Mowing robot recharging control method and device, mowing robot, and storage medium

Abstract

This application provides a method, apparatus, lawnmower, and storage medium for controlling the recharging of a lawnmower robot. The method includes: acquiring a preset global map in response to a recharging command for the lawnmower robot; controlling the lawnmower robot to enter a base station area based on the global map; when the lawnmower robot enters the base station area, sending a recharging request to a target device via a base station in the base station area; upon receiving an authorization from the target device in response to the recharging request, controlling the lawnmower robot to identify the base station identifier corresponding to the base station; comparing a pre-stored target identifier of the lawnmower robot with the base station identifier; and when the target identifier matches the base station identifier, controlling the lawnmower robot to dock with the base station. This application solves the problems of low docking accuracy and poor recharging reliability in current recharging control schemes.

Description

Technical Field

[0001] This application relates to the field of lawn mowing robot technology, and in particular to a lawn mowing robot recharging control method, device, lawn mowing robot and storage medium. Background Technology

[0002] With the rapid development of lawnmower technology, lawnmowers have gradually achieved autonomous operation and automatic recharging for extended battery life. Automatic recharging is a core component ensuring the continuous and stable operation of lawnmowers. Current lawnmower recharging solutions mostly rely on global map navigation to the vicinity of a base station, and then docking via visual markers.

[0003] However, when a lawnmower robot directly initiates docking upon entering a base station area, it is prone to false triggering and unauthorized devices illegally docking with the base station, resulting in poor security and controllability of base station use. In the base station identification process, traditional solutions rely solely on a single visual detection method. These either depend on dedicated identification detectors, which are susceptible to interference from complex backgrounds and lighting conditions, affecting the accuracy of the coded region positioning; or they fail to fully utilize the robustness of deep learning semantic segmentation, leading to poor identification feature extraction and coded region delineation, and are highly prone to identification recognition failures and decoding errors. Therefore, current lawnmower robot recharging control solutions suffer from low docking accuracy and poor recharging reliability. Summary of the Invention

[0004] This application provides a method, device, lawnmower robot, and storage medium for recharging control of a lawnmower robot, which can solve the problems of low docking accuracy and poor recharging reliability in current recharging control schemes.

[0005] Firstly, a method for controlling the recharging of a lawnmower robot is provided, the method comprising: In response to the recharge command for the lawnmower robot, a preset global map is obtained; The lawnmower robot is controlled to enter the base station area based on the global map. When the lawnmower robot enters the base station area, it sends a recharge request to the target device through the base station in the base station area. When the target device receives the authorization permission returned in response to the recharge request, the lawnmower robot is controlled to identify the base station identifier corresponding to the base station. The target identifier pre-stored by the lawnmower robot is compared with the base station identifier; When the target identifier matches the base station identifier, the lawnmower robot is controlled to dock with the base station.

[0006] Secondly, a recharging control device for a lawnmower robot is provided, the device comprising: The acquisition module is used to acquire a preset global map in response to the recharge command for the lawnmower robot; The first control module is used to control the lawnmower robot to enter the base station area according to the global map; The sending module is used to send a recharge request to the target device through the base station in the base station area when the lawnmower robot enters the base station area. The receiving module is used to control the lawnmower robot to identify the base station identifier corresponding to the base station when it receives the authorization permission returned by the target device in response to the recharge request. The comparison module is used to compare the target identifier pre-stored by the lawnmower robot with the base station identifier; The second control module is used to control the lawnmower robot to dock with the base station when the target identifier matches the base station identifier.

[0007] Thirdly, a lawnmower robot is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the lawnmower robot recharging control method described above.

[0008] Fourthly, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described lawnmower recharging control method.

[0009] This application provides a method, apparatus, lawnmower, and storage medium for controlling the recharging of a lawnmower robot. In response to a recharging command for the lawnmower robot, after acquiring a preset global map, the robot is controlled to enter a base station area based on the global map. When the lawnmower robot enters the base station area, it sends a recharging request to a target device via a base station in the area. Upon receiving authorization from the target device in response to the recharging request, the robot identifies the base station identifier corresponding to the base station. Then, the robot compares its pre-stored target identifier with the base station identifier. If the target identifier matches the base station identifier, the robot connects to the base station. In the solution provided in this application, after the lawnmower robot enters the base station area, the base station sends a recharging request to the target device and waits for authorization, adding a remote authorization verification step. This prevents accidental triggering of unauthorized connection to the base station by unauthorized devices from the source, significantly improving the security and controllability of base station use. During base station identification, a dual-path localization encoding region is employed, combining traditional contour detection with deep learning-based binarized pixel contour extraction. This effectively resists interference from complex backgrounds and lighting conditions, improving the accuracy of identification and decoding. Simultaneously, comparison and verification between pre-stored target and base station identifiers prevents the robot from mistakenly docking with non-target base stations, ensuring reliable recharging. During the docking phase, the robot's pose deviation relative to the base station is calculated based on identifier corner information, camera intrinsic parameter matrices, and camera type, achieving precise alignment control. This effectively addresses the low docking accuracy of traditional solutions, ultimately enabling the lawnmower robot to automatically recharge safely, accurately, and stably. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] in: Figure 1 This is an application environment diagram of the lawnmower recharging control method provided in the embodiments of this application; Figure 2 A flowchart illustrating the lawnmower robot recharging control method provided in this application embodiment; Figure 3 A schematic diagram of a scenario for the lawnmower robot recharging control method provided in this application embodiment. Figure 4 This is a structural schematic of the lawnmower robot recharging control device provided in the embodiments of this application; Figure 5This is a structural block diagram of the lawnmower robot provided in an embodiment of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating an application scenario for the lawnmower robot recharging control method provided in this application embodiment. The application scenario is a lawn environment, where both the intelligent lawnmower 1 and the charging base station 2 are located. The intelligent lawnmower 1 is equipped with a visual perception module (monocular or binocular camera), a navigation control module, and a communication module, and pre-stores a global map of the lawn and target identifiers corresponding to the target base station. The charging base station 2 is fixedly installed in a preset area of ​​the lawn, and its surface is marked with a base station identifier (such as an ArUco code or a dedicated QR code) for identification. It is also equipped with a communication module, enabling bidirectional communication with the intelligent lawnmower 1 and target devices (such as user mobile phones or tablets).

[0014] When the smart lawnmower 1 finishes mowing the lawn or detects that its own battery level is lower than a preset threshold, the user can send a return-to-charge command to the smart lawnmower 1 through the target device. The smart lawnmower 1 responds to the return-to-charge command, retrieves the pre-stored global map, and the navigation control module plans the optimal path according to the global map to control its own autonomous driving and gradually enter the base station area where the charging base station 2 is located.

[0015] When the intelligent lawnmower 1 enters the base station area, its communication module sends a trigger signal to the charging base station 2. Upon receiving this signal, the charging base station 2 generates a recharge request based on the intelligent lawnmower 1's robot identifier and sends it to the target device via its own communication module. After the user passes the verification and receives authorization from the target device, the intelligent lawnmower 1 activates its visual perception module, captures an image of the charging base station 2, and identifies the base station identifier. It compares the pre-stored target identifier with the identified base station identifier. After matching, it calculates its relative deviation from the charging base station 2 through pose calculation and ultimately precisely docks with the charging base station 2, completing the automatic recharge.

[0016] This application provides a method for controlling the recharging of a lawnmower robot, including: In response to a recharge command for the lawnmower robot, a preset global map is obtained; the lawnmower robot is controlled to enter the base station area according to the global map; when the lawnmower robot enters the base station area, a recharge request is sent to the target device through the base station in the base station area; when the target device receives an authorization permission in response to the recharge request, the lawnmower robot is controlled to identify the base station identifier corresponding to the base station; the target identifier pre-stored by the lawnmower robot is compared with the base station identifier; when the target identifier matches the base station identifier, the lawnmower robot is controlled to dock with the base station.

[0017] Please see Figure 2 As shown, Figure 2 The flowchart of the lawnmower robot recharging control method provided in this application embodiment can be specifically included in the following steps: 101. In response to the recharge command for the lawnmower robot, obtain the preset global map.

[0018] The recharge command refers to the instruction that triggers the lawnmower to start the automatic recharge process. For example, a recharge control command sent by the user to the lawnmower via a target device (such as a mobile app or tablet); or an internal recharge command automatically generated by the lawnmower itself after detecting that the battery level is below a preset threshold (e.g., ≤20%) or after completing a preset mowing task. The global map refers to a spatial map covering the entire lawn operation area, pre-stored in the lawnmower's local storage module. It is collected, drawn, and updated by the lawnmower during initial debugging or daily operation, and includes key information such as lawn boundaries, obstacle locations, precise coordinates of the base station area, and the operation path.

[0019] For example, specifically, when the lawnmower robot is in normal operation or standby mode, its communication module, storage module, and navigation module remain active with low power consumption, listening for recharging commands in real time. When the navigation control module receives a valid recharging command, it triggers the recharging process, pausing the current operation (if in operation mode) and switching to recharging navigation mode. The navigation control module sends a map retrieval command to the robot's local storage module, which then reads the pre-stored global map data according to the command. The retrieved global map is loaded into the navigation control module, and the map integrity is verified. If the map verification passes, step 102 is executed.

[0020] 102. Control the lawnmower robot to enter the base station area according to the global map.

[0021] The base station area refers to a specific spatial region centered on the charging base station, pre-defined on the global map. It is the trigger area for the lawnmower robot to initiate a recharging request. Its range is a circular or rectangular area with a preset radius, based on the physical location of the charging base station. This area corresponds to a clear coordinate range on the global map and includes key information such as the charging base station coordinates, surrounding safety passages, and obstacle-free areas. The purpose is to allow the lawnmower robot to quickly identify the base station and trigger the subsequent recharging request process upon arriving in this area, while also preventing the robot from wandering aimlessly near the base station.

[0022] After the lawnmower robot loads the preset global map, the navigation control module and its own positioning module (such as GPS and visual positioning) remain active, so as to know its real-time coordinates in the global map.

[0023] The navigation control module parses the loaded global map, extracting key parameters of the base station area, including the coordinate range, boundary coordinates, and precise coordinates of the base station itself, as well as the feasible paths and obstacle locations marked on the global map leading to the base station area. Then, based on the lawnmower robot's current real-time coordinates and the precise coordinates of the base station area, combined with the obstacles marked on the global map (such as trees, flower beds, and lawn edges), the navigation control module plans the optimal path from the current location to the base station area. According to the planned optimal path, the navigation control module sends speed and direction control commands to the robot's walking drive module, controlling the robot to move at a constant speed. During movement, the self-localization module collects the robot's position information in real time, compares it with the path on the global map, and corrects the driving direction in real time to avoid deviating from the path. Throughout the movement, the navigation control module continuously compares its own real-time coordinates with the coordinate range of the base station area defined on the global map to determine whether it has entered the base station area.

[0024] 103. When the lawnmower enters the base station area, it sends a recharge request to the target device through the base station in the base station area.

[0025] The target device refers to the terminal device used to manage the lawnmower's recharging process, receive recharging requests from the base station, and send authorization. It is mainly a smart device that users can directly operate, including smartphones and tablets. The device is equipped with a control APP that is compatible with the lawnmower and the base station. It is used to receive recharging requests sent by the base station for user review and confirmation, and then send authorization to the lawnmower to realize remote control of the recharging process, avoiding problems such as unauthorized recharging and accidental triggering. At the same time, it can receive status feedback from the robot and the base station (such as successful recharging and fault alarms).

[0026] For example, after the lawnmower enters the base station area, the navigation control module triggers the communication module to send a trigger signal to the base station in the base station area. This trigger signal contains the unique robot identifier corresponding to the lawnmower.

[0027] The base station receives the trigger signal sent by the lawnmower robot through its own communication module, verifies the validity of the signal, and confirms that the trigger signal comes from a legitimate lawnmower robot that has entered the base station area (verifying the validity of the robot's identifier). If the verification fails, the base station rejects subsequent operations and reports an alarm to the target device. If the verification passes, the base station generates standardized recharge request information based on the verified trigger signal, its own base station identifier, and the robot's identifier. The recharge request information includes at least key information such as the robot's identifier, the base station identifier, the recharge trigger time, and the base station's location, so that the user can confirm the recharge entity and scenario on the target device.

[0028] Next, the base station, through its own communication module (such as a wireless communication module), sends the generated recharge request to the target device according to a preset communication protocol. The accompanying APP on the target device listens to the signal sent by the base station in real time. After successfully receiving the recharge request, it can notify the user through pop-ups, message prompts, etc., displaying relevant information about the recharge request (such as which robot and which base station initiated the recharge), and wait for the user to review and send authorization. Optionally, in some embodiments of this application, if the base station does not receive the robot's trigger signal, the trigger signal verification fails, or the recharge request fails to be sent, the base station will continue to attempt to receive and send, and simultaneously report alarm information to the target device, indicating that the recharge request initiation is abnormal.

[0029] Optionally, in some embodiments of this application, the step "when the lawnmower enters the base station area, it sends a recharging request to the target device through the base station in the base station area" may specifically include: When the lawnmower enters the base station area, the control system sends a trigger signal to the base station in the area. The base station then generates a recharge request based on the trigger signal and the lawnmower's identifier, and sends the recharge request to the target device.

[0030] After the lawnmower falls within the coordinates of the base station area, it generates a trigger signal containing its unique robot identifier and transmits this signal wirelessly to the base station within the area. Upon receiving the trigger signal, the base station verifies its validity. For example, it parses the lawnmower identifier carried in the signal and compares it with a pre-stored list of valid robot identifiers to verify whether the lawnmower is an authorized rechargeable device. If the identifier verification fails, the base station rejects further operations. If the verification succeeds, the base station, based on the verified trigger signal, the parsed lawnmower identifier, and supplemented with its own base station identifier, recharge trigger time, and other key information, generates a complete recharge request in a preset format. Then, the base station sends the generated recharge request to the target device according to a preset communication protocol.

[0031] 104. When the target device receives the authorization permission returned for the recharge request, the lawnmower robot is controlled to identify the base station identifier corresponding to the base station.

[0032] Authorization refers to the process where the target device (such as the control app on a user's phone or tablet) sends a recharging request to the base station. After user verification and confirmation, the device sends a command to the lawnmower and the base station authorizing the recharging process. The base station identifier is a unique visual identifier (such as an ArUco code or a custom-designed QR code) permanently mounted on the surface of the charging base station to identify the base station. Each base station has a unique identifier. Its surface contains specific encoded information that can be recognized and decoded by the lawnmower's visual perception module.

[0033] For example, when the lawnmower receives an authorization certificate from the target device, it verifies the validity of the certificate, confirming that the certificate originates from the target device and that the instruction encoding is correct. Then, the lawnmower adjusts its camera angle to align with the pre-defined marking area on the base station surface. Next, it selects clear base station images without significant obstructions as the raw images for marking recognition. Following this, the lawnmower preprocesses the acquired raw base station images, which may include binarization denoising, image correction, and contrast enhancement. Afterward, based on the base station's pre-defined marking type (such as ARC000 code or dedicated QR code), it extracts marking features from the preprocessed image, determines the encoded area containing the base station marking, and then uses a decoding algorithm corresponding to the marking type to parse the information within the encoded area, ultimately obtaining the unique base station marking for the base station.

[0034] Optionally, in some embodiments of this application, the step "when receiving the authorization permission returned by the target device for the recharging request, controlling the lawnmower robot to identify the base station identifier corresponding to the base station" may specifically include: When an authorization is received from the target device, the lawnmower robot is controlled to identify the base station image corresponding to the base station. Preprocess the base station images; Based on the identifier type corresponding to the base station, the base station identifier corresponding to the base station is identified from the preprocessed base station image.

[0035] The identifier type refers to the specific category of the base station identifier used by the base station, which has uniqueness and exclusive adaptability. In some embodiments of this application, the base station identifier type may include the following two types: (1) ArUco code (suitable for fast corner point extraction, adaptable to pose calculation, and strong anti-light interference capability); (2) dedicated customized QR code (containing the unique identity code of the base station, adaptable to fast decoding, and customizable encoding rules). It should be noted that each identifier type corresponds to an exclusive feature extraction method and decoding algorithm, and the lawnmower robot pre-stores the recognition logic of various identifier types.

[0036] For example, specifically, upon receiving authorization from the target device, the lawnmower's navigation control module sends an image acquisition command to the vision perception module. Upon receiving this command, the vision perception module activates the camera and adjusts its shooting angle and focus to align with the pre-set marker placement location on the base station. Then, it performs initial screening of the acquired images, removing blurry, obstructed, excessively bright, or insufficiently bright images. Further, the lawnmower's AI processing module calls a preprocessing algorithm to binarize the screened base station images, separating the marker area from background noise. Additionally, it performs image distortion correction to fix image deformation caused by camera shooting angle and lens errors. Simultaneously, it performs contrast enhancement processing to improve the distinction between the marker code and the background, eliminating the impact of uneven outdoor lighting on the image.

[0037] Furthermore, the lawnmower robot retrieves pre-stored base station identifier type parameters to determine whether the current base station uses an ArUco code or a dedicated QR code. If it is an ArUco code, the corner features and coded borders of the identifier in the image are extracted; if it is a dedicated QR code, the QR code positioning points and coded texture in the image are extracted. Then, based on the feature extraction results, the complete coded region containing the base station identifier in the preprocessed image is determined. The dedicated decoding algorithm corresponding to the identifier type is called, and the coded region is parsed to obtain the unique identity information of the base station within the coded region, which is the base station identifier corresponding to the base station. Optionally, in some embodiments of this application, the obtained base station identifier is validated. If the validation passes, step 105 is executed.

[0038] Optionally, in some embodiments of this application, the step of "identifying the base station identifier corresponding to the base station from the preprocessed base station image according to the identifier type corresponding to the base station" may specifically include: Feature extraction is performed on the preprocessed base station image based on the identifier type corresponding to the base station. Based on the feature extraction results, the coded regions containing base station identifiers in the base station images are determined; The encoded region is parsed according to the decoding algorithm corresponding to the identifier type to obtain the base station identifier corresponding to the base station.

[0039] The coded region refers to the core area of ​​the base station identifier that contains the unique identification code information of the base station. It is a core component of the base station identifier and corresponds to a specific area in the preprocessed base station image where the coded information can be extracted and can be distinguished from the background, such as the black border and internal coded dot matrix area of ​​the ArUco code, or the positioning point and internal coded texture area of ​​a custom QR code.

[0040] Feature extraction results refer to the set of key feature data extracted from preprocessed base station images that can be used to distinguish base station identifiers from the background and locate coded regions. The feature extraction results differ for different identifier types. For example, the corner coordinates, border outline, and coded dot matrix features of ArUco codes; and the location point coordinates, texture features, and arrangement patterns of coded modules for custom-designed QR codes.

[0041] For example, specifically, the lawnmower calls pre-stored parameters to determine the identifier type used by the current base station. If it is an ArUco code, it calls corner detection and contour extraction algorithms to extract the corner coordinates, black border contour features, and grayscale features of the internal coding dot matrix of the identifier in the pre-processed image. If it is a custom QR code, it calls positioning point detection and texture extraction algorithms to extract the positioning point coordinates, positioning box contour, and arrangement features of the internal coding modules of the QR code in the image. Then, it removes noise features and background interference features generated during the extraction process, and matches the filtered effective feature extraction results with the feature templates of the corresponding identifier type pre-stored by the lawnmower to select feature areas with a matching degree higher than a preset threshold (such as 90%).

[0042] Furthermore, based on the successfully matched features, the complete boundary of the identifier is fitted, the specific range of the encoding region (such as the coordinate range) is determined, and the validity of the encoding region is verified, such as whether the encoding region is complete, unobstructed, and distortion-free. If the encoding region has partial obstruction or incomplete boundaries, the process returns to the feature extraction step. If the encoding region is valid, the encoding region is parsed point by point according to the decoding algorithm corresponding to the base station identifier type, thereby extracting the encoding information within the encoding region and converting the encoding information into readable numerical or character encoding.

[0043] Optionally, in some embodiments of this application, the parsed encoded information can be validated to confirm that the encoding format conforms to preset rules and that there are no decoding errors. If the decoding validation passes, the parsed encoded information is identified as the base station identifier corresponding to the base station and transmitted to the navigation control module of the lawnmower robot for subsequent comparison with the pre-stored target identifier. If the decoding fails, the visual perception module is controlled to re-acquire the base station image and all the above steps are repeated. After multiple failures, an alarm is reported to the target device to indicate that the base station identifier decoding has failed.

[0044] Optionally, in some embodiments of this application, the step "determining the coded region containing the base station identifier in the base station image based on the feature extraction results" may specifically include: Candidate contours that meet preset conditions are identified from the feature extraction results; The candidate contour is approximated by polygons, and multiple vertex coordinates are obtained by fitting. Based on the preset identifier structure rules and multiple vertex coordinates, the encoded region containing the base station identifier in the base station image is determined.

[0045] Polygon approximation processing refers to the algorithmic operation of simplifying candidate contours in the feature extraction results. The core is to use approximation algorithms such as Douglas-Peucker to remove redundant pixels on the candidate contours, retain the core shape features of the contours, and fit irregular continuous pixel contours into polygon contours composed of several line segments.

[0046] Identifier structure rules refer to fixed rules used to determine whether a candidate contour is a base station identifier encoding area. These rules correspond one-to-one with the base station identifier type and are the core basis for distinguishing base station identifiers from background clutter. Specifically, these rules may include: the contour shape being a standard rectangle, the number of vertices, and the side length ratio of adjacent vertices conforming to a preset range.

[0047] Vertex coordinates refer to the two-dimensional pixel coordinates of each vertex of the polygon contour obtained by polygon approximation in the preprocessed base station image. For example, with the upper left corner of the image as the origin, the horizontal axis is the X-axis and the vertical axis is the Y-axis. Optionally, in some embodiments of this application, the fitted polygon can be a rectangle, and the vertex coordinates are the coordinates of the four corners of the rectangle, such as the upper left corner (X1, Y1), the upper right corner (X2, Y2), the lower right corner (X3, Y3), and the lower left corner (X4, Y4).

[0048] For example, specifically, in the feature extraction results, all contour features are determined, and all contours are filtered according to preset conditions. The preset conditions may include contour area within a preset range, contour smoothness, and contour closure. Then, contours that meet the above preset conditions are determined as candidate contours. For each candidate contour, a polygon approximation algorithm is called to simplify the continuous pixels on the candidate contour, remove redundant pixels, and retain key pixels that can reflect the overall shape of the contour. Then, through algorithm fitting, each candidate contour is transformed into a polygon contour composed of several line segments. Combining the rectangular structure characteristics of the base station identifier, the focus is on fitting a quadrilateral contour. Then, all vertices of the fitted polygon contour are extracted, and the two-dimensional pixel coordinates (i.e., vertex coordinates) of each vertex are recorded. They can be sorted in clockwise or counterclockwise order to facilitate subsequent comparison with the identifier structure rules. If the number of vertices of the fitted polygon does not meet the preset number (e.g., 4), the candidate contour is removed.

[0049] According to the identification structure rules, for each fitted candidate contour, its vertex coordinates are checked one by one, such as checking the number of vertices, the side length ratio, and the contour shape.

[0050] Identify all candidate contours that conform to the label structure rules. If multiple contours meet the criteria, select the contour with the highest matching degree to the pre-stored label template. If no contours meet the criteria, return to the previous step to re-extract features and filter candidate contours. Finally, candidate contours that meet the conditions are determined, and the image region corresponding to the contour is determined by using its vertex coordinates as the boundary. That is, the encoded region containing the base station identifier in the base station image is recorded, and the coordinate range of the encoded region is recorded for subsequent decoding operations.

[0051] Optionally, in some embodiments of this application, the step "determining the coded region containing the base station identifier in the base station image based on the feature extraction results" may specifically include: The feature extraction results are binarized. Determine the pixel outline corresponding to the base station identifier from the binarization processing results; Based on pixel contours, the encoded regions containing base station identifiers in the base station image are determined.

[0052] For example, specifically, the grayscale image in the feature extraction result is traversed pixel by pixel to determine the grayscale value of each pixel: if the pixel grayscale value is within a preset threshold range, the pixel is set as the foreground color (e.g., black); if the pixel grayscale value exceeds the preset threshold, the pixel is set as the background color (e.g., white). After completing the full-image pixel binarization, the image is subjected to simple denoising processing to obtain a binarized image containing only the foreground color identifier and the white background, i.e., the binarization processing result. Subsequently, edge scanning is performed on the binarized image to identify all continuous foreground color (black) pixel sets in the image. Each continuous pixel set is a pixel contour. Then, pixel contours that meet the preset judgment conditions are retained, and the boundary of the retained pixel contours is located to extract the outermost pixel coordinates of the pixel contour and determine the minimum bounding rectangle of the contour (or the actual boundary that fits the contour).

[0053] Furthermore, the validity of the region corresponding to the pixel contour can be detected first. If the pixel contour is valid, the image region covered by the contour can be locked based on the boundary coordinates of the pixel contour. This region is the encoded region containing the base station identifier in the base station image.

[0054] 105. Compare the target identifiers pre-stored by the lawnmower robot with the base station identifiers.

[0055] The target identifier refers to the unique identifier that is pre-stored in the local storage module of the lawnmower robot and is uniquely associated with the target recharge base station.

[0056] For example, specifically, the pre-stored target identifier is read, and then the retrieved target identifier and the identified base station identifier are uniformly converted into the same encoding format (such as binary encoding or string encoding) to ensure the consistency of the comparison between the two. Furthermore, it is determined whether the target identifier and the base station identifier are of the same type. If the types are different, a mismatch is determined. If the types are the same, the encoded content of the two is compared bit by bit to confirm that each bit of the encoded data is exactly the same. If the comparison matches, step 106 is triggered. If the comparison does not match, the lawnmower robot is controlled to stop subsequent operations and an alarm message is reported to the target device, indicating: the base station identifier and the target identifier do not match and cannot be connected.

[0057] 106. When the target identifier matches the base station identifier, control the lawnmower robot to dock with the base station.

[0058] For example, specifically, when the target identifier matches the base station identifier, the corner point information corresponding to the base station identifier, the physical size of the base station identifier, the intrinsic parameter matrix of the camera carried by the robot, and the camera type corresponding to the robot are obtained. Then, based on the obtained information, the pose deviation of the robot relative to the base station is calculated. Finally, the lawnmower robot is controlled to dock with the base station based on the pose deviation.

[0059] Optionally, in some embodiments of this application, the step "when the target identifier matches the base station identifier, control the lawnmower robot to dock with the base station" may specifically include: Obtain the corner information corresponding to the base station identifier and the intrinsic parameter matrix corresponding to the lawnmower robot; Based on corner information, intrinsic parameter matrix, and the camera type corresponding to the lawnmower, calculate the pose deviation of the lawnmower relative to the base station; The lawnmower robot is docked to the base station based on pose deviation control.

[0060] Corner information refers to the pixel coordinates of key geometric corner points extracted from the coded area of ​​the base station identifier in the image. For example, if the base station identifier is rectangular with four characteristic corner points, the corner information is the two-dimensional pixel coordinates of these four corner points. The intrinsic parameter matrix refers to the inherent internal parameter matrix of the camera mounted on the lawnmower robot, calibrated at the factory and pre-stored in the robot, including parameters such as focal length, principal point coordinates, and distortion correction coefficients. Pose deviation refers to the deviation of the lawnmower robot's current position and posture relative to the standard docking position of the base station charging interface, including positional deviation and posture deviation.

[0061] For example, specifically, the corresponding pose calculation algorithm can be determined based on the camera type of the Gechao robot. Then, the corner information and intrinsic parameter matrix are input into the pose calculation algorithm to calculate the rotation matrix and translation vector of the robot relative to the base station. The angle deviation is then obtained from the rotation matrix, and the position deviation is obtained from the translation vector. These are integrated to obtain the pose deviation of the robot relative to the base station. Finally, based on the calculated pose deviation, control commands for attitude adjustment and movement are generated, and the robot is controlled to correct its position and angle to move closer to the base station charging interface based on these control commands. During the movement, corner information is continuously updated, pose deviations are repeatedly calculated and dynamically corrected until the robot's recharge interface and the base station's charging interface are fully aligned and physically connected, completing the docking and recharge.

[0062] It should be noted that, in some embodiments of this application, the lateral deviation e of the lawnmower robot relative to the base station can be extracted from the pose deviation. x The straight-line distance d from the base station and the yaw angle deviation e θ The PID controller in the navigation control module generates corresponding speed control commands based on the aforementioned deviation values. The angular velocity control quantity is calculated according to... Calculations are performed to correct the heading angle of the lawnmower robot and compensate for yaw angle deviations; the linear velocity control quantity is calculated according to v=Kp. d d calculates and controls the lawnmower robot's forward speed towards the base station, while simultaneously adjusting the lateral deviation e. xThe robot's left and right wheel speeds are adjusted in real time to achieve precise lateral position correction. During the robot's movement, following speed control commands, the visual perception module re-acquires base station images and updates the corner information of the base station markers after each small distance. The AI ​​processing module and navigation control module repeatedly execute the pose deviation calculation process. The PID controller updates the speed control commands in real time based on the new pose deviation, forming a closed-loop control process of perception, calculation, and control. This closed-loop process continues iteratively until all deviations of the robot relative to the base station fall within a preset small tolerance range, such as position deviation ≤ 0.5cm and angle deviation ≤ 1°.

[0063] Optionally, in some embodiments of this application, the step "calculating the pose deviation of the lawnmower robot relative to the base station based on corner information, intrinsic parameter matrix, and the camera type corresponding to the lawnmower robot" may specifically include: The pose calculation algorithm is determined based on the type of camera corresponding to the lawnmower robot; Based on the pose calculation algorithm, corner information, and intrinsic parameter matrix, the rotation matrix and translation vector of the lawnmower robot relative to the base station identifier are calculated. The pose deviation of the lawnmower robot relative to the base station is determined based on the rotation matrix and translation vector.

[0064] Among them, the pose calculation algorithm is a visual algorithm that uses visual feature points and camera internal parameters to calculate the robot's position and orientation in three-dimensional space relative to the base station marker. For example, the EPnP algorithm is used for monocular cameras, and the stereo vision triangulation algorithm is used for binocular cameras.

[0065] Specifically, the acquired base station identifier corner information and camera intrinsic parameter matrix are used as inputs and fed into the selected pose calculation algorithm. Through algorithm calculation, the rotation matrix and translation vector of the robot relative to the base station identifier are output. The rotation matrix represents the robot's attitude deflection relationship relative to the base station, and the translation vector represents the robot's spatial distance and orientation relationship relative to the base station. Next, the robot's angular deviations, including yaw angle, pitch angle, and roll angle deviations, are decomposed from the rotation matrix, and the robot's position deviation is extracted from the translation vector. The angular deviations and position deviations are then integrated to obtain the complete pose deviation of the robot relative to the base station.

[0066] To further understand the lawnmower robot recharging control scheme provided in this application, the following will be combined with... Figure 3To further explain, the application scenario of this embodiment is still a lawn operation environment. The intelligent lawnmower is equipped with a visual perception module (monocular or binocular camera), a navigation control module, a communication module, and an AI processing module. The visual perception module can achieve collaborative operation of alignment mark recognition and environmental perception. The AI ​​processing module is equipped with a lightweight convolutional neural network, which can complete semantic segmentation and target detection tasks. The lawnmower has a pre-stored global lawn map, target mark corresponding to the target base station, camera intrinsic parameters and distortion coefficients, and pre-stored spatial range parameters of the base station docking safety channel. The charging base station is fixedly set in a preset area of ​​the lawn, and its surface is equipped with an alignment mark for identification (i.e., base station mark, such as ArUco code or specially designed QR code). The base station is equipped with a communication module, which can achieve bidirectional communication with the lawnmower and the target device, as detailed below: The visual perception module performs image preprocessing operations on the acquired raw RGB images. The input is the raw RGB image obtained from a monocular or binocular camera. The specific operations include: performing distortion correction on the raw RGB image using pre-calibrated camera intrinsic parameters and distortion coefficients; adjusting the color space of the corrected image, such as converting it to HSV or grayscale to enhance the robustness of specific color features; and performing illumination normalization processing on the image, such as histogram equalization, to cope with different lighting conditions in the lawn environment.

[0067] The visual perception module performs alignment mark recognition on the preprocessed image. The goal is to quickly and accurately detect and decode alignment marks (such as ArUco codes or specially designed QR codes) on the base station. The specific steps are as follows: Use a detector specific to the identifier type to detect and extract the features of the identifier. For example, for ArUco codes, use their inherent black and white square boundary features to locate the identifier in the image through steps such as image contour search, polygon approximation and code region extraction. After the positioning marker is successfully located, the algorithm obtains the precise pixel coordinates of the four corner points (or more key points) of the marker in the image. These pixel coordinates are the core input for subsequent calculation of the pose of the lawnmower robot relative to the base station. The data embedded in the positioning identifier, such as the unique ID of the base station, is decoded and compared with the target base station ID (i.e. the target identifier in Example 1) stored in the memory of the lawnmower robot to complete the base station identity confirmation. If the two do not match, the homing process is immediately terminated and an "illegal base station" alarm is reported to the target device. If the two match, the subsequent environmental perception process is entered.

[0068] A lightweight object detection neural network, such as a miniature version of YOLO or SSD-Mobilenet, is run in parallel in the AI ​​processing module. This model takes the preprocessed whole image as input and directly outputs the bounding boxes and class confidence scores of all detected target objects in the image. The detection area of ​​the alignment marker is extended towards the lawnmower robot. Based on the spatial range parameters of the safe channel for base station docking stored in the lawnmower robot, a safe channel, namely the region of interest (ROI), is defined. If a potential obstacle is detected in the ROI, an alarm is reported to the target device and the recharging process is paused. If there is no potential obstacle in the ROI, the docking environment is determined to be safe and the subsequent process continues.

[0069] The AI ​​processing module performs image recognition operations on the preprocessed image. The core of this module is a lightweight convolutional neural network, primarily responsible for semantic segmentation and / or object detection tasks, to achieve pixel-level understanding of the lawn environment. Its algorithm principle and detailed process are as follows: This lightweight convolutional neural network employs a fully convolutional network with an encoder-decoder structure. The encoder, such as the backbone networks of MobileNetV2 and EfficientNet-Lite, is responsible for extracting multi-level features, from shallow edges and textures to deep semantic information. The decoder, through upsampling and skip connections, fuses deep semantic information with shallow detail information, gradually restoring the spatial resolution of the image. The network ultimately outputs a "segmentation map" of the same size as the input image, where each pixel is assigned a category label, such as background, alignment marker, potential obstacle, and passable ground.

[0070] The preprocessed image is scaled to a fixed size (e.g., 256x256 pixels) and normalized before being input into the lightweight convolutional neural network; The image is processed by the encoder, which generates a series of feature maps with decreasing size but increasing number of feature channels. These feature maps are fed into the decoder, which upsamples them through transposed convolution or interpolation. At the same time, it incorporates feature maps from the corresponding stages of the encoder through "skip connections" to supplement positional details. The last convolutional layer of the network outputs the probability distribution of each pixel belonging to each category. In this method, a threshold (e.g., 0.8) is applied to the probability map of the position identifier category to obtain a binary mask. Contour lookup is then performed on this binary mask to obtain the precise pixel-level contour and corner coordinates of the position identifier. This method is more robust to complex backgrounds than traditional methods. The output of the potential obstacle category can be connected to a region proposal network to generate the bounding boxes of the obstacles.

[0071] This lightweight convolutional neural network requires training with a large number of labeled images of lawnmower working scenes. The labels include the precise contours of the alignment markers and the bounding boxes of obstacles. After training, the model is pruned, quantized, and converted to an inference framework format supported by specific hardware (such as the lawnmower's main control chip) to enable real-time operation on embedded devices.

[0072] After the above image recognition operation, the corner coordinates of the alignment marker are updated as accurate input parameters for subsequent pose calculation.

[0073] The lawnmower performs pose calculations based on the updated corner coordinates. The purpose of pose calculation is to obtain the three-dimensional relative position (X, Y, Z deviations) and orientation angle (yaw angle, and pitch and roll angles if necessary) between the camera on the lawnmower and the alignment marker on the base station. Depending on whether the lawnmower is equipped with a monocular or binocular camera, different pose calculation algorithms are used. Before pose calculation, the pixel coordinates of the corner points are converted to camera coordinates. The core conversion formula is: Where (u, v) are the pixel coordinates of the corner points, and K is the camera intrinsic parameter matrix, and f x and f y These are the focal lengths of the camera on the x and y axes, respectively. x c y ) is the main pixel coordinate, (X) c Y c Z c () represents the three-dimensional coordinates of the corner point in the camera coordinate system.

[0074] Using triangulation, given the horizontal distance (baseline) between two cameras, the depth (distance Z) of a point is calculated by matching the horizontal pixel difference (disparity) of the same feature point in the left and right images. The depth calculation formula is as follows: Where Z is the depth (vertical distance) from the feature point to the camera, B is the baseline distance of the binocular cameras, f is the camera focal length (equivalent focal length), and d is the parallax (the difference in horizontal pixel coordinates of the same feature point in the left and right images, d=uL). uR, uL are the x-coordinates of the pixels in the left image, and uR is the x-coordinate of the pixels in the right image.

[0075] Specifically, the images captured by the left and right cameras are projected onto a coplanar, row-aligned coordinate system, ensuring that feature points differ only in the horizontal direction. Then, in the corner regions of the alignment markers, a stereo matching algorithm (such as Semi-Global Matching, SGM) is used to calculate a dense or sparse disparity map. The 3D depth Z of the corner is calculated using the aforementioned depth formula. Combined with the pixel-camera coordinate transformation formula, the complete 3D coordinates (X, Z) of the corner in the camera coordinate system are obtained. c Y c Z c) .

[0076] Furthermore, based on the 3D coordinates of the camera coordinate system at multiple corner points and the 3D coordinates of the world coordinate system of the alignment marker, the rotation matrix R and translation vector t of the lawnmower robot camera relative to the base station alignment marker are calculated, completing the initial pose calculation. Then, the angular deviation of the lawnmower robot (yaw angle, and pitch angle and roll angle if necessary) is decomposed from the calculated rotation matrix R, and the position deviation of the robot (X, Y, and Z axis deviations) is extracted from the translation vector t. After integration, the complete pose deviation of the lawnmower robot relative to the base station is obtained. If the lawnmower robot is equipped with a monocular camera, its pose calculation execution logic is the same as in Example 1. The pose deviation is extracted after calculating the rotation matrix R and translation vector t using the EPnP algorithm. The definitions of corner coordinates, intrinsic parameter matrix, rotation matrix, translation vector, and pose deviation are completely consistent with those in Example 1.

[0077] The navigation control module of the lawnmower sends speed and direction control commands to the walking drive module based on the calculated pose deviation, planning and controlling the lawnmower to slowly approach the base station charging interface. During the docking process, the vision perception module collects images of the base station, and the AI ​​processing module updates the corner coordinates in real time and repeatedly executes the above pose calculation formula and steps to dynamically correct the lawnmower's driving direction and posture. At the same time, it performs environmental perception. If an obstacle is detected in the docking channel, it controls the lawnmower to stop and reports an alarm to the target device. When the lawnmower's return charging interface and the base station's charging interface make physical contact, and the charging detection module detects that the charging circuit is conducting, it controls the lawnmower to switch to charging mode and sends feedback information that the return charging docking is successful to the target device, completing the entire automatic return charging process.

[0078] This application provides a method for controlling the recharging of a lawnmower robot. In response to a recharging command for the lawnmower robot, a preset global map is acquired, and the robot is controlled to enter a base station area based on the global map. When the lawnmower robot enters the base station area, a recharging request is sent to the target device via a base station in the area. Upon receiving authorization from the target device in response to the recharging request, the lawnmower robot is controlled to identify the base station identifier corresponding to the base station. Then, the target identifier stored in the lawnmower robot is compared with the base station identifier. When the target identifier matches the base station identifier, the lawnmower robot is controlled to dock with the base station. In the solution provided in this application, after the lawnmower robot enters the base station area, the base station sends a recharging request to the target device and waits for authorization. A remote authorization verification step is added to prevent accidental triggering and unauthorized docking with the base station by unauthorized devices, significantly improving the security and controllability of base station use. During the base station identifier recognition process, a dual-path localization encoding region is adopted using traditional contour detection and deep learning binarized pixel contour extraction, effectively resisting complex backgrounds and lighting interference, and improving the accuracy of identifier recognition and decoding. Simultaneously, by comparing and verifying the pre-stored target identifier with the base station identifier, the robot is prevented from mistakenly docking with non-target base stations, ensuring the reliability of recharging. During the docking phase, the robot's pose deviation relative to the base station is calculated based on the identifier corner information, camera intrinsic parameter matrix, and camera type, achieving precise alignment control. This effectively solves the problem of low docking accuracy in traditional solutions, ultimately enabling the lawnmower robot to automatically recharge safely, accurately, and stably.

[0079] To facilitate better implementation of the lawnmower recharging control method of this application embodiment, this application embodiment also provides a lawnmower recharging control device, wherein the meanings of the terms are the same as those of the lawnmower recharging control method described above, and specific implementation details can be found in the description of the system embodiment.

[0080] Please see Figure 4 , Figure 4 This is a schematic diagram of the structure of the lawnmower recharging control device provided in an embodiment of this application. Specifically, the lawnmower recharging control device may include an acquisition module 201, a first control module 202, a sending module 203, a receiving module 204, a comparison module 205, and a second control module 206, as follows: The acquisition module 201 is used to acquire a preset global map in response to the recharge command for the lawnmower robot; The first control module 202 is used to control the lawnmower robot to enter the base station area according to the global map; The sending module 203 is used to send a recharge request to the target device through the base station in the base station area when the lawnmower enters the base station area; The receiving module 204 is used to control the lawnmower robot to identify the base station identifier corresponding to the base station when it receives the authorization permission returned by the target device for the recharge request. The comparison module 205 is used to compare the target identifier pre-stored by the lawnmower robot with the base station identifier; The second control module 206 is used to control the lawnmower robot to dock with the base station when the target identifier matches the base station identifier.

[0081] Optionally, in some embodiments of this application, the receiving module 204 may specifically be used for: When the authorization permission is received from the target device, the lawnmower robot is controlled to identify the base station image corresponding to the base station. The base station image is preprocessed; Based on the identifier type corresponding to the base station, the base station identifier corresponding to the base station is identified from the preprocessed base station image.

[0082] Optionally, in some embodiments of this application, the receiving module 204 may specifically be used for: Feature extraction is performed on the preprocessed base station image based on the identifier type corresponding to the base station; Based on the feature extraction results, the coded region containing the base station identifier in the base station image is determined; The encoded region is parsed according to the decoding algorithm corresponding to the identifier type to obtain the base station identifier corresponding to the base station.

[0083] Optionally, in some embodiments of this application, the receiving module 204 may specifically be used for: Candidate contours that meet preset conditions are identified from the feature extraction results; The candidate contour is approximated by polygons, and multiple vertex coordinates are obtained by fitting. Based on the preset identifier structure rules and the coordinates of the multiple vertices, the encoded region containing the base station identifier in the base station image is determined.

[0084] Optionally, in some embodiments of this application, the receiving module 204 may be specifically used for The feature extraction results are binarized. The pixel outline corresponding to the base station identifier is determined from the binarization processing result; Based on the pixel contours, the coded region containing the base station identifier in the base station image is determined.

[0085] Optionally, in some embodiments of this application, the second control module 206 may specifically be used for: Obtain the corner information corresponding to the base station identifier and the intrinsic parameter matrix corresponding to the lawnmower robot; Based on the corner information, intrinsic parameter matrix, and the camera type corresponding to the lawnmower, calculate the pose deviation of the lawnmower relative to the base station; Based on the pose deviation, the lawnmower robot is controlled to dock with the base station.

[0086] Optionally, in some embodiments of this application, the second control module 206 may specifically be used for: The pose calculation algorithm is determined based on the camera type corresponding to the lawnmower robot; Based on the pose calculation algorithm, the corner information, and the intrinsic parameter matrix, the rotation matrix and translation vector of the lawnmower robot relative to the base station identifier are calculated. The pose deviation of the lawnmower robot relative to the base station is determined based on the rotation matrix and translation vector.

[0087] Optionally, in some embodiments of this application, the sending module 203 may specifically be used for: When the lawnmower enters the base station area, the system controls the lawnmower to send a trigger signal to the base station in the base station area. The base station then generates a recharge request based on the trigger signal and the lawnmower identifier corresponding to the lawnmower, and sends the recharge request to the target device.

[0088] This application provides a lawnmower robot recharging control device. In response to a recharging command for the lawnmower robot, the acquisition module 201 acquires a preset global map. The first control module 202 controls the lawnmower robot to enter a base station area based on the global map. When the lawnmower robot enters the base station area, the sending module 203 sends a recharging request to the target device through the base station in the area. When the receiving module 204 receives authorization from the target device in response to the recharging request, it controls the lawnmower robot to identify the base station identifier corresponding to the base station. Then, the comparison module 205 compares the target identifier pre-stored by the lawnmower robot with the base station identifier. When the target identifier matches the base station identifier, the second control module 206 controls the lawnmower robot to connect to the base station. In the solution provided in this application embodiment, after the lawnmower robot enters the base station area, the base station sends a recharging request to the target device and waits for authorization. This adds a remote authorization verification step, preventing accidental triggering of unauthorized connection to the base station by unauthorized devices from the source, significantly improving the security and controllability of base station use. During base station identification, a dual-path localization encoding region is employed, combining traditional contour detection with deep learning-based binarized pixel contour extraction. This effectively resists interference from complex backgrounds and lighting conditions, improving the accuracy of identification and decoding. Simultaneously, comparison and verification between pre-stored target and base station identifiers prevents the robot from mistakenly docking with non-target base stations, ensuring reliable recharging. During the docking phase, the robot's pose deviation relative to the base station is calculated based on identifier corner information, camera intrinsic parameter matrices, and camera type, achieving precise alignment control. This effectively addresses the low docking accuracy of traditional solutions, ultimately enabling the lawnmower robot to automatically recharge safely, accurately, and stably.

[0089] In one embodiment, a lawnmower robot is provided, the internal structure of which can be shown in the following diagram: Figure 5 As shown, the lawnmower robot includes a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile and / or volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The network interface of the lawnmower robot is used to communicate with external clients via a network connection. When the computer program is executed by the processor, it implements the methods described in any of the foregoing embodiments of this application.

[0090] This application also provides a computer-readable storage medium storing computer-executable instructions that, when executed by a processor, are used to implement the methods described in any of the foregoing embodiments of this application.

[0091] This application also provides a chip for executing instructions, which is used to perform the methods described in any of the foregoing embodiments executed by an electronic device as described in any of the foregoing embodiments of this application.

[0092] This application also provides a computer program product, which includes a computer program that, when executed by a processor, can implement the methods described in any of the foregoing embodiments executed by an electronic device as described in any of the foregoing embodiments of this application.

[0093] It should be noted that the functions or steps that the computer-readable storage medium or lawnmower robot can achieve are described in the relevant descriptions of the server side and client side in the aforementioned method embodiments. To avoid repetition, they will not be described one by one here.

[0094] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0095] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0096] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for controlling the recharging of a lawnmower robot, characterized in that, include: In response to the recharge command for the lawnmower robot, a preset global map is obtained; The lawnmower robot is controlled to enter the base station area based on the global map. When the lawnmower robot enters the base station area, it sends a recharge request to the target device through the base station in the base station area. When the target device receives the authorization permission returned in response to the recharge request, the lawnmower robot is controlled to identify the base station identifier corresponding to the base station. The target identifier pre-stored by the lawnmower robot is compared with the base station identifier; When the target identifier matches the base station identifier, the lawnmower robot is controlled to dock with the base station.

2. The lawnmower robot recharging control method according to claim 1, characterized in that, When the authorization permission returned by the target device in response to the recharge request is received, the lawnmower robot is controlled to identify the base station identifier corresponding to the base station, including: When the authorization permission is received from the target device, the lawnmower robot is controlled to identify the base station image corresponding to the base station. The base station image is preprocessed; Based on the identifier type corresponding to the base station, the base station identifier corresponding to the base station is identified from the preprocessed base station image.

3. The lawnmower robot recharging control method according to claim 2, characterized in that, The step of identifying the base station identifier corresponding to the base station from the preprocessed base station image based on the identifier type corresponding to the base station includes: Feature extraction is performed on the preprocessed base station image based on the identifier type corresponding to the base station; Based on the feature extraction results, the coded region containing the base station identifier in the base station image is determined; The encoded region is parsed according to the decoding algorithm corresponding to the identifier type to obtain the base station identifier corresponding to the base station.

4. The lawnmower robot recharging control method according to claim 3, characterized in that, The step of determining the encoded region containing the base station identifier in the base station image based on the feature extraction results includes: Candidate contours that meet preset conditions are identified from the feature extraction results; The candidate contour is approximated by polygons, and multiple vertex coordinates are obtained by fitting. Based on the preset identifier structure rules and the coordinates of the multiple vertices, the encoded region containing the base station identifier in the base station image is determined.

5. The lawnmower robot recharging control method according to claim 3, characterized in that, The step of determining the encoded region containing the base station identifier in the base station image based on the feature extraction results includes: The feature extraction results are binarized. The pixel outline corresponding to the base station identifier is determined from the binarization processing result; Based on the pixel contours, the coded region containing the base station identifier in the base station image is determined.

6. The lawnmower robot recharging control method according to claim 1, characterized in that, When the target identifier matches the base station identifier, the lawnmower robot is controlled to dock with the base station, including: Obtain the corner information corresponding to the base station identifier and the intrinsic parameter matrix corresponding to the lawnmower robot; Based on the corner information, intrinsic parameter matrix, and the camera type corresponding to the lawnmower, calculate the pose deviation of the lawnmower relative to the base station; Based on the pose deviation, the lawnmower robot is controlled to dock with the base station.

7. The lawnmower robot recharging control method according to claim 6, characterized in that, The step of calculating the pose deviation of the lawnmower relative to the base station based on the corner information, the intrinsic parameter matrix, and the camera type corresponding to the lawnmower includes: The pose calculation algorithm is determined based on the camera type corresponding to the lawnmower robot; Based on the pose calculation algorithm, the corner information, and the intrinsic parameter matrix, the rotation matrix and translation vector of the lawnmower robot relative to the base station identifier are calculated. The pose deviation of the lawnmower robot relative to the base station is determined based on the rotation matrix and translation vector.

8. The lawnmower robot recharging control method according to any one of claims 1 to 7, characterized in that, When the lawnmower robot enters the base station area, it sends a recharging request to the target device through the base station in the base station area, including: When the lawnmower enters the base station area, the system controls the lawnmower to send a trigger signal to the base station in the base station area. The base station then generates a recharge request based on the trigger signal and the lawnmower identifier corresponding to the lawnmower, and sends the recharge request to the target device.

9. A recharging control device for a lawnmower robot, characterized in that, include: The acquisition module is used to acquire a preset global map in response to the recharge command for the lawnmower robot; The first control module is used to control the lawnmower robot to enter the base station area according to the global map; The sending module is used to send a recharge request to the target device through the base station in the base station area when the lawnmower robot enters the base station area. The receiving module is used to control the lawnmower robot to identify the base station identifier corresponding to the base station when it receives the authorization permission returned by the target device in response to the recharge request. The comparison module is used to compare the target identifier pre-stored by the lawnmower robot with the base station identifier; The second control module is used to control the lawnmower robot to dock with the base station when the target identifier matches the base station identifier.

10. A lawnmower robot, characterized in that, The lawnmower robot includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the lawnmower robot recharging control method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the lawnmower robot recharging control method as described in any one of claims 1 to 8.

Images