WO2024045489A1 - 基于扫地机器人的用户相似度计算方法及装置、存储介质 - Google Patents

基于扫地机器人的用户相似度计算方法及装置、存储介质 Download PDF

Info

Publication number
WO2024045489A1
WO2024045489A1 PCT/CN2023/073863 CN2023073863W WO2024045489A1 WO 2024045489 A1 WO2024045489 A1 WO 2024045489A1 CN 2023073863 W CN2023073863 W CN 2023073863W WO 2024045489 A1 WO2024045489 A1 WO 2024045489A1
Authority
WO
WIPO (PCT)
Prior art keywords
trajectory
different users
rolling ball
data
sweeping robot
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2023/073863
Other languages
English (en)
French (fr)
Inventor
路瑶
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qingdao Haier Technology Co Ltd
Haier Smart Home Co Ltd
Haier Uplus Intelligent Technology Beijing Co Ltd
Original Assignee
Qingdao Haier Technology Co Ltd
Haier Smart Home Co Ltd
Haier Uplus Intelligent Technology Beijing Co Ltd
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 Qingdao Haier Technology Co Ltd, Haier Smart Home Co Ltd, Haier Uplus Intelligent Technology Beijing Co Ltd filed Critical Qingdao Haier Technology Co Ltd
Publication of WO2024045489A1 publication Critical patent/WO2024045489A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • G06V10/457Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components by analysing connectivity, e.g. edge linking, connected component analysis or slices
    • GPHYSICS
    • G06COMPUTING OR CALCULATING; COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/60Type of objects
    • G06V20/64Three-dimensional [3D] objects

Definitions

  • This application relates to the field of smart home technology. Specifically, it relates to a user similarity calculation method and device, a storage medium and an electronic device based on a sweeping robot.
  • smart home products such as sweeping robots support networking, remote control and personalized working modes. They have also accumulated a large amount of information related to users' daily life, such as preferred usage patterns and usage time. and information such as cleaning areas. Continuously improving smart home products and continuously improving smart home product services have become the focus of current smart home development research.
  • the existing methods for measuring user similarity are mostly based on user behavior data, such as browsing history, geographical location, etc., but there is a lack of data on the physical space where users actually live. Therefore, how to use data from the working process of smart home products to analyze users' living habits and personality preferences while ensuring user privacy, and to identify similar users to improve product design capabilities and personalized service levels is the current promotion and development of smart home appliances. A big challenge in using it.
  • This application provides a user similarity calculation method, storage medium and electronic device based on a sweeping robot.
  • the user similarity calculation method based on the sweeping robot provided by this application can obtain the working trajectory data of the sweeping robots of different users, generate map data of the ground area where the sweeping robot operates based on the working trajectory data, and calculate different data based on the map data of the ground area. Similarity between users, the calculated similarity results can support subsequent user classification and clustering of similar user habits, etc., providing accurate data for the design of smart home products, accurate advertising, and personalized recommendations.
  • this application provides a user similarity calculation method based on a sweeping robot.
  • the user similarity calculation method includes:
  • this application provides a user similarity calculation device based on a sweeping robot.
  • the user similarity calculation device includes:
  • the acquisition module is used to obtain the working trajectory data of different users' sweeping robots within a historical period
  • a generation module configured to generate map data of the ground area where the sweeping robot operates based on the work trajectory data
  • a similarity calculation module is used to calculate similarities between different users based on the map data of the ground area.
  • the present application provides a computer-readable storage medium.
  • the computer-readable storage medium includes a stored program, wherein when the program is run, the method described in the first aspect of the present application is executed.
  • the present application provides an electronic device, including a memory and a processor.
  • a computer program is stored in the memory, and the processor is configured to execute the method described in the first aspect of the present application through the computer program. .
  • a user similarity calculation method based on a sweeping robot is proposed. This method obtains the working trajectory data of the sweeping robots of different users and generates a map of the ground area where the sweeping robot operates based on the working trajectory data. Data, calculate the similarity between different users based on the map data of the ground area, and the calculated similarity results can support subsequent user classification and clustering of similar user habits.
  • Figure 1 is a schematic diagram of the hardware environment of a user similarity calculation method based on a sweeping robot according to an embodiment of the present application
  • Figure 2 is a schematic flowchart of the main steps of a user similarity calculation method based on a sweeping robot according to an embodiment of the present application;
  • FIG. 3 is a schematic flowchart of the main steps of step S102 according to the embodiment of the present application.
  • FIG. 4 is a schematic flowchart of the main steps of step S1024 according to the embodiment of the present application.
  • FIG. 5 is a schematic flowchart of the main steps of step S103 according to the embodiment of the present application.
  • Figure 6 is a schematic flowchart of the main steps of a user similarity calculation device based on a sweeping robot according to an embodiment of the present application
  • FIG. 7 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
  • module and “processor” may include hardware, software, or a combination of both.
  • a module can include hardware circuits, various suitable sensors, communication ports, and memory. It can also include software parts, such as program code, or it can be a combination of software and hardware.
  • the processor may be a central processing unit, a microprocessor, an image processor, a digital signal processor, or any other suitable processor. Processor with data and/or signal processing Function.
  • the processor can be implemented in software, hardware, or a combination of both.
  • Non-transitory computer-readable storage media include any suitable media that can store program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random access memory, etc.
  • a and/or B means all possible combinations of A and B, such as just A, just B, or A and B.
  • the terms "at least one A or B” or “at least one of A and B” have a similar meaning to “A and/or B” and may include just A, just B or A and B.
  • the singular forms “a,””the” and “the” may also include the plural form.
  • smart home products such as sweeping robots support networking, remote control and personalized working modes. They have also accumulated a large amount of information related to users' daily life, such as preferred usage patterns and usage time. and cleaning areas, etc. These data reflect the user’s behavioral habits, personality preferences and family status to a certain extent.
  • the service provider needs to understand the user's living habits and residential space characteristics, such as the residential area, furniture placement, etc., so as to facilitate Similar users are grouped so that accurate and personalized products and services can be provided based on the characteristics of different user groups.
  • the existing methods for measuring user similarity are mostly based on user behavior data, such as browsing history, geographical location, etc., but there is a lack of data on the physical space where users actually live.
  • the similarity of users is measured from the perspective of daily life space.
  • Metrics are also important to improve personalized design and personalized recommendations for smart home products. Therefore, how to use data from the working process of smart home products to analyze users’ living habits and personality preferences while ensuring user privacy, and identify similar users to improve product design capabilities and personalized service levels is the current promotion and development of smart home appliances. A big challenge in using it.
  • this application provides a user similarity calculation method based on a sweeping robot.
  • This method obtains the working trajectory data of different users' sweeping robots, and generates map data of the ground area where the sweeping robot operates based on the working trajectory data.
  • ground area ground Graph data calculates the similarity between different users. The calculated similarity results can be used to identify users with similar home living spaces. By grouping similar users, we can subsequently design products and provide personalized services for different groups of user groups. Base.
  • a user similarity calculation method based on a sweeping robot is provided.
  • This user similarity calculation method based on sweeping robots is widely used in whole-house intelligent digital control application scenarios such as smart home, smart home, smart home equipment ecology, and smart residence (Intelligence House) ecology.
  • the above-mentioned user similarity calculation method based on the sweeping robot can be applied to the hardware environment composed of the terminal device 102 and the server 104 as shown in FIG. 1 .
  • the server 104 is connected to the terminal device 102 through the network, and can be used to provide services (such as application services, etc.) for the terminal or the client installed on the terminal, wherein the client installed on the terminal can be a sweeping robot intelligent APP, on the sweeping robot smart APP, the working trajectory data of the sweeping robot during a cumulative working period, such as 20 hours, can be obtained.
  • the terminal device 102 uploads the obtained working trajectory data to the server 104.
  • the server 104 receives the working trajectory data and based on The working trajectory data generates map data of the floor area where the sweeping robot operates, and the server 104 calculates similarities between different users based on the acquired map data of the floor area where the sweeping robot operates.
  • the above-mentioned network may include but is not limited to at least one of the following: wired network, wireless network.
  • the above-mentioned wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, and local area network.
  • the above-mentioned wireless network may include at least one of the following: WIFI (Wireless Fidelity, Wireless Fidelity), Bluetooth.
  • the terminal device 102 may not be limited to a PC, a mobile phone, a tablet, an intelligent sweeping robot, etc.
  • the server 104 includes, but is not limited to, various personal computers, notebook computers, tablet computers, portable wearable devices, and the like.
  • Figure 2 is a schematic flowchart of the main steps of a user similarity calculation method based on a sweeping robot according to an embodiment of the present application. As shown in Figure 2, this application implements The user similarity calculation method based on the sweeping robot in the example mainly includes the following steps S101 to S103.
  • Step S101 Obtain the working trajectory data of different users' sweeping robots in any historical time.
  • the sweeping robot can be a laser navigation sweeping robot, a visual navigation sweeping robot and a gyro navigation sweeping robot.
  • the laser navigation sweeping robot emits a laser to form a light spot on the obstacle, and performs distance measurement according to the pixel number of the light spot, and Combined with the SLAM algorithm system to implement path planning, house mapping and other procedures.
  • the visual navigation sweeping robot observes and detects the surrounding environment through the panoramic camera on the body and internal sensors, uses the perceived environmental information to construct and draw a room cleaning map, and locates the body position through the algorithm system to design the cleaning route. and programs.
  • Gyroscope navigation sweeping robot is commonly known as inertial navigation. It obtains environmental information through gyroscopes and accelerators and calculates the position information of the sweeping robot.
  • this application does not limit the type of navigation of the sweeping robot, whether it is laser navigation, visual navigation or gyro navigation.
  • the next step can be entered based on the working trajectory data.
  • different users have different working trajectories of sweeping robots placed in their homes due to their different living habits and home layouts.
  • the sweeping robots of different users can be obtained through the intelligent sweeping robot APP.
  • Work trajectory data within a historical period can be determined based on the battery life, route error and reporting mechanism of the sweeping robot. For example, the selected length of a historical period within seven days from the current moment Work trajectory data for 20 hours of working time.
  • step S101 includes:
  • the sweeping robot will establish its own coordinate system. During the cleaning process, the sweeping robot will mark the coordinates of the route points and the coordinates of the obstacle points by itself, and generate a trajectory scatter diagram, which will be displayed on the intelligent sweeping robot APP. , after the intelligent sweeping robot APP obtains these trajectory scatter plots, it uploads them to the server.
  • Step S102 Generate map data of the ground area where the sweeping robot operates based on the working trajectory data.
  • Figure 3 is a schematic flowchart of the main steps of step S102 according to the embodiment of the present application. As shown in Figure 3, the step S102 includes:
  • Step S1021 Binarize the trajectory scatter diagram.
  • the trajectory scatter diagram generated based on the working trajectory data of the sweeping robot is graphically binarized, the background pixel gray value is set to 0, and the gray value of each scatter point pixel on the trajectory scatter diagram is set to 255.
  • Step S1022 Set an initial rolling ball radius value, and extract the initial contour map of the trajectory scatter diagram after graphic binarization based on the rolling ball method based on the initial rolling ball radius.
  • the initial contour map of the trajectory scatter diagram after graphic binarization is extracted.
  • an initial rolling ball radius value is determined through the maximum Euclidean distance between two points in the trajectory scatter diagram, and then According to the initial rolling ball radius, the boundary points of the trajectory scatter plot after graphic binarization are extracted based on the rolling ball method, and the area enclosed by the boundary points is obtained as the initial contour map of the trajectory scatter plot.
  • Step S1023 Obtain the number of connected domains in the area enclosed by the initial contour map.
  • a two-pass scanning method (Two-Pass) or a seed-filling method (Seed-Filling) may be used to obtain the number of connected domains in the area enclosed by the initial contour map through connected domain marking.
  • the two-pass scanning method is to scan the initial contour map twice, find and mark all connected domains existing in the initial contour map; obtain the individual connected domains of the area enclosed by the initial contour map according to the number of marks. number.
  • the valid value is The label of a pixel is assigned to the label value of the pixel; when the left neighbor pixel and the upper neighbor pixel of the pixel are both valid values, the smaller label value is selected and assigned to the label value of the pixel; during the second scan, the The label of each point is updated to the smallest label in its set. After completing two scans, pixels with the same label value in the initial contour map form the same connected area.
  • the seed filling method assumes that at least one pixel inside the polygon or region is known, and then tries to find all other pixels in the region and fill them.
  • a region can be defined internally or with a boundary; in the case of a boundary definition, all pixels on the boundary of the region have a specific value or color, and all pixels inside the region do not take on that specific value, however, pixels outside the boundary can have The same value as the border; if defined internally, all pixels inside the region have one color or value, and all pixels outside the region have another color or value.
  • the algorithm for filling the internal defined area is called Flood Fill Algorithm
  • the algorithm for filling the boundary defined area is called the boundary filling algorithm.
  • this application does not limit the method for obtaining the number of connected domains of the initial contour enclosing area. Whether it is a two-pass scanning method, a seed filling method or other algorithms, as long as the initial contour enclosing area can be obtained. The number of connected domains in the region is sufficient.
  • Step S1024 Based on the number of connected domains, extract the contour diagram of the binarized trajectory scatter diagram as the map data of the ground area where the sweeping robot operates.
  • Figure 4 is a schematic flowchart of the main steps of step S1024 according to the embodiment of the present application. As shown in Figure 4, the step S1024 includes:
  • Step S10241 When the number of connected domains is greater than the first preset threshold, update the rolling ball radius value. According to the updated rolling ball radius, extract a new value of the trajectory scatter diagram after the binarization of the graphic based on the rolling ball method. Contour map, calculate the number of connected domains in the area enclosed by the new contour map;
  • Step S10242 When the number of connected domains is less than or equal to the first preset threshold, the rolling ball radius value is no longer updated, and the last updated rolling ball radius value is used as the final rolling ball radius value. According to the final rolling ball radius value Based on the rolling ball method, the final contour map of the trajectory scatter plot after graphic binarization is extracted as the map data of the ground area where the sweeping robot operates.
  • the number of connected domains is set to 1, that is, the first preset threshold is 1.
  • the rolling ball radius is updated, and the updated rolling ball radius is used.
  • the rolling ball radius will be updated until the last updated rolling ball radius is used to extract the corresponding contour map of the binary trajectory scatter diagram after the graphic binarization based on the rolling ball method, and the two-pass scanning method will be used again.
  • the seed filling method obtains the number of connected domains corresponding to the enclosed area of the contour map to 1, and extracts the final contour map of the trajectory scatter diagram after graphic binarization based on the last updated rolling ball radius based on the rolling ball method as a sweeping robot. Map data for the ground area of the operation.
  • Step S103 Calculate the similarity between different users based on the map data of the ground area.
  • the floor area where the sweeping robot operates is the user's daily activity area.
  • Figure 5 is a schematic flowchart of the main steps of step S103 according to the embodiment of the present application. As shown in Figure 5, the step S103 includes:
  • Step S1031 Obtain trajectory point clouds of the ground areas where the sweeping robots of different users operate, where the trajectory point cloud represents multiple trajectory scatter points;
  • Step S1032 Perform random consistency sampling on the acquired trajectory point clouds of different users to obtain sampling point cloud data of different users;
  • Step S1034 Compare Li with the second preset threshold. When Li is greater than the second preset threshold, continue sampling and calculate the value of Li;
  • Step S1035 When Li is less than the second preset threshold, take the average value L_avg of all sampled Lis as the ground area trajectory point cloud distance of the two users' sweeping robots, and obtain the similarity of the two sweeping robots corresponding to the two users as 1 /L_avg.
  • the trajectory point clouds of the ground areas where the sweeping robots of User A and User B have been obtained have been obtained.
  • Random consistency sampling is performed on the trajectory point clouds of the ground areas where the sweeping robots of User A and User B operate respectively. Randomly select N points from the trajectory point clouds of the ground area where user A and user B's sweeping robots operate, and perform matching calculations on N randomly selected points from the trajectory point clouds of the ground area where user A and user B's sweeping robots operate.
  • Sampling is performed from N randomly selected trajectory point clouds in the ground area where the sweeping robots of user A and user B operate. Starting from the first sampling, the point cloud data of the sampling points of the sweeping robots between user A and user B are calculated.
  • this application obtains the working trajectory data of the sweeping robots of different users, generates map data of the ground area where the sweeping robot operates based on the working trajectory data, and calculates the relationship between different users based on the map data of the ground area.
  • the calculated similarity results can support subsequent user classification, clustering, etc., providing accurate data for the design of smart home products, accurate advertising, and personalized recommendations.
  • this application also provides a user similarity calculation device based on a sweeping robot.
  • FIG. 6 is a main structural block diagram of a user similarity calculation device based on a sweeping robot according to an embodiment of the present application.
  • the user similarity calculation device based on the sweeping robot in the embodiment of the present application mainly includes an acquisition module 11 , a generation module 12 and a similarity calculation module 13 .
  • one or more of the acquisition module 11, the generation module 12, and the similarity calculation module 13 can be combined into one module.
  • the acquisition module 11 and the generation module 12 can be two separate modules, or they can be combined.
  • the combined module is called a ground inference module.
  • the acquisition module 11 may be configured to acquire working trajectory data of different users' sweeping robots within a historical period.
  • the generation module 12 may be configured to generate map data of the ground area where the sweeping robot operates based on the work trajectory data.
  • the similarity calculation module 13 may be configured to calculate similarities between different users according to the map data of the ground area.
  • the combined ground inference module is configured to obtain working trajectory data of different users' sweeping robots within a historical period, and generate map data of the ground area where the sweeping robot operates based on the working trajectory data.
  • the computer program can be stored in a computer-readable file.
  • the computer program includes computer program code, which may be in the form of source code, object code, executable file or some intermediate form.
  • the computer-readable storage medium may include: any entity or device capable of carrying the computer program code, media, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier wave signals, telecommunications signals, and software distribution media, etc.
  • computer-readable storage media Storage media does not include electrical carrier signals and telecommunications signals.
  • this application also provides a computer-readable storage medium.
  • the computer-readable storage medium may be configured to store a program for executing the user similarity calculation method based on the sweeping robot of the above method embodiment, and the program may be loaded by a processor. And run to implement the above user similarity calculation method based on sweeping robot.
  • the computer-readable storage medium may be the device mentioned above in the above embodiment.
  • the non-volatile computer storage media included in the device may also be a non-volatile computer storage media that exists separately and is not assembled into the terminal.
  • the above-mentioned computer-readable storage medium stores one or more programs.
  • the above-mentioned device obtains the working trajectory data of different users' sweeping robots in any historical time;
  • the working trajectory data generates map data of the ground area where the sweeping robot operates; and the similarity between different users is calculated based on the map data of the ground area.
  • the computer-readable storage medium may be a memory device formed by various electronic devices.
  • the computer-readable storage medium is a non-transitory computer-readable storage medium.
  • the electronic device includes a processor and a memory
  • the memory may be configured to store a program that executes the user similarity calculation method based on the sweeping robot according to the above method embodiment
  • the processor may be configured to execute a program in the memory, which program includes but is not limited to a program that executes the user similarity calculation method based on the cleaning robot of the above method embodiment.
  • the memory can be used to store software programs and modules, such as the program instructions/modules corresponding to the user similarity calculation method and device based on the sweeping robot in the embodiment of the present invention.
  • the processor runs the software stored in the memory.
  • Memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory.
  • the memory may further include memory located remotely relative to the processor, and these remote memories may be connected to the terminal through a network. Examples of the above-mentioned networks include but are not limited to the Internet, intranets, local area networks, mobile communication networks and combinations thereof.
  • the above-mentioned memory may include, but is not limited to, the above-mentioned The acquisition module 11 in the user similarity calculation device of the sweeping robot.
  • it may also include but is not limited to other module units in the above-mentioned user similarity calculation device based on the sweeping robot, which will not be described again in this example.
  • the physical devices corresponding to these modules may be the processor itself, or a part of the software in the processor, a part of the hardware, or Part of the combination of software and hardware. Therefore, the number of individual modules in the figure is only illustrative.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Manipulator (AREA)

Abstract

本申请公开了一种基于扫地机器人的用户相似度计算方法及装置、存储介质及电子装置,涉及智能家居/智慧家庭技术领域,该基于扫地机器人的用户相似度计算方法包括:获取不同用户的扫地机器人在任意历史时间内的工作轨迹数据;根据工作轨迹数据生成扫地机器人作业的地面区域的地图数据;根据地面区域的地图数据计算不同的用户的相似度。本申请通过获取不同用户的扫地机器人的工作轨迹数据,根据工作轨迹数据生成扫地机器人作业的地面区域的地图数据,根据地面区域的地图数据计算不同用户的相似度,相似度的结果可用来识别居家生活空间相似的用户。

Description

基于扫地机器人的用户相似度计算方法及装置、存储介质
本申请要求2022年08月30日提交的、发明名称为“基于扫地机器人的用户相似度计算方法及装置、存储介质”的中国专利申请CN202211049711.9的优先权,上述中国专利申请的全部内容通过引用并入本申请中。
技术领域
本申请涉及智能家居技术领域,具体而言,涉及一种基于扫地机器人的用户相似度计算方法及装置、存储介质及电子装置。
背景技术
随着智能家居的普及,越来越多的智能家居产品例如扫地机器人支持联网、远程控制和个性化的工作模式,也积累了大量和用户日常生活相关的信息,如偏好的使用模式、使用时间以及清洁区域等信息。不断改进智能家居产品和不断提高智能家居产品的服务成为目前智能家居发展研究的重点部分。
现有技术中对于用户的相似性度量的方法多基于用户行为数据,如浏览记录、地理位置等,而对于用户实际生活的物理空间的数据比较缺乏。因此,如何在保证用户隐私的同时,利用智能家居产品工作过程的数据分析用户的生活习惯、个性偏好,识别相似用户,用以提高产品设计能力和个性化服务水平,是当前智能家电的推广和使用所面临的一大挑战。
发明内容
本申请提供了一种基于扫地机器人的用户相似度计算方法、存储介质及电子装置。本申请提供的基于扫地机器人的用户相似度计算方法可以通过获取不同的用户的扫地机器人的工作轨迹数据,根据工作轨迹数据生成扫地机器人作业的地面区域的地图数据,根据地面区域的地图数据计算不同的用户之间的相似度,计算的相似度的结果可支持后续的用户分类以及相似用户习惯的聚类等,为智能家居产品的设计、广告精准投放以及个性化推荐提供了精准数据。
在第一方面,本申请提供一种基于扫地机器人的用户相似度计算方法,该用户相似度计算方法包括:
获取不同的用户的扫地机器人在任意历史时间内的工作轨迹数据;
根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据;
根据所述地面区域的地图数据计算不同的用户之间的相似度。
在第二方面,本申请提供一种基于扫地机器人的用户相似度计算装置,该用户相似度计算装置包括:
获取模块,用于获取不同的用户的扫地机器人在一段历史时间内的工作轨迹数据;
生成模块,用于根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据;
相似度计算模块,用于根据所述地面区域的地图数据计算不同的用户之间的相似度。
在第三方面,本申请提供一种计算机可读的存储介质,所述计算机可读的存储介质包括存储的程序,其中,所述程序运行时执行本申请第一方面所述的方法。
在第四方面,本申请提供一种电子装置,包括存储器和处理器,所述存储器中存储有计算机程序,所述处理器被设置为通过所述计算机程序执行本申请第一方面所述的方法。
本申请上述一个或多个技术方案,至少具有如下一种或多种有益效果:
在实施本申请的技术方案中,提出一种基于扫地机器人的用户相似度计算方法,该方法通过获取不同的用户的扫地机器人的工作轨迹数据,根据工作轨迹数据生成扫地机器人作业的地面区域的地图数据,根据地面区域的地图数据计算不同的用户之间的相似度,计算的相似度的结果可支持后续的用户分类以及相似用户习惯的聚类等。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本申请的实施例,并与说明书一起用于解释本申请的原理。
为了更清楚地说明本申请实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,对于本领域普通技术人员而言,在不付出创造性劳动性的前提下,还可以根据这些附图获得其他的附图。
图1是根据本申请实施例的一种基于扫地机器人的用户相似度计算方法的硬件环境示意图;
图2是根据本申请实施例的基于扫地机器人的用户相似度计算方法的主要步骤流程示意图;
图3是根据本申请实施例的步骤S102的主要步骤流程示意图;
图4是根据本申请实施例的步骤S1024的主要步骤流程示意图;
图5是根据本申请实施例的步骤S103的主要步骤流程示意图;
图6是根据本申请实施例的基于扫地机器人的用户相似度计算装置的主要步骤流程示意图;
图7是根据本申请实施例的电子装置的结构示意图。
具体实施方式
为了使本技术领域的人员更好地理解本申请方案,下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本申请一部分的实施例,而不是全部的实施例。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都应当属于本申请保护的范围。
需要说明的是,本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”等是用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换,以便这里描述的本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施。此外,术语“包括”和“具有”以及他们的任何变形,意图在于覆盖不排他的包含,例如,包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元,而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
在本申请的描述中,“模块”、“处理器”可以包括硬件、软件或者两者的组合。一个模块可以包括硬件电路,各种合适的感应器,通信端口,存储器,也可以包括软件部分,比如程序代码,也可以是软件和硬件的组合。处理器可以是中央处理器、微处理器、图像处理器、数字信号处理器或者其他任何合适的处理器。处理器具有数据和/或信号处理 功能。处理器可以以软件方式实现、硬件方式实现或者二者结合方式实现。非暂时性的计算机可读存储介质包括任何合适的可存储程序代码的介质,比如磁碟、硬盘、光碟、闪存、只读存储器、随机存取存储器等等。术语“A和/或B”表示所有可能的A与B的组合,比如只是A、只是B或者A和B。术语“至少一个A或B”或者“A和B中的至少一个”含义与“A和/或B”类似,可以包括只是A、只是B或者A和B。单数形式的术语“一个”、“这个”也可以包含复数形式。
随着智能家电的普及,越来越多的智能家居产品例如扫地机器人支持联网、远程控制和个性化的工作模式,也积累了大量和用户日常生活相关的信息,如偏好的使用模式、使用时间以及清洁区域等。这些数据一定程度上反映了用户的行为习惯、个性喜好和家庭状况。服务方为了不断改进智能家居产品和不断提高智能家居产品的个性化设计和个性化推荐等服务,需要了解用户的生活习惯、住所空间特点,如住所使用面积、家具摆放方式等,以便于将相似用户分组,从而可以根据不同用户群体的特征提供精准的、个性化的产品和服务。
现有技术中对于用户的相似性度量的方法多基于用户行为数据,如浏览记录、地理位置等,而对于用户实际生活的物理空间的数据比较缺乏,但是从日常生活空间角度对用户进行相似度度量对于提高智能家居产品的个性化设计和个性化推荐也十分重要。因此,如何在保证用户隐私的同时,利用智能家居产品工作过程的数据分析用户的生活习惯、个性偏好,识别相似用户,用以提高产品设计能力和个性化服务水平,是当前智能家电的推广和使用所面临的一大挑战。
为此,本申请提供了一种基于扫地机器人的用户相似度计算方法,该方法通过获取不同的用户的扫地机器人的工作轨迹数据,根据工作轨迹数据生成扫地机器人作业的地面区域的地图数据,根据地面区域的地 图数据计算不同的用户之间的相似度,计算的相似度的结果可以用来识别居家生活空间相似的用户,通过将相似用户分组,为后续给不同分组的用户群体设计产品和个性化服务提供基础。
根据本申请实施例的一个方面,提供了一种基于扫地机器人的用户相似度计算方法。该基于扫地机器人的用户相似度计算方法广泛应用于智慧家庭(Smart Home)、智能家居、智能家用设备生态、智慧住宅(Intelligence House)生态等全屋智能数字化控制应用场景。可选地,在本实施例中,上述基于扫地机器人的用户相似度计算方法可以应用于如图1所示的由终端设备102和服务器104所构成的硬件环境中。如图1所示,服务器104通过网络与终端设备102进行连接,可用于为终端或终端上安装的客户端提供服务(如应用服务等),其中,终端上安装的客户端可以为扫地机器人智能APP,在扫地机器人智能APP上可获取扫地机器人在累计工作一段时间例如20小时的工作轨迹数据,终端设备102将获取的工作轨迹数据上传至服务器104,服务器104接收所述工作轨迹数据,并根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据,服务器104根据获取的不同的用户的扫地机器人作业的地面区域的地图数据计算不同的用户之间的相似度。
上述网络可以包括但不限于以下至少之一:有线网络,无线网络。上述有线网络可以包括但不限于以下至少之一:广域网,城域网,局域网,上述无线网络可以包括但不限于以下至少之一:WIFI(Wireless Fidelity,无线保真),蓝牙。终端设备102可以并不限定于为PC、手机、平板电脑、智能扫地机器人等。服务器104包括但不限于各种个人计算机、笔记本电脑、平板电脑和便携式可穿戴设备等。
参阅附图2-5,图2是根据本申请的一个实施例的基于扫地机器人的用户相似度计算方法的主要步骤流程示意图。如图2所示,本申请实施 例中的基于扫地机器人的用户相似度计算方法主要包括下列步骤S101-步骤S103。
步骤S101:获取不同的用户的扫地机器人在任意历史时间内的工作轨迹数据。
其中,扫地机器人可以为激光导航扫地机器人、视觉导航扫地机器人和陀螺仪导航扫地机器人,其中,激光导航扫地机器人是通过发射激光到障碍物上形成光斑,根据光斑的像素序号来进行测距,并结合SLAM算法系统来实施路径规划、房屋建图等程序。视觉导航扫地机器人是通过机身上的全景摄像头和内部的传感器来观察、探测周围环境,运用感知到的环境信息来构建、绘制房间清扫地图,在通过算法系统定位机身位置,以设计清洁路线和方案。陀螺仪导航扫地机器人就是俗说的惯性导航,它是通过陀螺仪和加速器来获取环境信息,并计算出扫地机器人的位置信息。
需要说明的是,本申请对扫地机器人是激光导航、视觉导航还是陀螺仪导航的导航种类不作限定,只要能获取扫地机器人在一段历史时间内的工作轨迹数据,根据该工作轨迹数据就可以进入下一步骤。
一个实施方式中,不同的用户由于其生活习惯以及家庭布局等情况的不同,放置在不同的用户的家中的扫地机器人的工作轨迹也不相同,通过智能扫地机器人APP可以获取不同的用户的扫地机器人在一段历史时间内的工作轨迹数据,其中,任意一段历史时间的选取长度可根据扫地机器人的续航时间、路线误差和报数机制等情况制定,例如一段历史时间选取距离当下时刻的七天内的累计工作时间20小时的工作轨迹数据。
一个实施方式中,所述步骤S101包括:
标记不同的用户的扫地机器人在任意历史时间内工作过程中经过的途径点和障碍点,并生成轨迹散点图。
一个实施方式中,扫地机器人会建立自己的坐标系,在扫地机器人做清扫工作的过程中,自行标记途径点的坐标和障碍点的坐标,并生成轨迹散点图,显示在智能扫地机器人APP上,智能扫地机器人APP获取到这些轨迹散点图后,将其上传至服务器。
步骤S102:根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据。
一个实施方式中,图3是根据本申请实施例的步骤S102的主要步骤流程示意图,如图3所示,所述步骤S102包括:
步骤S1021:将所述轨迹散点图进行图形二值化。
一个实施方式中,根据扫地机器人的工作轨迹数据生成的轨迹散点图进行图形二值化,将背景像素灰度值设置为0,将轨迹散点图上的各散点像素灰度值设置为255。
步骤S1022:设置初始滚球半径值,根据所述初始滚球半径基于滚球法提取图形二值化后的轨迹散点图的初始轮廓图。
一个实施方式中,基于滚球法提取图形二值化后的轨迹散点图的初始轮廓图,首先先通过轨迹散点图中两点之间的最大欧式距离确定一个初始滚球半径值,然后根据初始滚球半径基于滚球法提取图形二值化后的轨迹散点图的边界点,得到边界点合围的区域,作为轨迹散点图的初始轮廓图。
步骤S1023:获取初始轮廓图围合区域的连通域的个数。
一个实施方式中,可以采用两遍扫描法(Two-Pass)或者种子填充法(Seed-Filling)通过连通域标记的方式来获得初始轮廓图围合区域的连通域的个数。
其中,两遍扫描法为,扫描两遍所述初始轮廓图,将所述初始轮廓图中存在的所有连通域找出并标记;根据标记个数获得初始轮廓图围合区域的连通域的个数。两遍扫描法的大致流程为:第一次扫描时,从左上角开始遍历初始轮廓图的像素点,背景像素保持0不变,找到第一个像素为255的点,label=1;当该像素的左邻像素和上邻像素为无效值时,给该像素置一个新的label值,label++,记录集合;当该像素的左邻像素或者上邻像素有一个为有效值时,将有效值像素的label赋给该像素的label值;当该像素的左邻像素和上邻像素都为有效值时,选取其中较小的label值赋给该像素的label值;第二次扫描时,对每个点的label进行更新,更新为其对于其集合中最小的label,完成两次扫描后,初始轮廓图中具有相同label值的像素就组成了同一个连通区域。
种子填充法为,假设在多边形或区域内部至少有一个像素是已知的,然后设法找到区域内所有其他像素,并对它们进行填充。区域可以用内部定义或边界定义;如果是边界定义,那么区域边界上所有像素均具有特定的值或颜色,区域内部的所有像素均不取这一特定值,然而,边界外的像素则可具有与边界相同的值;如果是内部定义,那么,区域内部所有像素具有同一种颜色或值,而区域外的所有像素具有另一种颜色或值。相应的,填充内部定义区域的算法成为泛填充算法(Flood Fill Algorithm),填充边界定义区域的算法称为边界填充算法。
需要说明的是,本申请对于获得初始轮廓图围合区域的连通域的个数的方法不作限定,无论是采用两遍扫描法还是种子填充法或是其它算法,只要能获取初始轮廓图围合区域的连通域的个数即可。
步骤S1024:基于所述连通域的个数,提取图形二值化后的轨迹散点图的轮廓图作为扫地机器人作业的地面区域的地图数据。
一个实施方式中,图4是根据本申请实施例的步骤S1024的主要步骤流程示意图,如图4所示,所述步骤S1024包括:
步骤S10241:在连通域的个数大于第一预设阈值的情况下,更新滚球半径值,根据所述更新滚球半径基于滚球法再次提取图形二值化后的轨迹散点图的新轮廓图,计算新轮廓图围合区域的连通域的个数;
步骤S10242:在所述连通域的个数小于等于第一预设阈值的情况下,不再更新滚球半径值,将最后更新的滚球半径值作为最终滚球半径值,根据最终滚球半径值基于滚球法提取图形二值化后的轨迹散点图的最终轮廓图作为扫地机器人作业的地面区域的地图数据。
一个实施方式中,当获得连通域的个数后,比较连通域的个数是否满足本实施方式中需要的连通域的个数,为了一目了然地获得扫地机器人作业的地面区域,本实施方式中将连通域的个数定为1个,即第一预设阈值为1,当获得的初始轮廓图围合区域的连通域的个数大于1时,更新滚球半径,利用更新后的滚球半径,再次基于滚球法提取图形二值化后的轨迹散点图的新轮廓图,再次使用两遍扫描法或者种子填充法获取新轮廓图围合区域的连通域的个数,若此时连通域的个数还是大于1,则一直更新滚球半径,直到利用最后更新的滚球半径,基于滚球法提取图形二值化后的轨迹散点图的对应轮廓图,再次使用两遍扫描法或者种子填充法获取对应轮廓图围合区域的连通域的个数为1后,根据最后更新的滚球半径基于滚球法提取图形二值化后的轨迹散点图的最终轮廓图作为扫地机器人作业的地面区域的地图数据。
步骤S103:根据所述地面区域的地图数据计算不同的用户之间的相似度。
一个实施方式中,确定了不同的用户的扫地机器人作业的地面区域的地图数据后,扫地机器人作业的地面区域即为用户的日常活动区域, 想得到用户之间的相似度时,可以根据不同的用户的地面区域的地图数据计算不同的用户之间的相似度,接下来,以计算两个用户之间的相似度为例进行说明。
一个实施方式中,图5是根据本申请实施例的步骤S103的主要步骤流程示意图,如图5所示,所述步骤S103包括:
步骤S1031:获取不同的用户的扫地机器人作业的地面区域的轨迹点云,其中,所述轨迹点云表示多个轨迹散点;
步骤S1032:将获取的不同的用户的轨迹点云分别执行随机一致性采样,获得不同的用户的采样点点云数据;
步骤S1033:将不同的用户的采样点点云数据进行匹配计算,得到不同的用户的采样点点云数据的平移损失参数Rti和旋转损失参数Rri,计算不同的用户的扫地机器人的点云的距离Li为Li=Rti*Rri;
步骤S1034:比较Li与第二预设阈值的大小,当Li大于第二预设阈值,继续采样并计算Li的值;
步骤S1035:当Li小于第二预设阈值,取采样的所有Li的平均值L_avg作为两个用户的扫地机器人的地面区域轨迹点云距离,得到两个扫地机器人对应两个用户的相似度为1/L_avg。
一个实施方式中,用户A和用户B的扫地机器人作业的地面区域的轨迹点云已经获得,将用户A和用户B的扫地机器人作业的地面区域的轨迹点云分别执行随机一致性采样,在用户A和用户B的扫地机器人作业的地面区域的轨迹点云中随机选取N个点,将用户A和用户B的扫地机器人作业的地面区域的轨迹点云中随机选取的N个点进行匹配计算,从用户A和用户B的扫地机器人作业的地面区域的轨迹点云中随机选取的N个中进行采样,从第一次采样开始,计算用户A和用户B间的扫地机器人的采样点点云数据的平移损失参数Rti和旋转损失参数Rri,并计 算用户A和用户B间的扫地机器人的点云的距离Li为Li=Rti*Rri,比较Li与第二预设阈值的大小,当大于第二预设阈值,便继续采样并重复计算,直到第i次采样后,Li小于第二预设阈值,此时取采样并计算得到的所有Li的平均值L_avg作为两个用户的扫地机器人的地面区域轨迹点云距离,取得到两个扫地机器人对应两个用户的相似度为1/L_avg,其中,1≤i≤N。
基于上述步骤S101-步骤S103,本申请通过获取不同的用户的扫地机器人的工作轨迹数据,根据工作轨迹数据生成扫地机器人作业的地面区域的地图数据,根据地面区域的地图数据计算不同的用户之间的相似度,计算的相似度的结果可支持后续的用户分类、聚类等,为智能家居产品的设计、广告精准投放以及个性化推荐提供了精准数据。
需要指出的是,尽管上述实施例中将各个步骤按照特定的先后顺序进行了描述,但是本领域技术人员可以理解,为了实现本申请的效果,不同的步骤之间并非必须按照这样的顺序执行,其可以同时(并行)执行或以其他顺序执行,这些变化都在本申请的保护范围之内。
进一步,本申请还提供了一种基于扫地机器人的用户相似度计算装置。
参阅附图6,图6是根据本申请的一个实施例的基于扫地机器人的用户相似度计算装置的主要结构框图。如图6所示,本申请实施例中的基于扫地机器人的用户相似度计算装置主要包括获取模块11、生成模块12和相似度计算模块13。在一些实施例中,获取模块11、生成模块12和相似度计算模块13中的一个或多个可以合并在一起成为一个模块。例如,获取模块11和生成模块12可以是单独分开的两个模块,也可以组合,组合后的模块称为地面推断模块。在一些实施例中获取模块11可以被配置成获取不同的用户的扫地机器人在一段历史时间内的工作轨迹数据。 生成模块12可以被配置成根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据。相似度计算模块13可以被配置成根据所述地面区域的地图数据计算不同的用户之间的相似度。组合后的地面推断模块被配置为获取不同的用户的扫地机器人在一段历史时间内的工作轨迹数据,并根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据。
在一个实施方式中,具体实现功能的描述可以参见步骤S101-步骤S103所述。
本领域技术人员能够理解的是,本申请实现上述一实施例的方法中的全部或部分流程,也可以通过计算机程序来指令相关的硬件来完成,所述的计算机程序可存储于一计算机可读存储介质中,该计算机程序在被处理器执行时,可实现上述各个方法实施例的步骤。其中,所述计算机程序包括计算机程序代码,所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读存储介质可以包括:能够携带所述计算机程序代码的任何实体或装置、介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器、随机存取存储器、电载波信号、电信信号以及软件分发介质等。需要说明的是,所述计算机可读存储介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减,例如在某些司法管辖区,根据立法和专利实践,计算机可读存储介质不包括电载波信号和电信信号。
进一步,本申请还提供了一种计算机可读存储介质。在根据本申请的一个计算机可读存储介质实施例中,计算机可读存储介质可以被配置成存储执行上述方法实施例的基于扫地机器人的用户相似度计算方法的程序,该程序可以由处理器加载并运行以实现上述基于扫地机器人的用户相似度计算方法。该计算机可读存储介质可以是上述实施例中上述装 置中所包含的非易失性计算机存储介质,也可以是单独存在,未装配入终端中的非易失性计算机存储介质。上述计算机可读存储介质存储有一个或者多个程序,当上述一个或者多个程序被一个设备执行时,使得上述设备:获取不同的用户的扫地机器人在任意历史时间内的工作轨迹数据;根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据;根据所述地面区域的地图数据计算不同的用户之间的相似度。
为了便于说明,仅示出了与本申请实施例相关的部分,具体技术细节未揭示的,请参照本申请实施例方法部分。该计算机可读存储介质可以是包括各种电子设备形成的存储器设备,可选的,本申请实施例中计算机可读存储介质是非暂时性的计算机可读存储介质。
进一步,本申请还提供了一种电子装置。在根据本申请的一个电子装置实施例中,如图7所示,电子装置包括处理器和存储器,存储器可以被配置成存储执行上述方法实施例的基于扫地机器人的用户相似度计算方法的程序,处理器可以被配置成用于执行存储器中的程序,该程序包括但不限于执行上述方法实施例的基于扫地机器人的用户相似度计算方法的程序。在一个具体示例中,存储器可用于存储软件程序以及模块,如本发明实施例中的基于扫地机器人的用户相似度计算方法及装置对应的程序指令/模块,处理器通过运行存储在存储器内的软件程序以及模块,从而执行各种功能应用以及数据处理,即实现上述的基于扫地机器人的用户相似度计算方法。存储器可包括高速随机存储器,还可以包括非易失性存储器,如一个或者多个磁性存储装置、闪存、或者其他非易失性固态存储器。在一些实例中,存储器可进一步包括相对于处理器远程设置的存储器,这些远程存储器可以通过网络连接至终端。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。作为一种示例,如图6所示,上述存储器中可以但不限于包括上述 扫地机器人的用户相似度计算装置中的获取模块11。此外,还可以包括但不限于上述基于扫地机器人的用户相似度计算装置中的其他模块单元,本示例中不再赘述。进一步,应该理解的是,由于各个模块的设定仅仅是为了说明本申请的装置的功能单元,这些模块对应的物理器件可以是处理器本身,或者处理器中软件的一部分,硬件的一部分,或者软件和硬件结合的一部分。因此,图中的各个模块的数量仅仅是示意性的。
本领域技术人员能够理解的是,可以对装置中的各个模块进行适应性地拆分或合并。对具体模块的这种拆分或合并并不会导致技术方案偏离本申请的原理,因此,拆分或合并之后的技术方案都将落入本申请的保护范围内。
以上所述仅是本申请的优选实施方式,应当指出,对于本技术领域的普通技术人员来说,在不脱离本申请原理的前提下,还可以做出若干改进和润饰,这些改进和润饰也应视为本申请的保护范围。

Claims (16)

  1. 一种基于扫地机器人的用户相似度计算方法,包括:
    获取不同的用户的扫地机器人在任意历史时间内的工作轨迹数据;
    根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据;
    根据所述地面区域的地图数据计算不同的用户之间的相似度。
  2. 根据权利要求1所述的方法,其中,所述获取不同的用户的扫地机器人在一段历史时间内的工作轨迹数据,包括:
    标记不同的用户的扫地机器人在一段历史时间内工作过程中的途径点和障碍点,并生成轨迹散点图。
  3. 根据权利要求2所述的方法,其中,所述根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据,包括:
    将所述轨迹散点图进行图形二值化;
    设置初始滚球半径值,根据所述初始滚球半径基于滚球法提取图形二值化后的轨迹散点图的初始轮廓图;
    获取初始轮廓图围合区域的连通域的个数;
    基于所述连通域的个数,提取图形二值化后的轨迹散点图的轮廓图作为扫地机器人作业的地面区域的地图数据。
  4. 根据权利要求3所述的方法,其中,所述基于所述连通域的个数,提取图形二值化后的轨迹散点图的轮廓图作为扫地机器人作业的地面区域的地图数据,包括:
    在连通域的个数大于第一预设阈值的情况下,更新滚球半径值,根据所述更新滚球半径基于滚球法再次提取图形二值化后的轨迹散点图的新轮廓图,计算新轮廓图围合区域的连通域的个数;
    在所述连通域的个数小于等于第一预设阈值的情况下,不再更新滚球半径值,将最后更新的滚球半径值作为最终滚球半径值,根据最终滚球半径值基于滚球法提取图形二值化后的轨迹散点图的最终轮廓图作为扫地机器人作业的地面区域的地图数据。
  5. 根据权利要求4所述的方法,其中,所述根据所述地面区域的地图数据计算不同的用户之间的相似度,包括:
    获取不同的用户的扫地机器人作业的地面区域的轨迹点云,其中,所述轨迹点云表示多个轨迹散点;
    将获取的不同的用户的轨迹点云分别执行随机一致性采样,获得不同的用户的采样点点云数据;
    将不同的用户的采样点点云数据进行匹配计算,得到不同的用户的采样点点云数据的平移损失参数Rti和旋转损失参数Rri,计算不同的用户的扫地机器人的点云的距离Li为Li=Rti*Rri
    比较Li与第二预设阈值的大小,当Li大于第二预设阈值,继续采样并计算Li的值;
    当Li小于第二预设阈值,取采样的所有Li的平均值L_avg作为两个用户的扫地机器人的地面区域轨迹点云距离,得到两个扫地机器人对应两个用户的相似度为1/L_avg。
  6. 根据权利要求3或4所述的方法,其中,所述初始滚球半径值通过所述轨迹散点图中两点之间的最大欧式距离确定。
  7. 根据权利要求3或4所述的方法,其中,所述获取初始轮廓图围合区域的连通域的个数,包括:
    扫描两遍所述初始轮廓图,将所述初始轮廓图中存在的所有连通域找出并标记;
    根据标记个数获得初始轮廓图围合区域的连通域的个数。
  8. 一种基于扫地机器人的用户相似度计算装置,包括:
    获取模块,设置为获取不同的用户的扫地机器人在一段历史时间内的工作轨迹数据;
    生成模块,设置为根据所述工作轨迹数据生成扫地机器人作业的地面区域的地图数据;
    相似度计算模块,设置为根据所述地面区域的地图数据计算不同的用户之间的相似度。
  9. 根据权利要求8所述的装置,其中,所述获取模块进一步被设置为:
    标记不同的用户的扫地机器人在一段历史时间内工作过程中的途径点和障碍点,并生成轨迹散点图。
  10. 根据权利要求9所述的装置,其中,所述生成模块进一步被设置为:
    将所述轨迹散点图进行图形二值化;
    设置初始滚球半径值,根据所述初始滚球半径基于滚球法提取图形二值化后的轨迹散点图的初始轮廓图;
    获取初始轮廓图围合区域的连通域的个数;
    基于所述连通域的个数,提取图形二值化后的轨迹散点图的轮廓图作为扫地机器人作业的地面区域的地图数据。
  11. 根据权利要求10所述的装置,其中,所述生成模块进一步被设置为:
    在连通域的个数大于第一预设阈值的情况下,更新滚球半径值,根据所述更新滚球半径基于滚球法再次提取图形二值化后的轨迹散点图的新轮廓图,计算新轮廓图围合区域的连通域的个数;
    在所述连通域的个数小于等于第一预设阈值的情况下,不再更新滚球半径值,将最后更新的滚球半径值作为最终滚球半径值,根据最终滚球半径值基于滚球法提取图形二值化后的轨迹散点图的最终轮廓图作为扫地机器人作业的地面区域的地图数据。
  12. 根据权利要求11所述的装置,其中,所述相似度计算模块进一步被设置为:
    获取不同的用户的扫地机器人作业的地面区域的轨迹点云,其中,所述轨迹点云表示多个轨迹散点;
    将获取的不同的用户的轨迹点云分别执行随机一致性采样,获得不同的用户的采样点点云数据;
    将不同的用户的采样点点云数据进行匹配计算,得到不同的用户的采样点点云数据的平移损失参数Rti和旋转损失参数Rri,计算不同的用户的扫地机器人的点云的距离Li为Li=Rti*Rri
    比较Li与第二预设阈值的大小,当Li大于第二预设阈值,继续采样并计算Li的值;
    当Li小于第二预设阈值,取采样的所有Li的平均值L_avg作为两个用 户的扫地机器人的地面区域轨迹点云距离,得到两个扫地机器人对应两个用户的相似度为1/L_avg。
  13. 根据权利要求10或11所述的装置,其中,所述初始滚球半径值通过所述轨迹散点图中两点之间的最大欧式距离确定。
  14. 根据权利要求10或11所述的装置,其中,所述生成模块进一步设置为:
    扫描两遍所述初始轮廓图,将所述初始轮廓图中存在的所有连通域找出并标记;
    根据标记个数获得初始轮廓图围合区域的连通域的个数。
  15. 一种计算机可读的存储介质,所述计算机可读的存储介质包括存储的程序,其中,所述程序运行时执行权利要求1至7中任一项所述的方法。
  16. 一种电子装置,包括存储器和处理器,所述存储器中存储有计算机程序,所述处理器被设置为通过所述计算机程序执行权利要求1至7中任一项所述的方法。
PCT/CN2023/073863 2022-08-30 2023-01-30 基于扫地机器人的用户相似度计算方法及装置、存储介质 Ceased WO2024045489A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211049711.9 2022-08-30
CN202211049711.9A CN115456057A (zh) 2022-08-30 2022-08-30 基于扫地机器人的用户相似度计算方法及装置、存储介质

Publications (1)

Publication Number Publication Date
WO2024045489A1 true WO2024045489A1 (zh) 2024-03-07

Family

ID=84301879

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/073863 Ceased WO2024045489A1 (zh) 2022-08-30 2023-01-30 基于扫地机器人的用户相似度计算方法及装置、存储介质

Country Status (2)

Country Link
CN (1) CN115456057A (zh)
WO (1) WO2024045489A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115456057A (zh) * 2022-08-30 2022-12-09 海尔优家智能科技(北京)有限公司 基于扫地机器人的用户相似度计算方法及装置、存储介质
CN119337129B (zh) * 2023-09-19 2026-04-03 同济大学 一种基于移动场景的集群联邦学习方法及装置

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109000655A (zh) * 2018-06-11 2018-12-14 东北师范大学 机器人仿生室内定位导航方法
CN110362099A (zh) * 2018-03-26 2019-10-22 科沃斯机器人股份有限公司 机器人清扫方法、装置、机器人及存储介质
CN112487234A (zh) * 2020-11-25 2021-03-12 北京集光通达科技股份有限公司 基于分裂型层次聚类的轨迹分析方法、设备及存储介质
CN112562029A (zh) * 2020-12-07 2021-03-26 深圳市愚公科技有限公司 图像重构方法、系统、电子设备及计算机可读存储介质
CN113520233A (zh) * 2021-09-16 2021-10-22 深圳市元鼎智能创新有限公司 多打扫模式的扫地机器人的控制方法、装置及扫地机器人
WO2022051252A1 (en) * 2020-09-01 2022-03-10 Ceva Technologies, Inc. Planar robots dead-reckoning with optical flow, wheel encoder and inertial measurement unit
CN115456057A (zh) * 2022-08-30 2022-12-09 海尔优家智能科技(北京)有限公司 基于扫地机器人的用户相似度计算方法及装置、存储介质

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111133336B (zh) * 2017-09-04 2024-04-02 联邦科学工业研究组织 用于执行定位的方法和系统
CN109522803B (zh) * 2018-10-18 2021-02-09 深圳乐动机器人有限公司 一种室内区域划分和识别方法、装置及终端设备
CN111709965B (zh) * 2020-08-19 2020-12-22 浙江欣奕华智能科技有限公司 扫地机器人的地图的优化方法及装置
CN114676210B (zh) * 2022-03-07 2025-01-17 达闼机器人股份有限公司 一种机器人扫图方法、装置、设备及存储介质

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110362099A (zh) * 2018-03-26 2019-10-22 科沃斯机器人股份有限公司 机器人清扫方法、装置、机器人及存储介质
CN109000655A (zh) * 2018-06-11 2018-12-14 东北师范大学 机器人仿生室内定位导航方法
WO2022051252A1 (en) * 2020-09-01 2022-03-10 Ceva Technologies, Inc. Planar robots dead-reckoning with optical flow, wheel encoder and inertial measurement unit
CN112487234A (zh) * 2020-11-25 2021-03-12 北京集光通达科技股份有限公司 基于分裂型层次聚类的轨迹分析方法、设备及存储介质
CN112562029A (zh) * 2020-12-07 2021-03-26 深圳市愚公科技有限公司 图像重构方法、系统、电子设备及计算机可读存储介质
CN113520233A (zh) * 2021-09-16 2021-10-22 深圳市元鼎智能创新有限公司 多打扫模式的扫地机器人的控制方法、装置及扫地机器人
CN115456057A (zh) * 2022-08-30 2022-12-09 海尔优家智能科技(北京)有限公司 基于扫地机器人的用户相似度计算方法及装置、存储介质

Also Published As

Publication number Publication date
CN115456057A (zh) 2022-12-09

Similar Documents

Publication Publication Date Title
US20200327680A1 (en) Visual target tracking method and apparatus based on deep adversarial training
CN111709965B (zh) 扫地机器人的地图的优化方法及装置
CN109584302B (zh) 相机位姿优化方法、装置、电子设备和计算机可读介质
CN109522803A (zh) 一种室内区域划分和识别方法、装置及终端设备
WO2024045489A1 (zh) 基于扫地机器人的用户相似度计算方法及装置、存储介质
US20140280231A1 (en) Dynamically Expiring Crowd-Sourced Content
US10332291B2 (en) Content aware image editing
CN110598705B (zh) 图像的语义标注方法及装置
CN112270384B (zh) 一种回环检测方法、装置及电子设备和存储介质
JP2022168167A (ja) 画像処理方法、装置、電子機器及び記憶媒体
CN110677488A (zh) 物联网系统的路径规划方法和装置、存储介质及电子装置
CN115265520A (zh) 智能作业设备及其建图方法、装置和存储介质
CN115761529B (zh) 图像处理方法和电子设备
CN111854751A (zh) 导航目标位置确定方法、装置、可读存储介质及机器人
US20220011119A1 (en) Generating and improving upon agricultural maps
CN107948721A (zh) 推送信息的方法和装置
CN115979272B (zh) 一种基于关联概率语义地图的机器人目标搜索方法
CN111738447A (zh) 基于时空关系学习的移动社交网络用户关系推断方法
CN115670308A (zh) 一种扫地机器人的清扫方法、装置、设备及存储介质
CN112435274B (zh) 一种基于面向对象分割的遥感影像面状地物提取方法
CN114648814A (zh) 人脸活体检测方法及模型的训练方法、装置、设备及介质
Miao et al. Road centreline extraction from classified images by using the geodesic method
Wang et al. Edge detection and simulation analysis of multimedia images based on intelligent monitoring robot
CN115191213B (zh) 机器人定位方法、装置、割草机器人以及存储介质
CN113672193B (zh) 音频数据播放方法与装置

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: 23858532

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 23858532

Country of ref document: EP

Kind code of ref document: A1