Robot and map building and positioning method thereof

CN119072669APending Publication Date: 2024-12-03SZ SENHE TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380012059.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-03-31
Filing Date
2023-09-28
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

Existing robot positioning methods rely on expensive GPS, RTK or beacon technology, and have high signal interference and environmental dependence, or the accuracy of the three-dimensional environment information obtained using cameras is not high, resulting in low accuracy of map construction, especially if they are not trained. Poor performance in scenarios.

Method used

Using a robot design that includes motion sensors, vision sensors and lidar, by acquiring motion parameters, front images and object information, a point cloud map is constructed to determine the position of the robot and the location of obstacles, achieving accurate three-dimensional map construction and path planning.

Benefits of technology

It improves the accuracy of the robot's position and obstacle information, achieves precise path planning and obstacle avoidance capabilities, reduces dependence on expensive equipment, and maintains efficient positioning performance in a variety of environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119072669A_ABST
    Figure CN119072669A_ABST
Patent Text Reader

Abstract

The invention provides a robot and a map building and positioning method thereof. The robot comprises a motion sensor, a visual sensor, a laser radar and a processor, the motion sensor is used for acquiring motion parameters of the robot, the visual sensor is used for acquiring images in front of the robot, and the laser radar is used for acquiring information of objects in front of the robot. The information of the object in front of the robot comprises the distance and the angle of the object relative to the robot, and the processor is used for determining the pose information of the robot according to the motion parameters acquired by the motion sensor and the images acquired by the visual sensor; a point cloud map is constructed at least according to the pose information and information, obtained by the laser radar, of an object in front of the robot, and the position of the robot and the position of an obstacle in front of the robot are determined at least according to the point cloud map. According to the robot provided by the invention, an accurate three-dimensional map can be constructed, and accurate robot position information and obstacle information can be acquired.
Need to check novelty before this filing date? Find Prior Art

Description

Robot and map construction and positioning method thereof

[0001] This application claims priority to the Chinese patent application filed with the China Patent Office on March 31, 2023, with application number 202310379347.0 and application name “Robot and its map construction and positioning method”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of robotics, and in particular to a robot and a method for mapping and positioning the robot. Background Art

[0003] Currently, robots generally perceive the environment through GPS, RTK, beacon technology, or use cameras to obtain environmental information, and then locate and build maps based on this environmental information. However, GPS, RTK, or beacon technology is expensive and highly dependent on GPS signals. Interference with the GPS signal will greatly reduce positioning reliability. Regarding beacon technology, the environment must be arranged in advance, and interference with beacon signals will affect positioning reliability. The three-dimensional environmental information obtained from images taken by the camera is not very accurate, and the constructed map is not very precise. It also needs to rely on big data for operation and performs poorly in untrained scenarios.

[0004] Summary of the Invention

[0005] In order to solve the above technical problems, the present application provides a robot and its map construction and positioning method, which can construct an accurate three-dimensional map, which is conducive to the robot obtaining accurate robot position information and obstacle information, and then performing accurate path planning.

[0006] On the one hand, the present application provides a robot, which includes a motion sensor, a visual sensor, a laser radar and a processor, wherein the motion sensor is used to obtain motion parameters of the robot, and the motion parameters include acceleration and angular velocity; the visual sensor is used to obtain an image in front of the robot; the laser radar is used to obtain information about objects in front of the robot, and the information about objects in front of the robot includes the distance and angle of the objects relative to the robot; the processor is used to determine the posture information of the robot based on the motion parameters obtained by the motion sensor and the image obtained by the visual sensor, and to construct a point cloud map based on at least the posture information and the information about the object in front of the robot obtained by the laser radar, and to determine the position of the robot and the position of obstacles in front of the robot based on at least the point cloud map.

[0007] The robot provided in this application obtains the distance and angle of the object in front relative to the robot through a laser radar, and obtains the robot's posture information through a visual sensor and a motion sensor. Based on the obtained distance and angle of the object in front relative to the robot and the robot's posture information, an accurate three-dimensional map can be constructed, which is beneficial for the robot to obtain accurate robot position information and obstacle information, and then perform accurate path planning and obstacle avoidance.

[0008] On the other hand, the present application provides a map construction and positioning method, which is applied to a robot, and the method includes: obtaining motion parameters of the robot, the motion parameters including acceleration and angular velocity; obtaining an image in front of the robot; obtaining information about objects in front of the robot, the information about objects in front of the robot including the distance and angle of the objects relative to the robot; determining the posture information of the robot based on the obtained motion parameters and the obtained image; constructing a point cloud map based on at least the posture information and the obtained information about the objects in front of the robot; and determining the position of the robot and the position of obstacles in front of the robot based on at least the point cloud map.

[0009] The map construction and positioning method provided in this application can construct an accurate three-dimensional map based on the distance and angle of the front object relative to the robot and the robot's posture information, which is beneficial for the robot to obtain accurate robot position information and obstacle information, and then perform accurate path planning and obstacle avoidance. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] FIG1 is a structural block diagram of a robot provided in an embodiment of the present application;

[0011] FIG2 is a schematic structural diagram of a robot provided in one embodiment of the present application;

[0012] FIG3 is a top view of the robot shown in FIG2 ;

[0013] FIG4 is a schematic structural diagram of a robot provided by another embodiment of the present application;

[0014] FIG5 is a top view of the robot shown in FIG4 ;

[0015] FIG6 is a side view of the robot shown in FIG4 ;

[0016] FIG7 is a flowchart of a map construction and positioning method provided in one embodiment of the present application;

[0017] FIG8 is a flowchart of a map construction and positioning method provided by another embodiment of the present application.

[0018] Description of reference numerals:

[0019] 100-robot; 10-motion sensor; 20-visual sensor; 30-laser radar; 31-first laser radar; 32-second laser radar; 81-first plane; 82-second plane; 91-first intersection; 92-second intersection; 101-first side; 102-second side; 83-third plane; 84-fourth plane; 93-third intersection; 40-processor; 50-cutting component; 60-millimeter wave radar; 70-ultrasonic radar. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] In the description of this application, the terms "first", "second", etc. are used to distinguish different objects rather than to describe a specific order. In addition, the terms "upper", "lower", "inner", "outer", etc. indicate directions or positional relationships based on the directions or positional relationships shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction. Therefore, they cannot be understood as limitations on this application.

[0022] In the description of this application, unless otherwise specified or limited, the term "connection" should be understood in a broad sense. For example, it can mean a fixed connection, a detachable connection, or an integral connection; it can mean a direct connection, an indirect connection through an intermediate medium, or internal communication between two components; it can mean a communication connection; or it can mean an electrical connection. Those skilled in the art will understand the specific meanings of the above terms in this application based on the specific circumstances.

[0023] Please refer to Figure 1, which is a block diagram of the structure of a robot 100 provided in an embodiment of the present application. As shown in Figure 1, the robot 100 includes a motion sensor 10, a visual sensor 20, a laser radar 30, and a processor 40. The motion sensor 10 is used to obtain motion parameters of the robot 100, and the motion parameters include acceleration and angular velocity. The visual sensor 20 is used to obtain an image in front of the robot 100. The laser radar 30 is used to obtain information about objects in front of the robot 100, and the information about objects in front of the robot 100 includes the distance and angle of the objects relative to the robot 100. The processor 40 is used to determine the posture information of the robot 100 based on the motion parameters obtained by the motion sensor 10 and the image obtained by the visual sensor 20, and to construct a point cloud map based on at least the posture information and the information about objects in front of the robot 100 obtained by the laser radar 30, and to determine the position of the robot 100 and the position of obstacles in front of the robot 100 based on at least the point cloud map.

[0024] The robot 100 provided in the embodiment of the present application obtains the distance and angle of the front object relative to the robot 100 through the laser radar 30, and obtains the posture information of the robot 100 through the visual sensor 20 and the motion sensor 10. According to the obtained distance and angle of the front object relative to the robot 100 and the posture information of the robot 100, an accurate three-dimensional map can be constructed, which is beneficial for the robot 100 to obtain accurate robot 100 position information and obstacle information, and then perform accurate path planning and obstacle avoidance.

[0025] Among them, the posture information of the robot 100 includes the mileage and direction angle of the current position of the robot 100 relative to the position at the previous moment, that is, the path length from the position of the robot 100 at the previous moment to the current moment and the change value of the direction angle at the current moment relative to the direction angle at the previous moment.

[0026] Among them, the visual sensor 20 sends the image in front of the robot 100 obtained at the current moment to the processor 40, and the motion sensor 10 sends the motion parameters of the robot 100 obtained at the current moment to the processor 40. When the processor 40 receives the image in front of the robot 100 at the current moment and the motion parameters of the robot 100 at the current moment, it can perform feature point processing on the image, match the feature points according to the motion parameters to obtain visual mileage data, and use the Kalman filter method to process the visual mileage data to obtain the posture information of the robot 100 at the current moment.

[0027] Among them, the information about objects in front of the robot 100 obtained by the laser radar 30 may include ground information and obstacle information in front, that is, the distance and angle of each position of the ground in front relative to the robot 100 and the distance and angle of the obstacle in front relative to the robot 100.

[0028] Among them, the laser radar 30 sends the information of the object in front of the robot 100 to the processor 40. When the processor 40 receives the information, it obtains a laser point cloud based on the information, and combines the laser point cloud and the current posture information of the robot 100 to construct a point cloud map at the current moment. The obtained point cloud map is a three-dimensional map.

[0029] In which, the robot 100 may include a memory (not shown in the figure), and the memory may store the global map of the environment in which the robot 100 is located that was constructed by the robot 100 last time. The processor 40 is also used to match and compare the aforementioned point cloud map at the current moment and the global map constructed last time, so as to determine the current position of the robot 100 and the information of the obstacles in front of the robot 100 at the current moment. The information of the obstacles in front of the robot 100 may at least include the position and size of the obstacle.

[0030] The motion sensor 10 may be an inertial measurement unit (IMU), which may include a gyroscope, an accelerometer, a magnetometer, or the like. The visual sensor 20 may be a camera, such as a binocular camera or a monocular camera. The lidar 30 may be a single-line lidar or a multi-line lidar. The processor 40 may include a processing chip such as a single-chip microcomputer, a microcontroller, or a digital signal processor. The memory may be a non-volatile memory. The robot 100 may be a sweeping robot, a lawn mowing robot, a delivery robot, a disinfection robot, or the like.

[0031] In some embodiments, the image acquired by the visual sensor 20 includes an RGB image, and the processor 40 is used to construct the point cloud map based on at least the posture information, the RGB image and the information of the object in front of the robot acquired by the lidar 30, and perform semantic recognition on the point cloud map to obtain semantic information of the obstacle in front of the robot 100.

[0032] Among them, the point cloud map is constructed according to the RGB image, so that the point cloud map includes color information of the ground and obstacles in front of the robot 100, and then the point cloud map can be semantically recognized, so as to identify the types of various obstacles in front of the robot 100, for example, the obstacles are trees, walls, stones, etc., so that accurate information of various obstacles in front can be obtained, which is conducive to path planning and obstacle avoidance.

[0033] In some embodiments, the laser radar 30 is used to obtain the laser reflectivity of the ground in front of the robot 100; the processor 40 is further used to determine whether the ground in front of the robot 100 includes a lawn area based on the laser reflectivity obtained by the laser radar 30 and / or the image obtained by the visual sensor 20, and to determine the boundary of the lawn area if the ground in front of the robot 100 includes a lawn area. The processor 40 is also used to construct the point cloud map based on at least the posture information, the information of the object in front of the robot 100 obtained by the laser radar 30, and the determined boundary of the lawn area, and to determine the position of the robot 100 and the position of the obstacle in front of the robot 100 based on the point cloud map.

[0034] Among them, the robot 100 can be a lawn mowing robot, as shown in Figure 1, which includes a cutting component 50, and the cutting component 50 is used for mowing grass. The processor 40 is also used to control the robot 100 to move in the lawn area and along the boundary line of the lawn area and control the cutting component 50 to mow grass while the robot 100 is moving, thereby avoiding mowing non-lawn areas, improving mowing efficiency, and extending the service life of the cutting component 50.

[0035] Among them, determining the boundary line of the lawn area and constructing the point cloud map at least based on the boundary line is beneficial for the robot 100 to plan the driving path when mowing according to the specific position of the lawn area in the constructed point cloud map, and then mow the lawn area in a targeted manner, thereby improving mowing efficiency.

[0036] In some embodiments, the processor 40 determines whether the ground in front of the robot 100 includes a lawn area based on the laser reflectivity obtained by the laser radar 30, and determines the boundary of the lawn area if the ground in front includes a lawn area. Specifically, the processor 40 is configured to determine that the ground in front of the robot 100 includes a lawn area and determine the boundary of the lawn area when the laser reflectivity obtained by the laser radar 30 is within a preset laser reflectivity range.

[0037] The preset laser reflectivity range may be stored in the memory.

[0038] Among them, the laser radar 30 includes a laser transmitter, a laser receiver and a processing unit. The laser transmitter is used to transmit a laser in front of the robot 100. The transmitted laser is reflected on the object in front, and the reflected laser is received by the laser receiver. The processing unit calculates the laser reflectivity according to the laser intensity of the transmitted laser and the laser intensity of the reflected laser, and sends the calculated laser reflectivity to the processor 40. When receiving the laser reflectivity, the processor 40 determines whether the laser reflectivity is within a preset laser reflectivity range. When the laser reflectivity is within the preset laser reflectivity range, it determines that the object in front of the robot 100 is a blade of grass. When the laser reflectivity is outside the preset laser reflectivity range, it determines that the object in front of the robot 100 is not a blade of grass, thereby determining the lawn area and the boundary line of the lawn area.

[0039] In some embodiments, the processor 40 determines whether the ground in front of the robot 100 includes a lawn area based on the image acquired by the visual sensor 20 and determines a boundary line of the lawn area when the ground in front includes a lawn area.

[0040] Among them, the image acquired by the visual sensor 20 includes an RGB image, and the processor 40 can perform semantic recognition on the front image according to the RGB image acquired by the visual sensor 20 to determine whether the ground in front of the robot 100 includes a lawn area and determine the boundary line of the lawn area when the ground in front includes a lawn area.

[0041] In some embodiments, the processor 40 combines the laser reflectivity obtained by the lidar 30 and the image obtained by the visual sensor 20 to determine whether the ground in front of the robot 100 includes a lawn area and determines the boundary line of the lawn area when the ground in front includes a lawn area.

[0042] In some embodiments, the detection range of the laser radar 30 and the detection range of the visual sensor 20 may partially overlap or not overlap. By combining the laser reflectivity and the image to determine the boundary line of the lawn area, a more complete and accurate boundary line of the lawn area can be obtained.

[0043] Among them, the processor 40 can determine the first boundary line of the lawn area based on the laser reflectivity obtained by the laser radar 30, and determine the second boundary line of the lawn area based on the image obtained by the visual sensor 20, and compare or merge the first boundary line and the second boundary line to obtain a more accurate boundary line of the lawn area.

[0044] In some embodiments, the laser radar 30 is at least disposed on the top of the robot 100, and the laser radar 30 is used to at least obtain information about objects in front of the robot 100. By disposing the laser radar 30 on the top of the robot 100, the laser radar 30 can be prevented from being blocked, thereby preventing the laser emitted by the laser radar 30 from being blocked, which would affect the acquisition of information about objects in front of the robot 100.

[0045] In some embodiments, the visual sensor 20 is disposed at least at the front end of the robot 100. The visual sensor 20 is used to capture images directly in front of the robot 100. The front end is the frontmost portion of the robot 100 in its normal travel direction. By locating the visual sensor 20 at the front end of the robot 100, images directly in front of the robot can be captured, including images of the ground and obstacles in front of the robot.

[0046] The laser radar 30 may be disposed on the top of the robot 100 and close to the front end, so that the detection range of the laser radar 30 is as far as possible in front of the robot 100 .

[0047] Among them, by arranging the laser radar 30 on the top of the robot 100 and the visual sensor 20 at the front end of the robot 100, that is, arranging the laser radar 30 and the visual sensor 20 at different positions of the robot 100, the robot 100's perception range of environmental information can be increased, and more comprehensive environmental information can be obtained.

[0048] In some embodiments, the laser radar 30 and the visual sensor 20 may also be arranged on the side, front end and tail of the robot 100, and the tail is the tail end of the robot 100 in the direction of travel when it is driving normally.

[0049] Please refer to Figures 2 and 3. Figure 2 is a schematic structural diagram of a robot 100 provided in an embodiment of the present application, and Figure 3 is a top view of the robot 100 shown in Figure 2. In some embodiments, as shown in Figures 2 and 3, the robot 100 includes at least a first laser radar 31 and a second laser radar 32. The first laser radar 31 and the second laser radar 32 are respectively arranged on opposite sides of the top of the robot 100 along a preset direction (direction A as shown in Figure 3). The laser emitted by the first laser radar 31 is located on a first plane 81, and the laser emitted by the second laser radar 32 is located on a second plane 82. The first intersection line 91 of the first plane 81 and the ground and the second intersection line 92 of the second plane 82 and the ground are both arranged at an angle to the direction of travel of the robot 100 during normal driving. The preset direction is perpendicular to the direction of travel of the robot 100 during normal driving (direction B as shown in Figure 3). The lasers emitted by the first laser radar 31 and the second laser radar 32 will not be blocked by the robot 100.

[0050] The first laser radar 31 and the second laser radar 32 are respectively arranged on a first side 101 and a second side 102 along a preset direction of the top of the robot 100 , and the first side 101 and the second side 102 are opposite to each other.

[0051] The angle may be greater than 0° and less than or equal to 90°.

[0052] The first laser radar 31 and the second laser radar 32 may both be single-line laser radars. The first laser radar 31 includes a first laser emitter, and the second laser radar 32 includes a second laser emitter. The first laser emitter can rotate around a first preset center line and emit a single-line laser during rotation. The plane where the single-line laser emitted by the first laser emitter when rotating around the first preset center line is located is the first plane 81. The first preset center line is perpendicular to the first plane 81. The intersection of the first plane 81 and the ground is the first intersection line 91. The first intersection line 91 is set at an angle to the direction of travel. The angle between the first intersection line 91 and the direction of travel is shown as α in Figure 3. The second laser emitter can rotate around the second preset center line and emit a single-line laser during rotation. The plane where the single-line laser emitted by the second laser emitter when rotating around the second preset center line is the second plane 82. The second preset center line is perpendicular to the second plane 82. The intersection of the second plane 82 and the ground is the second intersection line 92. The second intersection line 92 is set at an angle to the direction of travel. The angle between the second intersection line 92 and the direction of travel is shown as β in Figure 3. The first preset center line extends outward from the first side 101 in a direction away from the robot 100 and is set at an angle to the top plane of the robot 100. The second preset center line extends outward from the second side 102 in a direction away from the robot 100 and is set at an angle to the top plane of the robot 100.

[0053] The first plane 81 and the ground have a first intersection line 91, meaning the first plane 81 is set at an angle to the ground. The second plane 82 and the ground have a second intersection line 92, meaning the second plane 82 is set at an angle to the ground. By setting the first plane 81 and the second plane 82 at an angle to the ground, the first laser radar 31 and the second laser radar 32 can acquire information about the ground in front of the robot 100, which helps the first laser radar 31 and the second laser radar 32 acquire information about the lawn area in front of the robot 100.

[0054] Among them, the intersection of the first plane 81 and the ground or obstacle is the first scanning line of the first laser radar 31, and the first laser radar 31 is used to obtain information about the part of the front object corresponding to the first scanning line, that is, to obtain information about the part of the front object intersecting with the first plane 81, and the information includes the distance and angle of the part of the front object corresponding to the first scanning line relative to the robot 100; the intersection of the second plane 82 and the ground or obstacle is the second scanning line of the second laser radar 32, and the second laser radar 32 is used to obtain information about the part of the front object corresponding to the second scanning line, that is, to obtain information about the part of the front object intersecting with the second plane 82, and the information includes the distance and angle of the part of the front object corresponding to the second scanning line relative to the robot 100.

[0055] Among them, by arranging the first laser radar 31 and the second laser radar 32 to be located on opposite sides of the top of the robot 100 along a preset direction, the detection range of the robot 100 to the front environment can be increased, and the first plane 81 where the laser emitted by the first laser radar 31 is located and the first intersection line 91 of the ground and the second plane 82 where the laser emitted by the second laser radar 32 is located and the second intersection line 92 of the ground are both set at an angle to the moving direction of the robot 100, that is, the first plane 81 is set at an angle to the moving direction, and the second plane 82 is set at an angle to the moving direction, so that the first laser radar 31 and the second laser radar 32 can not only detect the environment directly in front of the robot 100 but also detect the front area outside the front, such as information about objects in front of the left and right of the robot 100, which can further increase the detection range of the robot 100 to the front environment, thereby obtaining more environmental information.

[0056] In some embodiments, the first plane 81 intersects with the second plane 82, and the intersection line of the first plane 81 and the second plane 82 is close to the front end of the robot 100, so that the detection range of the first laser radar 31 and the second laser radar 32 is as far as possible in front of the robot 100.

[0057] In some embodiments, the first laser radar 31 and the second laser radar 32 are symmetrically arranged on opposite sides of the top of the robot 100, and the first plane 81 intersects the second plane 82, so that the first laser radar 31 at least obtains information about objects directly in front of, to the left of, or to the right of the robot 100 in the forward direction, and the second laser radar 32 at least obtains information about objects directly in front of, to the right of, or to the left of the robot 100 in the forward direction. The first plane 81 and the second plane 82 can be symmetrically arranged relative to a reference plane, which is perpendicular to the robot 100, passes through the center of the robot 100, and is parallel to the forward direction. The forward direction is the forward direction.

[0058] Among them, when the first laser radar 31 is arranged on the top left side of the robot 100 and the second laser radar 32 is arranged on the top right side of the robot 100, the first laser radar 31 is used to at least obtain information about objects directly in front and in front of the left in the direction of movement of the robot 100, and obtain information about objects on the side of the second side 102 of the robot 100, and the second laser radar 32 is used to at least obtain information about objects directly in front and in front of the right in the direction of movement of the robot 100, and obtain information about objects on the side of the first side 101 of the robot 100; when the second laser radar 32 is arranged on the top left side of the robot 100 and the first laser radar 31 is arranged on the top right side of the robot 100, the second laser radar 32 is used to at least obtain information about objects directly in front and in front of the left in the direction of movement of the robot 100, and obtain information about objects on the side of the first side 101 of the robot 100, and the first laser radar 31 is used to at least obtain information about objects directly in front and in front of the right in the direction of movement of the robot 100, and obtain information about objects on the side of the second side 102 of the robot 100.

[0059] In some embodiments, the angle between the first intersection line 91 and the direction of travel of the robot during normal driving, and the angle between the second intersection line 92 and the direction of travel of the robot during normal driving, are both values ​​between 10° and 80°, that is, α and β are both values ​​between 10° and 80°, so that the detection range of the first laser radar 31 and the second laser radar 32 is large, and they can detect the environment in front of the robot 100. When the angle between the first intersection line 91 and the direction of travel of the robot during normal driving is less than 10°, the detection range of the first laser radar 31 is small, and the detection range of the first laser radar 31 is mainly the environment on the side of the first side 101 of the robot 100. When the angle between the first intersection line 91 and the direction of travel of the robot during normal driving is greater than 80°, the detection range of the first laser radar 31 is mainly the environment on the side of the second side 102 of the robot 100, which is not conducive to obtaining information about objects in front of the robot 100. When the angle between the second intersection line 92 and the direction of travel of the robot during normal travel is outside the range of 10°-80°, similarly, no further details will be given.

[0060] Among them, when the angle is within the above-mentioned angle range, the larger the angle between the first intersection line 91 and the direction of travel of the robot during normal driving, the larger the angle between the second intersection line 92 and the direction of travel of the robot during normal driving, and the larger the detection range of the first laser radar 31 and the second laser radar 32.

[0061] In some embodiments, the angle between the first intersection line 91 and the direction of travel of the robot during normal driving and the angle between the second intersection line 92 and the direction of travel of the robot during normal driving may be equal.

[0062] Please refer to Figures 4 to 6. Figure 4 is a structural schematic diagram of a robot 100 provided in another embodiment of the present application, Figure 5 is a top view of the robot 100 shown in Figure 4, and Figure 6 is a side view of the robot 100 shown in Figure 4. In some embodiments, as shown in Figure 4, the first laser radar 31 is arranged at the top of the robot 100, and the second laser radar 32 is arranged at the front end of the robot 100. As shown in Figures 4 to 6, the laser emitted by the first laser radar 31 is located in the third plane 83, and the laser emitted by the second laser radar 32 is located in the fourth plane 84. The third plane 83 is parallel to the ground, and the fourth plane 84 intersects with the ground. The third intersection line 93 of the fourth plane 84 and the ground is at an angle to the forward direction of the robot 100 during normal driving (direction B as shown in Figure 4). The first laser radar 31 and the second laser radar 32 can be single-line laser radars.

[0063] Among them, the first laser emitter can rotate 360° around a third preset center line and emit a single-line laser during rotation. The third preset center line is perpendicular to the top of the robot 100 and perpendicular to the ground. The plane where the single-line laser emitted by the first laser emitter when rotating around the third preset center line is located is the third plane 83, and the third plane 83 is parallel to the ground, as shown in Figures 4 to 6.

[0064] The second laser emitter can rotate 360° around a fourth preset center line and emit a single-line laser during rotation. There is an acute angle between the fourth preset center line and the direction of travel of the robot 100 during normal driving, and the fourth preset center line is set at an angle to the front end surface of the robot 100. The plane where the single-line laser emitted by the second laser emitter when rotating around the fourth preset center line is located is the fourth plane 84, and the intersection of the fourth plane 84 and the ground is the third intersection line 93. The third intersection line 93 is set at an angle to the direction of travel, and the angle between the third intersection line 93 and the direction of travel is shown as γ in Figure 5.

[0065] The processor 40 is used to determine the posture information of the robot 100 based on the motion parameters obtained by the motion sensor 10 and the image obtained by the visual sensor 20, and to construct a point cloud map based on at least the posture information, the information of objects around the robot 100 obtained by the first laser radar 31, and the information of objects close to the ground in front of the robot 100 obtained by the second laser radar 32, and to determine the position of the robot 100 and the positions of obstacles around the robot 100 based on at least the point cloud map.

[0066] By setting the third plane 83 on which the laser emitted by the first laser radar 31 is located to be parallel to the ground, the first laser radar 31 can detect information about objects around the robot 100, thereby increasing the detection range of the first laser radar 31 and obtaining more environmental information. By setting the fourth plane 84 on which the laser emitted by the second laser radar 32 is located to intersect with the ground, the second laser radar 32 can at least detect information about objects close to the ground in front of the robot 100, including information about the lawn area on the ground in front. This is conducive to constructing a more accurate point cloud map that includes lawn area information, and is further conducive to obtaining the positioning information, obstacle information and lawn area information of the robot 100.

[0067] Among them, the intersection of the third plane 83 and the objects around the robot 100 is the third scanning line of the first laser radar 31, and the first laser radar 31 is used to obtain information about the parts of the objects around the robot 100 that intersect with the third plane 83, and the information includes the distance and angle of the parts of the objects around the robot 100 corresponding to the third scanning line relative to the robot 100; the intersection of the fourth plane 84 and the ground or objects close to the ground is the fourth scanning line of the second laser radar 32, and the second laser radar 32 is used to obtain information about the parts of the objects on the ground or close to the ground that intersect with the fourth plane 84, and the information includes the distance and angle of the parts of the objects on the ground or close to the ground corresponding to the second scanning line relative to the robot 100.

[0068] In some embodiments, the angle between the third intersection line 93 and the forward direction of the robot 100 during normal travel is a value between 45° and 135°, that is, γ is a value between 45° and 135°.

[0069] By setting γ to a value between 45° and 135°, it is possible to avoid the angle between the third intersection line 93 and the forward direction of the robot 100 during normal driving being too large or too small. If the angle between the third intersection line 93 and the forward direction of the robot 100 during normal driving is too large, the detection range of the second laser radar 32 will mainly be the area directly in front and to the right of the front, and there will be more blind spots on the left side of the front, which is not conducive to the robot 100 avoiding obstacles in the left front. If the angle between the third intersection line 93 and the forward direction of the robot 100 during normal driving is too small, the detection range of the second laser radar 32 will mainly be the area directly in front and to the left of the front, and there will be more blind spots on the right side of the front, which is not conducive to the robot 100 avoiding obstacles in the right front.

[0070] In some embodiments, the angle between the fourth plane 84 and the ground is between 10° and 40°. The angle between the fourth plane 84 and the ground is shown as θ in FIG6 , where θ is between 10° and 40°.

[0071] By setting θ to a value between 10° and 40°, it is beneficial for the second laser radar 32 to detect information about the ground or objects close to the ground, and to avoid the angle between the fourth plane 84 and the ground being too large or too small. If the angle between the fourth plane 84 and the ground is too large, the object detected by the second laser radar 32 will be close to the robot 100, which is not conducive to obstacle avoidance; if the angle between the fourth plane 84 and the ground is too small, the detection object of the second laser radar 32 will be close to the detection object of the first laser radar 31, which is not conducive to the second laser radar 32 detecting information about the ground and objects close to the ground, nor is it conducive to detecting information about the lawn area.

[0072] In some embodiments, γ is 90° and θ is 20°.

[0073] In some embodiments, the scanning angle of the second laser radar 32 is a value between 180° and 360°.

[0074] By setting the scanning angle of the second laser radar 32 to a value between 180° and 360°, the range in front of the robot 100 detected by the second laser radar 32 can be larger, and more comprehensive information about objects on the ground or close to the ground in front of the robot 100 can be detected, which is conducive to constructing a point cloud map and then obtaining the position information and obstacle information of the robot 100.

[0075] In some embodiments, the scanning angle of the second laser radar 32 is 360° (as shown in Figures 4 and 5). The position of the second laser radar 32 on the front end of the robot 100 can be set so that the fourth plane 84 where the laser emitted by the second laser radar 32 is located does not intersect with the robot 100, that is, the laser emitted by the second laser radar 32 is not blocked by the robot 100, thereby making the scanning angle of the second laser radar 32 360°.

[0076] In some embodiments, the scanning angle of the second laser radar 32 is greater than or equal to 180° and less than 360°. The position of the second laser radar 32 on the front end of the robot 100 can be set so that the part of the fourth plane 84 where the laser emitted by the second laser radar 32 is located away from the ground intersects with the robot 100, that is, the laser emitted by the second laser radar 32 is blocked by the robot 100 when it rotates away from the ground, thereby making the scanning angle of the second laser radar 32 greater than or equal to 180° and less than 360°.

[0077] In some embodiments, the robot 100 further includes a millimeter-wave radar 60, which is configured to obtain at least information about objects obscured by blades of grass in the lawn area. The processor 40 is configured to construct the point cloud map based on at least the pose information, information about objects in front of the robot 100 obtained by the lidar 30, and information about objects obscured by blades of grass obtained by the millimeter-wave radar 60.

[0078] The millimeter wave radar 60 obtains information about objects blocked by grass leaves in the lawn area, so that the robot 100 can avoid objects blocked by grass leaves when mowing the lawn area.

[0079] In some embodiments, the processor 40 is also used to determine whether the object obscured by the grass leaves is an obstacle based on the information of the object obscured by the grass leaves obtained by the millimeter wave radar 60, and when it is determined that the object obscured by the grass leaves is an obstacle, control the robot 100 to avoid the position of the object obscured by the grass leaves, or, when the robot 100 drives to a position close to the object obscured by the grass leaves, control the cutting component 50 to be raised until the distance between the bottom of the cutting component 50 and the ground is greater than the height of the object obscured by the grass leaves, and control the cutting component 50 to mow the grass.

[0080] The information of the object blocked by the grass includes the distance and angle of the object blocked by the grass relative to the robot 100 and the size of the object blocked by the grass.

[0081] The millimeter-wave radar 60 transmits a millimeter-wave signal toward the lawn area. The millimeter-wave signal penetrates the grass blades and is projected onto the object blocked by the grass blades, and generates an echo signal on the object blocked by the grass blades. The echo signal penetrates the grass blades and is received by the millimeter-wave radar 60. The millimeter-wave radar 60 images the object blocked by the grass blades based on the transmitted millimeter-wave signal and the received echo signal, and determines the distance and angle of the object blocked by the grass blades relative to the robot 100 and the size of the object blocked by the grass blades.

[0082] The robot 100 may be a lawn mowing robot, and the cutting assembly 50 of the robot 100 may include a rotary cutter module and / or a rolling cutter module. When the robot 100 travels in a lawn area, the cutting assembly 50 is used to cut grass blades.

[0083] Among them, the processor 40 is used to receive information about objects obscured by grass leaves in the lawn area obtained by the millimeter-wave radar 60, and when receiving the information about objects obscured by grass leaves in the lawn area, compare the size of the object obscured by grass leaves with a preset size, and when the size of the object obscured by grass leaves is greater than or equal to the preset size, determine that the object obscured by grass leaves is an obstacle; otherwise, determine that the object obscured by grass leaves is not an obstacle.

[0084] Among them, the size of the object obscured by the grass leaves may be the outer contour size of the object obscured by the grass leaves, which may specifically include height and width, and the preset size may include a preset height and a preset width. When the height of the object obscured by the grass leaves is greater than or equal to the preset height and / or the width is greater than or equal to the preset width, the processor 40 determines that the object obscured by the grass leaves is an obstacle; when the height of the object obscured by the grass leaves is less than the preset height and the width is less than the preset width, the processor determines that the object obscured by the grass leaves is not an obstacle.

[0085] The preset width can be set according to the actual size of the robot 100 to prevent the robot 100 from being stuck by obstacles or hindering the movement of the robot 100. For example, the preset width can be the distance between the two front wheels of the robot 100. The preset height can be the shortest distance between the bottom of the robot 100 and the ground.

[0086] In some embodiments, the height of the object obscured by the grass leaves is greater than or equal to the preset height and / or the width is greater than or equal to the preset width. The processor 40 determines that the object obscured by the grass leaves is an obstacle and controls the robot 100 to avoid the object obscured by the grass leaves during driving.

[0087] In other embodiments, when the cutting assembly 50 is cutting normally, the distance between the bottom of the cutting assembly 50 and the ground is equal to the shortest distance between the robot and the ground. The preset height is the distance between the bottom of the cutting assembly 50 and the ground when the cutting assembly 50 is cutting normally. The distance between the cutting assembly 50 and the ground when cutting normally can be set according to the actual design requirements of the robot 100. The height of the cutting assembly 50 is adjustable, and the processor 40 can control the adjustment of the height of the bottom of the cutting assembly 50 to be greater than the distance between the bottom of the cutting assembly 50 and the ground when the cutting assembly 50 is cutting normally. The processor 40 is further configured to, when the width of the object blocked by the grass is less than the preset width, determine whether the height of the object blocked by the grass is greater than the distance between the bottom of the cutting assembly 50 and the ground when the cutting assembly 50 is cutting normally, and, when the height of the object blocked by the grass is greater than the distance between the bottom of the cutting assembly 50 and the ground when the cutting assembly 50 is cutting normally and less than the distance between the bottom of the cutting assembly 50 and the ground when the cutting assembly 50 is adjusted to the highest position, control the cutting assembly 50 to be raised until the distance between the bottom of the cutting assembly 50 and the ground is greater than the height of the object blocked by the grass. The height of the object blocked by the grass leaves allows the cutting component 50 to cut the grass leaves located above the object blocked by the grass leaves, thereby not only cutting off the grass leaves located above the object blocked by the grass leaves to achieve a trimming effect, but also avoiding the object blocked by the grass leaves from affecting the cutting of the cutting component 50, for example, the object blocked by the grass leaves is drawn into the cutting component 50 and damages the internal structure of the cutting component 50, or the object blocked by the grass leaves is stuck between the blades of the cutting component 50 and damages the blades, thereby protecting the cutting component 50 and extending its service life.

[0088] Among them, when the width of the object blocked by the grass leaves is smaller than the preset width, and the height of the object blocked by the grass leaves is greater than or equal to the distance between the bottom of the cutting component 50 and the ground when the cutting component 50 is adjusted to the highest position, the robot 100 is controlled to avoid the position of the object blocked by the grass leaves.

[0089] In other embodiments, when the millimeter-wave radar 60 determines that there is an object obscured by grass leaves in the lawn area, it sends information about the existence of the object obscured by grass leaves to the processor 40. When the processor 40 receives the information about the existence of the object obscured by grass leaves, it controls the robot 100 to avoid the location of the object obscured by grass leaves.

[0090] In some embodiments, when the processor 40 determines that the object obscured by the grass is not an obstacle, the processor 40 controls the robot 100 to travel normally in the lawn area without avoiding the object obscured by the grass.

[0091] In some embodiments, the millimeter-wave radar 60 may be located in front of the cutting assembly 50 in the direction of travel. When the robot 100 is normally traveling and mowing in a lawn area, the millimeter-wave radar 60 obtains information about objects obscured by grass leaves in a partial area in front of the robot 100 in the lawn area and sends the information to the processor 40. When the processor 40 receives the information about objects obscured by grass leaves in a partial area in front of the robot 100 in the lawn area, it determines whether the obstacles obscured by grass leaves in the partial area in front of the robot 100 in the lawn area are obstacles. When it is determined that the objects obscured by grass leaves are obstacles, the processor 40 controls the robot 100 to travel in a manner that avoids the position of the objects obscured by grass leaves. Alternatively, when the robot 100 travels to a position close to the objects obscured by grass leaves, the processor controls the cutting assembly 50 to be raised until the distance between the bottom of the cutting assembly 50 and the ground is greater than the height of the objects obscured by grass leaves.

[0092] In some embodiments, the cutting assembly 50 includes a roller module, which includes a roller, a roller blade and a fixed blade. The roller blade extends along the circumference of the roller and is fixed to the outer circumference of the roller. The fixed blade is spaced apart from the roller blade to form a shear gap. The processor 40 can control the rotation of the roller to drive the roller blade to rotate along the circumference of the roller. The rotating roller blade draws the grass blades between the roller blade and the fixed blade, and cuts the drawn grass blades by forming a shear gap with the fixed blade. The distance between the bottom of the cutting assembly 50 and the ground can be the distance between the roller blade and the ground when it rotates to the lowest point.

[0093] In some embodiments, the cutting assembly 50 includes a rotary blade module, which includes a rotating shaft and multiple rotary blades. The multiple rotary blades are fixed to the outer periphery of the rotating shaft. The multiple rotary blades are in the same plane and parallel to the ground. The processor 40 can control the rotation of the rotating shaft to drive the multiple rotary blades to rotate along the circumference of the rotating shaft. The rotating rotary blades cut the grass blades, wherein the distance between the bottom of the cutting assembly 50 and the ground can be the distance between the multiple rotary blades and the ground.

[0094] Among them, the millimeter-wave radar 60 has a strong penetrating ability and can detect objects blocked by grass leaves through grass leaves. The embodiment of the present application sets the millimeter-wave radar 60 and determines whether the object blocked by grass leaves is an obstacle based on the information of the object blocked by grass leaves detected by the millimeter-wave radar 60. When the object blocked by grass leaves is an obstacle, the robot 100 is controlled to avoid the object blocked by grass leaves or to raise the cutting component 50, so that the mowing work can be carried out smoothly and the cutting component 50 can be avoided from being damaged.

[0095] The millimeter-wave radar 60 may also be located at other locations of the robot 100 according to actual needs.

[0096] In some embodiments, the robot 100 further includes an ultrasonic radar 70, which is disposed at least on the side and / or rear end of the robot 100 and is configured to acquire information about objects on the side and / or rear end of the robot 100. The rear end is the rearmost end of the robot 100 in its normal travel direction. The processor 40 is configured to construct the point cloud map based on at least the posture information, information about objects in front of the robot 100 acquired by the laser radar, and information about objects on the side and / or rear end of the robot 100 acquired by the ultrasonic radar 70.

[0097] By obtaining information about objects on the side and / or rear of the robot 100, the robot 100 can avoid obstacles on the side and / or rear of the robot 100 when performing path planning, thereby preventing the robot 100 from colliding with obstacles on the side and / or rear.

[0098] The ultrasonic radar 70 may also be located at other locations of the robot 100 according to actual needs.

[0099] Please refer to Figure 7, which shows a map construction and positioning method provided in one embodiment of the present application, which is applied to the robot 100 provided in any of the aforementioned embodiments. As shown in Figure 7, the map construction and positioning method includes the following steps:

[0100] S10: Obtain motion parameters of the robot 100, where the motion parameters include acceleration and angular velocity.

[0101] S20: Acquire an image in front of the robot 100.

[0102] S30: Acquire information about an object in front of the robot 100 , where the information about the object in front of the robot 100 includes a distance and an angle of the object relative to the robot 100 .

[0103] S40: Determine the position and posture information of the robot 100 according to the acquired motion parameters and the acquired image.

[0104] S50: Constructing a point cloud map at least based on the posture information and the acquired information of the object in front of the robot 100.

[0105] S60: Determine the position of the robot 100 and the position of the obstacle in front of the robot 100 at least based on the point cloud map.

[0106] The map construction and positioning method provided in the embodiment of the present application can construct an accurate three-dimensional map based on the distance and angle of the front object relative to the robot 100 and the posture information of the robot 100, which is beneficial for the robot 100 to obtain accurate robot 100 position information and obstacle information, and then perform accurate path planning and obstacle avoidance.

[0107] Please refer to FIG8 , which shows a map construction and positioning method provided by another embodiment of the present application. In some embodiments, as shown in FIG8 , the map construction and positioning method further includes the following steps:

[0108] S70: Obtaining the laser reflectivity of the ground in front of the robot 100.

[0109] S80: Determine whether the ground in front of the robot 100 includes a lawn area based on the acquired laser reflectivity, and determine the boundary line of the lawn area when the ground in front includes a lawn area; and / or determine whether the ground in front of the robot 100 includes a lawn area based on the acquired image, and determine the boundary line of the lawn area when the ground in front includes a lawn area.

[0110] In some embodiments, judging whether the ground in front of the robot 100 includes a lawn area based on the acquired laser reflectivity includes: when the acquired laser reflectivity is within a preset laser reflectivity range, determining that the ground in front of the robot 100 includes a lawn area.

[0111] In some embodiments, the map construction and positioning method further includes: obtaining information about objects in the lawn area that are obscured by grass leaves; and constructing the point cloud map based at least on the posture information, the obtained information about objects in front of the robot, and the obtained information about objects obscured by grass leaves.

[0112] In some embodiments, the map construction and positioning method also includes: obtaining information about objects on the side and / or rear of the robot 100, where the rear is the rear in the direction of travel of the robot 100 during normal driving; and constructing the point cloud map based on at least the posture information, the information obtained about objects in front of the robot 100, and the information obtained about objects on the side and / or rear of the robot 100.

[0113] In some embodiments, the image acquired in front of the robot 100 includes an RGB image, and the map construction and positioning method further includes: constructing the point cloud map based on at least the posture information, the RGB image and the acquired information of the object in front of the robot 100, and performing semantic recognition on the point cloud map to obtain semantic information of the obstacle in front of the robot 100.

[0114] An embodiment of the present application also provides a robot control method, which is applied to the robot 100. The robot control method includes the map construction and positioning method provided by any of the aforementioned embodiments, and the control method also includes: judging whether the object obscured by the grass leaves is an obstacle based on the information of the object obscured by the grass leaves obtained; when it is determined that the object obscured by the grass leaves is an obstacle, controlling the robot 100 to avoid the position of the object obscured by the grass leaves, or, when the robot 100 drives to a position close to the object obscured by the grass leaves, controlling the cutting component 50 to be raised until the distance between the bottom of the cutting component 50 and the ground is greater than the height of the object obscured by the grass leaves, and controlling the cutting component 50 to mow the grass.

[0115] Among them, the map construction and positioning method and the robot control method correspond to the aforementioned robot 100. For more detailed descriptions, please refer to the contents of the various embodiments of the aforementioned robot 100. The map construction and positioning method and the contents of the aforementioned robot 100 can also be referenced to each other.

[0116] It should be noted that for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that this application is not limited by the order of the actions described, because according to this application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0117] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0118] The above is an implementation method of the embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the embodiment of the present application. These improvements and modifications are also considered to be within the scope of protection of the present application.

Claims

1. A robot, characterized in that: The robot comprises: A motion sensor, used to obtain motion parameters of the robot, wherein the motion parameters include acceleration and angular velocity; A visual sensor, used to obtain an image in front of the robot; a laser radar, used to obtain information about an object in front of the robot, wherein the information about the object in front of the robot includes a distance and an angle of the object relative to the robot; and A processor is used to determine the posture information of the robot based on the motion parameters obtained by the motion sensor and the image obtained by the visual sensor, and to construct a point cloud map based on at least the posture information and the information of the object in front of the robot obtained by the laser radar, and to determine the position of the robot and the position of the obstacle in front of the robot based on at least the point cloud map.

2. The robot according to claim 1, characterized in that: The laser radar is used to obtain the laser reflectivity of the ground in front of the robot; the processor is also used to determine whether the ground in front of the robot includes a lawn area and to determine the boundary line of the lawn area when the ground in front includes a lawn area based on the laser reflectivity obtained by the laser radar and / or the image obtained by the visual sensor; the processor is also used to construct the point cloud map based on at least the posture information, the information of the object in front of the robot obtained by the laser radar, and the determined boundary line of the lawn area.

3. The robot according to claim 2, characterized in that: The processor is used to determine that the ground in front of the robot includes a lawn area and determine a boundary line of the lawn area when the laser reflectivity acquired by the laser radar is within a preset laser reflectivity range.

4. The robot according to claim 1, characterized in that: The laser radar is at least arranged on the top of the robot; the visual sensor is at least arranged on the front end of the robot, and the front end is the frontmost end in the traveling direction of the robot when the robot is driving normally.

5. The robot according to claim 4, characterized in that: The robot includes at least a first laser radar and a second laser radar, which are respectively arranged on opposite sides of the top of the robot along a preset direction, the laser emitted by the first laser radar is located in a first plane, and the laser emitted by the second laser radar is located in a second plane, and a first intersection line of the first plane and the ground and a second intersection line of the second plane and the ground are both set at an angle to the direction of travel of the robot during normal driving, and the preset direction is perpendicular to the direction of travel of the robot during normal driving.

6. The robot according to claim 5, characterized in that: The angle between the first intersection line and the direction of travel of the robot during normal travel and the angle between the second intersection line and the direction of travel of the robot during normal travel are both values ​​between 10° and 80°.

7. The robot according to claim 4, characterized in that: The robot includes at least a first laser radar and a second laser radar, the first laser radar is arranged on the top of the robot, and the second laser radar is arranged on the front end of the robot, the laser emitted by the first laser radar is located in a third plane, and the laser emitted by the second laser radar is located in a fourth plane, the third plane is parallel to the ground, the fourth plane intersects with the ground, and the third intersection line of the fourth plane and the ground forms an angle with the forward direction of the robot during normal driving.

8. The robot according to claim 7, characterized in that: The angle between the third intersection line and the forward direction of the robot during normal driving is a value between 45° and 135°, and the angle between the fourth plane and the ground is a value between 10° and 40°.

9. The robot according to claim 2 or 3, characterized in that: The robot further includes a millimeter wave radar, which is used at least to obtain information about objects blocked by grass leaves in the lawn area; The processor is used to construct the point cloud map based on at least the posture information, the information of the object in front of the robot acquired by the laser radar, and the information of the object blocked by the grass leaves acquired by the millimeter wave radar.

10. The robot according to claim 1, characterized in that: The robot also includes an ultrasonic radar, which is at least arranged on the side and / or rear end of the robot, and is used to obtain information about objects on the side and / or rear of the robot, and the rear end is the rearmost end in the direction of travel of the robot during normal driving; the processor is used to construct the point cloud map based on at least the posture information, the information about the object in front of the robot obtained by the laser radar, and the information about the object on the side and / or rear of the robot obtained by the ultrasonic radar.

11. The robot according to claim 1, characterized in that: The image acquired by the visual sensor includes an RGB image. The processor is used to construct the point cloud map based on at least the posture information, the RGB image and the information of the object in front of the robot acquired by the lidar, and perform semantic recognition on the point cloud map to obtain the semantic information of the obstacle in front of the robot.

12. A map construction and positioning method, applied to a robot, characterized in that: The map construction and positioning method comprises: Acquiring motion parameters of the robot, wherein the motion parameters include acceleration and angular velocity; Acquire an image in front of the robot; Acquire information about an object in front of the robot, wherein the information about the object in front of the robot includes a distance and an angle of the object relative to the robot; Determining the position and posture information of the robot according to the acquired motion parameters and the acquired image; constructing a point cloud map based at least on the position information and the acquired information of the object in front of the robot; and The position of the robot and the position of an obstacle in front of the robot are determined at least according to the point cloud map.

13. The map construction and positioning method according to claim 12, characterized in that: The method further comprises: Obtaining laser reflectivity of the ground in front of the robot; According to the acquired laser reflectivity, determining whether the ground in front of the robot includes a lawn area and determining a boundary line of the lawn area when the ground in front includes the lawn area; And / or, judging whether the ground in front of the robot includes a lawn area according to the acquired image, and determining a boundary line of the lawn area when the ground in front includes the lawn area.

14. The map construction and positioning method according to claim 13, characterized in that: The determining, based on the acquired laser reflectivity, whether the ground in front of the robot includes a lawn area includes: When the acquired laser reflectivity is within a preset laser reflectivity range, it is determined that the ground in front of the robot includes a lawn area.

15. The map construction and positioning method according to claim 12, characterized in that: The method further comprises: Acquire information of objects blocked by grass leaves in the lawn area; The point cloud map is constructed at least according to the position information, the acquired information of the object in front of the robot and the acquired information of the object blocked by the grass leaves.

16. The map construction and positioning method according to claim 12, characterized in that: The method further comprises: Obtaining information about objects on the side and / or behind the robot; The point cloud map is constructed at least according to the position information, the acquired information of the object in front of the robot, and the acquired information of the object on the side and / or behind the robot.

17. The map construction and positioning method according to claim 12, characterized in that: The image in front of the robot obtained includes an RGB image, and the method further includes: The point cloud is constructed at least according to the posture information, the RGB image and the information of the object in front of the robot. A map is generated, and semantic recognition is performed on the point cloud map to obtain semantic information of obstacles in front of the robot.