Control method for gardening robot, and gardening robot

Abstract

The present disclosure relates to the technical field of intelligent control, and particularly relates to a control method for a gardening robot, and a gardening robot. A positioning sensing system of the gardening robot consists of a camera, an inertial navigation module and an odometer. The control method comprises: using images collected by a camera, inertial navigation information acquired by an inertial navigation module and a movement distance acquired by an odometer to perform fused positioning, so as to determine positioning information of a gardening robot; in a mapping mode, controlling the gardening robot to move along the boundary of a working area, and during the movement, using the images collected by the camera and the positioning information of the gardening robot to generate a working map corresponding to the working area; and in a working mode, using the positioning information of the gardening robot and the working map corresponding to the working area to control the gardening robot to work in the working area, wherein the working area is located in an open space. By means of the present application, the implementation cost can be greatly reduced, thereby widely improving the applicability.

Description

A control method of a gardening robot and the gardening robot TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of intelligent control, and particularly relates to a control method of a gardening robot and the gardening robot. BACKGROUND

[0002] With the gradual popularization of intelligent devices in daily life, a gardening robot can autonomously complete automatic work, such as planting, mowing, fertilizing, and irrigation work tasks. Taking an automatic mower as an example, the automatic mower can travel on the lawn of a user's home and perform cutting work, thereby realizing automatic cutting of the lawn and greatly saving manpower.

[0003] Existing gardening robots that realize automatic work are often based on sensors such as a GPS (Global Positioning System), an RTK (Real-time kinematic), and a laser radar. These sensors are costly and have poor applicability. SUMMARY

[0004] Therefore, the present application provides a control method of a gardening robot and the gardening robot to reduce the implementation cost and improve the applicability.

[0005] In a first aspect, a control method of a gardening robot is provided. The positioning sensor system of the gardening robot is composed of a camera, an inertial navigation module, and an odometer. The control method comprises the following steps.

[0006] Fusing positioning is performed by using images collected by the camera, inertial navigation information obtained by the inertial navigation module, and a moving distance obtained by the odometer to determine the positioning information of the gardening robot.

[0007] In a mapping mode, the gardening robot is controlled to move at the boundary of a working area. During the movement, images collected by the camera and the positioning information of the gardening robot are used to generate a working map corresponding to the working area.

[0008] In a working mode, the positioning information of the gardening robot and the working map corresponding to the working area are used to control the gardening robot to work in the working area.

[0009] The working area is located in an open space.

[0010] Optionally, the control method further comprises the following steps.

[0011] Obstacle recognition is performed by using images collected by the camera, and the gardening robot is controlled to perform corresponding obstacle avoidance processing according to the type of the recognized obstacle.

[0012] Optionally, the camera is a downward camera.

[0013] Optionally, the image captured by the camera carries a timestamp.

[0014] The fusion positioning using the image captured by the camera, the inertial navigation information obtained by the inertial navigation module and the moving distance obtained by the odometer comprises:

[0015] Aligning the image captured by the camera, the inertial navigation information obtained by the inertial navigation module and the moving distance obtained by the odometer according to the timestamp.

[0016] Fusion positioning using the aligned image and inertial navigation information.

[0017] Optionally, the control method further comprises:

[0018] Switching to the mapping mode in response to the first start of the gardening robot, entering a working area not contained in the working map or receiving a mapping instruction; or,

[0019] Switching to the working mode in response to the completion of the establishment of the working map corresponding to the working area, entering a working area already contained in the working map or receiving a working instruction.

[0020] Optionally, in the mapping mode, controlling the gardening robot to move on the boundary of the working area, and generating the working map corresponding to the working area using the image captured by the camera and the positioning information of the gardening robot during the movement comprises:

[0021] In the mapping mode, controlling the gardening robot to move on the boundary of the working area, and generating an initial map corresponding to the working area using the image captured by the camera on the boundary of the working area and the positioning information of the gardening robot.

[0022] According to the initial map, controlling the gardening robot to move, and updating the initial map to obtain the working map using the image captured at a position on or within a preset distance range from the boundary of the working area during the movement.

[0023] Optionally, the controlling the gardening robot to move on the boundary of the working area in the mapping mode comprises:

[0024] In the mapping mode, in response to a first control instruction of a user, controlling the gardening robot to move along the boundary of the working area; and / or,

[0025] The image captured by the camera is used to identify the boundary of the working area, and the robot is controlled to move along the boundary of the working area according to the identification result.

[0026] Optionally, the working area includes a first working area, a second working area and a passageway, and the start position and the end position of the passageway are located in the first working area and the second working area, respectively.

[0027] Optionally, the working map corresponding to the working area includes a passageway map.

[0028] The generation of the passageway map includes:

[0029] In response to a first instruction of a user, the start position of the passageway is determined, and the distance between the start position of the passageway and the boundary of the first working area is less than or equal to a first threshold value.

[0030] In response to a second control instruction of a user, the robot is controlled to move from the start position of the passageway to the end position of the passageway, and a passageway map is established by using the image captured by the camera during the movement.

[0031] Optionally, the control of the robot to work in the working area includes:

[0032] A working path is planned according to the working map, the robot is controlled to move and work according to the working path, and the working path includes a path through the inside of the working area; or

[0033] The robot is controlled to move and work through the inside of the working area according to the working map, and the robot is turned according to the working coverage of the working area after reaching the boundary of the working area; or

[0034] A working path is planned according to the working map, the robot is controlled to move and work according to the working path, and in response to reaching a preset working time, a working sub-area is determined according to the working coverage of the working area, and the robot is controlled to work in the working sub-area.

[0035] Optionally, the control of the robot to work in the working area further includes:

[0036] The attributes of each boundary segment in the boundary of the working area are determined.

[0037] A boundary path is determined according to the attributes of each boundary segment, and the boundary path includes a boundary segment of a non-preset type attribute and a path obtained by extending a boundary segment of a preset type attribute by a preset distance.

[0038] The robot is controlled to move and work along the boundary path.

[0039] Optionally, the controlling the gardening robot to work in the working area comprises:

[0040] loop detection using images captured by the camera;

[0041] if a time length from a last time of detecting a loop is greater than or equal to a preset time length, controlling the gardening robot to move along the working area boundary until detecting a loop, and then controlling the gardening robot to work in the working area.

[0042] Optionally, the preset time length is less than or equal to a maximum loop time, and the maximum loop time is determined according to an area of the working area.

[0043] Optionally, the control method further comprises:

[0044] recording a charging station position;

[0045] in a charging mode, using the positioning information of the gardening robot and the charging station position to control the gardening robot to move to the charging station position;

[0046] recognizing a charging station according to images captured by the camera and controlling the gardening robot to perform charging docking.

[0047] In a second aspect, a gardening robot is provided. The gardening robot comprises a positioning sensing system, the positioning sensing system being composed of a camera, an inertial navigation module and an odometer, and the gardening robot further comprises:

[0048] one or more processors; and

[0049] a memory associated with the one or more processors, the memory being used to store program instructions and a working map, the program instructions being read and executed by the one or more processors to perform the steps of the control method of any one of the first aspect.

[0050] Optionally, the camera is a camera facing obliquely downward.

[0051] Optionally, an included angle between a visual field center line of the camera and a horizontal direction is in a range of 3 degrees to 20 degrees.

[0052] Optionally, the gardening robot is an automatic lawn mower.

[0053] The working area comprises a lawn.

[0054] In a third aspect, a gardening robot is provided. The gardening robot comprises a positioning sensing system, a position detection system, a boundary mapping system, a navigation working system and a driver.

[0055] The positioning sensing system is composed of a camera, an inertial navigation module and an odometer;

[0056] The position detection system is connected with the camera, the inertial navigation module and the odometer respectively, and is configured to perform fusion positioning by using images collected by the camera, inertial navigation information obtained by the inertial navigation module and a moving distance obtained by the odometer, and output positioning information of the gardening robot;

[0057] The boundary mapping system is connected with the camera and the driver respectively, and is configured to send a first instruction to the driver in a mapping mode; in the process that the gardening robot moves at the boundary of the working area, images collected by the camera and positioning information output by the position detection system are obtained, and a working map corresponding to the working area is output;

[0058] The navigation working system is connected with the position detection system, and is configured to send a second instruction to the driver by using the positioning information output by the position detection system and the working map in a working mode;

[0059] The driver is configured to control the gardening robot to move at the boundary of the working area according to the first instruction, and control the gardening robot to work in the working area according to the second instruction.

[0060] The working area is located in an open space.

[0061] Optionally, the gardening robot further comprises a first memory.

[0062] The boundary mapping system is configured to write the working map into the first memory.

[0063] The navigation working system is configured to read the working map from the first memory.

[0064] Optionally, the gardening robot further comprises an obstacle avoidance detection system.

[0065] The obstacle avoidance detection system is connected with the camera, and is configured to identify an obstacle by using images collected by the camera and control the gardening robot to perform obstacle avoidance processing.

[0066] Optionally, the camera is a camera facing obliquely downward.

[0067] Optionally, an included angle between a visual field center line of the camera and a horizontal direction is in a range of 3 degrees to 20 degrees.

[0068] Optionally, the image captured by the camera carries a timestamp.

[0069] The position detection system is configured to align the image captured by the camera, the inertial navigation information obtained by the inertial navigation module and the moving distance obtained by the odometer according to the timestamp.

[0070] The aligned image and inertial navigation information are fused for positioning.

[0071] Optionally, the boundary mapping system is configured to:

[0072] Switch to the mapping mode in response to the first start of the gardening robot, entering a working area not contained in the working map or receiving a mapping instruction.

[0073] Optionally, the navigation working system is configured to switch to the working mode in response to the establishment of the working map corresponding to the working area, entering a working area already contained in the working map or receiving a working instruction.

[0074] Optionally, the working area includes a first working area, a second working area and a passageway, the start position and the end position of the passageway being located in the first working area and the second working area respectively.

[0075] Optionally, the working map corresponding to the working area includes a passageway map.

[0076] The boundary mapping system is configured to:

[0077] In response to a first instruction of the user, determine the start position of the passageway, the distance between the start position of the passageway and the boundary of the first working area being less than or equal to a first threshold value.

[0078] In response to a second control instruction of the user, control the gardening robot to move from the start position of the passageway to the end position of the passageway, wherein the passageway map is established by using the image captured by the camera during the movement.

[0079] Optionally, the gardening robot further includes an instruction receiving system.

[0080] The instruction receiving system is configured to obtain a first control instruction of the user, and output the first control instruction to the boundary mapping system, the first control instruction being used to control the gardening robot to move at the boundary of the working area.

[0081] Optionally, the gardening robot further includes a boundary recognition system.

[0082] The boundary recognition system is connected with the camera, and the boundary recognition system is configured to recognize the working area boundary by using the image collected by the camera and output information of the working area boundary to the boundary mapping system.

[0083] Optionally, the working area includes a first working area, a second working area and a passageway, and the start position and the end position of the passageway are located in the first working area and the second working area respectively.

[0084] The instruction receiving system is configured to obtain a second operation instruction of the user and output the second operation instruction to the boundary mapping system.

[0085] The boundary mapping system is configured to send a fifth instruction to the driver according to the second operation instruction to control the horticultural robot to move from the start position of the passageway to the end position, and obtain a working map of the passageway by using the image collected by the camera during the movement.

[0086] Optionally, the navigation and working system is configured to:

[0087] plan a working path according to the working map, control the horticultural robot to move and work according to the working path, and the working path includes a path through the inside of the working area; or

[0088] control the horticultural robot to move and work through the inside of the working area according to the working map, and turn according to the working coverage of the working area when reaching the working area boundary; or

[0089] plan a working path according to the working map, control the horticultural robot to move and work according to the working path, and in response to reaching a preset working time length, determine a working sub-area according to the working coverage of the working area, and control the horticultural robot to work in the working sub-area.

[0090] Optionally, the navigation and working system is further configured to:

[0091] determine the attribute of each boundary segment in the working area boundary;

[0092] determine a boundary path according to the attribute of each boundary segment, and the boundary path includes a boundary segment of a non-preset type attribute and a path obtained by extending a boundary segment of a preset type attribute by a preset distance;

[0093] control the horticultural robot to move and work along the boundary path.

[0094] Optionally, the navigation and working system is further configured to:

[0095] perform loop detection using images captured by the camera;

[0096] If the time since the last detection of a loop is greater than or equal to a preset time, the robot is controlled to move along the boundary of the work area until a loop is detected, and then the robot is controlled to work in the work area.

[0097] Optionally, the preset time is less than or equal to a maximum loop time, which is determined according to the area of the work area.

[0098] Optionally, the robot further comprises a return charging system and a second memory;

[0099] The second memory records charging station position information;

[0100] The return charging system is connected to the second memory and the driver, respectively, and is configured to read the charging station position information from the second memory in a charging mode, and to send a third instruction to the driver using the position information output by the position detection system and the charging station position information.

[0101] The driver is configured to control the robot to move to the charging station position according to the third instruction.

[0102] Optionally, the return charging system is connected to the camera and the driver, and is configured to identify a charging station according to images captured by the camera and to send a fourth instruction to the driver;

[0103] The driver is configured to control the robot to charge and dock according to the fourth instruction.

[0104] Optionally, the robot is an automatic lawn mower, and the work area comprises a lawn.

[0105] As can be seen from the above technical solutions, the present application can use only a camera, an inertial navigation module, and an odometer to achieve the establishment of a work map for a robot and the control of the robot in a work area, without the need for GPS, RTK, laser radar, and other high-cost sensors, greatly reducing the implementation cost and greatly improving the applicability. BRIEF DESCRIPTION OF DRAWINGS

[0106] The above-mentioned objects, technical solutions, and advantages of the present disclosure can be clearly obtained through the following detailed description of specific embodiments that can implement the present disclosure, in conjunction with the accompanying drawings.

[0107] The same reference numerals and symbols in the drawings and the specification are used to represent the same or equivalent elements.

[0108] Fig. 1 is a schematic structural diagram of an autonomous mobile machine according to an example embodiment of the present application;

[0109] Fig. 2 is another schematic structural diagram of a gardening robot according to an example embodiment of the present application;

[0110] Fig. 3 is a system composition structural diagram of a gardening robot according to an example embodiment of the present application;

[0111] Fig. 4 is a schematic diagram of establishing a working map according to an example embodiment of the present application;

[0112] Fig. 5 is a schematic diagram of establishing a passage map according to an example embodiment of the present application;

[0113] Fig. 6a is a schematic diagram of one working mode according to an example embodiment of the present application;

[0114] Fig. 6b is a schematic diagram of another working mode according to an example embodiment of the present application;

[0115] Fig. 6c is a schematic diagram of still another working mode according to an example embodiment of the present application;

[0116] Fig. 7 is a schematic block diagram of a gardening robot according to an example embodiment of the present application. DETAILED DESCRIPTION

[0117] In order to facilitate the understanding of the present disclosure, the present disclosure will be described more fully below with reference to the accompanying drawings. The preferred embodiments of the present disclosure are shown in the drawings. However, the present disclosure can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present disclosure more thorough and comprehensive. The embodiments provided in the specification can be combined with each other.

[0118] In the present disclosure, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting", "fixing" and the like should be understood broadly, for example, can be fixedly connected, or can be detachably connected, or can be integrated; can be mechanically connected, or can be electrically connected; can be directly connected, or can be indirectly connected through an intermediate medium, can be the internal communication of two elements or the interaction relationship between two elements, unless otherwise explicitly limited. For those of ordinary skill in the art, the specific meaning of the above terms in the present disclosure can be understood according to the specific circumstances.

[0119] The terms "first", "second", etc. are used only for descriptive purposes and are not to be construed as indicating or implying relative importance or an indicated number of technical features. Thus, features defined with "first", "second" can include at least one of the features explicitly or implicitly. In the description of the present disclosure, the meaning of "a plurality of" is at least two, for example, two, three, etc., unless otherwise explicitly and specifically limited.

[0120] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. The terminology used in the description of the present disclosure herein is only for the purpose of describing specific embodiments thereof and is not intended to be limiting of the present disclosure. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0121] Generally, the garden robot needs to work in the working area set by the user. Taking the intelligent mower as an example, for a long time, the intelligent mower sets the boundary through the cable, and the magnetic field is generated in the cable through the current, so that the intelligent mower can distinguish the inside and outside of the working area. Because the work of arranging the cable is tedious and time-consuming, in order to solve this pain point, with the technological innovation, the setting method of the working area of the intelligent mower has different attempts in different directions. For example, setting the boundary through absolute positioning, such as GPS (Global Positioning System, Global Satellite Positioning); or setting the boundary through relative positioning, such as UWB (Ultra-Wideband, Ultra-Wideband); or recognizing the boundary without positioning, such as visual recognition of the lawn boundary. However, these single ways of boundary setting or recognition have certain limitations, such as GPS, which will be affected by the environment and have different positioning accuracy, UWB, which needs to set signal stakes in a specific area in advance, and visual recognition, which strongly depends on the clarity of the lawn boundary, and it is difficult to work in multiple areas, and the charging efficiency is low. With the realization of the limitations of the above products, the current mainstream products tend to achieve through the combination of different types of sensors, such as combining GPS positioning with visual recognition, so that visual recognition can compensate for the instability of GPS positioning. However, these sensors themselves have high costs, and the processing and calculation of signals after adding sensors need to further increase the cost of chips, which will inevitably increase the production cost of the whole machine. For some consumers with small lawn area or complex lawn environment, they hope to participate in the boundary setting work of the intelligent mower as little as possible before work, and hope to obtain a better cutting effect and control the cost of consumption. Therefore, the existing intelligent mower may not be able to meet their needs well.

[0122] The embodiments provide a control method of a gardening robot and the gardening robot. The gardening robot can be an automatic mower or an automatic planter or an automatic fertilizer applicator or an automatic irrigator, etc., which can automatically complete gardening work. The gardening robot can automatically move and perform corresponding work in a specified working area, and can also return to a parking station along the boundary of the working area to park or charge.

[0123] The embodiments provide a gardening robot 10, as shown in FIG. 1 and FIG. 2, which includes a body 100, a positioning sensing system 200, a memory 400, and a controller 600.

[0124] Specifically, the gardening robot 10 includes a driving device 700 arranged on the body 100, which is used to drive the body 100 to move on a working surface according to a received driving instruction. The driving device 700 usually includes a roller and a motor for driving the roller to rotate. The roller can include a driving wheel and a driven wheel. The roller can be distributed on both sides of the body 100, and the number of rollers on each side can be one or two, etc.

[0125] The gardening robot 10 also includes a working module, which is used to perform specific work tasks. For example, the gardening robot 10 is an automatic mower, and the working module includes a grass cutting blade, a cutting motor, etc., and can also include a grass cutting height adjusting mechanism and other auxiliary components for optimizing or adjusting the grass cutting effect. For example, the gardening robot 10 is an automatic planter, and the working module includes a seed box, a conveyor belt, a vibrator, a ditching mechanism, a soil covering mechanism, etc., which are used to perform planting tasks.

[0126] The gardening robot 10 can also include an energy module, which is used to provide energy for various works of the gardening robot 10. The energy module can include a rechargeable battery and a charging connection structure, wherein the charging connection structure is usually a charging electrode sheet which can be used with a charging electrode sheet arranged on a parking station to charge the gardening robot 10.

[0127] The gardening robot 10 also includes a memory 400, which is used to store data generated by the positioning sensing system 200 or the controller 600, or to pre-store data for use by the controller 600.

[0128] In the gardening robot 10 provided by some embodiments of the present application, the positioning sensing system is composed of a camera, an IMU (inertial navigation module), and an ODO (odometer).

[0129] In addition to the above modules, the gardening robot 10 can also include a shell for accommodating and installing various modules, a control panel for user operation, etc.

[0130] The controller 600 is a core component of the gardening robot 10, used to control the automatic movement and work of the gardening robot 10, and the functions executed by the controller 600 include controlling the work module to start or stop work, controlling the driving device 700 to move, judging the power of the energy module and timely controlling the gardening robot 10 to return to the docking station for automatic docking and charging, executing corresponding programs in combination with the data of the positioning sensing system, and the like.

[0131] Referring to FIGS. 1 and 2, the positioning sensing system of the gardening robot 10 can include a camera 220 connected to the body 100, used to collect images in the forward direction of the body 100, which are at least partially images of the work surface in the forward direction. The collected images are located within the field of view range 210 of the camera 220.

[0132] In some embodiments of the present application, the camera 220 is installed at an upper position of the front of the body 100, preferably centrally arranged, with a viewing angle directed obliquely downward to facilitate the collection of images of the work surface. The size of the field of view range 210 can be adjusted according to actual needs, and the larger the field of view range 210, the more images in the forward direction of the body 100 are collected, and vice versa. The forward direction of the body 100 can be various, such as normal forward movement, backward movement, turning, and the like. In the present embodiment, the forward direction of the body 100 refers to the direction of normal forward movement, i.e. the direction of the central axis of the body 100.

[0133] To solve the technical problems of high cost and poor applicability of existing gardening robots, the present embodiment provides a control method of a gardening robot, which is configured to move and / or work in a work area. The positioning sensing system 200 of the gardening robot 10 is shown in FIG. 3, which is composed of a camera 220, an inertial navigation module 230 and an odometer 240. The gardening robot 10 further includes a position detection system 510, a boundary mapping system 520 and a navigation work system 530. The above control method mainly includes:

[0134] The images collected by the camera 220, the inertial navigation information obtained by the inertial navigation module 230 and the movement distance obtained by the odometer 240 are fused to determine the positioning information of the gardening robot 10. The processing can be performed by the above-mentioned position detection system 510.

[0135] In the mapping mode, the gardening robot 10 is controlled to move along the boundary of the work area, and the images collected by the camera 220 and the positioning information of the gardening robot 10 are used to generate a work map corresponding to the work area. The processing can be performed by the above-mentioned boundary mapping system 520.

[0136] In the working mode, the positioning information of the gardening robot 10 and the working map corresponding to the working area are utilized to control the gardening robot 10 to work in the working area; wherein, the working area is located in an open space. The processing can be performed by the navigation working system 530 described above.

[0137] The above scheme utilizes the position sensor system composed of the camera 220, the inertial navigation module 230 and the odometer 240 to realize positioning and establish the boundary of the working area of the gardening robot. Since the environment of the home lawn changes less, this positioning manner can reduce the influence of environmental factors on the positioning accuracy, thereby outputting stable positioning information and forming an accurate boundary to avoid the gardening robot 10 from going out of the boundary during the working process. Further, the gardening robot 10 described above can further include an obstacle detection system 540, which utilizes the image collected by the camera to recognize obstacles and controls the gardening robot to perform corresponding obstacle avoidance processing according to the type of the recognized obstacles.

[0138] Further, the gardening robot 10 described above can further include a regression charging system 550, which utilizes the image collected by the camera to recognize the features of the charging station, for controlling the gardening robot 10 to dock with the charging station for charging.

[0139] The above two schemes utilize the image and position detected by the position sensor system to realize the boundary recognition, obstacle avoidance and regression charging of the intelligent mower through the continuous reuse of the camera 200, on the basis of extremely simple hardware.

[0140] The above processes will be described in detail in combination with specific embodiments. First, the processing of the position detection system 510 will be described in detail.

[0141] In some embodiments of the present application, the camera 220 can adopt a camera facing obliquely downward to facilitate the collection of the image of the working surface. Wherein, “downward” refers to the direction pointing to the ground. As one of the implementable manners, the included angle between the visual center line of the camera 220 and the horizontal direction is in the range of 3 degrees to 20 degrees, so that the field of view angle of the camera 220 can cover enough environmental information to support the AI (Artificial Intelligence, artificial intelligence) algorithm (such as machine learning model) to perform target detection, and can also provide the visual features required by the VSLAM (Simultaneous Localization and Mapping, simultaneous localization and mapping) algorithm to perform positioning and map construction. In some embodiments of the present application, the camera 220 collects the image of the surrounding environment at an obliquely downward angle during the movement of the gardening robot 10.

[0142] In addition, the camera 220 needs to have a high resolution and frame rate to support the target detection, pose estimation and other tasks required by the AI algorithm. At the same time, for the VSLAM algorithm, a higher resolution can provide more visual features and map construction information, and the frame rate will affect the real-time performance and the accuracy of motion tracking.

[0143] In some embodiments of the present application, the camera 220 can adopt a monocular camera, which only needs one lens and a corresponding photosensitive element, thus having lower cost, lower power consumption and being easier to integrate into the gardening robot 10.

[0144] The inertial navigation unit (IMU) includes a gyroscope and an accelerometer. The gyroscope is mainly used to measure angular acceleration, while the accelerometer is used to measure the acceleration of the motion of an object. When given the initial conditions, the inertial navigation module can determine the current position, yaw angle (mainly representing the direction) and speed without the need for external reference, mainly representing the pose of the autonomous working robot 10.

[0145] The odometry can record and estimate the moving distance of the moving object (the gardening robot in some embodiments of the present application) in real time, and can further estimate the direction information.

[0146] As shown in FIG. 3, the position detection system 510 is connected with the camera 200, the inertial navigation module 230 and the odometry 240. The position detection system 510 fuses the image collected by the camera 220, the inertial navigation information obtained by the inertial navigation module 230 and the moving distance obtained by the odometry 240 to determine and output the positioning information of the gardening robot 10. The positioning information mainly includes position information and attitude information. As one of the implementable ways for fusion positioning, the image collected by the camera 220 can be used to calculate the motion change between adjacent image frames through feature matching and motion estimation, which involves the matching of image features and inter-frame features. Then, the continuous estimation of the attitude, speed and position of the gardening robot 10 is obtained through integral operation using the inertial navigation information, and the state update can be performed in combination with the kinematic model. The moving distance measured by the odometry is used as an observation value to correct the estimation result of the inertial navigation. The inertial navigation information and the moving distance of the odometry can be fused through a Kalman filter or other fusion algorithm, so as to obtain more accurate positioning information. In addition, after the image collected by the camera 220 is obtained, the image can be preprocessed before feature matching and motion estimation, including filtering (such as median filtering, Gaussian filtering, etc.) and distortion removal, so as to remove the noise in the image and improve the accuracy of subsequent processing.

[0147] Further, the position detection system 510 can identify the positions that have been visited through loop closure detection, and optimize the positioning accuracy in combination with VSLAM, RANSAC (Random Sample Consensus) and other algorithms.

[0148] Further, the images captured by the camera 220 can carry timestamps. Accordingly, the position detection system 510 can align the images captured by the camera 220, the inertial navigation information obtained by the inertial navigation module 230 and the moving distance obtained by the odometer 240 according to the timestamps; and perform fusion positioning using the aligned images and inertial navigation information.

[0149] The processing of the boundary mapping system 520 will be described in detail below in conjunction with embodiments.

[0150] In the mapping mode, the boundary mapping system 520 controls the gardening robot 10 to move along the boundary of the working area, and generates a working map corresponding to the working area using the images captured by the camera 220 and the positioning information of the gardening robot 10 during the movement.

[0151] As shown in FIG. 3, the gardening robot 10 further includes a driver 700. The boundary mapping system 520 is connected with the camera 220 and the driver 700 respectively. The position detection system 510 can send a first instruction to the driver 700 in the mapping mode. The images captured by the camera 220 and the positioning information of the gardening robot 10 during the movement of the gardening robot 10 along the boundary of the working area are used to output a working map corresponding to the working area.

[0152] The driver 700 controls the gardening robot 10 to move along the boundary of the working area according to the first instruction.

[0153] For the gardening robot 10, as one of the implementable manners, the mapping mode can be switched in response to the first start of the gardening robot 10, i.e., the gardening robot 10 needs to establish a map before it can work by default when it is started for the first time.

[0154] As another implementable manner, the mapping mode can be switched in response to the entry of the gardening robot 10 into a working area that is not contained in the working map. When the gardening robot 10 enters a new working area, it needs to establish a map for the new working area before working.

[0155] As a further realizable manner, the mapping mode can be switched to in response to the gardening robot 10 receiving a mapping instruction. The mapping instruction can be sent by a user through a user terminal bound to the gardening robot 10, a remote control device, a physical button on the gardening robot 10, or a virtual button on a display screen.

[0156] In some embodiments of the present application, the boundary mapping system 520 can control the gardening robot 10 to move along the boundary of the working area in the mapping mode, generate an initial map corresponding to the working area by using the images captured by the camera 220 at the boundary of the working area and the positioning information of the gardening robot 10. Then, the gardening robot 10 is controlled to move according to the initial map, and the initial map is updated to obtain a working map by using the images captured at positions on or within a preset distance range from the boundary of the working area during the movement.

[0157] In some embodiments of the present application, the working area is located in an open space. The open space refers to a space without a roof and not completely surrounded by physical boundaries, which is mainly used to distinguish indoor spaces and outdoor spaces completely surrounded by physical boundaries. The physical boundaries refer to walls, shrubs, fences, etc.

[0158] The type of the working area can be determined according to the type of the gardening robot 10. For example, when the gardening robot 10 is an automatic mower, the working area is actually a lawn (which can also be referred to as a grass field, etc.), and the lawn usually does not have physical boundaries or has partial physical boundaries.

[0159] In some embodiments of the present application, the boundary of the working area is mainly used to distinguish the working area and the non-working area. In some embodiments of the present application, the boundary of the working area can include one or more of the peripheral boundary of the working area, the boundary of a passageway in the working area, the boundary of an obstacle in the working area, etc. Taking an automatic mower as an example, the boundary of the working area is mainly used to distinguish grass and non-grass.

[0160] In the process of generating the initial map, as one of the realizable manners, the gardening robot 10 can be controlled to move along the boundary of the working area in the mapping mode in response to a first control instruction of the user. That is, the gardening robot 10 is controlled to move along the boundary of the working area under the control of the user.

[0161] As shown in FIG. 3, in some embodiments of the present application, the gardening robot 10 can further include an instruction receiving system 500. The instruction receiving system 500 obtains a first control instruction of the user and outputs the first control instruction to the boundary mapping system 520. The first control instruction is used to control the gardening robot 10 to move along the boundary of the working area.

[0162] As another implementable manner, the image captured by the camera can be used to identify the boundary of the working area, and the identified result is used to control the movement of the gardening robot along the boundary of the working area. The boundary of the working area can be automatically identified by using AI (Artificial Intelligence).

[0163] As shown in FIG. 3, in some embodiments of the present application, the gardening robot 10 can further include a boundary identification system 560. The boundary identification system 560 is connected with the camera 220, and uses the image captured by the camera 220 to identify the boundary of the working area, and outputs the information of the boundary of the working area to the boundary mapping system 520.

[0164] The above-mentioned process of generating the initial map can include: obtaining the image captured by the camera 220 at the boundary of the working area and the positioning information of the gardening robot 10, extracting the visual feature from the image captured by the camera 220 at the boundary of the working area, establishing the correspondence between the visual feature and the positioning information, and thus generating the initial map.

[0165] The visual feature extracted from the image in some embodiments of the present application can include but is not limited to one or any combination of the following: feature points, descriptor information of the feature points, position information of the feature points in the image, three-dimensional coordinate information of the feature points, edges, textures, histograms, etc.

[0166] Further, in the above-mentioned process of generating the initial map, an image can also be captured at the mapping starting position on the boundary, which represents the image corresponding to the surrounding environment of the mapping starting position; and the above-mentioned process of generating the initial map is ended in response to the gardening robot 10 returning to the mapping starting position. Based on the image captured by the camera 220 in real time and the image captured at the mapping starting position, if the visual loop can be successfully performed, it indicates that the gardening robot 10 returns to the mapping starting position along the boundary of the working area, forming a closed boundary track, and the mapping is ended.

[0167] The image is captured when the mapping is started at the mapping starting position, so that more visual features near the mapping starting position are obtained, which can improve the success rate of visual loop when returning to the mapping starting position along the boundary, and improve the mapping accuracy.

[0168] Of course, in addition to the above-mentioned manner of generating the initial map, the initial map can also be received from the server or the user end, or the initial map is downloaded according to the map download instruction received from the server or the user end, etc.

[0169] In the process of generating the initial map, the gardening robot 10 moves along the boundary smoothly, and the time taken to generate the initial map is short, and the mapping is completed quickly. Also due to the short time, the number of images in the initial map is small, and the corresponding visual features are also small. Therefore, on the basis of the initial map, the gardening robot 10 can be controlled to move according to the initial map, and in the process of moving, images collected at positions on or within a preset distance range from the boundary of the working area are used to update the initial map to obtain a working map.

[0170] As one of the implementable modes, in the process of controlling the gardening robot 10 to move according to the initial map, the shooting angle of the images collected by the camera 220 can be different from the shooting angle of the images collected in the process of generating the initial map, so as to obtain more other visual features, so as to make the obtained working map more abundant.

[0171] For the gardening robot 10, the information related to the boundary on the map is crucial, and the accuracy of the boundary-related information will affect the positioning accuracy, movement accuracy and movement safety of the gardening robot 10. The initial map can have the problem of less or inaccurate visual features, so images need to be collected while the gardening robot 10 moves according to the initial map subsequently, so as to supplement the visual features of the environment around the boundary to the initial map according to the collected images, that is, to perform feature supplement collection, so that the visual features of the initial map are more abundant, and the map accuracy is better, which can improve the positioning accuracy and movement accuracy of the gardening robot 10 when moving / working based on the VSLAM technology and the first map, avoid out-of-bound, and improve safety.

[0172] After obtaining the initial map, the gardening robot 10 moves along the boundary according to the initial map, and the feature supplement collection is performed while the gardening robot 10 moves along the boundary, so that the feature supplement collection of the boundary can be more targeted, and the supplement collection efficiency is high. In some embodiments, in response to the gardening robot 10 returning to the starting position of the movement, or the distance between the machine position of the gardening robot 10 and the starting position being less than a distance threshold, the feature supplement collection is ended.

[0173] In some embodiments, the moving direction of the gardening robot 10 in the process of generating the initial map and the feature supplement collection is the same.

[0174] Illustratively, the gardening robot 10 moves along the boundary in a counterclockwise direction in the process of generating the initial map, and the gardening robot 10 also moves along the boundary in a counterclockwise direction in the process of feature supplement collection.

[0175] In some embodiments, the moving direction of the gardening robot 10 in the process of generating the initial map and the feature supplement collection is opposite.

[0176] Exemplarily, the machine moves at least one circle along the boundary in a counterclockwise direction in the initial map generation process, and moves at least one circle along the boundary in a clockwise direction in the feature supplement process; or the machine moves at least one circle along the boundary in a clockwise direction in the initial map generation process, and moves at least one circle along the boundary in a counterclockwise direction in the feature supplement process.

[0177] As shown in FIG. 4, after the generation of the initial map is completed, the automatic mower obtains the initial map of the boundary of the working area A. The automatic mower starts from the boundary position a of the working area A, moves along the boundary, and performs feature supplement every 2 meters during the movement until it returns to the boundary position a again. In FIG. 4, the arrows on the boundary represent the movement path corresponding to the feature supplement, and the circles on the boundary represent the position of the feature supplement, i.e., the target position.

[0178] In the process of generating the initial map and the feature supplement, the camera 220 can collect images in the following ways, but is not limited thereto:

[0179] The first way is to control the horticultural robot 10 to rotate at the target position, and during the rotation, at least two images of different shooting angles are collected based on the camera 220 of the horticultural robot 10.

[0180] The adjacent target positions can be spaced apart by a preset distance, the collection time instants of the adjacent target positions can be spaced apart by a preset time length, and the like.

[0181] For example, during the movement of the automatic mower along the boundary, the camera remains stationary while the automatic mower rotates at the current position after moving a certain distance, so as to realize image collection in multiple directions at the position.

[0182] The second way is to control the camera 220 to rotate relative to the body of the horticultural robot 10, and during the rotation of the camera 220, at least two images of different shooting angles are collected based on the camera of the horticultural robot 10.

[0183] For example, the camera is controlled to rotate after the automatic mower moves a certain distance, and the automatic mower remains stationary, so as to realize image collection in multiple directions.

[0184] The third way is to control the panoramic camera of the horticultural robot 10 to collect a panoramic image, and the first updated image is the panoramic image.

[0185] For example, a panoramic camera is provided on the automatic mower, and the panoramic camera collects a panoramic image after the automatic mower moves a certain distance, so as to realize image collection in multiple directions.

[0186] In addition to the above three ways, other ways can also be used to realize, which are not listed here.

[0187] In some scenarios, it is necessary to switch between multiple work areas, i.e., to move from one work area to another work area, and therefore it is necessary to establish a passage map between the work areas. For example, the above work areas include a first work area, a second work area, and a passage, the start position and the end position of the passage being located in the first work area and the second work area, respectively.

[0188] As shown in FIG. 5, when generating the passage map, the start position of the passage can be determined in response to a first passage instruction of the user, the distance between the start position of the passage and the boundary of the first work area being less than or equal to a first threshold value; the movement of the gardening robot 10 from the start position of the passage to the end position of the passage can be controlled in response to a second operation instruction of the user, and the passage map can be established by using the images captured by the camera 220 during the movement.

[0189] As shown in FIG. 3, in some embodiments of the present application, the gardening robot 10 can further include an instruction receiving system 500. The instruction receiving system 500 can obtain the second operation instruction of the user and output the second operation instruction to the boundary mapping system 520.

[0190] The boundary mapping system 520 can send a fifth instruction to the driver 700 according to the second operation instruction to control the movement of the gardening robot 10 from the start position of the passage to the end position of the passage, and the work map of the passage can be obtained by using the images captured by the camera 220 during the movement.

[0191] Since the work map of the first work area is usually established by controlling the gardening robot 10 to capture images at the boundary of the first work area, and the actual work of the gardening robot 10 is usually performed after the work in the first work area is completed and then the gardening robot 10 is switched to another work area, i.e., the second work area, through the passage. Therefore, the start position of the passage should usually be set near the boundary of the first work area to achieve fast switching. Further, the start position of the passage is thus set to ensure that the gardening robot 10 can be repositioned at this position.

[0192] It should be noted that the meaning of "repositioning" involved in some embodiments of the present application at least includes satisfying the visual loop condition, and can further include being able to determine the current pose of the gardening robot 10 according to the visual features of the images. The visual loop condition refers to the fact that the gardening robot 10 recognizes that it has arrived at this position, so that the map is closed. When determining whether a position satisfies the visual loop condition, images can be captured at the position, and the captured images are matched with the images at the position in the work map of the first work area (e.g., the similarity is greater than or equal to a preset similarity threshold value). If the matching is successful, it is considered that the visual loop condition is satisfied, wherein the visual features extracted from the images are mainly used in the matching.

[0193] Therefore, in the embodiments of the present disclosure, the distance between the determined passage starting position and the boundary of the first working area is less than or equal to a first threshold value. The first threshold value can adopt an empirical value or an experimental value, for example, 2 meters.

[0194] As one of the implementable manners, the garden robot 10 can be controlled to collect images near the boundary of the first working area, and if the confidence of the images is greater than or equal to a confidence threshold value, it is considered that the distance between the position and the boundary of the first working area is less than or equal to the first threshold value.

[0195] For example, the user can control the garden robot 10 to move near the boundary of the first working area, and the garden robot 10 collects images during the movement and continuously judges the confidence of the images. If the confidence is greater than or equal to the confidence threshold value, a prompt information can be issued. The prompt information can be issued by the garden robot 10 through its own screen, loudspeaker, etc., or can be sent to the user terminal and displayed on the screen or loudspeaker, etc. of the user terminal.

[0196] For example, the user can control the garden robot 10 to move near the boundary of the first working area, and the garden robot 10 collects images during the movement and continuously judges the confidence of the images. If the confidence is greater than or equal to the confidence threshold value, a prompt information can be issued. The prompt information can be issued by the garden robot 10 through its own screen, loudspeaker, etc., or can be sent to the user terminal and displayed on the screen or loudspeaker, etc. of the user terminal.

[0197] The movement of the garden robot 10 between the first working area and the second working area is usually a non-working area between the two working areas, and cannot be self-moved by using the boundary of the working area. Therefore, as one of the implementable manners, after the garden robot 10 reaches the passage starting position, the user can send a control instruction to the garden robot 10 through a remote controller, a button or a controller on the garden robot 10, a user terminal bound with the garden robot 10, etc. to control the moving direction of the garden robot 10, or the user can control the moving direction of the garden robot 10 by means of a detachable handrail mounted on the garden robot 10 or riding the garden robot 10, etc. so that the garden robot 10 moves between the first working area and the second working area from the passage starting position.

[0198] In the process of moving of the gardening robot 10 from the passage starting position, the real-time position and the moving path of the gardening robot 10 are acquired and sent to the user terminal bound with the gardening robot 10, so as to display the real-time position and the moving path in the mapping interface of the user terminal.

[0199] The passage ending position is located in the second working area. As one of the implementable manners, the user can send a second passage instruction to the gardening robot 10 when the gardening robot 10 moves to the second working area. In response to the second passage instruction, the current position of the gardening robot 10 is determined as the passage ending position.

[0200] As one of the implementable manners, the user can send the above-mentioned second passage instruction through the user terminal bound with the gardening robot 10, and the second passage instruction is sent to the gardening robot 10 via the server end.

[0201] In addition, in addition to determining whether the gardening robot 10 reaches the second working area by the user, it can also be automatically recognized by the gardening robot 10 through the image collected by the camera 220 whether it reaches the second working area, and a prompt can be issued after recognizing that it reaches the second working area. The prompt can be presented by the gardening robot 10, for example, by the screen display of the gardening robot 10, the flash of the gardening robot 10, the sound prompt of the gardening robot 10, etc. The prompt can be sent to the user terminal via the server end and presented on the user terminal, which can be prompted by text, animation, sound, etc. After the user obtains the prompt, the sending of the above-mentioned second passage instruction is triggered.

[0202] In the above-mentioned moving process, the gardening robot 10 is controlled to collect images at each target position. The target position can include the passage starting position and the passage ending position. That is, the gardening robot 10 is controlled to collect images in different directions at the passage starting position and to collect images in different directions at the passage ending position.

[0203] Similar to the process of establishing the working map of the first working area, in the process of establishing the passage map, the gardening robot 10 also acquires the positioning information of the gardening robot 10 when collecting images. The visual features are extracted from the images, and then the positioning information (i.e. the pose) and the visual features can obtain the passage map. The passage map can actually be regarded as the correspondence between the pose of the gardening robot 10 and the visual features.

[0204] If the passage is short, images can be captured at the passage start position and the passage end position respectively to establish the passage map. But if the passage is long, a second image can be further captured at a preset time interval or a preset distance (e.g. every 3 meters) between the passage start position and the passage end position until the passage end position is reached.

[0205] Further, after the passage end position is determined, the passage map is obtained. The work map establishment of the second work area can be continued. The horticultural robot 10 can be controlled to move to the start position of the second work area; the horticultural robot 10 can be controlled to move along the boundary from the boundary start position of the second work area to the boundary end position of the second work area, and images can be captured at each target position during the movement, the distance between the boundary start position and the boundary end position being less than or equal to the third threshold value. It can be regarded as moving along the boundary of the second work area for one round (i.e. the boundary start position and the boundary end position are very close), and images are captured during the movement to establish the map of the second work area.

[0206] The processing of the navigation work system 530 will be described in detail below in combination with an embodiment.

[0207] In the working mode, the navigation work system 530 controls the horticultural robot 10 to work in the work area by using the positioning information of the horticultural robot 10 and the work map corresponding to the work area.

[0208] As shown in FIG. 3, the navigation work system 530 is connected with the driver 700. In the working mode, the navigation work system 530 sends a second instruction to the driver 700 by using the positioning information of the horticultural robot 10 and the work map corresponding to the work area.

[0209] The driver 700 controls the horticultural robot 10 to work in the work area according to the second instruction.

[0210] Further, the memory 400 in the horticultural robot 10 can include a first memory 410. The boundary mapping system 520 can write the work map into the first memory 410, so that the navigation work system 530 reads the work map from the first memory 410.

[0211] The work mode can include but is not limited to the following modes:

[0212] The first mode: a work path is planned according to the work map, and the horticultural robot 10 is controlled to move and work according to the work path, wherein the work path includes a path through the inside of the work area.

[0213] This way is based on the planning of the work mode, for example, the automatic mower, as shown in FIG. 6a, the automatic mower plans the working path based on the working area A corresponding to the working map, the working path forms an "arch", so the "arch cutting" mode is adopted to carry out the cutting operation in the working area A, and the cutting path is represented by the arrowed line segment in the figure. For example, after moving along the boundary for 0.4 meters, turn to the direction inside the working area and cut through the inside of the working area, after reaching the opposite boundary, turn to move along the boundary for 0.4 meters, continue to turn to the direction inside the working area and cut through the inside of the working area, and so on, so as to complete the mowing work of the working area. Wherein, the coordinate information and visual features of the boundary position in the working map of the working area A are used to determine whether the boundary is reached and moved along the boundary during the working process.

[0214] The second way: according to the working map, control the garden robot 10 to move through the inside of the working area and work, wherein after reaching the boundary of the working area, turn according to the working coverage condition of the working area.

[0215] This way adopts a semi-random working mode, which determines the turning according to the actual cutting state. For example, the automatic mower, as shown in FIG. 6b, after the automatic mower reaches the position b on the boundary of the working area A, moves along the boundary for 0.4 meters to reach the position c, and turns to the direction of the area with poor working coverage according to the working coverage condition of the working area, for example, the right side of the working area A has lower working coverage rate, so it turns to the right side of the direction and moves through the inside of the working area and works.

[0216] The third way: according to the working map to plan the working path, control the garden robot 10 to move and work according to the working path, in response to reaching the preset working time, determine the working sub-area according to the working coverage condition of the working area, and control the garden robot 10 to work in the working sub-area.

[0217] This way also adopts a semi-random working mode, that is, first plan the working path, move and work according to the working path for a period of time, then determine a sub-area according to the working coverage condition of the working area, for example, a sub-area with lower working coverage rate, and control the garden robot 10 to work in the sub-area.

[0218] For example, as shown in FIG. 6c, the automatic mower plans a working path based on the working map corresponding to the working area A, the working path forms an "arch", and thus the automatic mower performs cutting work in the working area A in an "arch cutting" manner. It is assumed in the figure that the cutting path is represented by a line segment with an arrow from position h to position j after the working time is preset. Then, according to the working coverage of the working area, a sub-area A1 with low working coverage is determined. The automatic mower is controlled to move along the edge to position k, and then performs arch cutting in the sub-area A1.

[0219] After completing the work in the sub-area, a further sub-area can be determined according to the working coverage of the working area, and the garden robot 10 is controlled to work in the sub-area. Alternatively, after completing the work in the sub-area, the garden robot 10 can be controlled to work based on the planned working path according to the remaining area, i.e., the area that has not been worked.

[0220] During the working process of the above-mentioned modes, due to the structural problem of the garden robot 10, the garden robot 10 can not be able to cut the working area boundary cleanly, for example, there can be a distance between the cutter head of the automatic mower and the frontmost side of the machine body, which can cause the automatic mower to be unable to cut the working area boundary cleanly without going out of the boundary. Therefore, a further edge working mode can be further adopted.

[0221] The above-mentioned edge working mode can include: determining the attribute of each boundary segment in the working area boundary; determining a boundary path according to the attribute of each boundary segment, wherein the boundary path includes: a boundary segment with a non-preset type attribute and a path obtained by extending a boundary segment with a preset type attribute by a preset distance; and controlling the garden robot 10 to move along the boundary path and work.

[0222] The attribute of each boundary segment in the working boundary can be defined by the user when establishing the working map, or can be defined by the user after establishing the working map. For example, the user can define the attribute of each boundary segment in the working area boundary on the mapping interface presented by the client on the user terminal. Alternatively, the attribute of each boundary segment in the working area boundary can be determined by performing semantic recognition on the image captured by the camera through an AI algorithm during the establishment of the working map.

[0223] As one of the implementable manners, the attribute can include a safe boundary segment and a non-safe boundary segment. The outside of the safe boundary segment can be a cobblestone road, a lawn, etc. These safe boundary segments mean that the gardening robot is not a problem even if it moves out of the boundary, so the safe boundary segment can be expanded outward by a preset distance to obtain a boundary path. The outside of the non-safe boundary segment can be a pond, mud bath, water pit, etc. These non-safe boundary segments mean that the gardening robot 10 can cause damage or affect normal work if it moves out of the boundary, so the non-safe boundary segment is taken as the boundary path and cannot be expanded outward.

[0224] When controlling the gardening robot 10 to work in the working area, loop detection can be performed by using the image collected by the camera; if the time interval from the last time loop detection is greater than or equal to a preset time interval, the gardening robot 10 is controlled to move along the boundary of the working area until the loop is detected, and then the gardening robot 10 is controlled to work in the working area. That is, if the gardening robot 10 cannot successfully detect the loop for a long time, it means that the positioning of the gardening robot 10 is lost, and since the boundary position in the working map corresponds to an image feature, it is beneficial for the gardening robot 10 to re-detect the loop, so the gardening robot 10 can be controlled to move along the boundary of the working area for a period of time first, and then continue to work in the working area after detecting the loop.

[0225] The above-mentioned preset time interval is less than or equal to a maximum loop time, which is determined according to the area of the working area. This is because if the area of the working area is larger, the gardening robot 10 will move inside the working area for a longer time, i.e., it will take longer to return to the boundary, so the time interval for detecting the loop will also be longer.

[0226] The processing of the obstacle detection system 540 will be described below.

[0227] As shown in FIG. 3, the obstacle detection system 540 is connected with the camera 220. The obstacle detection system 540 performs obstacle recognition by using the image collected by the camera, and controls the gardening robot 10 to perform corresponding obstacle avoidance processing according to the type of the recognized obstacle.

[0228] In some embodiments of the present application, an AI algorithm can be used to identify the type of the obstacle in the image collected by the camera, for example, a deep learning model (such as a YOLO model) is used to extract features from the image collected by the camera, and the extracted features are used to identify the category of the obstacle. The specific manner is not limited in the present application.

[0229] For example, for the obstacle type of small area such as fallen leaves, mud, and stone pavement, the obstacle avoidance process can not be performed. For the obstacle type of static obstacles such as stones and road signs, the obstacle avoidance process of bypassing the obstacles can be adopted. For the obstacle type of dynamic obstacles such as animals, humans, and other robots, the working can be paused and the working can be continued after the dynamic obstacles are removed. Other obstacle avoidance processes can also be adopted, and some embodiments of the present application do not exhaustively list all the obstacle avoidance processes.

[0230] The process of the return charging system 550 is described in detail below in combination with an embodiment.

[0231] As one of the implementable modes, the return charging system 550 can record the charging station position; in the charging mode, the positioning information of the gardening robot 10 and the charging station position are used to control the gardening robot 10 to move to the charging station position. Then, the charging station is identified according to the image collected by the camera 220 and the gardening robot 10 is controlled to perform the charging docking.

[0232] As shown in FIG. 3, the above-mentioned memory 400 can include a second memory 420.

[0233] The second memory 420 records the charging station position information.

[0234] The return charging system 550 is connected with the second memory 420 and the driver 700, and in the charging mode, the charging station position information is read from the second memory 420, the positioning information output by the position detection system 510 and the charging station position information are used to send a third instruction to the driver 700.

[0235] The driver 700 controls the gardening robot 10 to move to the charging station position according to the third instruction.

[0236] The interior of the gardening robot 10 is usually managed by a battery management system (not shown in the figure), which can monitor the remaining power of the battery in real time. When the power is lower than a preset threshold, the system triggers the decision of returning to charging, so as to enter the charging mode. In some embodiments, when the gardening robot 10 monitors that the remaining power of the battery is lower than the preset threshold, the current working state and the surrounding environment can be evaluated, so as to determine whether to immediately stop the current work and enter the charging mode, or to enter the charging mode after completing all or part of the current work.

[0237] After the return charging system 550 enters the charging mode, the positioning information of the gardening robot 10 and the charging station position can be used for path planning to control the gardening robot 10 to move to the charging station position. In the process of moving to the charging station position, the obstacle avoidance detection system 540 also identifies the obstacles and performs the corresponding obstacle avoidance process.

[0238] In some embodiments of the present application, the charging station is arranged within a certain range of the boundary of the working area, so that the gardening robot 10 can still have a high positioning accuracy at the charging station position.

[0239] When the gardening robot 10 approaches the charging station, precise docking can be performed by using the image collected by the camera 220. The so-called precise docking includes: identifying the charging station marker, using the positioning information of the gardening robot 10 and the charging station marker, and adjusting the pose of the gardening robot 10 by the VSLAM algorithm to ensure that the gardening robot 10 can be docked with the charging station (which can be a charging seat, a charging pile, etc.). The charging station marker can include but is not limited to a two-dimensional code, a special pattern or a special marker, etc.

[0240] As shown in FIG. 3, the return charging system 550 can also be connected with the camera 220, identify the charging station according to the image collected by the camera 220, and send a fourth instruction to the driver 700. The driver 700 controls the gardening robot 10 to perform charging docking according to the fourth instruction.

[0241] In the case of successful docking, full battery, charging abnormality, etc., the return charging system 550 can also prompt by voice, change of indicator light, sending a message to the user terminal bound with the gardening robot 10, etc.

[0242] The above embodiments only express several implementation manners of the present disclosure, and the description is relatively specific and detailed, but it cannot be understood as a limitation on the scope of the patent of the present disclosure. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present disclosure, several modifications and improvements can be made, which are within the protection scope of the present disclosure.

[0243] Based on the same inventive concept as the foregoing embodiments, the gardening robot 10 provided by the embodiments of the present disclosure, as shown in FIG. 7, includes a processor 700 and a memory 400 storing a computer program; wherein the controller 600 shown in FIG. 2 can be one or more processing units contained in the processor 700; similarly, the memory 400 shown in FIG. 2 also has the same meaning, that is, only used to indicate the positional relationship of the memory 400 relative to other devices, and in actual application, the number of the memory 400 can be one or more. When the processor 700 runs the above computer program, the control method of the above embodiments is realized.

[0244] In some embodiments of the present application, the position detection system 510, the boundary mapping system 520, the navigation working system 530, the obstacle avoidance detection system 540 and the return charging system 550 can be programs running in the processor 700, or can be control logic, control circuit or control chip contained in the controller 600.

[0245] The position detection system 510, the boundary mapping system 520, the navigation working system 530, the obstacle avoidance detection system 540 and the returning charging system 550 can be independently arranged, for example, respectively implemented as programs executed by different processing units in the processor 700, or respectively implemented as different control logics, different control circuits or different control level chips in the controller 600. The position detection system 510, the boundary mapping system 520, the navigation working system 530, the obstacle avoidance detection system 540 and the returning charging system 550 can also be arranged as programs executed by the same processing unit, or respectively implemented as the same control circuit or the same control level chip in the controller 600.

[0246] The horticultural robot 10 can further include an input device 710 and an output device 720. The processor 700, the memory 400, the input device 710 and the output device 720 can be connected by a bus or other means, and in FIG. 7, an example of connection by a bus is shown.

[0247] The input device 710 can receive inputted digital or character information, and generate signal input related to user settings and function control of the horticultural robot 10, such as a touch screen, a joystick, etc. The output device 720 can include a display device, an auxiliary lighting device (for example, an LED), a tactile feedback device (for example, a vibration motor), etc. In some embodiments, the display device can be a touch screen.

[0248] The various components in the horticultural robot 10 are coupled together by a bus. It can be understood that the bus is used to realize the connection communication between the components. In addition to including a data bus, the bus also includes a power supply bus, a control bus and a state signal bus.

[0249] The memory 400 can be a volatile memory or a non-volatile memory, and can include both a volatile memory and a non-volatile memory. The non-volatile memory can be a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a ferromagnetic random access memory (FRAM), a Flash Memory, a magnetic surface memory, an optical disc, or a Compact Disc Read-Only Memory (CD-ROM). The magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a Random Access Memory (RAM) used as an external cache. By way of example and not limitation, many forms of RAM can be used, such as a Static Random Access Memory (SRAM), a Synchronous Static Random Access Memory (SSRAM), a Dynamic Random Access Memory (DRAM), a Synchronous Dynamic Random Access Memory (SDRAM), a Double Data Rate Synchronous Dynamic Random Access Memory (DDR SDRAM), an Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), a SyncLink Dynamic Random Access Memory (SLDRAM), a Direct Rambus Random Access Memory (DRRAM).The memory 400 described in some embodiments of the present application is intended to include, but is not limited to, these and any other suitable type of memory.

[0250] The memory 400 in some embodiments of the present application is used to store various types of data to support the operation of the gardening robot 10. Examples of these data include: any computer programs used to operate on the gardening robot 10, such as operating systems and application programs; contact data; phonebook data; messages; pictures; videos; and the like. Among them, the operating system contains various system programs, such as framework layer, core library layer, driver layer, etc., for implementing various basic services and processing hardware-based tasks. The application program can contain various application programs, such as media player (Media Player), browser (Browser), etc., for implementing various application services. Here, the program for implementing the method of some embodiments of the present application can be contained in the application program.

[0251] Based on the same inventive concept as the foregoing embodiments, the present embodiment also provides a computer storage medium having a computer program stored therein, which can be a ferromagnetic random access memory (FRAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM), etc. The computer storage medium can also be various devices including one or any combination of the above memories, such as a mobile phone, a computer, a tablet device, a personal digital assistant, etc. The computer program stored in the computer storage medium is run by the processor to implement the control method applied to the gardening robot 10 described above. The specific step flow implemented by the computer program executed by the processor is described in the method embodiment and will not be described here.

[0252] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combinations of the technical features do not contradict, they should be considered as the scope of the present disclosure.

[0253] In this document, the terms "comprise", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0254] The above description is only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present disclosure, which should be covered by the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

1. A control method of a gardening robot, wherein, The positioning sensor system of the gardening robot is composed of a camera, an inertial navigation module and an odometer; the control method comprises: fusion positioning is performed by using the image collected by the camera, the inertial navigation information obtained by the inertial navigation module and the moving distance obtained by the odometer to determine the positioning information of the gardening robot; in the mapping mode, the gardening robot is controlled to move along the boundary of the working area, and in the process of the movement, the image collected by the camera and the positioning information of the gardening robot are used to generate a working map corresponding to the working area; in the working mode, the positioning information of the gardening robot and the working map corresponding to the working area are used to control the gardening robot to work in the working area; wherein, the working area is located in an open space.

2. The control method according to claim 1, further comprising: obstacle recognition is performed by using the image collected by the camera, and according to the type of the recognized obstacle, the gardening robot is controlled to perform corresponding obstacle avoidance processing.

3. The control method according to claim 1 or 2, wherein, The camera is a camera facing obliquely downward.

4. The control method according to any one of claims 1 to 3, wherein The image collected by the camera carries a time stamp; the fusion positioning by using the image collected by the camera, the inertial navigation information obtained by the inertial navigation module and the moving distance obtained by the odometer comprises: the image collected by the camera, the inertial navigation information obtained by the inertial navigation module and the moving distance obtained by the odometer are aligned according to the time stamp; fusion positioning is performed by using the aligned image and inertial navigation information.

5. The control method according to any one of claims 1-4, further comprising: in response to the first start of the gardening robot, entering a working area not contained in the working map or receiving a mapping instruction, switching to the mapping mode; or, in response to the completion of the establishment of the working map corresponding to the working area, entering a working area already contained in the working map or receiving a working instruction, switching to the working mode.

6. The control method according to any one of claims 1-5, wherein, in the mapping mode, the gardening robot is controlled to move along the boundary of the working area, and in the process of the movement, the image collected by the camera and the positioning information of the gardening robot are used to generate a working map corresponding to the working area; in the mapping mode, the gardening robot is controlled to move along the boundary of the working area, and the image collected by the camera along the boundary of the working area and the positioning information of the gardening robot are used to generate an initial map corresponding to the working area; according to the initial map, the gardening robot is controlled to move, and in the process of the movement, the image collected at a position on or within a preset distance range from the boundary of the working area is used to update the initial map to obtain the working map.

7. The control method according to any one of claims 1-6, wherein, in the mapping mode, the gardening robot is controlled to move along the boundary of the working area, and in the process of the movement, the image collected by the camera and the positioning information of the gardening robot are used to generate a working map corresponding to the working area; in the mapping mode, the gardening robot is controlled to move along the boundary of the working area in response to a first control instruction of a user; and / or, the image collected by the camera is used to recognize the boundary of the working area, and the gardening robot is controlled to move along the boundary of the working area according to the recognition result.

8. The control method according to any one of claims 1-7, wherein, The working area includes a first working area, a second working area, and a passageway, starting and ending positions of the passageway being located in the first working area and the second working area respectively.

9. The control method according to claim 8, wherein The working map corresponding to the working area includes a passageway map; The generation of the passageway map includes: in response to a first instruction of a user, determining the starting position of the passageway, the distance between the starting position of the passageway and the boundary of the first working area being less than or equal to a first threshold value; in response to a second instruction of the user, controlling the horticultural robot to move from the starting position of the passageway to the ending position of the passageway, wherein the passageway map is established by using the images captured by the camera during the movement.

10. The control method according to any one of claims 1-9, wherein, Controlling the horticultural robot to work in the working area includes: planning a working path according to the working map, and controlling the horticultural robot to move and work according to the working path, the working path including a path passing through the inside of the working area; or, controlling the horticultural robot to move and work in the working area according to the working map, wherein the horticultural robot turns according to the working coverage of the working area after reaching the boundary of the working area; or, planning a working path according to the working map, and controlling the horticultural robot to move and work according to the working path, in response to reaching a preset working time, determining a working sub-area according to the working coverage of the working area, and controlling the horticultural robot to work in the working sub-area.

11. The control method according to any one of claims 1-10, wherein, Controlling the horticultural robot to work in the working area further includes: determining the properties of each boundary segment in the boundary of the working area; determining a boundary path according to the properties of each boundary segment, the boundary path including a boundary segment of a non-predefined type of property and a path obtained by extending a boundary segment of a predefined type of property by a preset distance; and controlling the horticultural robot to move and work along the boundary path.

12. The control method according to any one of claims 1-11, wherein, Controlling the horticultural robot to work in the working area includes: performing loop detection by using the images captured by the camera; if the time since the last detection of a loop is greater than or equal to a preset time, controlling the horticultural robot to move along the boundary of the working area until a loop is detected, and then controlling the horticultural robot to work in the working area.

13. The control method according to claim 12, wherein The preset time is less than or equal to a maximum loop time, and the maximum loop time is determined according to the area of the working area.

14. The control method according to any one of claims 1-13, wherein, The control method further includes: recording the position of a charging station; in a charging mode, controlling the horticultural robot to move to the position of the charging station by using the positioning information of the horticultural robot and the position of the charging station; identifying the charging station by using the images captured by the camera and controlling the horticultural robot to perform charging docking.

15. A horticultural robot, comprising a positioning sensing system composed of a camera, an inertial navigation module, and an odometer, the horticultural robot further comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions and a working map, the program instructions being read and executed by the one or more processors to perform the steps of the control method of any one of claims 1-14.

16. The horticulture robot according to claim 15, wherein, The camera is a camera facing obliquely downward.

17. The horticulture robot according to claim 15 or 16, wherein, An included angle between a visual field center line of the camera and a horizontal direction is in a range of 3 degrees to 20 degrees.

18. The horticulture robot according to any one of the claims 15-17, wherein, The gardening robot is an automatic mower. The working area includes a lawn.

Images