WO2022143285A1 - 扫地机器人及其测距方法、装置以及计算机可读存储介质 - Google Patents

扫地机器人及其测距方法、装置以及计算机可读存储介质 Download PDF

Info

Publication number
WO2022143285A1
WO2022143285A1 PCT/CN2021/139948 CN2021139948W WO2022143285A1 WO 2022143285 A1 WO2022143285 A1 WO 2022143285A1 CN 2021139948 W CN2021139948 W CN 2021139948W WO 2022143285 A1 WO2022143285 A1 WO 2022143285A1
Authority
WO
WIPO (PCT)
Prior art keywords
depth information
obstacle
depth
sweeping robot
target
Prior art date
Application number
PCT/CN2021/139948
Other languages
English (en)
French (fr)
Inventor
杨勇
宫海涛
罗志康
Original Assignee
深圳市杉川机器人有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 深圳市杉川机器人有限公司 filed Critical 深圳市杉川机器人有限公司
Publication of WO2022143285A1 publication Critical patent/WO2022143285A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
    • A47L11/40—Parts or details of machines not provided for in groups A47L11/02 - A47L11/38, or not restricted to one of these groups, e.g. handles, arrangements of switches, skirts, buffers, levers
    • A47L11/4061—Steering means; Means for avoiding obstacles; Details related to the place where the driver is accommodated
    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L11/00—Machines for cleaning floors, carpets, furniture, walls, or wall coverings
    • A47L11/24—Floor-sweeping machines, motor-driven
    • A—HUMAN NECESSITIES
    • A47—FURNITURE; DOMESTIC ARTICLES OR APPLIANCES; COFFEE MILLS; SPICE MILLS; SUCTION CLEANERS IN GENERAL
    • A47L—DOMESTIC WASHING OR CLEANING; SUCTION CLEANERS IN GENERAL
    • A47L2201/00—Robotic cleaning machines, i.e. with automatic control of the travelling movement or the cleaning operation
    • A47L2201/04—Automatic control of the travelling movement; Automatic obstacle detection

Definitions

  • the present application relates to the technical field of visual matching, and in particular, to a cleaning robot and a distance measurement method, device and computer-readable storage medium thereof.
  • the sweeping robot In the process of cleaning tasks, the sweeping robot needs to obtain the object information in the cleaning environment to control its walking path, so as to ensure that the sweeping robot will not collide during cleaning.
  • the present application proposes a cleaning robot and a distance measuring method, device, and computer-readable storage medium to solve the problem that the cleaning robot in the prior art cannot accurately and accurately identify the distance between the cleaning robot and the obstacle.
  • the present application proposes a distance measurement and obstacle avoidance method for a sweeping robot, wherein the sweeping robot is equipped with dual TOF cameras, and the distance measurement and obstacle avoidance method for the sweeping robot includes: when the sweeping robot performs a cleaning task, The dual TOF cameras obtain depth information of obstacles; perform a binocular stereo matching operation based on the depth information to obtain target depth information; and calculate the distance between the obstacle and the sweeping robot through the target depth information.
  • the depth information of the obstacle is obtained based on the dual TOF cameras, and then the depth information is performed by performing a binocular stereo matching operation to obtain the target depth information,
  • the distance between the obstacle and the sweeping robot can be determined through the target depth information.
  • the information of the obstacles in the cleaning task of the sweeping robot is obtained through the dual TOF cameras, and the distance between the camera and the obstacle is obtained by using the binocular stereo matching operation, which can solve the problem that the binocular RGB cannot be accurately determined in the case of exposure.
  • the distance between the sweeping robot and the obstacle, and the single TOF has a large error when calculating the distance between the sweeping robot and the obstacle at a close distance.
  • the obstacle information is improved by the dual TOF cameras.
  • the calculation of the sweeping robot and the obstacle The precision and accuracy of distances between objects.
  • the present application proposes a cleaning robot device, comprising: an acquisition module for acquiring a depth map of obstacles based on the dual TOF cameras when the cleaning robot performs a cleaning task; a matching module for acquiring a depth map of obstacles based on the dual TOF cameras; The depth information in the depth map is subjected to binocular stereo matching to obtain target depth information; the determination module is used for calculating the distance between the obstacle and the sweeping robot through the target depth information.
  • the depth information of the obstacle is obtained based on the dual TOF cameras, and then the depth information performs a binocular stereo matching operation to obtain the target depth information.
  • the distance between the obstacle and the cleaning robot may be determined.
  • the information of the obstacles in the cleaning task of the sweeping robot is obtained through the dual TOF cameras, and the distance between the camera and the obstacle is obtained by using the binocular stereo matching operation, which can solve the problem that the binocular RGB cannot be accurately determined in the case of exposure.
  • the distance between the sweeping robot and the obstacle, and the single TOF has a large error when calculating the distance between the sweeping robot and the obstacle at a close distance.
  • the obstacle information is improved by the dual TOF cameras.
  • the calculation of the sweeping robot and the obstacle The precision and accuracy of distances between objects.
  • the present application provides a cleaning robot, which includes a processor, a memory, and a distance measuring and obstacle avoidance program of the cleaning robot that is stored in the memory and can run on the processor.
  • the distance measuring and obstacle avoidance program of the sweeping robot is executed by the processor, the steps of the above-mentioned method for measuring the distance and obstacle avoidance of the sweeping robot are realized.
  • the depth information of the obstacle is obtained based on the dual TOF cameras, and then the depth information performs a binocular stereo matching operation to obtain the target depth information, and the target depth information can be obtained.
  • the information of the obstacles in the cleaning task of the sweeping robot is obtained through the dual TOF cameras, and the distance between the camera and the obstacle is obtained by using the binocular stereo matching operation, which can solve the problem that the binocular RGB cannot be accurately determined in the case of exposure.
  • the distance between the sweeping robot and the obstacle, and the single TOF has a large error when calculating the distance between the sweeping robot and the obstacle at a close distance.
  • the obstacle information is improved by the dual TOF cameras.
  • the calculation of the sweeping robot and the obstacle The precision and accuracy of distances between objects.
  • the present application provides a computer-readable storage medium on which a ranging and obstacle-avoidance program of a sweeping robot is stored, and when the ranging and obstacle-avoidance program of the sweeping robot is executed by a processor The steps of implementing the above method for distance measurement and obstacle avoidance of the sweeping robot.
  • the depth information of the obstacle is obtained based on the dual TOF cameras, and then the depth information performs a binocular stereo matching operation to obtain the target depth information.
  • the information can be used to determine the distance between the obstacle and the cleaning robot.
  • the information of the obstacles in the cleaning task of the sweeping robot is obtained through the dual TOF cameras, and the distance between the camera and the obstacle is obtained by using the binocular stereo matching operation, which can solve the problem that the binocular RGB cannot be accurately determined in the case of exposure.
  • the distance between the sweeping robot and the obstacle, and the single TOF has a large error when calculating the distance between the sweeping robot and the obstacle at a close distance.
  • the obstacle information is improved by the dual TOF cameras.
  • the calculation of the sweeping robot and the obstacle The precision and accuracy of distances between objects.
  • FIG. 1 is a schematic structural diagram of a terminal of a hardware operating environment involved in a solution according to an embodiment of the present application
  • FIG. 2 is a schematic flowchart of an embodiment of a method for distance measurement and obstacle avoidance of a sweeping robot according to the present application;
  • FIG. 3 is a schematic flowchart of another embodiment of the method for distance measurement and obstacle avoidance of the sweeping robot according to the present application;
  • FIG. 4 is a schematic flowchart of another embodiment of the method for distance measurement and obstacle avoidance of the sweeping robot according to the present application;
  • FIG. 5 is a schematic diagram of functional modules of the method for distance measurement and obstacle avoidance of a sweeping robot according to the present application.
  • the main solutions of the embodiments of the present application are: when the sweeping robot performs the cleaning task, the depth information of the obstacle is obtained based on the dual TOF cameras; the target depth information is obtained by performing a binocular stereo matching operation based on the depth information; The target depth information is calculated to obtain the distance between the obstacle and the sweeping robot.
  • the distance between the sweeping robot and the obstacle is usually calculated by using dual RGB cameras, general structured light or TOF cameras.
  • TOF cameras have problems such as relatively low resolution, which is difficult to improve, systematic errors and random errors have a significant impact on the results, and inaccurate distance measurement at close range. .
  • the existence of the binocular camera requires high computing resources, the calculation is complex, the real-time performance is poor, and it is affected by the illumination and the texture of the object, but the advantage lies in the accurate close-range accuracy. It can be seen that in the prior art, the distance between the sweeping robot and the obstacle cannot be calculated accurately and accurately.
  • FIG. 1 is a schematic structural diagram of a terminal of a hardware operating environment involved in the solution of an embodiment of the present application.
  • the terminal may include: a processor 1001 , such as a CPU, a network interface 1004 , a user interface 1003 , a memory 1005 , and a communication bus 1002 .
  • the communication bus 1002 is used to realize the connection and communication between these components.
  • the user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.
  • the network interface 1004 may include a standard wired interface and a wireless interface (eg, a WI-FI interface).
  • the memory 1005 may be high-speed RAM memory, or may be non-volatile memory, such as disk memory.
  • the memory 1005 may also be a storage device independent of the aforementioned processor 1001 .
  • the terminal may further include a camera, an RF (Radio Frequency, radio frequency) circuit, a sensor, a remote control, an audio circuit, a WiFi module, a detector, and the like.
  • RF Radio Frequency, radio frequency
  • the terminal may also be configured with other sensors such as a gyroscope, a barometer, a hygrometer, a temperature sensor, etc., which will not be repeated here.
  • the terminal structure shown in FIG. 1 does not constitute a limitation on the terminal device, and may include more or less components than the one shown, or combine some components, or arrange different components.
  • the memory 1005 which is a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and a control program of a home appliance.
  • the network interface 1004 is mainly used to connect to the background server and perform data communication with the background server;
  • the user interface 1003 is mainly used to connect to the client (client) and perform data communication with the client;
  • the processor 1001 can be used to call the control program of the home appliance stored in the memory 1005, and perform the following operations:
  • the depth information of the obstacle is obtained based on the dual TOF cameras
  • the distance between the obstacle and the sweeping robot is obtained by calculating the target depth information.
  • FIG. 2 is a schematic flowchart of the first embodiment of the method for distance measurement and obstacle avoidance of the sweeping robot of the present application.
  • This embodiment of the present application provides an embodiment of a method for measuring distance and obstacle avoidance for a cleaning robot. It should be noted that although a logical sequence is shown in the flowchart, in some cases, it may be executed in a sequence different from that here. steps shown or described.
  • the ranging and obstacle avoidance methods of the sweeping robot include:
  • Step S10 when the sweeping robot performs the cleaning task, obtain the depth information of the obstacle based on the dual TOF cameras;
  • the TOF cameras are installed on the cleaning robot based on the direction facing to perform the cleaning task, and the TOF cameras include sensors and light sources.
  • the light source emits light to illuminate the object in front, and the refracted light refracted after being irradiated to the object is obtained through the sensor, and the depth information is formed according to the time of refraction.
  • the depth information can be calculated according to the time difference from the light source emitted by the TOF camera to the reception of the refracted light refracted by the object, that is, the depth information can reflect the distance between the obstacle and the sweeping robot.
  • Step S20 performing a binocular stereo matching operation based on the depth information to obtain target depth information
  • the refracted light is acquired by the sensor, it is converted into depth information, and then the depth information of the target is obtained by performing a binocular stereo matching operation through the depth information.
  • the distance between the sweeping robot and the obstacle is usually calculated by using dual RGB cameras, general structured light or TOF cameras.
  • TOF cameras have problems such as relatively low resolution, which is difficult to improve, systematic errors and random errors have a significant impact on the results, and inaccurate distance measurement at close range. .
  • the existence of the binocular camera requires high computing resources, the calculation is complex, the real-time performance is poor, and it is affected by the illumination and the texture of the object, but the advantage lies in the accurate close-range accuracy.
  • the target depth information is obtained by using dual TOF cameras and binocular stereo matching operations.
  • the binocular stereo matching includes matching cost calculation and cost aggregation calculation, and the depth information includes the first depth information obtained in the first depth map based on the first TOF camera and the second depth map obtained based on the second TOF camera.
  • the second depth information the step of performing a binocular stereo matching operation based on the depth information to obtain a depth difference value, comprising:
  • Step S21 matching the first depth information and the second depth information based on matching cost calculation, and acquiring each cost matrix obtained in the matching process;
  • the first depth information is information of each depth point in the first depth map, which includes depth values and coordinates.
  • the purpose of the matching cost calculation is to measure the correlation between the depth information to be matched and the depth information of candidates.
  • the matching cost can be calculated by the matching cost function whether the two depth information is the same name or not. The smaller the cost, the greater the correlation, and the greater the probability of being the same name.
  • Step S22 establishing the connection between the first depth information, optimizing the cost matrix, and obtaining the aggregated cost matrix of each depth information
  • the fundamental purpose of cost aggregation is to make the cost value obtained through cost matching accurately reflect the correlation between depth information.
  • the matching cost calculation often only considers local information, and calculates the cost value through the depth information in a certain size window in the two depth information neighborhoods, which is easily affected by image noise, and when the image is in a weak texture or repeated texture area. , this cost value may not accurately reflect the correlation between depth information, and the direct manifestation is that the cost value of the real point of the same name is not the smallest.
  • the cost aggregation is to optimize the cost matrix by establishing the connection between the adjacent depth information, with certain criteria, such as the adjacent depth information should have continuous disparity values.
  • the cost value is recalculated according to the cost value of its adjacent depth information under the same disparity value or nearby disparity value, and a new DSI is obtained, which is represented by a matrix S.
  • Step S23 determining the depth difference of each depth value according to the minimum cost value in the aggregation cost matrix
  • Step S24 deeply optimizing the depth difference to obtain a target depth map
  • Step S25 obtaining the target depth information according to the target depth map.
  • the depth difference of each depth value is determined by the cost matrix S after cost aggregation, which is usually calculated by the winner-takes-all algorithm (WTA, Winner-Takes-All).
  • the Left-Right Check algorithm is used to remove the wrong depth difference caused by occlusion and noise, the small connected area removal algorithm is used to remove isolated abnormal points, and the Median Filter and Bilateral Filter are used. ) and other smoothing algorithms to smooth the target depth map, and finally obtain the target depth information.
  • the binocular stereo matching operation is performed based on the depth information obtained by the dual TOF cameras, the target depth map is obtained by optimization, and finally the target depth information is obtained according to the target depth map, which improves the accuracy of obtaining the target depth information.
  • Step S30 calculating the distance between the obstacle and the sweeping robot through the target depth information.
  • the target depth information is calculated to determine the distance between the obstacle and the sweeping robot.
  • the depth information of the obstacle is obtained based on the dual TOF cameras, and then the depth information performs a binocular stereo matching operation to obtain the target depth information, and the target depth information can be used to determine the target depth information.
  • the information of the obstacles in the cleaning task of the sweeping robot is obtained through the dual TOF cameras, and the distance between the camera and the obstacle is obtained by using the binocular stereo matching operation, which can solve the problem that the binocular RGB cannot be accurately determined in the case of exposure.
  • the distance between the sweeping robot and the obstacle, and the single TOF has a large error when calculating the distance between the sweeping robot and the obstacle at a close distance.
  • the obstacle information is improved by the dual TOF cameras.
  • the calculation of the sweeping robot and the obstacle The precision and accuracy of distances between objects.
  • FIG. 3 is a schematic flowchart of another embodiment of the present application.
  • the step of performing matching on the first depth information and the second depth information based on the matching cost, and acquiring each matching cost matrix obtained in the matching process includes:
  • Step S211 obtaining second depth information of a preset range in the second depth map according to the determined first depth information
  • the second depth information is obtained by matching through a local matching method.
  • the step of obtaining the second depth information of a preset range in the second depth map according to the determined first depth information includes:
  • Step S2111 obtaining the coordinates of the first depth information, and obtaining the depth information adjacent to the abscissa of the coordinates in the second depth map as the depth information to be determined;
  • Step S2112 Acquire a target number of depth information from the to-be-determined depth information based on the ordinate of the first depth information as the second depth information.
  • the first depth information is depth information in the first depth map captured by the TOF camera in the dual TOF camera
  • the second depth information is depth information in the second depth map captured by the right TOF camera in the dual TOF camera.
  • the depth information adjacent to the abscissa is to search for the abscissa with the first depth information in the second depth map as a reference, and to search for the depth information of the adjacent abscissa. For example, when the coordinates of the first depth information in the first depth map are (2, 3), the depth information whose abscissas are 2 and 3 are searched for the depth information to be determined in the second depth map.
  • the coordinates of the first depth information are used to obtain the depth information whose abscissa is adjacent to the abscissa of the first depth information coordinate in the second depth map as the depth information to be determined.
  • the second depth there are multiple pieces of second depth information, and a preset range of second depth information is obtained for calculation to obtain a matching cost matrix.
  • the ordinate of the first depth information is determined as the reference, and the left and right three to-be-determined depth information are used as the second depth information, so as to obtain the second depth information matching the first depth information.
  • Step S212 performing a difference operation on the second depth information to obtain the matching cost matrix.
  • the preset range is to control the value range of the second depth information that is equal to the abscissa of the first depth information in the second depth map.
  • second depth information within a preset range is acquired in the second depth map.
  • FIG. 4 is a schematic flowchart of another embodiment of the present application. After the step of optimizing the depth difference to obtain a target depth map, the steps include:
  • Step 26 obtaining a plurality of first target depth information in the target depth map
  • Step 27 connecting the plurality of first target depth information to form a target depth area
  • Step 28 Determine the type of the obstacle according to the ratio of the target depth area to the area of the first depth map.
  • the first depth map is an image captured by the left camera in the binocular TOF camera.
  • the first depth information is depth information obtained in the target depth map.
  • Obtain a plurality of first target depth information in the target depth map connect the plurality of first target depth information to form a target depth area, and then compare the target depth area with the area of the first depth map to obtain the ratio between the two , to determine the type of the obstacle.
  • the size of the preset ratio may be used to determine the type, and the type of the obstacle includes a large obstacle and a small obstacle. For example, when the ratio of the target depth area to the area of the first depth map is greater than 50%, the type of the obstacle is determined to be a large obstacle; otherwise, it is determined to be a small obstacle. By judging the size of the obstacle, it provides a judgment basis for the sweeping robot to avoid obstacles.
  • Step 29 Determine the forward path of the cleaning robot according to the type of the obstacle and the distance.
  • the type of obstacles in front of the sweeping robot when performing the cleaning task can be obtained, and a planned path to avoid obstacles can be reasonably made.
  • the sweeping robot can avoid touching the edge of large obstacles by adopting a decelerating and forward cleaning method. Make sure that the robot vacuum cleaner does not touch obstacles when cleaning, resulting in damage and loss.
  • the method further includes:
  • Step S30 when it is determined that the obstacle is a large obstacle, a threshold control method is used to determine the physical volume of the obstacle;
  • Step S31 when it is determined that the obstacle is a large obstacle, a method of fitting with depth information is used to determine the physical volume of the obstacle.
  • the depth value data calculated by the two TOFs can be fitted, and the depth value with little depth change is used for fitting.
  • the edge depth is controlled by threshold value, the edge depth of objects with large targets changes drastically, and threshold value control can be used to reduce the amount of algorithm calculation and quickly determine the physical volume of obstacles, so that the sweeping robot can make a reasonable path planning.
  • FIG. 5 is a schematic diagram of a module of the present application.
  • the present application also provides a cleaning robot device, including:
  • the acquisition module 10 is used to acquire the depth map of the obstacle based on the dual TOF cameras when the sweeping robot performs the cleaning task;
  • a matching module 20 configured to perform binocular stereo matching based on the depth information in the depth map to obtain target depth information
  • the determining module 30 is configured to calculate the distance between the obstacle and the sweeping robot through the target depth information.
  • the present application further provides a cleaning robot, which includes a processor, a memory, and a distance measurement and obstacle avoidance of the cleaning robot that is stored in the memory and can run on the processor.
  • the program when the distance measuring and obstacle avoidance program of the sweeping robot is executed by the processor, implements the steps of any of the above methods for measuring distance and obstacle avoidance of the sweeping robot.
  • the present application also provides a computer-readable storage medium, where a ranging and obstacle-avoidance program of a sweeping robot is stored on the computer-readable storage medium, and the ranging and obstacle-avoidance program of the sweeping robot is When executed by the processor, the steps of any one of the above methods for distance measurement and obstacle avoidance for a sweeping robot are implemented.
  • the embodiments of the present application may be provided as a method, a system, or a computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
  • computer-usable storage media including, but not limited to, disk storage, CD-ROM, optical storage, etc.
  • These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory result in an article of manufacture comprising instruction means, the instructions
  • the apparatus implements the functions specified in the flow or flow of the flowcharts and/or the block or blocks of the block diagrams.
  • any reference signs placed between parentheses shall not be construed as limiting the claim.
  • the word “comprising” does not exclude the presence of elements or steps not listed in a claim.
  • the word “a” or “an” preceding an element does not preclude the presence of a plurality of such elements.
  • the present application may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by one and the same item of hardware.
  • the use of the words first, second, and third, etc. do not denote any order. These words can be interpreted as names.

Landscapes

  • Length Measuring Devices By Optical Means (AREA)
  • Manipulator (AREA)
  • Image Processing (AREA)
  • Measurement Of Optical Distance (AREA)

Abstract

一种扫地机器人及其测距方法、装置以及计算机可读存储介质。扫地机器人安装有双TOF相机，扫地机器人的测距避障方法包括：在扫地机器人执行清扫任务时，基于双TOF相机获取障碍物的深度信息（S10）；基于深度信息执行双目立体匹配操作得到目标深度信息（S20）；通过目标深度信息计算得到障碍物与扫地机器人之间的距离（S30）。

Description

扫地机器人及其测距方法、装置以及计算机可读存储介质
相关申请的交叉引用
本申请要求于2020年12月31日提交的申请号为202011645269.7，名称为“扫地机器人及其测距方法、装置以及计算机可读存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及视觉匹配技术领域，特别涉及一种扫地机器人及其测距方法、装置以及计算机可读存储介质。
背景技术
随着科技的发展，越来越多智能设备出现在人们的生活中，例如扫地机器人。在扫地机器人执行清洁任务的过程中需要通过获取清洁环境中的物体信息来控制其行走路径，确保扫地机器人在清扫时不会发生碰撞等问题。
现有技术中通常采用双RGB摄像头和可选的照明系统匹配，或者TOF相机的方法实现扫地机器人确定与障碍物之间存在的距离，但TOF相机存在分辨率相对较低，系统误差及随机误差对结果影响明显，近距离测量距离不精确等问题，但优势在于不受光照变化和物体纹理影响，能实时计算深度信息。而双目相机存在需要很高的计算资源，计算复杂，实时性差，受光照、物体纹理性质的影响，但优势在于近距离精度精确。即，现有技术中仍无法准确且精确地确定扫地机器人与障碍物之间的距离。
公开内容
本申请提出了一种扫地机器人及其测距方法、装置以及计算机可读存储介质，以解决现有技术中的扫地机器人无法精确且准确地识别扫地机器人与障碍物之间的距离的问题。
第一方面，本申请提出了一种扫地机器人的测距避障方法，所述扫地机器人安装有双TOF相机，所述扫地机器人的测距避障方法包括：在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息；基于所述深度信息执行双目立体匹配操作得到目标深度信息；通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
根据本申请实施例的扫地机器人的测距避障方法，当扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息，进而深度信息执行双目立体匹配操作得到目标深度信息，通过目标深度信息即可确定所述障碍物与扫地机器人之间的距离。实现了通过双 TOF相机获取扫地机器人在执行清洁任务前进方向的障碍物的信息，利用双目立体匹配操作得到相机与障碍物之间的距离，能够解决双目RGB在曝光情况下无法实现准确确定扫地机器人与障碍物之间的距离，以及单TOF在近距离计算扫地机器人与障碍物之间的距离时存在较大误差的问题，通过双TOF相机对障碍物信息进行提高了计算扫地机器人与障碍物之间距离的精度和准确性。
第二方面，本申请提出了一种扫地机器人装置，包括：获取模块，用于在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度图；匹配模块，用于基于所述深度图中的深度信息进行双目立体匹配得到目标深度信息；确定模块，用于通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
根据本申请实施例的扫地机器人装置，当扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息，进而深度信息执行双目立体匹配操作得到目标深度信息，通过目标深度信息即可确定所述障碍物与扫地机器人之间的距离。实现了通过双TOF相机获取扫地机器人在执行清洁任务前进方向的障碍物的信息，利用双目立体匹配操作得到相机与障碍物之间的距离，能够解决双目RGB在曝光情况下无法实现准确确定扫地机器人与障碍物之间的距离，以及单TOF在近距离计算扫地机器人与障碍物之间的距离时存在较大误差的问题，通过双TOF相机对障碍物信息进行提高了计算扫地机器人与障碍物之间距离的精度和准确性。
第三方面，本申请提出了一种扫地机器人，所述扫地机器人包括处理器、存储器以及存储在所述存储器上并可在所述处理器上运行的扫地机器人的测距避障程序，所述扫地机器人的测距避障程序被所述处理器执行时实现上述扫地机器人的测距避障方法的步骤。
根据本申请实施例的扫地机器人，当扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息，进而深度信息执行双目立体匹配操作得到目标深度信息，通过目标深度信息即可确定所述障碍物与扫地机器人之间的距离。实现了通过双TOF相机获取扫地机器人在执行清洁任务前进方向的障碍物的信息，利用双目立体匹配操作得到相机与障碍物之间的距离，能够解决双目RGB在曝光情况下无法实现准确确定扫地机器人与障碍物之间的距离，以及单TOF在近距离计算扫地机器人与障碍物之间的距离时存在较大误差的问题，通过双TOF相机对障碍物信息进行提高了计算扫地机器人与障碍物之间距离的精度和准确性。
第四方面，本申请提出了一种计算机可读存储介质，所述计算机可读存储介质上存储有扫地机器人的测距避障程序，所述扫地机器人的测距避障程序被处理器执行时实现上述扫地机器人的测距避障方法的步骤。
根据本申请实施例的计算机可读存储介质，当扫地机器人执行清扫任务时，基于所述双 TOF相机获取障碍物的深度信息，进而深度信息执行双目立体匹配操作得到目标深度信息，通过目标深度信息即可确定所述障碍物与扫地机器人之间的距离。实现了通过双TOF相机获取扫地机器人在执行清洁任务前进方向的障碍物的信息，利用双目立体匹配操作得到相机与障碍物之间的距离，能够解决双目RGB在曝光情况下无法实现准确确定扫地机器人与障碍物之间的距离，以及单TOF在近距离计算扫地机器人与障碍物之间的距离时存在较大误差的问题，通过双TOF相机对障碍物信息进行提高了计算扫地机器人与障碍物之间距离的精度和准确性。
本申请的附加方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本申请的实践了解到。
附图说明
本申请的上述和/或附加的方面和优点从结合下面附图对实施例的描述中将变得明显和容易理解，其中：
图1为本申请实施例方案涉及的硬件运行环境的终端结构示意图；
图2为本申请扫地机器人的测距避障方法一实施例的流程示意图；
图3为本申请扫地机器人的测距避障方法又一实施例的流程示意图；
图4为本申请扫地机器人的测距避障方法另一实施例的流程示意图；
图5为本申请扫地机器人的测距避障方法的功能模块示意图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
具体实施方式
应当理解，此处所描述的具体实施例仅用以解释本申请，并不用于限定本申请。
本申请实施例的主要解决方案是：在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息；基于所述深度信息执行双目立体匹配操作得到目标深度信息；通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
在现有技术中，通常在计算扫地机器人与障碍物之间的距离时，一般是通过双RGB摄像头、一般结构光或者TOF相机。但是TOF相机存在分辨率相对较低，很难提高，系统误差及随机误差对结果影响明显，近距离测量距离不精确等问题，但优势在于不受光照变化和物体纹理影响，能实时计算深度信息。而双目相机存在需要很高的计算资源，计算复杂，实时性差，受光照，物体纹理性质影响，但优势在于近距离精度精确。可见，现有技术中在计算扫地机器人与障碍物之间的距离时无法实现精确且准确地计算。
如图1所示，图1为本申请实施例方案涉及的硬件运行环境的终端结构示意图。
如图1所示，该终端可以包括：处理器1001，例如CPU，网络接口1004，用户接口1003，存储器1005，通信总线1002。其中，通信总线1002用于实现这些组件之间的连接通信。用户接口1003可以包括显示屏(Display)、输入单元比如键盘(Keyboard)，可选用户接口1003还可以包括标准的有线接口、无线接口。网络接口1004可选的可以包括标准的有线接口、无线接口(如WI-FI接口)。存储器1005可以是高速RAM存储器，也可以是稳定的存储器(non-volatile memory)，例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的存储装置。
终端还可以包括摄像头、RF(Radio Frequency，射频)电路，传感器、遥控器、音频电路、WiFi模块、检测器等等。当然，终端还可配置陀螺仪、气压计、湿度计、温度传感器等其他传感器，在此不再赘述。
本领域技术人员可以理解，图1中示出的终端结构并不构成对终端设备的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
如图1所示，作为一种计算机可读存储介质的存储器1005中可以包括操作系统、网络通信模块、用户接口模块以及家电设备的控制程序。
在图1所示的终端中，网络接口1004主要用于连接后台服务器，与后台服务器进行数据通信；用户接口1003主要用于连接客户端(用户端)，与客户端进行数据通信；而处理器1001可以用于调用存储器1005中存储的家电设备的控制程序，并执行以下操作：
在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息；
基于所述深度信息执行双目立体匹配操作得到目标深度信息；
通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
参考图2，图2为本申请扫地机器人的测距避障方法第一实施例的流程示意图。
本申请实施例提供了扫地机器人的测距避障方法的实施例，需要说明的是，虽然在流程图中示出了逻辑顺序，但是在某些情况下，可以以不同于此处的顺序执行所示出或描述的步骤。
扫地机器人的测距避障方法包括：
步骤S10，在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息；
在本申请中扫地机器人基于面向执行清洁任务方向上安装有两个TOF相机，所述TOF相机包括传感器和光源。在扫地机器人执行清洁任务的过程中光源发出光照射前方的物体，并通过传感器获取在照射到物体后折射回来的折射光，根据折射的时间形成深度信息。可 以理解的是，根据TOF相机原理，可根据从TOF相机发出光源至接收到经过物体折射的折射光的时间差距计算得到深度信息，即所述深度信息可反映障碍物与扫地机器人的距离。
步骤S20，基于所述深度信息执行双目立体匹配操作得到目标深度信息；
通过传感器获取到折射光后，将其转换成深度信息，进而通过深度信息执行双目立体匹配操作得到目标深度信息。
在现有技术中，通常在计算扫地机器人与障碍物之间的距离时，一般是通过双RGB摄像头、一般结构光或者TOF相机。但是TOF相机存在分辨率相对较低，很难提高，系统误差及随机误差对结果影响明显，近距离测量距离不精确等问题，但优势在于不受光照变化和物体纹理影响，能实时计算深度信息。而双目相机存在需要很高的计算资源，计算复杂，实时性差，受光照，物体纹理性质影响，但优势在于近距离精度精确。
在本申请中基于现有技术中双RGB摄像头利用双目匹配的原理，使用双TOF相机和双目立体匹配操作得到目标深度信息。
所述双目立体匹配包括匹配代价计算和代价聚合计算，所述深度信息包括基于第一TOF相机获取到第一深度图中的第一深度信息和基于第二TOF相机获取到第二深度图中的第二深度信息，所述基于所述深度信息执行双目立体匹配操作得到深度差值的步骤，包括：
步骤S21，基于匹配代价计算对所述第一深度信息和所述第二深度信息进行匹配，获取匹配过程中得到的每一代价矩阵；
所述第一深度信息为第一深度图中各深度点的信息，其包括深度值以及坐标。
匹配代价计算的目的是衡量待匹配深度信息与候选深度信息之间的相关性。两个深度信息无论是否为同名点，都可以通过匹配代价函数计算匹配代价，代价越小则说明相关性越大，是同名点的概率也越大。
步骤S22，建立所述第一深度信息之间的联系，对代价矩阵进行优化，得到每一深度信息的聚合代价矩阵；
代价聚合的根本目的是让通过代价匹配得到的代价值能够准确的反映深度信息之间的相关性。匹配代价计算往往只会考虑局部信息，通过两个深度信息邻域内一定大小的窗口内的深度信息信息来计算代价值，这很容易受到影像噪声的影响，而且当影像处于弱纹理或重复纹理区域，这个代价值极有可能无法准确的反映深度信息之间的相关性，直接表现就是真实同名点的代价值非最小。
而代价聚合是通过建立邻接深度信息之间的联系，以一定的准则，如相邻深度信息应该具有连续的视差值，来对代价矩阵进行优化，每个深度信息在某个视差下的新代价值都会根据其相邻深度信息在同一视差值或者附近视差值下的代价值来重新计算，得到新的DSI，用矩阵S来表示。
步骤S23，根据所述聚合代价矩阵中根据最小代价值确定每个深度值的深度差；
步骤S24，深度优化所述深度差得到目标深度图；
步骤S25，根据所述目标深度图得到所述目标深度信息。
即通过代价聚合之后的代价矩阵S来确定每个深度值的深度差，通常使用赢家通吃算法(WTA，Winner-Takes-All)来计算。
采用左右一致性检查(Left-Right Check)算法剔除因为遮挡和噪声而导致的错误深度差，采用剔除小连通区域算法来剔除孤立异常点，采用中值滤波(Median Filter)、双边滤波(Bilateral Filter)等平滑算法对目标深度图进行平滑，最后得到目标深度信息。
在本实施例中，基于双TOF相机获取到的深度信息进行双目立体匹配操作，优化得到目标深度图，最后根据目标深度图得到目标深度信息，提高了目标深度信息获取的准确度。
步骤S30，通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
将所述目标深度信息进行计算确定所述障碍物与扫地机器人之间的距离。
在本实施例中，当扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息，进而深度信息执行双目立体匹配操作得到目标深度信息，通过目标深度信息即可确定所述障碍物与扫地机器人之间的距离。实现了通过双TOF相机获取扫地机器人在执行清洁任务前进方向的障碍物的信息，利用双目立体匹配操作得到相机与障碍物之间的距离，能够解决双目RGB在曝光情况下无法实现准确确定扫地机器人与障碍物之间的距离，以及单TOF在近距离计算扫地机器人与障碍物之间的距离时存在较大误差的问题，通过双TOF相机对障碍物信息进行提高了计算扫地机器人与障碍物之间距离的精度和准确性。
参照图3，图3为本申请的又一实施例流程示意图。所述基于所述匹配代价对所述第一深度信息和所述第二深度信息进行匹配，获取匹配过程中得到的每一匹配代价矩阵的步骤，包括：
步骤S211，根据确定的第一深度信息在第二深度图中获取预设范围的第二深度信息；
在本实施例中通过局部匹配的方法匹配得到第二深度信息。所述根据确定的第一深度信息在第二深度图中获取预设范围的第二深度信息的步骤，包括：
步骤S2111，获取所述第一深度信息的坐标，在所述第二深度图中获取与所述坐标的横坐标相邻的深度信息作为待确定深度信息；
步骤S2112，基于第一深度信息的纵坐标在所述待确定深度信息中获取目标数量的深度信息作为所述第二深度信息。
所述第一深度信息为双TOF相机中做TOF相机拍摄的第一深度图中的深度信息，所述第二深度信息为双TOF相机中右TOF相机拍摄的第二深度图中的深度信息。
横坐标相邻的深度信息为在第二深度图中查找与第一深度信息的横坐标为参照物，查找与其相邻的横坐标坐标的深度信息。例如，当第一深度信息在第一深度图中的坐标为(2,3)，则在第二深度图中查找横坐标为2和3的深度信息为待确定深度信息。
在本申请中以第一深度信息的坐标，在第二深度图中获取横坐标与第一深度信息坐标的横坐标相邻的深度信息作为待确定深度信息，可以理解的是，在第二深度图中所述第二深度信息存在多个，获取预设范围的第二深度信息进行计算，得到匹配代价矩阵。例如，在得到待确定深度信息后，确定与第一深度信息纵坐标为基准，左右3个待确定深度信息作为第二深度信息，以实现获取与第一深度信息匹配的第二深度信息。
步骤S212，将所述第二深度信息进行差值运算得到所述匹配代价矩阵。
所述预设范围即为控制第二深度图中与第一深度信息的横坐标相等的第二深度信息的取值范围。
在本实施例中基于在第一深度图中的第一深度信息，在第二深度图中获取预设范围内的第二深度信息。避免了单一匹配第二深度信息时由于匹配不准确导致双目立体匹配结果存在较大误差的问题。
参照图4，图4为本申请的又一实施例流程示意图。所述深度优化所述深度差得到目标深度图的步骤之后，包括：
步骤26，在所述目标深度图中获取多个第一目标深度信息；
步骤27，连接所述多个第一目标深度信息，形成目标深度面积；
步骤28，根据所述目标深度面积与所述第一深度图的面积的比例确定所述障碍物的类型。
所述第一深度图为双目TOF相机中左相机拍摄的图像。
所述第一深度信息为在目标深度图中获得的深度信息。
在目标深度图中获取多个第一目标深度信息，连接多个第一目标深度信息形成目标深度面积，进而根据目标深度面积与第一深度图的面积进行比对，获取两者之间的比例，确定所述障碍物的类型。
可以理解的是，在确定障碍物的类型是，可以将预设比例的大小确定类型，所述障碍物的类型包括大型障碍物和小型障碍物。例如，当目标深度面积与第一深度图的面积的比例大于50％，则确定所述障碍物的类型为大型障碍物，反之，则判断其为小型障碍物。通过判断障碍物的大小为扫地机器人进行障碍物避让提供了判断基础。
其还可结合扫地机器人所在的环境不同来确定障碍物的类型。
所述根据所述目标深度面积与所述第一深度图的面积的比例确定所述障碍物的类型的 步骤之后，包括：
步骤29，根据所述障碍物的类型和所述距离确定扫地机器人的前进路径。
在本申请中能够基于使用双目TOF相机获取到的深度信息，并根据深度信息得到扫地机器人在执行清洁任务时前方障碍物的类型，合理作出避让障碍物的规划路径。例如，对于大型障碍物扫地机器人可在通过采用减速前进的清扫方式，避免触碰到大型障碍物的边缘。确保扫地机器人在执行清洁时不会触碰到障碍物，导致发生损坏，造成损失。
根据所述目标深度面积与所述第一深度图的面积的比例确定所述障碍物的类型的步骤之后，所述方法还包括：
步骤S30，当确定所述障碍物为大型障碍物时，采用阈值控制方法确定所述障碍物的物理体积；
步骤S31，当确定所述障碍物为大型障碍物时，采用深度信息进行拟合的方法确定所述障碍物的物理体积。
在本实施例中，对于小型障碍物，例如如细线等，可在两个TOF计算出来的深度值数据拟合，使用深度变化不大的深度值进行拟合，对于对于小型障碍物，例如，冰箱，床等，进行边缘的深度进行阈值控制，目标大的物体边缘深度变化剧烈，采取阈值控制，可减少算法计算量和快速确定障碍物的物理体积，使扫地机器人合理作出路径规划。
此外，参照图5，图5为本申请的模块示意图。为实现上述实施例，本申请还提供一种扫地机器人装置，包括：
获取模块10，用于在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度图；
匹配模块20，用于基于所述深度图中的深度信息进行双目立体匹配得到目标深度信息；
确定模块30，用于通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
此外，为实现上述实施例，本申请还提供一种扫地机器人，所述扫地机器人包括处理器、存储器以及存储在所述存储器上并可在所述处理器上运行的扫地机器人的测距避障程序，所述扫地机器人的测距避障程序被所述处理器执行时实现如上任一项扫地机器人的测距避障方法的步骤。
此外，为实现上述实施例，本申请还提供一种计算机可读存储介质，所述计算机可读存储介质上存储有扫地机器人的测距避障程序，所述扫地机器人的测距避障程序被处理器执 行时实现如上中任一项扫地机器人的测距避障方法的步骤。
本领域内的技术人员应明白，本申请的实施例可提供为方法、系统、或计算机程序产品。因此，本申请可采用完全硬件实施例、完全软件实施例、或结合软件和硬件方面的实施例的形式。而且，本申请可采用在一个或多个其中包含有计算机可用程序代码的计算机可用存储介质(包括但不限于磁盘存储器、CD-ROM、光学存储器等)上实施的计算机程序产品的形式。
本申请是参照根据本申请实施例的方法、设备(系统)、和计算机程序产品的流程图和/或方框图来描述的。应理解可由计算机程序指令实现流程图和/或方框图中的每一流程和/或方框、以及流程图和/或方框图中的流程和/或方框的结合。可提供这些计算机程序指令到通用计算机、专用计算机、嵌入式处理机或其他可编程数据处理设备的处理器以产生一个机器，使得通过计算机或其他可编程数据处理设备的处理器执行的指令产生用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的装置。
这些计算机程序指令也可存储在能引导计算机或其他可编程数据处理设备以特定方式工作的计算机可读存储器中，使得存储在该计算机可读存储器中的指令产生包括指令装置的制造品，该指令装置实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能。
这些计算机程序指令也可装载到计算机或其他可编程数据处理设备上，使得在计算机或其他可编程设备上执行一系列操作步骤以产生计算机实现的处理，从而在计算机或其他可编程设备上执行的指令提供用于实现在流程图一个流程或多个流程和/或方框图一个方框或多个方框中指定的功能的步骤。
应当注意的是，在权利要求中，不应将位于括号之间的任何参考符号构造成对权利要求的限制。单词“包含”不排除存在未列在权利要求中的部件或步骤。位于部件之前的单词“一”或“一个”不排除存在多个这样的部件。本申请可以借助于包括有若干不同部件的硬件以及借助于适当编程的计算机来实现。在列举了若干装置的单元权利要求中，这些装置中的若干个可以是通过同一个硬件项来具体体现。单词第一、第二、以及第三等的使用不表示任何顺序。可将这些单词解释为名称。
尽管已描述了本申请的可选实施例，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例作出另外的变更和修改。所以，所附权利要求意欲解释为包括可选实施例以及落入本申请范围的所有变更和修改。
显然，本领域的技术人员可以对本申请进行各种改动和变型而不脱离本申请的精神和范 围。这样，倘若本申请的这些修改和变型属于本申请权利要求及其等同技术的范围之内，则本申请也意图包含这些改动和变型在内。

Claims (10)

  1. 一种扫地机器人的测距避障方法，所述扫地机器人安装有双TOF相机，所述扫地机器人的测距避障方法包括：
    在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度信息；
    基于所述深度信息执行双目立体匹配操作得到目标深度信息；
    通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
  2. 如权利要求1所述的扫地机器人的测距避障方法，其中，所述深度信息包括基于第一TOF相机获取到第一深度图中的第一深度信息和基于第二TOF相机获取到第二深度图中的第二深度信息，所述基于所述深度信息执行双目立体匹配操作得到目标深度信息的步骤，包括：
    基于匹配代价对所述第一深度信息和所述第二深度信息进行匹配，获取匹配过程中得到的每一代价矩阵；
    建立所述第一深度信息之间的联系，对代价矩阵进行优化，得到每一深度信息的聚合代价矩阵；
    根据所述聚合代价矩阵中根据最小代价值确定各第一深度信息的深度值的深度差；
    深度优化所述深度差得到目标深度图；
    根据所述目标深度图得到所述目标深度信息。
  3. 如权利要求2所述的扫地机器人的测距避障方法，其中，所述基于所述匹配代价对所述第一深度信息和所述第二深度信息进行匹配，获取匹配过程中得到的每一匹配代价矩阵的步骤，包括：
    根据确定的第一深度信息在第二深度图中获取预设范围的第二深度信息；
    将所述第二深度信息进行差值运算得到所述匹配代价矩阵。
  4. 如权利要求3所述的扫地机器人的测距避障方法，其中，所述根据确定的第一深度信息在第二深度图中获取预设范围的第二深度信息的步骤，包括：
    获取所述第一深度信息的坐标，在所述第二深度图中获取与所述坐标的横坐标相邻的深度信息作为待确定深度信息；
    基于第一深度信息的纵坐标在所述待确定深度信息中获取目标数量的深度信息作为所述第二深度信息。
  5. 如权利要求2-4任一项所述的扫地机器人的测距避障方法，其中，所述深度优化所述深度差得到目标深度图的步骤之后，包括：
    在所述目标深度图中获取多个第一目标深度信息；
    连接所述多个第一目标深度信息，形成目标深度面积；
    根据所述目标深度面积与所述第一深度图的面积的比例确定所述障碍物的类型。
  6. 如权利要求5所述的扫地机器人的测距避障方法，其中，所述根据所述目标深度面积与所述第一深度图的面积的比例确定所述障碍物的类型的步骤之后，包括：
    根据所述障碍物的类型和所述距离确定扫地机器人的前进路径。
  7. 如权利要求5-6任一项所述的扫地机器人的测距避障方法，其中，所述根据所述目标深度面积与所述第一深度图的面积的比例确定所述障碍物的类型的步骤之后，所述方法还包括：
    当确定所述障碍物为大型障碍物时，采用阈值控制方法确定所述障碍物的物理体积；
    当确定所述障碍物为大型障碍物时，采用深度信息进行拟合的方法确定所述障碍物的物理体积。
  8. 一种扫地机器人装置，包括：
    获取模块，用于在扫地机器人执行清扫任务时，基于所述双TOF相机获取障碍物的深度图；
    匹配模块，用于基于所述深度图中的深度信息进行双目立体匹配得到目标深度信息；
    确定模块，用于通过所述目标深度信息计算得到所述障碍物与扫地机器人之间的距离。
  9. 一种扫地机器人，所述扫地机器人包括处理器、存储器以及存储在所述存储器上并可在所述处理器上运行的扫地机器人的测距避障程序，所述扫地机器人的测距避障程序被所述处理器执行时实现如权利要求1-7任一项扫地机器人的测距避障方法的步骤。
  10. 一种计算机可读存储介质，所述计算机可读存储介质上存储有扫地机器人的测距避障程序，所述扫地机器人的测距避障程序被处理器执行时实现如权利要求1至7中任一项扫地机器人的测距避障方法的步骤。
PCT/CN2021/139948 2020-12-31 2021-12-21 扫地机器人及其测距方法、装置以及计算机可读存储介质 WO2022143285A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011645269.7A CN112842180A (zh) 2020-12-31 2020-12-31 扫地机器人及其测距避障方法、装置及可读存储介质
CN202011645269.7 2020-12-31

Publications (1)

Publication Number Publication Date
WO2022143285A1 true WO2022143285A1 (zh) 2022-07-07

Family

ID=76001140

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/139948 WO2022143285A1 (zh) 2020-12-31 2021-12-21 扫地机器人及其测距方法、装置以及计算机可读存储介质

Country Status (2)

Country Link
CN (1) CN112842180A (zh)
WO (1) WO2022143285A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112842180A (zh) * 2020-12-31 2021-05-28 深圳市杉川机器人有限公司 扫地机器人及其测距避障方法、装置及可读存储介质
CN113598653B (zh) * 2021-08-05 2023-03-31 北京石头世纪科技股份有限公司 清洁方法及装置、设备、计算机可读存储介质
CN115267825A (zh) * 2022-06-24 2022-11-01 奥比中光科技集团股份有限公司 基于tof传感器的扫地机避障和导航方法、装置及存储介质

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005093536A1 (de) * 2004-03-22 2005-10-06 BSH Bosch und Siemens Hausgeräte GmbH Flächenbearbeitungssytem
KR20180024326A (ko) * 2016-08-29 2018-03-08 엘지전자 주식회사 이동 로봇 및 그 제어방법
CN108245099A (zh) * 2018-01-15 2018-07-06 深圳市沃特沃德股份有限公司 机器人移动方法和装置
CN108888187A (zh) * 2018-05-31 2018-11-27 四川斐讯信息技术有限公司 一种基于深度相机的扫地机器人
CN110974088A (zh) * 2019-11-29 2020-04-10 深圳市杉川机器人有限公司 扫地机器人控制方法、扫地机器人及存储介质
CN210803719U (zh) * 2019-08-27 2020-06-19 浙江舜宇光学有限公司 深度图像成像装置、系统和终端
CN111609854A (zh) * 2019-02-25 2020-09-01 北京奇虎科技有限公司 基于多个深度相机的三维地图构建方法及扫地机器人
CN112842180A (zh) * 2020-12-31 2021-05-28 深圳市杉川机器人有限公司 扫地机器人及其测距避障方法、装置及可读存储介质

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110427968B (zh) * 2019-06-28 2021-11-02 武汉大学 一种基于细节增强的双目立体匹配方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005093536A1 (de) * 2004-03-22 2005-10-06 BSH Bosch und Siemens Hausgeräte GmbH Flächenbearbeitungssytem
KR20180024326A (ko) * 2016-08-29 2018-03-08 엘지전자 주식회사 이동 로봇 및 그 제어방법
CN108245099A (zh) * 2018-01-15 2018-07-06 深圳市沃特沃德股份有限公司 机器人移动方法和装置
CN108888187A (zh) * 2018-05-31 2018-11-27 四川斐讯信息技术有限公司 一种基于深度相机的扫地机器人
CN111609854A (zh) * 2019-02-25 2020-09-01 北京奇虎科技有限公司 基于多个深度相机的三维地图构建方法及扫地机器人
CN210803719U (zh) * 2019-08-27 2020-06-19 浙江舜宇光学有限公司 深度图像成像装置、系统和终端
CN110974088A (zh) * 2019-11-29 2020-04-10 深圳市杉川机器人有限公司 扫地机器人控制方法、扫地机器人及存储介质
CN112842180A (zh) * 2020-12-31 2021-05-28 深圳市杉川机器人有限公司 扫地机器人及其测距避障方法、装置及可读存储介质

Also Published As

Publication number Publication date
CN112842180A (zh) 2021-05-28

Similar Documents

Publication Publication Date Title
WO2022143285A1 (zh) 扫地机器人及其测距方法、装置以及计算机可读存储介质
US11074706B2 (en) Accommodating depth noise in visual slam using map-point consensus
CN107909612B (zh) 一种基于3d点云的视觉即时定位与建图的方法与系统
EP3401815B1 (en) Determining an architectural layout
US20200011668A1 (en) Simultaneous location and mapping (slam) using dual event cameras
WO2020206666A1 (zh) 基于散斑图像的深度估计方法及装置、人脸识别系统
WO2018205803A1 (zh) 位姿估计方法和装置
WO2020228643A1 (zh) 交互控制方法、装置、电子设备及存储介质
US10288425B2 (en) Generation of map data
WO2020118545A1 (en) Time-aware occupancy grid mapping for robots in dynamic environments
CN107504917B (zh) 一种三维尺寸测量方法及装置
US10613546B2 (en) Stochastic map-aware stereo vision sensor model
WO2023155387A1 (zh) 多传感器目标检测方法、装置、电子设备以及存储介质
US20170108338A1 (en) Method for geolocating a carrier based on its environment
US20190164296A1 (en) Systems and methods for determining a confidence measure for a motion vector
Xu et al. FollowUpAR: Enabling follow-up effects in mobile AR applications
Zhou et al. Semi-dense visual odometry for RGB-D cameras using approximate nearest neighbour fields
JP7166446B2 (ja) ロボットの姿勢を推定するシステムおよび方法、ロボット、並びに記憶媒体
CN113850859A (zh) 用于增强图像深度置信度图的方法、系统、制品和装置
WO2021114026A1 (zh) 一种基于局部参考坐标系的3d形状匹配方法及装置
US10215858B1 (en) Detection of rigid shaped objects
CN113177980B (zh) 用于自动驾驶的目标对象速度确定方法、装置及电子设备
CN117437348A (zh) 计算装置及模型生成方法
US11068739B1 (en) Method and system for comparison of object contours
US10012729B2 (en) Tracking subjects using ranging sensors

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21914024

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 13/11/2023)