Control method of cleaning robot, robot and computer-readable storage medium

Abstract

This invention provides a control method for a cleaning robot. The cleaning robot is equipped with a LiDAR and a depth camera. The control method includes: acquiring point cloud data collected by the LiDAR; detecting wall features using the point cloud data collected by the LiDAR; acquiring image data collected by the depth camera; fusing the image data collected by the depth camera with the point cloud data collected by the LiDAR to obtain fused data; comparing the differences between the point cloud data collected by the LiDAR and the fused data; obtaining perception data based on wall features, the differences between the point cloud data collected by the LiDAR and the fused data; and controlling the robot to execute corresponding cleaning motion strategies based on the perception data. Using the technical solution of this invention, the robot has a stronger adaptability to complex environments and can better meet the requirements of detailed wall edge cleaning and collision-free operation during edge cleaning.

Description

Technical Field

[0001] This invention relates generally to the field of robotics, and more particularly to a control method for a cleaning robot, a robot, and a computer-readable storage medium. Background Technology

[0002] Existing cleaning robots, when performing edge-to-wall cleaning on solid walls, mostly employ simple sensing and detection solutions such as edge-mounted infrared sensors and contact sensors. The resulting motion control strategies for edge-to-wall cleaning are relatively simplistic. During cleaning, collisions are required to achieve wall-hugging cleaning and obstacle avoidance, which is unreliable and unsuitable for large commercial cleaning robots. Furthermore, there is a lack of effective multi-sensor fusion perception solutions for solid walls, making them unsuitable for relatively complex wall edge environments with obstacles. A single motion control strategy cannot meet the requirements for close-range cleaning in complex wall boundary environments. Therefore, there is an urgent need for a technical solution that can meet the requirements of close-range wall-hugging cleaning and collision-free operation in various complex solid wall environments.

[0003] The contents of the background technology section are merely the technologies known to the inventors and do not necessarily represent the existing technologies in this field. Summary of the Invention

[0004] To address one or more of the problems existing in the prior art, this invention provides a control method for a cleaning robot. This method, based on a fusion perception scheme and combined with a comprehensive cleaning motion control strategy, enables the robot to perform close-range, collision-free cleaning in various complex solid wall environments. The cleaning robot is equipped with a LiDAR and a depth camera, and the control method includes:

[0005] Acquire the point cloud data collected by the lidar;

[0006] Wall features are detected using point cloud data collected by the lidar.

[0007] Acquire image data captured by the depth camera;

[0008] The image data acquired by the depth camera is fused with the point cloud data acquired by the lidar to obtain fused data;

[0009] Compare the differences between the point cloud data collected by the lidar and the fused data;

[0010] Perception data is obtained based on the wall features, the differences between the point cloud data collected by the lidar, and the fused data; and

[0011] The robot is controlled to execute corresponding cleaning motion strategies based on the perceived data.

[0012] According to one aspect of the present invention, the step of detecting wall features using point cloud data acquired by the lidar includes: detecting the wall features by performing feature matching with preset structured features in the point cloud acquired by the lidar.

[0013] According to one aspect of the invention, the step of comparing the difference between the point cloud data acquired by the lidar and the fused data includes: determining a first distance between the robot and the wall based on the point cloud data at a desired pose, determining a second distance between the robot and the wall based on the fused data, and determining a first difference between the first distance and the second distance.

[0014] According to one aspect of the present invention, the step of comparing the difference between the point cloud data acquired by the lidar and the fused data further includes: obtaining a first current pose of the robot based on the point cloud data, and determining the heading difference between a first heading angle of the first current pose and a second heading angle of the desired pose.

[0015] According to one aspect of the present invention, the step of obtaining perception data based on the wall features, the point cloud data collected by the lidar, and the fused data includes: determining whether the wall features are preset features, determining whether the first difference is less than a first threshold, and determining whether the heading difference is less than a second threshold, obtaining a determination result, and obtaining the perception data based on the determination result.

[0016] According to one aspect of the present invention, the step of obtaining the sensing data based on the determination result includes: when it is determined that the wall feature is the continuous straight line feature or the continuous curve feature, the first difference is less than the first threshold, and the heading difference is less than the second threshold, the point cloud data collected by the lidar is used as the sensing data; when it is determined that the wall feature does not satisfy the continuous straight line feature or the continuous curve feature, the first difference is not less than the first threshold, or the heading difference is not less than the second threshold, the fused data is used as the sensing data.

[0017] According to one aspect of the present invention, the step of controlling the robot to perform a corresponding cleaning motion strategy based on the sensing data includes: adaptively adjusting the robot's pose and / or speed based on the sensing data to control the robot to maintain a first preset distance between its wall-side and the wall while performing the cleaning task.

[0018] According to one aspect of the present invention, the step of controlling the robot to execute a corresponding cleaning motion strategy based on the perception data further includes: detecting whether there are obstacles in front of the robot, to the side front, and to the opposite side of the wall, respectively, based on the perception data; if there are obstacles, controlling the robot to maintain a second preset distance from the obstacle in front, a third preset distance from the obstacle to the side front, and a fourth preset distance from the obstacle to the opposite side of the wall, wherein the fourth preset distance is greater than the first preset distance.

[0019] The present invention also provides a robot, comprising:

[0020] case;

[0021] Mobile chassis with a walking mechanism;

[0022] Sensors, including a lidar and a depth camera, are mounted on the robot and configured to collect information about the robot's surrounding environment;

[0023] Cleaning devices for sweeping; and

[0024] The processing unit, coupled to the walking mechanism, the lidar, and the depth camera, is configured to execute the control method described above.

[0025] The present invention also provides a computer-readable storage medium including computer-executable instructions stored thereon, wherein the executable instructions, when executed by a processor, implement the control method described above.

[0026] By adopting the technical solution of this invention, the robot can perform close-range wall cleaning on various complex solid walls, has a very good adaptability to environmental changes, can complete high-quality edge cleaning in different environments, and can achieve good detection and obstacle avoidance effects on surrounding obstacles. This better meets the requirements of wall cleaning details and collision-free operation in the edge cleaning process, and further improves the robot's safety. Attached Figure Description

[0027] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0028] Figure 1 A flowchart of a control method for a cleaning robot according to an embodiment of the present invention is shown;

[0029] Figure 2 A schematic diagram of a robot according to an embodiment of the present invention is shown;

[0030] Figure 3a and Figure 3b Schematic diagrams of lidar point clouds showing wall features as continuous straight lines and continuous curves according to an embodiment of the present invention are shown respectively.

[0031] Figure 4 A schematic diagram illustrating a robot constraining safe distances to multiple directions around it according to a preferred embodiment of the present invention is shown; and

[0032] Figure 5 A schematic diagram of a control method according to a preferred embodiment of the present invention is shown. Detailed Implementation

[0033] In the following description, only certain exemplary embodiments are briefly described. As those skilled in the art will recognize, the described embodiments can be modified in various ways without departing from the spirit or scope of the invention. Therefore, the drawings and description are considered to be exemplary in nature and not restrictive.

[0034] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention 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 the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0035] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection, an electrical connection, or a connection that allows for communication; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0036] In this invention, unless otherwise explicitly specified and limited, "above" or "below" the second feature can include direct contact between the first and second features, or contact between the first and second features through another feature between them. Furthermore, "above," "over," and "on top" the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature includes the first feature directly above or diagonally above the second feature, or simply indicates that the first feature is at a lower horizontal level than the second feature.

[0037] The following disclosure provides many different embodiments or examples for implementing various structures of the invention. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the invention. Furthermore, reference numerals and / or letters may be repeated in different examples; such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed. In addition, examples of various specific processes and materials are provided in this invention, but those skilled in the art will recognize the application of other processes and / or the use of other materials.

[0038] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0039] This invention provides a control method for a cleaning robot. This control method is based on a fusion perception scheme and combined with a complete cleaning motion control strategy, which enables the robot to perform close-range, collision-free cleaning in various complex solid wall environments. It can also achieve good detection and obstacle avoidance effects for surrounding obstacles, better meeting the requirements of the robot for wall-hugging cleaning details and collision-free cleaning during the cleaning process. The details are described below.

[0040] Figure 1 A flowchart of a control method 100 for a cleaning robot according to an embodiment of the present invention is shown. Figure 2 A schematic diagram of a robot according to an embodiment of the present invention is shown. Figure 5 A schematic diagram of a control method 100 according to a preferred embodiment of the present invention is shown below; please refer to it as well. Figure 1 , Figure 2 and Figure 5 The robot is equipped with a lidar and a depth camera. The control method 100 includes steps S101 to S107. The steps of the method 100 are described in detail below.

[0041] In step S101, the point cloud data collected by the lidar is acquired.

[0042] LiDAR is a commonly used ranging sensor, which includes a transmitting unit, a receiving unit, and a processing unit. The transmitting unit is configured to emit a laser beam to detect surrounding objects. The receiving unit is used to receive the echo reflected from the laser beam incident on the obstacle and convert it into an electrical signal. The electrical signal is processed by the processing unit to generate several points, each of which carries various data of the detected object, such as reflectivity and coordinate information. The data set of several points constitutes point cloud data.

[0043] According to a preferred embodiment of the present invention, the lidar can be a single-line lidar, which includes only a single transmitting unit and a single receiving unit, offering both high sensitivity and low cost. When the robot receives an edge-cleaning task, it can determine its current location based on a known map and navigate to a preset position near the target wall. While moving along the wall within a certain range, it can emit a laser beam to detect the wall and obtain point cloud data of the wall. It should be noted that the present invention does not limit the type of lidar; in other embodiments, the lidar can also be a multi-line lidar, and the specific choice can be made according to the actual situation.

[0044] In step S102, wall features are detected using point cloud data collected by the lidar.

[0045] According to a preferred embodiment of the present invention, wall features can be detected in the point cloud acquired by lidar by feature matching with preset structured features. The present invention does not limit the specific method of feature matching. In some preferred embodiments, the ICP algorithm can be used for feature matching to determine whether the wall features are continuous straight lines or continuous curves with a certain curvature. It should be understood that before feature matching, the point cloud can be preprocessed, such as by filtering, to remove noise, which is beneficial for obtaining more accurate judgment results and improving processing efficiency. Figure 3a and Figure 3b The following are exemplary schematic diagrams of lidar point clouds showing wall features as continuous straight lines and continuous curves according to a preferred embodiment of the present invention.

[0046] In step S103, image data acquired by the depth camera is obtained.

[0047] In some preferred embodiments, the lidar of the present invention can be a single-line lidar. A single-line lidar does not have a longitudinal field of view. When it scans and detects along the horizontal direction, the longitudinal coordinates of the point cloud data obtained are basically the same. By acquiring images of the detected object through a depth camera, image data can be obtained. The three-dimensional coordinates of the detected object can be obtained through the image data, which helps to make up for the lack of longitudinal field of view of the single-line lidar. Furthermore, the depth of the detected object can also be obtained through the image data, which is convenient for subsequent processing.

[0048] In step S104, the image data acquired by the depth camera and the point cloud data acquired by the lidar are fused to obtain fused data.

[0049] In some preferred embodiments, an extended Kalman filter algorithm can be used to perform post-fusion of the image data acquired by the depth camera and the point cloud data acquired by the lidar. Specifically, the image data acquired by the depth camera and the point cloud data acquired by the lidar can be unified into the robot's own coordinate system for data statistics to obtain fused data, thereby obtaining more comprehensive information about the detected object. It should be understood that, in addition to using the extended Kalman filter algorithm for post-fusion, other algorithms can also be used for pre-fusion. This invention does not impose limitations and depends on the actual situation.

[0050] In step S105, the difference between the point cloud data collected by the lidar and the fused data is compared.

[0051] According to a preferred embodiment of the present invention, at the desired pose P0(X0,Y0,θ0), a first distance d1 between the robot and the wall can be determined based on the point cloud data, and a second distance d2 between the robot and the wall can be determined based on the fused data. A first difference Δd = |d1-d2| between the first distance d1 and the second distance d2 is also determined. The desired pose can be a point set during the mapping process. The cleaning robot obtains its current location based on real-time environmental features and the global map, and uses the desired pose point as the target point to move forward and reach it. Due to the error and accuracy range of electronic components, there is a deviation between the actual point reached by the robot and the desired point. Therefore, by using the wall distance measurement results from sensors as a basis for implementing a wall-hugging cleaning strategy, the dependence on positioning can be greatly reduced.

[0052] Furthermore, based on the current positioning, the robot's first current pose P1(X1,Y1,θ1) can be obtained from the point cloud data, and the heading difference Δθ between the first heading angle θ1 of the first current pose and the second heading angle θ0 of the desired pose P0(X0,Y0,θ0) can be determined. The adjustment of the robot's posture needs to consider the desired wall-hugging distance and the robot's desired posture relative to the wall shape, i.e., the second heading angle θ0. The desired posture refers to the optimal desired wall-hugging posture obtained by detecting the wall shape along the robot's side. Optionally, the desired posture can be the same as the tangent orientation of the nearest wall segment of a preset length, thereby matching different shape features.

[0053] In step S106, perception data is obtained based on the wall features, the differences between the point cloud data collected by the lidar and the fused data.

[0054] According to a preferred embodiment of the present invention, it can be determined whether the wall feature is a preset feature, whether the first difference is less than a first threshold, and whether the heading difference is less than a second threshold, to obtain a determination result, and the sensing data is obtained based on the determination result. Specifically, when it is determined that the wall feature is the continuous straight line feature or the continuous curve feature, the first difference Δd is less than the first threshold (e.g., 5cm), and the heading difference Δθ is less than the second threshold (e.g., 5°), it indicates that the point cloud data collected by the lidar is relatively accurate, and the point cloud data collected by the lidar can be used as the sensing data; conversely, when it is determined that the wall feature does not satisfy the continuous straight line feature or the continuous curve feature, the first difference Δd is not less than the first threshold (e.g., 5cm), or the heading difference Δθ is not less than the second threshold (e.g., 5°), it indicates that the point cloud data collected by the lidar is not accurate enough, and in this case, the fused data can be used as the sensing data.

[0055] In step S107, the robot is controlled to execute a corresponding cleaning motion strategy based on the perceived data.

[0056] According to a preferred embodiment of the present invention, the robot's pose and / or speed can be adaptively adjusted based on the perceived data to control the robot to maintain a first preset distance between its wall-side and the wall while performing a cleaning task. Specifically, for example, when point cloud data collected by a LiDAR is used as perceived data, the LiDAR point cloud data can be directly output to the motion control layer to control the robot to maintain a first preset distance S1 (e.g., 10cm) between its wall-side and the wall while performing a cleaning task. Conversely, when fused data is used as perceived data, the fused data is output to the motion control layer to perform real-time obstacle avoidance based on the fused data, and to control the robot to perform a cleaning task based on the cleanable area defined by the cost map. The advantage of the cost map output data is that it can be easily integrated with virtual walls and no-go zones, thereby reliably defining the cleanable area. Figure 4 As exemplified, taking a straight wall as an example, the robot maintains a first preset distance S1 (e.g., 10cm) from the wall while performing an edge cleaning task. It should be understood that although not shown in the accompanying drawings, the process is similar when the wall is curved; the robot also maintains a first preset distance S1 (e.g., 10cm) from the wall while performing the edge cleaning task.

[0057] In some preferred embodiments, the robot's edge-cleaning function can be implemented using a control method based on lateral offset (including but not limited to PID control algorithms). Wall-following cleaning motion control refers to adjusting the robot's posture and speed to achieve real-time adjustment of the robot's distance from the wall without collisions during movement. It should be noted that the lateral offset refers to the distance control between the robot's side (e.g., left or right side) and obstacles, walls, etc., that is, the control of the robot's lateral distance along the edge. In some preferred embodiments, a cleaning trajectory can also be planned based on a known map, and the robot can be controlled to follow the trajectory for edge-cleaning. This trajectory can be referenced... Figure 4 The dashed line in the middle.

[0058] Furthermore, adjusting the robot's posture requires considering the desired wall-hugging distance and the desired posture of the robot relative to the wall shape. The desired posture refers to the optimal wall-hugging posture obtained by detecting and matching the robot's position against the wall shape along its edge. Therefore, the constraints for posture adjustment are the wall-hugging distance deviation and its variation, the deviation between the desired posture and the current posture and its variation, thereby ensuring that the robot always maintains a preset distance while cleaning along the wall and keeps the posture output stable.

[0059] In some preferred embodiments, the robot's speed needs to be constrained during wall-following cleaning to ensure its safety. For example, when the robot performs edge-cleaning along a wall with straight lines, its maximum linear speed v is limited. maxThe speed can be 0.2 m / s. When the robot turns (e.g., makes a right-angle turn), the penalty for the distance between the robot and the wall will be increased, thus constraining the linear velocity. This ensures that the robot will not scrape against the wall due to being too close during the turn, thereby guaranteeing the robot's operational safety. According to a preferred embodiment of the present invention, the range of the robot's turning speed v can be a weight multiplied by the maximum linear velocity, i.e., v = k * v max The weight k is related to the distance between the robot and the wall when the robot turns. When the robot is close to the wall when turning, the weight k should be adjusted as small as possible to minimize the robot's turning speed and ensure safe operation. Conversely, when the robot is far from the wall when turning, the weight k can be appropriately increased. It should be noted that the weight k ranges from 0 to 1. According to a preferred embodiment of the present invention, when k = 0.1, the turning speed v is minimized, i.e., the minimum turning speed v0. min =0.1*v max It should be understood that the above embodiments are for illustrative purposes only and do not constitute a limitation of the present invention. In practical applications, appropriate adjustments can be made according to the actual situation, and these adjustments are all within the protection scope of the present invention.

[0060] The above embodiments illustrate how to improve robot operation safety by constraining the robot's travel speed. To further enhance robot operation safety, when the robot is performing a cleaning task along a wall, the presence of obstacles in multiple directions around the robot should also be considered. This ensures that the robot maintains a safe distance from obstacles in multiple directions during the cleaning process to prevent collisions and damage. These multiple directions include, but are not limited to, the robot's direct front, its front side, and the opposite side from the wall. This will be described in detail below.

[0061] According to a preferred embodiment of the present invention, the presence of obstacles in front of the robot, to its front side, and on the opposite side of the wall can be detected based on the perception data. If an obstacle exists, refer to... Figure 4 The robot can be controlled to maintain a second preset distance S2 from an obstacle directly in front, a third preset distance S3 from an obstacle to the side in front, and a fourth preset distance S4 from an obstacle on the opposite side of the wall. The fourth preset distance S4 (e.g., 12cm to 15cm) is greater than the first preset distance S1 (e.g., 10cm). In addition, there is no necessary relationship between the other preset distances, and they can all be adjusted appropriately according to the actual situation.

[0062] In some preferred embodiments, the fused data should also fully consider the blind spots of sensors such as depth cameras and implement appropriate strategies. For example, by comprehensively analyzing the detection results of multiple sensors, corresponding decisions can be made.

[0063] In some other preferred embodiments, when it is detected that the safe distance is not met in one or more directions around the robot, such as when the safe distance S4 is not met on the opposite side of the wall, the collision condition is met, and the robot can be triggered to hug the wall and bypass the obstacle or re-plan the path to avoid the obstacle in order to prevent a collision.

[0064] In some preferred embodiments, the control method 100 further includes: determining virtual walls and no-sweep areas through a cost map, so as to control the robot to avoid the virtual walls and no-sweep areas when performing tasks, so as to ensure the safety of the robot.

[0065] The control method 100 has been described in detail above. This invention also provides a robot 20, such as... Figure 2 As shown, the robot 20 includes a housing 21; a mobile chassis having a walking mechanism 210; sensors 22 including a lidar 221 and a depth camera 223, the lidar 221 and the depth camera 223 being mounted on the robot 20 and configured to collect information about the robot's surrounding environment; a cleaning device 30, such as a cleaning brush, for cleaning; and a processing unit (not shown) coupled to the walking mechanism 210, the lidar 221 and the depth camera 223, configured to execute the control method 100 as described above.

[0066] According to a preferred embodiment of the present invention, the lidar 221 can be disposed at the opening of the robot shell, thereby facilitating the emission of laser signals to detect surrounding objects.

[0067] According to a preferred embodiment of the present invention, the walking mechanism 210 is provided with at least two sets of drive wheels 211. Optionally, the robot may also include at least two sets of driven wheels, with one set of drive wheels corresponding to one set of driven wheels. At least one set of driven wheels is used as a left driven wheel, and at the same time, at least one set of driven wheels is used as a right driven wheel. The left and right driven wheels are used to assist the left and right drive wheels in driving the robot to move, so as to reduce the load pressure on the drive wheels 211.

[0068] According to a preferred embodiment of the present invention, the sensor 22 further includes an odometer 222, which can obtain the linear velocity and angular velocity of the robot using the rotational speed values ​​of at least two sets of drive wheels 211, substitute them into the kinematic model of the mobile robot, deduce the current pose of the robot, that is, the position and heading angle information, thereby controlling the speed and / or pose of the robot.

[0069] This invention does not limit the specific type of robot. For example, for differential motion robots, the wall-hugging distance for wall cleaning can be directly achieved by adjusting the robot's posture and longitudinal speed in real time. For omnidirectional wheel robots, the control of the robot's lateral speed also needs to be considered, and specific adjustments can be made according to the actual situation.

[0070] It should be noted that the sensors of this invention are not limited to lidar, depth cameras, and odometers, but may also include other sensors, all of which are within the scope of protection of this invention.

[0071] The present invention also provides a computer-readable storage medium including computer-executable instructions stored thereon, wherein the executable instructions, when executed by a processor, implement the control method 100 as described above.

[0072] In summary, the technical solution of this invention has been described in detail. Employing this technical solution enables the robot to perform edge cleaning effectively without collisions, achieving the expected cleaning effect for detailed cleaning of solid walls. Furthermore, the edge cleaning function exhibits excellent adaptability to environmental changes, enabling high-quality edge cleaning in various environments. It can detect obstacles in multiple directions around the robot and maintain a safe distance from them. Additionally, it restricts cleaning of virtual walls, further enhancing the safety of the edge cleaning function. In conclusion, compared to existing technologies, the technical solution of this invention can meet the requirements of close-range, collision-free wall cleaning in various complex solid wall environments, and its rich motion strategies improve the robot's adaptability and robustness.

[0073] In some preferred embodiments, the computer-readable storage medium may take the form of any combination of one or more computer-readable media. The computer-readable storage medium may be, for example, but not limited to, electrical, magnetic, optical, or semiconductor forms or devices, and more specific examples (a non-exhaustive list) include: electrical connections having one or more wires, portable computer hard disks, hard disks, random access memory (RAM), non-volatile random access memory (NVRAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0074] In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used or combined with an instruction execution system, apparatus, or device. The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. This invention is not limited to any particular type; the specific definition depends on the circumstances.

[0075] It should be noted that this specification provides the operational steps of the methods described in the embodiments or diagrams, but based on conventional or non-inventive labor, more or fewer operational steps may be included. The order of steps listed in the embodiments is merely one possible execution order among many steps and does not represent the only execution order. In actual system or device products, the methods shown in the embodiments or flowcharts can be executed sequentially or in parallel.

[0076] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A control method for a cleaning robot, the cleaning robot being equipped with a lidar and a depth camera, the control method comprising: Acquire the point cloud data collected by the lidar; Wall features are detected using point cloud data collected by the lidar. Acquire image data captured by the depth camera; The image data acquired by the depth camera is fused with the point cloud data acquired by the lidar to obtain fused data; Compare the differences between the point cloud data collected by the lidar and the fused data; Perception data is obtained based on the wall features, the differences between the point cloud data collected by the lidar and the fused data, including: obtaining perception data based on whether the wall features are preset features, the first difference determined by the point cloud data collected by the lidar and the fused data at the desired pose, and the difference in heading determined by the point cloud data at the robot's first current pose and the desired pose. and The robot is controlled to execute corresponding cleaning motion strategies based on the perceived data.

2. The control method according to claim 1, wherein the step of detecting wall features using point cloud data acquired by the lidar includes: The wall features are detected by matching the point cloud acquired by the lidar with preset structured features.

3. The control method according to claim 1, wherein the step of comparing the difference between the point cloud data acquired by the lidar and the fused data includes: At the desired pose, a first distance between the robot and the wall is determined based on the point cloud data, a second distance between the robot and the wall is determined based on the fused data, and a first difference between the first distance and the second distance is determined.

4. The control method according to claim 3, wherein the step of comparing the difference between the point cloud data acquired by the lidar and the fused data further includes: The robot's first current pose is obtained based on the point cloud data, and the heading difference between the first heading angle of the first current pose and the second heading angle of the desired pose is determined.

5. The control method according to any one of claims 2-4, wherein the step of obtaining perception data based on the wall features, the point cloud data acquired by the lidar, and the fused data includes: Determine whether the wall feature is a preset feature, determine whether the first difference is less than a first threshold, and determine whether the heading difference is less than a second threshold to obtain a determination result, and obtain the perception data based on the determination result.

6. The control method according to claim 5, wherein the step of obtaining the sensing data based on the determination result includes: When the wall feature is determined to be a continuous straight line feature or a continuous curve feature, the first difference is less than the first threshold, and the heading difference is less than the second threshold, the point cloud data collected by the lidar is used as the perception data; when the wall feature is determined not to satisfy the continuous straight line feature or the continuous curve feature, the first difference is not less than the first threshold, or the heading difference is not less than the second threshold, the fused data is used as the perception data.

7. The control method according to claim 6, wherein the step of controlling the robot to execute a corresponding cleaning motion strategy based on the perceived data includes: Based on the perceived data, the robot's pose and / or speed are adaptively adjusted to control the robot to maintain a first preset distance from the wall side while performing the cleaning task.

8. The control method according to claim 7, wherein the step of controlling the robot to execute a corresponding cleaning motion strategy based on the perceived data further includes: Based on the perception data, the robot is detected to have obstacles in front of it, to its side, and to the opposite side of the wall. If obstacles are present, the robot is controlled to maintain a second preset distance from the obstacle in front of it, a third preset distance from the obstacle to its side, and a fourth preset distance from the obstacle to the opposite side of the wall, wherein the fourth preset distance is greater than the first preset distance.

9. A robot, comprising: case; Mobile chassis with a walking mechanism; Sensors, including a lidar and a depth camera, are mounted on the robot and configured to collect information about the robot's surrounding environment; Cleaning equipment used for sweeping; and The processing unit, coupled to the walking mechanism, the lidar, and the depth camera, is configured to perform the control method as described in any one of claims 1-8.

10. A computer-readable storage medium comprising computer-executable instructions stored thereon, the executable instructions, when executed by a processor, implementing the control method as described in any one of claims 1-8.

Images