot 2 is driven to perform rotation and / or translation trigger actions, so that the cleaning robot 2 is closer to the base station 1, and / or, the situation where the cleaning robot 2 and the base station 1 are obstructed by obstacles is eliminated, and / or, the cleaning robot 2 is freed from the trapped state, thereby fully ensuring that the lidar 22 can scan the base station 1, ensuring that the cleaning robot 2 can identify the base station 1 based on the laser point cloud set scanned by the lidar 22, and ensuring the identification effect of the base station 1.

[0116] In another embodiment, if the cleaning robot 2 still cannot identify base station 1 based on the newly acquired laser point cloud set after performing the above-mentioned triggering action, the control module 23 saves the message to the user, indicating that base station 1 cannot be found and the operation to return to base station 1 cannot be completed. For example, the alarm can be triggered by a horn, voice prompt, vibration, or text prompt.

[0117] In one embodiment, the lidar 22 on the cleaning robot 2 has a relatively low cost but limited ranging range and accuracy. It can only scan the contour information of the positioning component 10 with sufficient accuracy at a distance close to the base station 1. To address this issue, the control module 23 can mark a first target position at a preset distance from the base station 1 when the cleaning robot 2 leaves the base station 1 to perform a cleaning task. Then, when the cleaning robot 2 needs to return to the base station 1, the control module 23 first drives the cleaning robot 2 to the first target position. After moving to the first target position, it then identifies and locates the base station 1. The control module 23 internally stores an environmental map of the user's home floor area. The control module 23 controls the cleaning robot 2 to perform cleaning work based on this environmental map, which marks the location of the base station 1. When the cleaning robot 2 leaves the base station 1, it can mark the first target position on the environmental map. Then, when the cleaning robot 2 needs to return to the base station 1, the control module 23 can drive the cleaning robot 2 to the first target position based on the coordinates of the first target position on the environmental map. For example, the first target position is a distance from the base station 1.

[0118] Through the above measures, the cleaning robot 2 is first moved to the first target position at a preset distance from the base station 1, thereby improving the accuracy of the LiDAR 22 scan and ensuring the accuracy of the identification and positioning of the base station 1.

[0119] In one embodiment, the control module 23 can control the cleaning robot 2 to travel back to the base station 1 along the movement path in the following manner: First, the control module 23 can control the cleaning robot 2 to move to a second target position along the movement path; wherein, at the second target position, the central axis of the cleaning robot 2 is substantially coincident with the central axis of the receiving cavity 40. Then, after moving into position, the cleaning robot 2 is driven to rotate by a preset angle threshold, so that when the cleaning robot 2 returns to the base station 1 along the movement path after rotation, the cleaning component 24 is placed on the cleaning component 51. For example, the preset angle threshold can be 175-180°.

[0120] In this embodiment, when the cleaning robot 2 is at the first target position, it is far from the base station 1, and the determined position information of the central axis of the receiving cavity 40 will have a large error. That is, the determined position information of the central axis of the receiving cavity 40 will deviate from the actual position information of the central axis of the receiving cavity 40 in the actual structure of the base station 1. In this case, when the cleaning robot 2 moves to the second target position according to the determined position information of the central axis of the receiving cavity 40, the above-mentioned deviation problem still exists. At this time, if the cleaning robot 2 moves according to the planned movement path, the central axis of the cleaning robot 2 will only coincide with the determined position information of the central axis of the receiving cavity 40. However, because the determined position information of the central axis of the receiving cavity 40 will deviate from the actual position information of the central axis of the receiving cavity 40 in the actual structure of the base station 1, the central axis of the cleaning robot 2 cannot coincide with the actual central axis of the receiving cavity 40, and deviation will occur. As the deviation continues, the cleaning robot 2 will be unable to return to the base station 1. More seriously, the cleaning robot 2 will collide with the base station 1. Therefore, in order to improve positioning accuracy, after the cleaning robot 2 moves to the second target position, the cleaning robot 2 is rotated by a preset angle threshold. After rotation, the cleaning robot 2 is controlled to move. During the movement, the central axis of the receiving cavity 40 is determined in real time according to the method mentioned in the above embodiment. During the determination process, the movement path of the cleaning robot 2 is adjusted in real time. By adjusting the movement path, the central axis of the cleaning robot 2 gradually coincides with the central axis of the actual receiving cavity 40. The movement path of the cleaning robot 2 is continuously adjusted in this way until the cleaning robot 2 enters the base station 1.

[0121] For example, such as Figure 13 As shown, upon returning to base station 1, control module 23 drives cleaning robot 2 to first move to the first target position corresponding to point F. Then, control module 23 initially determines the position information of the central axis of the receiving cavity 40 at point F; specifically, the dashed lines P3 and P4 in the figure represent the initially determined position information of the central axis of the receiving cavity 40. Next, control module 23 plans the movement path of cleaning robot 2. After successful planning, it drives cleaning robot 2 to move along the aforementioned path to the second target position corresponding to point P. After successful movement, control module 23 controls cleaning robot 2 to rotate 180°. After rotation, cleaning robot 2 repositions the position information of the central axis of the receiving cavity 40. After successful positioning, it adjusts the movement path of cleaning robot 2 in real time, gradually aligning the central axis of cleaning robot 2 with the actual central axis of the receiving cavity 40. This process of continuously adjusting the movement path of cleaning robot 2 continues until cleaning robot 2 enters base station 1.

[0122] It is worth noting that the purpose of the aforementioned rotating cleaning robot is to ensure that when the cleaning robot 2 enters the base station 1, the cleaning component 24 can be placed on the cleaning component 51, so as to ensure that the cleaning component can clean the cleaning component 24 and ensure the smooth progress of the cleaning work.

[0123] Through the above measures, the cleaning robot 2 is first moved to the first target position, where the position of the central axis of the receiving cavity 40 is initially located. After initial positioning, the cleaning robot 2 is rotated. While rotating, the cleaning robot 2 moves and precisely positions the central axis of the receiving cavity 40, adjusting its movement path simultaneously. As the cleaning robot 2 gets closer to the base station 1, the accuracy of the located central axis of the receiving cavity 40 becomes higher. When it moves to a certain extent, the determined position of the central axis of the receiving cavity 40 will coincide with the actual central axis of the receiving cavity 40. At this point, the movement path adjusted by the cleaning robot 2 is also the most accurate. When the cleaning robot 2 moves along the highly accurate movement path, its central axis will accurately coincide with the actual central axis of the receiving cavity 40. Based on this, the cleaning robot 2 can accurately return to the base station 1. The above method first uses a combination of preliminary positioning and precise positioning to ensure the positioning accuracy of the central axis of the receiving cavity 40. This ensures that when the cleaning robot 2 moves according to the planned path, the central axis of the cleaning robot 2 can accurately coincide with the central axis of the actual receiving cavity 40, thus ensuring that the cleaning robot 2 can accurately enter the base station 1 and avoiding deviation and collision events.

[0124] In another embodiment, the control module 23 can move the cleaning robot 2 to the third target position after the cleaning robot 2 rotates. After reaching the third target position, the control module 23 will not adjust the movement path of the cleaning robot 2, and will directly make the cleaning robot 2 enter the base station 1 according to the last adjusted movement path. In this embodiment, the positioning accuracy of the lidar 22 at the third target position is relatively high, and the position of the central axis of the receiving cavity 40 located by the lidar 22 is basically coincident with the actual central axis of the receiving cavity 40. Therefore, the movement path is already relatively accurate, and the cleaning robot 2 can accurately enter the base station 1 when it moves according to the above movement path.

[0125] like Figure 14 As shown, it is a schematic diagram of an optical theoretical model provided in an embodiment of this application. Figure 14 As shown, b is the actual distance between the inner sides of the two positioning tags; e is the actual distance between the outer sides of the two positioning tags; d is the vertical distance from the center of the lidar 22 to the plane of the positioning tag, that is, the distance between the cleaning robot 2 and the base station 1; α is the laser rotation angle required to cover a single positioning tag; β is the laser rotation angle required to cover the feature part 112 in the positioning component 10; θ is the rotation angle of the lidar 22 required to cover the entire positioning component 10.

[0126] The aforementioned angle parameters can be determined using the following formulas (1), (2), and (3).

[0127]

[0128]

[0129] α = (θ – β) / 2 (3)

[0130] After determining the above parameters, the number of laser points in angle α and the number of laser points in angle β can be determined. The number of laser points in angle α is N*α / 360°, and the number of laser points in angle β is N*β / 360°; where N is the number of points in a single frame (one rotation) of the lidar 22.

[0131] Table 1. Relationship between LiDAR range and number of LiDAR points

[0132]

[0133]

[0134] Referring to Table 1, the number of points falling within the two positioning tags will differ depending on the distance *d* between the cleaning robot 2 and base station 1. When the LiDAR 22 has 400 points per revolution and the distance *d* between the cleaning robot 2 and base station 1 is 1m, the number of laser points within each positioning tag is 2. This ensures a certain level of positioning accuracy, but the accuracy is relatively low. When the distance *d* between the cleaning robot 2 and base station 1 is 0.6m, the number of laser points within each positioning tag is 4, and the positioning accuracy of the LiDAR 22 is doubled, resulting in higher positioning accuracy. Therefore, when the LiDAR 22 has 400 points per revolution, the position at the distance from base station 1 can be set as the first target position. In this case, the control module 23 first controls the cleaning robot 2 to move to the first target position 1m away from the base station, then rotates the cleaning robot 2. After rotation, the control module 23 adjusts the movement path of the cleaning robot 2 in real time, causing the cleaning robot 2 to move to the third target position 0.6m to 0.5m in front of base station 1. After successful movement, the control module 23 controls the cleaning robot 2 to enter base station 1.

[0135] As can be seen from Table 1, when the number of points in one ring of the lidar 22 is 800, the positioning accuracy of the lidar 22 at 1m is also relatively high. Therefore, the first target position can be larger. For example, the first target position can be a location 2m away from the base station 1. The specific principle is the same as in the above embodiment, and will not be repeated here.

[0136] In one embodiment, the first positioning part 111 and the second positioning part 113 have identical dimensional parameters. In this way, while ensuring the above-mentioned dimensional parameter settings, the feature part 112 is set as an arc groove structure, realizing a symmetrical shape of "an arc + two straight lines" after Hough transform. Because this symmetrical shape is unique in size and shape, and has significant features, it is difficult to find objects with similar shape and size in a home environment other than base station 1, thus fully ensuring the accuracy of identifying and locating base station 1 based on the above-mentioned symmetrical shape.

[0137] Because the position of the laser point is unstable and may fluctuate, in the above embodiments of this application, the first position is respectively set on the first positioning part 111 and the second positioning part 113. The structural feature that the first positioning part 111 and the second positioning part 113 are far apart and on the same plane ensures the accuracy of the position and angle of the straight line feature generated after the Hough transform.

[0138] In the above embodiments of this application, the first positioning tag and the second positioning tag are disposed on the first positioning part 111 and the second positioning part 113, so that the positioning tag is in a relatively conspicuous position on the base station 1, which makes it easy to find when it is dirty and to wipe the positioning tag. In addition, even when the cleaning robot 2 enters the base station 1, the positioning tag is still exposed, which avoids the cleaning liquid from the base station 1 from contaminating and corroding the positioning tag when the cleaning robot 2 is cleaning it, thus fully ensuring the cleanliness of the positioning tag and further ensuring the identification and positioning accuracy of the base station 1.

[0139] The following details the method by which the cleaning robot 2 returns to base station 1 as provided in this application.

[0140] Working principle:

[0141] like Figure 15 As shown, when cleaning robot 2 leaves base station 1 to perform cleaning tasks, control module 23 marks the location of the first target location V1 on the environmental map. After successful marking, control module 23 controls cleaning robot 2 to perform cleaning tasks. Figure 16As shown, during or after the cleaning task, the cleaning robot 2 needs to return to the base station 1. At this time, the control module 23 first drives the cleaning robot 2 to the first target position V1. After successful movement, the control module 23 receives the laser point cloud set obtained by the LiDAR 22 scan and identifies the base station 1 based on the laser point cloud set. If the base station 1 is not successfully identified, the control module 2 executes translational and / or rotational trigger actions. After execution, it re-acquires the laser point cloud set obtained by the LiDAR scan and identifies the base station 1 based on the laser point cloud set. If the base station 1 is successfully identified, the control module 23 extracts the feature point cloud set corresponding to the positioning component 10 from the laser point cloud set. After successful extraction, the control module 23 identifies the contour information of the positioning component 10 from the feature point cloud set. After successful identification, the position information of the positioning component 10 is determined based on the contour information. After the position information of the positioning component 10 is determined, the position information of the central axis of the receiving cavity 40 is initially determined based on the position information of the positioning component 10. After successful determination, a movement path is generated, and then the cleaning robot 2 is controlled to move to the second target position according to the movement path. In this case, the central axis of the cleaning robot 2 at the second target position is substantially coincident with the central axis of the receiving cavity 40. If the central axis of the cleaning robot 2 at the first target position V1 is already substantially coincident with the central axis of the receiving cavity 40, then no movement is required. Then, as... Figure 17 As shown, the cleaning robot 2 is controlled to rotate at a preset angle threshold in the direction indicated by the arrow in the figure. After rotation, the control module 23 re-determines the position of the central axis of the receiving cavity 40 based on the feature point cloud set, and adjusts the movement path of the cleaning robot 2 according to the positioning result. Figure 18 As shown, when the cleaning robot 2 moves to the third target position V2, the adjustment of the cleaning robot 2's movement path is stopped, allowing the cleaning robot 2 to proceed into the base station 1 according to the final adjusted movement path. Figure 19 As shown, when the cleaning robot 2 enters the base station 1, the cleaning robot 2 can be controlled to stop moving.

[0142] In existing technologies, infrared positioning schemes are also used to locate the position of base station 1. Specifically, the principle of infrared positioning is as follows: an infrared transmitting module is set on base station 1 to emit infrared signals outward, and multiple infrared receiving modules are set at different angles around the cleaning robot 2 to receive the infrared signals emitted by base station 1. The cleaning robot 2 compares the intensity of the signals received by the infrared receiving modules, and the location of base station 1 is the location of the part with the stronger signal intensity.

[0143] The advantage of infrared positioning is that the signal transmission distance is relatively long, and it can cover a single room in a typical home scenario. Its disadvantage is that it requires one or more infrared transmitting modules and multiple infrared receiving modules, which takes up a certain amount of structural space and brings certain hardware costs. At the same time, it requires that base station 1 must be powered on.

[0144] However, the base station 1 positioning method provided in the above embodiments of this application locates and identifies the base station 1 through positioning tags, without the need to set up infrared transmitting modules and infrared receiving modules, resulting in low hardware costs, small space occupation, and no requirement that the base station 1 must be in a powered-on state, thus having a wide range of applications.

[0145] The apparatuses and methods disclosed in the several embodiments provided in this application can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatuses, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0146] In addition, the functional modules in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0147] If a function is implemented as a software module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

Claims

1. A method for a cleaning robot to return to a base station, characterized in that, The cleaning robot is equipped with a lidar, and the base station has an opening with a positioning component at the opening, forming a receiving cavity for accommodating the cleaning robot. The method for the cleaning robot to return to the base station includes: The base station is identified based on the laser point cloud set obtained by the lidar scan; If the identification is successful, geometric features are extracted from the laser point cloud set, and the position information of the positioning component is determined based on the geometric features; The position information of the central axis of the receiving cavity is determined based on the position information of the positioning component, and a movement path is generated based on the position information of the central axis of the receiving cavity; Control the cleaning robot to travel back to the base station along the stated path; Before extracting geometric features from the laser point cloud set, the method further includes: The feature point cloud set corresponding to the base station is selected from the laser point cloud set, so as to extract the geometric features from the feature point cloud set; The surface of the positioning component is provided with a first positioning part, a feature part and a second positioning part in sequence. The first positioning part and the feature part intersect at a first straight line, and the second positioning part and the feature part intersect at a second straight line. The step of extracting geometric features from the laser point cloud set and determining the position information of the positioning component based on the geometric features includes: Geometric features are extracted from the feature point cloud set, and the position information of the first straight line and the second straight line is determined based on the geometric features; The first positioning part and the second positioning part are planar structures and are located on the same plane; Extracting geometric features from the feature point cloud set, and determining the position information of the first straight line and the second straight line based on the geometric features, including: Extract the straight line features corresponding to the first positioning part and the second positioning part from the feature point cloud set; Extract the arc feature corresponding to the feature part from the feature point cloud set; Determine the first and second intersection points of the arc feature and the straight line feature, and determine the position information of the first and second straight lines based on the first and second intersection points; Determining the position information of the central axis of the receiving cavity based on the position information of the positioning component includes: Based on the first intersection point, the second intersection point, and the straight line feature, the position information of the central axis of the receiving cavity is determined.

2. The method for a cleaning robot to return to a base station according to claim 1, characterized in that, Before identifying the base station based on the laser point cloud set obtained from the lidar scan, the method further includes: Control the cleaning robot to move to a first target location at a preset distance from the base station.

3. The method for a cleaning robot to return to a base station according to claim 1, characterized in that, The positioning component is equipped with a positioning tag; The step of identifying the base station based on the laser point cloud set obtained by the lidar scan includes: Determine whether the target point cloud set corresponding to the positioning tag exists in the laser point cloud set; If the target point cloud set exists, the base station has been successfully identified; If the target point cloud set does not exist, the base station was not successfully identified.

4. The method for a cleaning robot to return to a base station according to claim 3, characterized in that, The location tag includes a first location tag and a second location tag; The positioning component has a first positioning part and a second positioning part spaced apart on its surface. The first positioning label is disposed on the first positioning part, and the second positioning label is disposed on the second positioning part. The step of determining whether the laser point cloud set contains a target point cloud set corresponding to the positioning tag includes: Determine whether there exists a first point cloud set and a second point cloud set that meet preset conditions in the laser point cloud set; Wherein, the first point cloud set and the second point cloud set are the point cloud sets corresponding to the first positioning tag and the second positioning tag.

5. The method for a cleaning robot to return to a base station according to claim 1 or 3, characterized in that, The method for the cleaning robot to return to the base station also includes: If the base station is not successfully identified, the cleaning robot is controlled to perform rotation and / or translation trigger actions within a preset range; During or after the triggering action is executed, the laser point cloud set scanned by the lidar is reacquired, and the base station is identified based on the reacquired laser point cloud set.

6. The method for a cleaning robot to return to a base station according to claim 1, characterized in that, The opening is provided with a cleaning component, and the cleaning component and the positioning component form the receiving cavity; the cleaning robot is provided with a cleaning component for wiping the surface to be cleaned, and the cleaning component is provided with a cleaning component for cleaning the cleaning component; The control of the cleaning robot to travel back to the base station according to the movement path includes: The cleaning robot is controlled to move to the second target position according to the moving path; wherein, at the second target position, the central axis of the cleaning robot is substantially coincident with the central axis of the receiving cavity; The cleaning robot is driven to rotate at a preset angle threshold so that when the rotated cleaning robot returns to the base station along the moving path, the cleaning component is placed on the cleaning part.

Images