Method for robot motion control for laser scanning of a support of a furniture bottom

Abstract

The application discloses a robot motion control method for laser scanning of a support part of a furniture bottom, comprising: when a robot walks into a detectable area of a target furniture without colliding with a support part of the target furniture bottom, the robot turns around the target furniture while keeping a laser sensor of the robot scanning the support part, and meanwhile, the robot distinguishes different support parts of the target furniture bottom according to ranging data and azimuth characteristic feedback from the laser sensor; whenever the robot turns around the target furniture by a preset angle, the laser sensor scans a local contour line of a side of all the support parts of the target furniture bottom; until the total angle of the robot turning around the target furniture is greater than or equal to 360 degrees, the robot connects each local contour line scanned on the same side of a support part into a closed contour of the same side of the support part, so as to connect closed contours of the sides of each support part of the target furniture bottom.

Description

Technical Field

[0001] This application relates to the field of obstacle perception and navigation planning technology, specifically to a robot motion control method for laser scanning the support portion at the bottom of furniture. Background Technology

[0002] Chinese invention patent application CN202011336271.6 discloses a robotic vacuum cleaner that uses four single-point ranging sensors to measure the distance information of the supporting components at the bottom of furniture while moving straight through the hollow area at the bottom of furniture. Although the robotic vacuum cleaner does not need to control the rotation of the robotic vacuum cleaner to drive the single-point ranging sensors to scan and obtain distance information, it can also obtain the outline of the bottom of the sofa and coffee table that meets the positioning requirements (corresponding to the outline boundary of the supporting component on one side of the bottom of the furniture) while moving straight. However, it still has blind spots in the field of vision because it only obtains the distance between the robot and obstacles in four different directions at the same time and converts these distances into coordinate information on the same map. The robotic vacuum cleaner cannot scan and obtain the complete edge of the supporting part at the bottom of the furniture at a single location.

[0003] Chinese invention patent application number CN202011041184.8 discloses a method for marking dense obstacle points based on a grid map. This method sets a movable rectangle on a map with known grid information. It identifies dense obstacle points and constructs a dense obstacle area by the relationship between the number of grids of obstacles distributed at intervals within the rectangle and the area covered by the rectangle. However, it does not control laser sensor scanning for accurate obstacle identification and positioning. It only uses very few resources (including hardware and software resources) to enable the robot vacuum cleaner to identify narrow passable areas between chair legs and table legs. What is identified are multiple densely distributed obstacles that are discretely distributed, without any closed obstacles of regular shape occupying the area. Summary of the Invention

[0004] This application discloses a robot motion control method for laser scanning of the support portion at the bottom of furniture, and proposes the following technical solution:

[0005] A robot motion control method for laser scanning the support parts at the bottom of furniture, the robot motion control method comprising: when the robot walks into the detectable area of ​​the target furniture without colliding with the support parts at the bottom of the target furniture, while keeping the robot's laser sensor scanning the support parts, the robot turns around the target furniture, and at the same time, the robot distinguishes different support parts at the bottom of the target furniture based on the ranging data and orientation features fed back by the laser sensor; wherein, the area where the laser sensor keeps detecting the support parts is the detectable area of ​​the target furniture; whenever the robot turns around the target furniture by a preset angle, the laser sensor scans the local contour lines around all the support parts at the bottom of the target furniture; until the total angle of the robot turning around the target furniture is greater than or equal to 360 degrees, the robot connects the local contour lines scanned around the same support part to form a closed contour around the same support part, so as to connect the closed contours around each support part at the bottom of the target furniture respectively; wherein, the target furniture is furniture with multiple support parts installed on the bottom and multiple support parts in contact with the ground, and the ground is the walking surface of the robot. In summary, by executing the robot motion control method described above, this application identifies the closed contours around each support of the target furniture when the robot walks into the detectable area of ​​the target furniture without colliding with the support at the bottom of the target furniture. The closed contours around each support form a warning area, which solves the technical problem of poor detection effect before the robot walks into the support at the bottom of the target furniture. It provides a warning area to solve technical problems such as the robot being trapped between supports and the possibility of collision accidents, prompting the robot not to approach the warning area. It can also prompt the robot not to miss the narrow area around the warning area when navigating.

[0006] Furthermore, when the robot turns around the target furniture at a preset angle, the robot is located on the central axis of the target furniture. The robot simultaneously detects the support parts on both sides of the central axis of the bottom of the target furniture through its laser sensor. Among the support parts on each side of the central axis, the number of local contour lines scanned by the laser sensor in the support part closest to the robot is greater than or equal to the number of local contour lines scanned by the laser sensor in the support part farthest from the robot. In the bottom of the target furniture, half of the support parts are symmetrically installed with the other half of the support parts based on the central axis. This application identifies support parts at different distances and orientations based on the central axis of the target furniture, and analyzes in detail the relationship between the number of local contour lines in the support parts and the distance of the support parts, forming the position and peripheral contour information of each support part collected within a certain distance range in front of the robot each time the robot turns around the target furniture at the preset angle.

[0007] Furthermore, the method by which the robot distinguishes different support parts at the bottom of the target furniture based on the ranging data and orientation features fed back by the laser sensor includes: when the robot turns around the target furniture at a preset angle, the robot distinguishes the support parts located on the left side of the central axis of the target furniture and the support parts located on the right side of the central axis of the target furniture based on the direction of the ranging data obtained by its laser sensor. On each side of the central axis, based on the ranging data of each local contour line fed back by the laser sensor, each support part is distinguished one by one from near to far along the central axis, including distinguishing the support part closest to the robot and its local contour line scanned by the laser sensor and the support part farthest from the robot and its local contour line scanned by the laser sensor.

[0008] Furthermore, within the detectable area of ​​the target furniture, among the support parts scanned by the robot before turning around the target furniture at the preset angle and the support parts scanned by the robot after turning around the target furniture at the preset angle, there are instances of the same support part being scanned repeatedly. Within the repeatedly scanned same support part, the local contour lines scanned by the robot after turning around the target furniture at the preset angle include either parts of the local contour lines not scanned before turning around the target furniture at the preset angle or parts of the local contour lines scanned by the robot before turning around the target furniture at the preset angle. In summary, by repeatedly scanning the same support part while turning around the target furniture, the robot can use the local contour information scanned by the optical sensor when at the bottom of the target furniture to correct the optimal viewing angle direction of the laser sensor or align the line of sight with the direction of travel of the planned route along which the robot is moving. This results in a high detection rate of the laser sensor on the sides of each support part directly in front, improving the efficiency of detecting the peripheral contours of the support parts, reducing the occurrence of scanning blind spots, improving navigation smoothness, and reducing collisions.

[0009] Furthermore, the robot turns around the target furniture by rotating clockwise or counterclockwise around the center of the furniture. The robot's rotation radius is greater than half the maximum outline of the target furniture, ensuring that the robot reaches the central axis of the target furniture at each preset turning angle. Each time the robot turns around the target furniture at the preset angle, it updates the central axis set in the direction the target furniture points towards the robot's current position to the central axis of the target furniture at the robot's current location. Before turning around the target furniture, the robot is located on the central axis of one direction of the target furniture. The robot does not collide with the support parts at the bottom of the target furniture before or after turning, and the laser sensor scans the local outlines of all the support parts around the bottom of the target furniture. The central axes in each direction of the target furniture and the center of the target furniture are pre-marked by the robot. This expands the laser sensor's field of view of the bottom of the target furniture, improving the robot's detection effect on the support parts. It also increases the distance between the robot and the support parts, avoiding sudden braking and jamming when moving towards the bottom of the target furniture, thus improving the safety and smoothness of the robot's movement over the target furniture.

[0010] Furthermore, each time the robot turns at the preset angle around the target furniture, it rotates its body to align its front with the furniture, ensuring that the robot's central axis is parallel to the central axis of the target furniture. The laser sensor is mounted on the top front of the robot. This allows the optimal viewing angle or line of sight of the laser sensor to align with the central axis of the target furniture, improving the coverage of the laser sensor's detection of the sides of the supporting parts directly in front and reducing blind spots.

[0011] Furthermore, the projected shape of the target furniture on the ground is rectangular; four support parts are symmetrically installed on the bottom of the target furniture, located at the four corner points of the rectangle; wherein, the closed contour of the periphery of the support part is represented by the circumscribed rectangle that encloses the periphery of the support part, so that each local contour line continuously scanned from the periphery of the same support part is the four sides of the rectangle.

[0012] Furthermore, the preset angle is 90 degrees, so that after the robot makes four consecutive turns around the target furniture with the preset angle as the rotation step size, the total angle turned is equal to 360 degrees.

[0013] Furthermore, the robot rotates clockwise or counterclockwise around the center of the target furniture, simultaneously scanning with a laser sensor at least one local contour line aligned with the robot in each support section on both sides of the central axis of the target furniture. On each side of the central axis of the target furniture, each support section includes a pair of perpendicularly aligned local contour lines scanned by the laser sensor in the support section closest to the robot, and a local contour line perpendicular to the central axis of the machine in the support section furthest from the robot. Thus, the two local contour lines in the support section closest to the robot and the one local contour line in the support section furthest from the robot on each side of the central axis of the target furniture all fall within the detection area of ​​the laser sensor, determining the reasonable construction range of the periphery contour of each support section, making the contour lines formed by the robot on the periphery of the support sections in different directions at the bottom of the target furniture environmentally adaptable.

[0014] Furthermore, the number of local contour lines scanned by the robot each time it turns the preset angle around the target furniture is equal; the robot repeatedly scans the same support portion before and after turning the preset angle around the target furniture; in the support portion closest to the robot's current position in the repeated scan, the local contour line scanned after the current turn of the preset angle is a local contour line that was not scanned before the current turn of the preset angle; in the support portion farthest from the robot's current position in the repeated scan, the local contour lines scanned before the current turn of the preset angle include the local contour lines scanned after the current turn of the preset angle. In summary, by turning at the preset angle, the robot simultaneously scans multiple different support parts at two different locations and repeatedly scans the corresponding local contour lines of the same support part. This ensures that while the contour lines of some support parts are completely scanned, unscanned local contour lines of other support parts are left for subsequent scanning when turning at the preset angle, until the total angle of the robot's turns around the target furniture is greater than or equal to 360 degrees. The robot then connects the local contour lines scanned around the same support part to form a closed contour around the same support part. Thus, by rotating and repeatedly scanning, the robot expands the viewing angle of the laser sensor on the bottom of the target furniture, ensuring the completeness of the robot's scan of support parts in different directions and improving the robot's adaptability to the regularly distributed support parts on the bottom of the target furniture. Attached Figure Description

[0015] Figure 1 A flowchart of a robot motion control method for laser scanning the support portion at the bottom of furniture is provided as an embodiment of this application.

[0016] Figure 2 This application provides a schematic diagram of the movement of a robot scanning four support parts at the bottom of furniture during a turn, as part of another embodiment of the present application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be described in detail below with reference to the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other. In this application, it should be understood that the terms "center," "middle position," "central axis," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and 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 orientation or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. If terms such as "first," "second," and "third" appear in the embodiments, they are for the convenience of distinguishing related features and should not be construed as indicating or implying their relative importance, order, or number of technical features.

[0018] Robotic vacuum cleaners are designed to inevitably come into contact with the supports under furniture when cleaning, or to pre-scan the outlines of the supports in the corresponding locations. However, in order to achieve a certain scanning accuracy and outline integrity, when the passable area between the supports is large enough, the robot often repeatedly circles between the supports, getting stuck under the furniture. It may scan the complete edges of the supports, but may not be able to escape from the hollow areas between different supports. If it only scans the narrow passable areas between chair legs and table legs, it identifies multiple densely distributed obstacles, without forming regular closed obstacles for the robot to effectively avoid, causing the robotic vacuum cleaner to fail to maintain the preset cleaning mode.

[0019] To address the aforementioned technical deficiencies, this application discloses a robot motion control method for laser scanning of the support parts at the bottom of furniture. The robot motion control method is executed by a robot with a fixedly mounted laser sensor. Considering that the robot is used to scan the support parts at the bottom of furniture, such as navigating multiple upright table or chair legs, the robot can be configured as a circular robot, such as a robotic vacuum cleaner, that operates in indoor environments. The laser sensor mounted on the robot contains a rotating laser probe. During rotation, one end of the laser beam emitted by the laser probe forms an arc, which is formed in front of the robot through the laser sensor's exit aperture, sampling the contour of obstacles in front. Preferably, the angle between the laser beams emitted from both sides of the exit aperture is the detection area of ​​the laser sensor, corresponding to the following: Figure 2The circular robot successively emitted angular regions from position points O1, O2, and O3. The boundaries of each angular region covered part of the outline of support A, support B, support C, and support D. All of these were considered to be within the detectable area of ​​the target furniture.

[0020] In this application, a robot can be any mechanical device capable of highly autonomous spatial movement within its operating environment, such as unmanned vehicles, robots, air purifiers, etc. The robot can be various types of robots, including cleaning robots and other service robots. Cleaning robots refer to robots capable of autonomously performing cleaning tasks within their operating environment, including sweeping robots and window-cleaning robots. Other service robots refer to robots capable of autonomously moving within their operating environment and providing non-cleaning services, including air-purifying robots, home companion robots, and welcoming robots. Of course, the shape of the robot will vary depending on its implementation. This embodiment does not limit the shape; taking the outer contour shape of the robot as an example, the outer contour shape of the robot device can be irregular or regular. For example, the outer contour shape of the robot can be a circle, ellipse, square, triangle, teardrop shape, or D-shape, etc. Correspondingly, shapes other than regular shapes are called irregular shapes, such as the outer contour of a humanoid robot, the outer contour of an unmanned vehicle, and the outer contour of a drone. Similarly, the outline shape of the support can be irregular or regular. For example, the outer outline of the support formed on the ground when viewed from above can be a regular shape such as a circle, ellipse, square, triangle, teardrop, or D-shape. Figure 2 The peripheral contour shapes of support A, support B, support C, and support D are all rectangular. The peripheral shape referred to in this application is defined as the side edges of the support distributed in the horizontal plane or the robot's walking surface. These side edges can surround the side surfaces of the support to form a closed contour, which can be divided into multiple local contour lines.

[0021] like Figure 1As shown, the robot motion control method includes: when the robot walks into the detectable area of ​​the target furniture but does not collide with the support part at the bottom of the target furniture, it can be understood that there is a safe distance between the robot and the support part at the bottom of the target furniture, but the laser sensor is allowed to scan each support part in the corresponding direction. Then, while keeping the robot's laser sensor scanning the support part, the robot turns around the target furniture. Generally, the robot turns around the center of the target furniture by a certain radius and a certain rotation angle step. At the same time, the robot distinguishes different support parts at the bottom of the target furniture according to the ranging data and orientation characteristics fed back by the laser sensor. This can be done by distinguishing the support parts at different directions and different distances at the bottom of the target furniture each time a certain rotation angle step is taken.

[0022] In this application, the robot walks into the detectable area of ​​the target furniture without colliding with the support at the bottom of the furniture. It only starts turning around the target furniture when the robot's laser sensor scans the support. On the one hand, this can expand the laser sensor's field of view of the bottom of the target furniture and improve the detection effect of the support. On the other hand, it can increase the distance between the robot and the support, avoiding the jamming caused by sudden braking when walking towards the bottom of the target furniture, and improving the safety and smoothness of the robot walking over the target furniture.

[0023] In this application, the area detected by the laser sensor is the detectable area of ​​the target furniture, including the area where the robot collides with the support. The detectable area of ​​the target furniture covers at least all the support parts on the bottom of the target furniture, and there are gaps between the various support parts on the bottom of the target furniture to provide laser incident space and reflection space for the laser sensor. It should be noted that the target furniture is furniture with multiple support parts installed on the bottom and multiple support parts in contact with the ground, and the ground is the walking surface of the robot. When the robot is applied to the bottom of furniture as a vacuum cleaner, the vacuum cleaner can clean in a connected area where the height of the hollow part at the bottom is greater than the height of the vacuum cleaner body (e.g., a passable area formed between different support parts at the bottom of a dining table or desk, or a non-passable area formed between different support legs at the bottom of a chair). The vacuum cleaner can move in the connected area by planning the cleaning method, or walk around the edge of the support to scan the specific position and outline size of the support. The vacuum cleaner can scan and supplement the local outline lines of the support in various directions in real time.

[0024] Each time the robot turns around the target furniture at a preset angle, the laser sensor scans the local contour lines of all the support parts around the bottom of the furniture. This means that after the robot turns around the target furniture at the preset angle, the laser sensor has a scanning blind spot due to the scanning angle, so it cannot scan the complete contour line of a single support part at the same location. However, the positional advantage gained from turning around the target furniture at the preset angle allows the robot to scan the local contour lines of all the support parts around the bottom of the same furniture. The local contour lines scanned in a single scan can be obtained by fitting laser points collected by the laser sensor at the corresponding support part, or they can be pre-marked based on a map. The actual object corresponding to the local contour line can be the contour line of the support part's perimeter.

[0025] In this application, the robot obtains the ranging data fed back by the laser sensor by calculating the distance between the laser point collected by the laser sensor and the center of the laser sensor; and obtains the orientation features fed back by the laser sensor by calculating the angle between the laser point collected by the laser sensor and a pre-set coordinate axis. Based on this, the pose relationship between the local contour lines of each support part on the bottom of the target furniture and the robot is determined.

[0026] It is worth noting that when the robot's laser sensor detects highly reflective or low-reflective support parts such as glass and shiny metal, the detection rate will be relatively low, resulting in poor object detection and the appearance of scanning blind spots. In addition, if a person or other object suddenly appears from the scanning blind spot, it is easy to collide with the robot.

[0027] Therefore, the robot needs to turn around the target furniture multiple times at the preset angle, scanning and distinguishing the local contour lines around each support part on the bottom of the target furniture in real time. Each scanned local contour line can be a local contour line located in the scanning blind zone that existed in the previous scan, until the total angle of the robot turning around the target furniture is greater than or equal to 360 degrees. The robot will connect the local contour lines scanned around the same support part to form a closed contour around the same support part, thereby realizing the continuous scanning of the complete contour line of a single support part in multiple turns. The complete contour line of a single support part can be the closed contour of a single support part, and then connecting them to form the closed contours around each support part on the bottom of the target furniture.

[0028] In summary, by executing the robot motion control method described above, this application identifies the closed contours around each support of the target furniture when the robot walks into the detectable area of ​​the target furniture without colliding with the support at the bottom of the target furniture. The closed contours around each support form a warning area, which solves the technical problem of poor detection effect before the robot walks into the support at the bottom of the target furniture. It provides a warning area to solve technical problems such as the robot being trapped between supports and the possibility of collision accidents, prompting the robot not to approach the warning area. It can also prompt the robot not to miss the narrow area around the warning area when navigating.

[0029] As one embodiment, when the robot turns around the target furniture at the preset angle, the robot is located on the central axis of the target furniture. The robot simultaneously detects the support parts on both sides of the central axis at the bottom of the target furniture using its laser sensor. Specifically, in each support part on each side of the central axis, the number of local contour lines scanned by the laser sensor in the support part closest to the robot is greater than or equal to the number of local contour lines scanned by the laser sensor in the support part farthest from the robot. In the bottom of the target furniture, one half of the support parts is symmetrically installed with the other half based on the central axis. This embodiment identifies support parts at different distances and orientations based on the central axis of the target furniture and analyzes in detail the relationship between the number of local contour lines in the support parts and the distance between the support parts. This forms the position and peripheral contour information of each support part collected within a certain distance range in front of the robot each time it turns around the target furniture at the preset angle.

[0030] In this embodiment, the area within a certain distance in front of the robot can be a fan-shaped or circular region, which is related to the robot's current orientation ( Figure 2 The small dashed circle in the circular robot shown is positioned within the robot body (direction) and has a field of view angle ( Figure 2 The angle of the radiating area (shown as the angle of the included area of ​​the small dashed circle); the specific value of the certain distance range and the preset angle can also be reasonably set according to the size and shape of the robot, the spacing between different support parts in the target furniture, etc.

[0031] Indicatively, such as Figure 2As shown, the support parts of the target furniture include support parts A, B, C, and D. Support parts A, B, C, and D together lift the target furniture above the ground to form four legs. The robot first turns clockwise from position O1 to position O2 by the preset angle, and then turns clockwise from position O2 to position O3 by the preset angle. The robot is located on the horizontal central axis of the target furniture at both positions O1 and O3. The vertical central axis of the target furniture is parallel to the line connecting support parts A and B or the line connecting support parts D and C. The horizontal central axis of the target furniture is also parallel to the line connecting support parts A and D or the line connecting support parts B and C.

[0032] exist Figure 2 In the diagram, when the robot is at position O3, for ease of description and differentiation, the supports above the horizontal central axis of the target furniture are designated as support C and support B from near to far, and the supports below the horizontal central axis of the target furniture are designated as support D and support A from near to far. Supports D and C are symmetrically installed on the bottom of the target furniture based on the horizontal central axis; supports B and A are also symmetrically installed on the bottom of the target furniture based on the horizontal central axis. Among the supports above the horizontal central axis of the target furniture, the local contour lines scanned by the laser sensor in the supports closest to the robot are local contour lines c2c4 and c3c4, respectively. The local contour line scanned by the laser sensor in the support farthest from the robot is local contour line b2b4. Due to the limitation of the maximum detection range of the laser sensor, the local contour line b3b4 of support B is not scanned by the laser sensor. In the support section located below the horizontal central axis of the target furniture, the local contour lines scanned by the laser sensor in the support section closest to the robot are local contour lines d2d4 and d2d1, respectively. The local contour line scanned by the laser sensor in the support section farthest from the robot is local contour line a2a4. Due to the limitation of the maximum detection range of the laser sensor, the local contour lines a1a2 of support section A were not scanned by the laser sensor.

[0033] Therefore, the peripheral local contour lines or closed contours mentioned in the foregoing embodiments are all represented as the sides that enclose the top view shape of the support portion on the ground. Based on this, it is convenient to subsequently distinguish between the unscanned and scanned local contour lines on the periphery of each support portion.

[0034] As one embodiment, the method by which the robot distinguishes different support parts at the bottom of a target furniture based on ranging data and orientation features fed back by a laser sensor includes: when the robot turns around the target furniture at a preset angle, the robot distinguishes the support parts located to the left of the central axis and the support parts located to the right of the central axis of the target furniture based on the direction of the ranging data obtained by its laser sensor; and on each side of the central axis, based on the ranging data of each local contour line fed back by the laser sensor, the robot distinguishes each support part one by one from near to far along the central axis, including distinguishing the support part closest to the robot and its local contour line scanned by the laser sensor, and the support part farthest from the robot and its local contour line scanned by the laser sensor. It should be noted that the direction of the central axis of the target furniture can be a direction perpendicular to the plane where two adjacent support parts are located, and the form in which the direction of the central axis of the target furniture is represented can vary depending on the orientation of the environmental information scanned by the robot after turning around the target furniture at a preset angle. Within the detection area of ​​the laser sensor, the identified local contour lines are distributed as continuous line segments around the same support and as discontinuous line segments between different supports (i.e., two discontinuous local contour lines located on two different support sections). Both are obtained by curve fitting of the laser points, providing the robot with contour information of the chair legs or table legs at the bottom of the target furniture.

[0035] Indicatively, such as Figure 2 As shown, when the robot is at position O2, the support parts on the left side of the vertical central axis of the target furniture are distributed sequentially from near to far as support part B and support part A, and the support parts on the right side of the vertical central axis of the target furniture are distributed sequentially from near to far as support part C and support part D. Support parts B and C are symmetrically installed on the bottom of the target furniture based on the vertical central axis; support parts A and support parts D are symmetrically installed on the bottom of the target furniture based on the vertical central axis. In the support parts located on the left side of the vertical central axis of the target furniture, the local contour lines distinguished one by one along the vertical central axis from near to far are local contour lines b1b2, b2b4, and a1a2, respectively; local contour lines b2b4 and a1a2 are considered as two discontinuous local contour lines, located on the periphery of support part B and the periphery of support part A, respectively. In the support part located to the right of the vertical central axis of the target furniture, the local contour lines that are distinguished one by one from near to far along the vertical central axis are local contour lines c1c2, c1c3, and d1d2, respectively; local contour lines c1c3 and d1d2 are regarded as two discontinuous local contour lines, located on the periphery of support part C and support part D, respectively.

[0036] Based on the aforementioned embodiment, where the robot needs to make multiple turns around the target furniture at the preset angle, within the detectable area of ​​the target furniture, among the support parts scanned by the robot before and after making the turns around the target furniture at the preset angle, there are instances of the same support part being scanned repeatedly. That is, the robot scans the same support part both times it makes the turns around the target furniture at the preset angle. For example, (illustrated). Figure 2 As shown, when the robot starts from position O1 and turns clockwise along the arc path L1 to position O2, the robot scans support parts A, B, C and D through the laser sensor at both positions O1 and O2; when the robot starts from position O2 and turns clockwise along the arc path L2 to position O3, the robot scans support parts A, B, C and D through the laser sensor at both positions O2 and O3.

[0037] In the same support section that is repeatedly scanned, the local contour line scanned by the robot after it turns around the target furniture at the preset angle includes either a portion of the local contour line that was not scanned before the robot turned around the target furniture at the preset angle (forming a scanning blind spot before the robot turned around the target furniture at the preset angle) or a portion of the local contour line that was scanned before the robot turned around the target furniture at the preset angle, so that the local contour line scanned by the robot after it turns around the target furniture at the preset angle supplements the scanning blind spot before the robot turned around the target furniture at the preset angle. (Illustratively, as...) Figure 2 As shown, the robot scans support C using laser sensors at both positions O2 and O3. The local contour lines of support C scanned by the robot at position O3 include local contour lines c2c4 and c4c3. Both local contour lines c2c4 and c4c3 are local contour lines of support C that the robot did not scan at position O2. Therefore, by turning around the target furniture once at the preset angle, the robot completes the contour of the repeatedly scanned same support. Additionally, the robot scans support B using laser sensors at both positions O2 and O3. The local contour lines of support B scanned by the robot at position O3 include local contour lines b2b4. Local contour lines b2b4 are local contour lines of support B that the robot has already scanned at position O2.

[0038] In summary, by repeatedly scanning the same support part by turning around the target furniture, the robot can use the local contour information scanned by the optical sensor when it is at the bottom of the target furniture to correct the optimal viewing angle of the laser sensor or make the line of sight consistent with the direction of travel of the robot along the planned route. This makes the laser sensor have a high detection rate of the side of each support part in front, improves the efficiency of detecting the periphery of the support part, reduces the occurrence of scanning blind spots, improves navigation smoothness, and reduces collisions.

[0039] In one embodiment, the robot turns around the target furniture by rotating clockwise or counterclockwise around the center of the target furniture. The robot's rotation radius is greater than half the maximum outline of the target furniture, ensuring that the robot reaches the central axis of the target furniture with each turn at the preset angle. In this embodiment, the distance between the robot's body center and the center of the target furniture is greater than half the maximum outline of the target furniture, but the laser sensor can still detect the support parts during the robot's turn. Therefore, the robot's rotation radius can be selected as the length of the maximum outline of the target furniture or the maximum length of the line connecting the two support parts, preventing the robot from colliding with either support part. The size of the preset angle depends on the angle formed by the robot's position before starting to turn around the target furniture with respect to the central axis of the target furniture. The side of the target furniture scanned by the robot at its initial position before starting to turn around the target furniture differs from the side of the target furniture that the robot needs to traverse to reach the central axis of the target furniture. This embodiment can, on the one hand, expand the laser sensor's field of view of the bottom of the target furniture, improve the robot's detection effect on the support, and on the other hand, increase the distance between the robot and the support, avoiding the jamming caused by sudden braking when walking towards the bottom of the target furniture, thus improving the safety and smoothness of the robot walking over the target furniture.

[0040] It should be noted that when planning the robot's turning route, to prevent collisions during the turn, the turning radius closest to the support should be at least half the length of the target furniture's maximum outline. Of course, the minimum turning radius may vary depending on the robot's size and shape, control strategy, and application scenario. For larger or longer target furniture, a larger turning radius may be needed to safely bypass the support. Therefore, the minimum turning radius should be set as needed, and this application does not impose any restrictions on it. Furthermore, since the robot reaches one of the target furniture's directional central axes with each turn, the preset angle for the robot's turn around the target furniture can be the angle between the target furniture's horizontal and vertical central axes. In addition, the preset angle may also vary depending on the robot's size and shape, control strategy, and application scenario; it should be set as needed, and this application does not impose any restrictions on it.

[0041] Generally speaking, a larger turning radius usually means smoother and safer turns. The smaller the angle between the turning direction of the robot around the target furniture and the central axis of the target furniture to be reached, the better. This can reduce the scanning blind zone and effectively improve the scanning efficiency of the laser sensor. In other words, it allows the robot to identify the peripheral contours of each support at a higher speed, thereby improving the robot's operating efficiency.

[0042] Each time the robot turns the preset angle around the target furniture, it updates the centerline of the target furniture pointing towards the robot's current position to the centerline of the target furniture where the robot is currently located. That is, when the robot moves from position O1 to position O2 by turning the preset angle around the target furniture, the centerline of the target furniture where the robot is currently located becomes... Figure 2 The dashed rectangle connecting support parts A and D points to position O2. Before turning around the target furniture, the robot is positioned on the central axis of the target furniture in one direction, without colliding with the support parts at the bottom of the target furniture. The robot's laser sensor scans the local contour lines around all the support parts at the bottom of the target furniture. Schematic, when the robot is at position O1, position O1 is also located on the central axis along the direction from the target furniture to the robot's current position O1, i.e., on the horizontal central axis of the target furniture. The robot does not collide with the support parts at the bottom of the target furniture before or after turning around it, and the laser sensor scans the local contour lines around all the support parts at the bottom of the target furniture. The central axes in each direction of the target furniture and the center of the target furniture are pre-marked by the robot, specifically within a map built by the robot in real time.

[0043] Indicatively, such as Figure 2 As shown, when the robot is at position O2, the central axis of the target furniture where the robot is currently located is the vertical central axis of the target furniture; to the left of the vertical central axis of the target furniture, the scanned support parts are distributed from near to far as support part B and support part A; to the right of the vertical central axis of the target furniture, the scanned support parts are distributed from near to far as support part C and support part D. Figure 2 In the process, when the robot turns from position O2 to position O3, the central axis of the target furniture where the robot is currently located is updated to the horizontal central axis of the target furniture; on the upper side of the horizontal central axis of the target furniture, the scanned support parts are distributed from near to far as support part C and support part B; on the lower side of the horizontal central axis of the target furniture, the scanned support parts are distributed from near to far as support part D and support part A.

[0044] In one example, each time the robot turns the preset angle around the target furniture, it rotates its body to align its front with the furniture, making the robot's central axis parallel to the central axis of the target furniture. A laser sensor is mounted on the top front of the robot. This allows the optimal viewing angle or line of sight of the laser sensor to align with the central axis of the target furniture, improving the coverage of the laser sensor's detection of the sides of the supports directly in front and reducing blind spots.

[0045] The robot adjusts its front to face the target furniture by rotating its body; therefore, the robot's current orientation is... Figure 2 The small dashed circle in the circular robot shown is positioned within the robot body. The robot's current orientation at position O2 is parallel to the vertical central axis of the target furniture. The scanning angle of the laser sensor is the angle of the area emanating from the small dashed circle in the robot at position O2. To the left of the vertical central axis of the target furniture, it covers the upper partial contour lines b1b2, the right partial contour line b2b4, and the upper partial contour line a1a2 of support part B. To the right of the vertical central axis of the target furniture, it covers the upper partial contour lines c1c2, the left partial contour line c1c3, and the upper partial contour line d1d2 of support part D. This constitutes the detection results of the corresponding directions of the partial contour lines of all the supports at the bottom of the target furniture when the robot is facing the target furniture.

[0046] As one embodiment, the projected shape of the target furniture on the ground is rectangular; four support parts are symmetrically installed on the bottom of the target furniture, located at the four corner points of the rectangle, and these four support parts are respectively denoted as... Figure 2 Support parts A, B, C, and D are shown. The closed contour of the periphery of each support part is represented by the circumscribed rectangle enclosing the periphery of the support part, such that each continuously scanned local contour line around the same support part is one of the four sides of the rectangle. Specifically, as shown... Figure 2 As shown, support A is represented by a rectangle formed by connecting local contour lines a1a2, a2a4, a4a3, and a3a1 in sequence; support B is represented by a rectangle formed by connecting local contour lines b1b2, b2b4, b4b3, and b3b1 in sequence; support C is represented by a rectangle formed by connecting local contour lines c1c2, c2c4, c4c3, and c3c1 in sequence; and support D is represented by a rectangle formed by connecting local contour lines d1d2, d2d4, d4d3, and d3d1 in sequence.

[0047] Preferably, the preset angle is 90 degrees, corresponding to Figure 2 In the process, the robot starts from position O1 and turns clockwise along the arc path L1 to position O2 by a preset angle of 90 degrees. Similarly, the robot starts from position O2 and turns clockwise along the arc path L2 to position O3 by a preset angle of 90 degrees. And so on. The robot then continues to turn clockwise from position O3 by the preset angle to a position below the vertical center axis of the target furniture, and then continues to turn clockwise from that position to position O1. The robot's turning radius does not change. After the robot turns around the target furniture 4 times with the preset angle as the rotation step, the total angle turned is equal to 360 degrees.

[0048] As one embodiment, in order to identify the circumferential closed contours of each support part, the robot rotates clockwise or counterclockwise around the center of the target furniture. Specifically, each time the robot turns around the target furniture at the preset angle, the laser sensor simultaneously scans at least one local contour line aligned with the robot in each support part on both sides of the central axis of the target furniture. On each side of the central axis of the target furniture, the at least one local contour line aligned with the robot in each support part includes a pair of perpendicularly arranged local contour lines scanned by the laser sensor in the support part closest to the robot, and a local contour line perpendicular to the central axis of the machine in the support part farthest from the robot. The pair of perpendicularly arranged local contour lines scanned by the laser sensor in the support part closest to the robot includes a local contour line perpendicular to the central axis of the machine and a local contour line parallel to the central axis of the machine. Both are considered as local contour lines aligned with the robot and scanned in the support part closest to the robot. Therefore, the local contour line perpendicular to the central axis of the machine in the support part farthest from the robot can be considered as local contour lines aligned with the robot and scanned. Thus, the two local contour lines of the support closest to the robot and the one local contour line of the support farthest from the robot on each side of the central axis of the target furniture fall within the detection area of ​​the laser sensor, and the reasonable construction range of the periphery contour of each support is determined, so that the contour lines formed by the robot on the periphery of the support in different directions at the bottom of the target furniture have environmental adaptability.

[0049] Indicatively, such as Figure 2As shown, when the robot is in position O2, the robot's central axis is parallel to the vertical central axis of the target furniture. Among the support parts to the left of the vertical central axis of the target furniture, at least one local contour line aligned with the robot includes local contour lines b1b2, b2b4, and a1a2. Local contour lines b1b2 and b2b4 form a pair of perpendicularly arranged local contour lines scanned by the laser sensor in the support part closest to the robot (corresponding to support part B). Local contour lines a1a2 are local contour lines perpendicular to the central axis of the robot, scanned by the laser sensor in the support part furthest from the robot (corresponding to support part A). Additionally, due to the limitation of the maximum detection range of the laser sensor, local contour lines b1b3 and b3b4 in support part B, and local contour lines a1a3, a3a4, and a4a2 in support part A, constitute the local contour lines not scanned in the support parts to the left of the vertical central axis of the target furniture. This constitutes the effective scanning angle range of the laser sensor for the support parts to the left of the vertical central axis of the target furniture, improving the contour scanning effect of the support parts to the left of the vertical central axis of the target furniture.

[0050] For ease of description and distinction, it is also listed that when the robot is in position O2, the robot's body central axis is parallel to the vertical central axis of the target furniture. At least one local contour line aligned with the robot in each support part to the right of the vertical central axis of the target furniture includes local contour lines c1c2, c1c3, and d1d2. Local contour lines c1c2 and c1c3 form a pair of perpendicularly arranged lines scanned by the laser sensor in the support part closest to the robot (corresponding to support part C). Local contour lines, d1d2, are local contour lines that are perpendicular to the central axis of the robot and scanned by the laser sensor in the support part farthest from the robot (corresponding to support part D). Additionally, due to the limitation of the maximum detection range of the laser sensor, local contour lines c2c4 and c3c4 in support part C, and local contour lines d1d3, d3ad4, and d4d2 in support part D, constitute the local contour lines that are not scanned in the support parts to the right of the vertical central axis of the target furniture. This constitutes the effective scanning angle range of the laser sensor for the support parts to the right of the vertical central axis of the target furniture, improving the contour scanning effect of the support parts.

[0051] Based on the above embodiments, the number of local contour lines scanned by the robot each time it turns the preset angle around the target furniture is equal, combined with... Figure 2It can be seen that the number of local contour lines scanned by the robot when it turns around the target furniture to positions O1, O2, and O3 is equal. The preferred number of local contour lines scanned in each scan is 6. (Illustratively, as shown...) Figure 2 As shown, during the process of the robot scanning the support parts at positions O2 and O3 using the laser sensor, the robot scanned the local contour lines b1b2 and b2b4 of support part B and the local contour line a1a2 of support part A at position O2. It also scanned the local contour lines c1c2 and c1c3 of support part C and the local contour line d1d2 of support part D at position O2, thus accumulating 6 local contour lines scanned at position O2. Similarly, the robot scanned the local contour lines c2c4 and c3c4 of support part C and the local contour line b2b4 of support part B at position O3. It also scanned the local contour lines d1d2 and d2d4 of support part D and the local contour line a2a4 of support part A at position O3, thus accumulating 6 local contour lines scanned at position O3.

[0052] Specifically, before and after the robot turns around the target furniture by the preset angle, it repeatedly scans the same support portion. Each scan of the same support portion includes the support portion closest to the robot's current position and the support portion farthest from the robot's current position. In the repeatedly scanned support portion closest to the robot's current position, the local contour line scanned after the current turn of the preset angle is a local contour line that was not scanned before the current turn of the preset angle. (Illustratively, as shown...) Figure 2 As shown, the robot scans support C (the closest support to the robot's current position) at positions O2 and O3 using a laser sensor. The local contour lines of support C scanned by the robot at position O3 include local contour lines c2c4 and c4c3. Both local contour lines c2c4 and c4c3 are local contour lines of support C that the robot did not scan at position O2. Therefore, by turning around the target furniture once at the preset angle, the robot repeatedly scans the complete contour of the same support closest to the robot. Furthermore, in the support farthest from the robot's current position during repeated scanning, the local contour lines scanned before the current turn at the preset angle include the local contour lines scanned after the current turn at the preset angle. (Illustratively, as shown...) Figure 2As shown, the robot scans support B (the support furthest from the robot's current position) using its laser sensor at positions O2 and O3. The local contour lines of support B scanned by the robot at position O3 include local contour lines b2b4. Local contour lines b2b4 are the local contour lines of support B already scanned by the robot at position O2. However, local contour lines b3b4 are local contour lines not scanned after the current turn at the preset angle. Therefore, the sum of the number of local contour lines repeatedly scanned for the same support farthest from the robot and the number of local contour lines repeatedly scanned for the same support closest to the robot is equal to half the number of local contour lines scanned each time the robot turns around the target furniture at the preset angle.

[0053] In summary, by turning at the preset angle, the robot simultaneously scans multiple different support parts at two different locations and repeatedly scans the corresponding local contour lines of the same support part. This ensures that while the contour lines of some support parts are completely scanned, unscanned local contour lines of other support parts are left for subsequent scanning when turning at the preset angle, until the total angle of the robot's turns around the target furniture is greater than or equal to 360 degrees. The robot then connects the local contour lines scanned around the same support part to form a closed contour around the same support part. Thus, by rotating and repeatedly scanning, the robot expands the viewing angle of the laser sensor on the bottom of the target furniture, ensuring the completeness of the robot's scan of support parts in different directions and improving the robot's adaptability to the regularly distributed support parts on the bottom of the target furniture.

[0054] Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications can still be made to the specific embodiments of the present invention or equivalent substitutions can be made to some technical features without departing from the spirit of the technical solution of the present invention, and all such modifications and substitutions should be covered within the scope of the technical solution claimed in the present invention.

Claims

1. A robot motion control method for laser scanning the support portion at the bottom of furniture, characterized in that, The robot motion control method includes: When the robot walks into the detectable area of ​​the target furniture without colliding with the support at the bottom of the furniture, it turns around the target furniture while keeping the robot's laser sensor scanning the support. At the same time, the robot distinguishes different support parts at the bottom of the target furniture based on the ranging data and orientation features fed back by the laser sensor. The area where the laser sensor keeps detecting the support is the detectable area of ​​the target furniture. Whenever the robot turns around the target furniture at a preset angle, the laser sensor scans the local contour lines around all the support parts on the bottom of the target furniture; until the total angle of the robot turning around the target furniture is greater than or equal to 360 degrees, the robot connects the local contour lines scanned around the same support part to form a closed contour around the same support part, so as to connect the closed contours around each support part on the bottom of the target furniture respectively. The target furniture is furniture with multiple support parts installed on the bottom and multiple support parts in contact with the ground, and the ground is the walking surface of the robot. When the robot turns around the target furniture at a preset angle, the robot is located on the central axis of the target furniture. The robot uses its laser sensor to simultaneously detect the support parts on both sides of the central axis of the bottom of the target furniture. Among the support parts on each side of the central axis, the number of local contour lines scanned by the laser sensor in the support part closest to the robot is greater than or equal to the number of local contour lines scanned by the laser sensor in the support part farthest from the robot. At the bottom of the target furniture, one half of the support part and the other half of the support part are symmetrically installed based on the central axis.

2. The robot motion control method according to claim 1, characterized in that, The method by which the robot distinguishes different support parts at the bottom of the target furniture based on ranging data and orientation features fed back by the laser sensor includes: When the robot turns around the target furniture at a preset angle, it uses the direction of the ranging data obtained by its laser sensor to distinguish the support part located on the left side of the central axis of the target furniture and the support part located on the right side of the central axis of the target furniture. On each side of the central axis, based on the ranging data of each local contour line fed back by the laser sensor, the robot distinguishes each support part one by one from near to far along the central axis, including distinguishing the support part closest to the robot and its local contour line scanned by the laser sensor and the support part farthest from the robot and its local contour line scanned by the laser sensor.

3. The robot motion control method according to claim 1, characterized in that, Within the detectable area of ​​the target furniture, among the support parts scanned by the robot before turning around the target furniture at the preset angle and the support parts scanned by the robot after turning around the target furniture at the preset angle, there are identical support parts that are scanned repeatedly; in the identical support parts that are scanned repeatedly, the local contour lines scanned by the robot after turning around the target furniture at the preset angle include the local contour lines that were not scanned by the robot before turning around the target furniture at the preset angle or the local contour lines that were scanned by the robot before turning around the target furniture at the preset angle.

4. The robot motion control method according to claim 1, characterized in that, The robot turns around the target furniture by rotating clockwise or counterclockwise around the center of the target furniture. The robot's rotation radius is greater than half of the maximum outline of the target furniture, so that the robot reaches the central axis of the target furniture every time it turns around the target furniture at the preset angle. Each time the robot turns around the target furniture at the preset angle, the robot updates the central axis set in the direction of the target furniture pointing to the robot's current position to the central axis of the target furniture in which the robot is currently located. Before the robot turns around the target furniture, it is positioned on the central axis of the target furniture in one direction. The robot did not collide with the support parts at the bottom of the target furniture before or after turning around it, and used laser sensors to scan the local contour lines around all the support parts at the bottom of the target furniture; the central axis of the target furniture in each direction and the center of the target furniture were pre-marked by the robot.

5. The robot motion control method according to claim 4, characterized in that, Each time the robot turns around the target furniture at the preset angle, the robot rotates its body to adjust its front side to face the target furniture, so that the robot's central axis is parallel to the central axis of the target furniture in which the robot is currently located; The laser sensor is mounted on the top of the front side of the robot's body.

6. The robot motion control method according to claim 1, characterized in that, The target furniture has a rectangular projection shape on the ground; four support parts are symmetrically installed on the bottom of the target furniture, located at the four corner points of the rectangle respectively; The closed contour around the support portion is represented by the circumscribed rectangle that encloses the periphery of the support portion, so that each local contour line continuously scanned around the same support portion is the four sides of the rectangle.

7. The robot motion control method according to claim 6, characterized in that, The preset angle is 90 degrees. After the robot makes four consecutive turns around the target furniture with the preset angle as the rotation step size, the total angle turned is equal to 360 degrees.

8. The robot motion control method according to claim 6, characterized in that, The robot rotates clockwise or counterclockwise around the center of the target furniture, and simultaneously scans at least one local contour line aligned with the robot in each support part on both sides of the central axis of the target furniture through a laser sensor; in each side of the central axis of the target furniture, the at least one local contour line aligned with the robot in each support part includes a pair of perpendicularly arranged local contour lines scanned by the laser sensor in the support part closest to the robot, and a local contour line perpendicular to the central axis of the machine body scanned by the laser sensor in the support part farthest from the robot.

9. The robot motion control method according to claim 8, characterized in that, The number of local contour lines scanned by the robot at each preset angle when it turns around the target furniture is equal; Before and after the robot turns around the target furniture by the preset angle, it repeatedly scans the same support part. In the support part closest to the robot's current position in the repeated scan, the local contour line scanned after the current turn of the preset angle is a local contour line that was not scanned before the current turn of the preset angle. In the support part farthest from the robot's current position in the repeated scan, the local contour line scanned before the current turn of the preset angle includes the local contour line scanned after the current turn of the preset angle.

Images