A pose determination method and apparatus, autonomous mobile device, and storage medium

Abstract

Embodiments of the present application disclose a pose determination method and device, an autonomous mobile device and a storage medium. Specifically, the method comprises: acquiring an original image including a stereoscopic label collected by the autonomous mobile device; determining a region image of a region where the stereoscopic label is located in the original image; and determining a pose of the autonomous mobile device according to the region image. The technical solution of the embodiments of the present application identifies the position and pixel information of the stereoscopic label in the original image acquired by the autonomous mobile device, determines the pose of the autonomous mobile device based on the identification of the stereoscopic label, solves the problem of high identification requirement and poor identification accuracy of a planar label, and improves the positioning accuracy of the autonomous mobile device in a label identification scene.

Description

Technical Field

[0001] This application relates to the field of robotics, and more particularly to a pose determination method, apparatus, autonomous mobile device, and storage medium. Background Technology

[0002] With the continuous upgrading of the automation industry, more and more industries are moving towards intelligentization, with the most obvious example being the use of robots to replace human labor in production or services. For autonomously moving robots, accurate positioning is crucial.

[0003] Currently, many service robots rely on environmental landmarks for self-localization during autonomous movement. For example, preset graphic labels can be fixed in the robot's working environment. The robot can identify these graphic labels to determine its position and posture in the working environment. However, this method of identifying graphic labels is easily affected by environmental interference and limitations of the robot's own identification equipment, resulting in poor positioning accuracy. Summary of the Invention

[0004] This application provides a pose determination method, apparatus, autonomous mobile device, and storage medium to improve the pose determination accuracy of autonomous mobile devices.

[0005] In a first aspect, embodiments of this application provide a pose determination method, including:

[0006] Acquire raw images, including stereo tags, from autonomous mobile devices;

[0007] Determine the region image of the area where the stereo label is located in the original image;

[0008] The pose of the autonomous mobile device is determined based on the regional image.

[0009] Secondly, embodiments of this application also provide a pose determination device, including:

[0010] The raw image acquisition module is used to acquire raw images, including stereo tags, collected by the autonomous mobile device;

[0011] The region image determination module is used to determine the region image of the area where the stereo label is located in the original image;

[0012] The pose determination module is used to determine the pose of the autonomous mobile device based on the positioning image.

[0013] Thirdly, embodiments of this application also provide an autonomous mobile device, including:

[0014] One or more processors;

[0015] Memory, used to store one or more programs;

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement a pose determination method as described in the first aspect of the present application.

[0017] Fourthly, embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a pose determination method as described in the first aspect of this application.

[0018] The technical solution of this application embodiment identifies the position and pixel information of the 3D label in the original image acquired by the autonomous mobile device, and determines the pose of the autonomous mobile device based on the recognition of the 3D label. This solves the problems of high requirements and poor recognition accuracy of planar label recognition, and improves the positioning accuracy of the autonomous mobile device in the label recognition scenario. Attached Figure Description

[0019] Figure 1A This is a flowchart of a pose determination method provided in Embodiment 1 of this application;

[0020] Figure 1A This is a schematic diagram of a highlight filling process provided in Embodiment 1 of this application;

[0021] Figure 2A This is a schematic diagram of a three-dimensional label provided in Embodiment 2 of this application;

[0022] Figure 2A This is a schematic diagram of a three-dimensional label provided in Embodiment 2 of this application;

[0023] Figure 2A This is a schematic diagram of the installation of a three-dimensional label provided in Embodiment 2 of this application;

[0024] Figure 3 This is a flowchart of a pose determination method provided in Embodiment 3 of this application;

[0025] Figure 3 This is a structural diagram of a pose determination device provided in Embodiment 4 of this application;

[0026] Figure 4 This is a structural diagram of an autonomous mobile device provided in Embodiment 5 of this application. Detailed Implementation

[0027] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the application and not intended to limit its scope. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present application, not the entire structure.

[0028] Example 1

[0029] Figure 1A This is a flowchart of the pose determination method provided in Embodiment 1 of this application. This embodiment is applicable to autonomous positioning using stereo tags. The method can be executed by a pose determination device, which can be implemented in software and / or hardware and specifically configured in an autonomous mobile device.

[0030] refer to Figure 1A The pose determination method shown includes the following steps:

[0031] S110. Acquire the original image, including the stereo tag, collected by the autonomous mobile device.

[0032] The autonomous mobile device can be a device that moves according to a preset program, such as a planar mobile robot, which can be a service robot, food delivery robot, logistics robot, etc. in different application scenarios. The 3D tag can be an auxiliary device for providing positioning for the autonomous mobile device. Compared with planar tags, 3D tags allow the autonomous mobile device to recognize the tag content even at more oblique angles, improving the recognition success rate and efficiency. The original image including the 3D tag can be an image captured by the autonomous mobile device using its built-in camera or other equipment, and this image includes the 3D tag used for positioning.

[0033] For example, assuming the 3D tag is set on the ceiling of the target room, the autonomous mobile device uses its own camera, which is set to shoot towards the ceiling, to capture the original information image of the ceiling containing the 3D tag information, thereby providing a basis for determining the pose of the autonomous mobile device based on the original image.

[0034] S120. Determine the region image of the area where the 3D label is located in the original image.

[0035] The region image can be the pixel portion of the original image that includes the 3D label. For example, in the original image captured by the autonomous mobile device, the 3D label only occupies a portion of the pixels. The pixel portion containing the 3D label can be cropped to obtain the region image. The current region image can be an image of a regular graphic including the pixel information of the 3D label, or it can be an image of an irregular graphic including the pixel information of the 3D label. This application embodiment does not limit the specific form of the region image. It is understood that when the planar label and the autonomous mobile device deviate in the horizontal direction, the original image of the planar label obtained by the autonomous mobile device is stored in perspective distortion, which will greatly reduce the image recognition accuracy. However, compared with planar labels, 3D labels allow the autonomous mobile device to recognize more angles, thereby better helping the autonomous mobile device to determine its pose. The region image can be determined using any image processing method in the prior art, and this application embodiment does not limit it.

[0036] S130. Determine the pose of the autonomous mobile device based on the regional image.

[0037] The aforementioned steps define a region image including 3D tags. Based on the information from the 3D tags in the region image, the pose of the autonomous mobile device is determined. The method for determining the pose of the autonomous mobile device based on the region image can employ any existing pose determination algorithm, such as a 3D pose estimation algorithm (e.g., SolvePnP), and this application embodiment does not limit this approach.

[0038] In one optional implementation, determining the pose of the autonomous mobile device based on the region image may include: denoising the region image to update the region image; and determining the pose of the autonomous mobile device based on the updated region image.

[0039] The region image obtained in the aforementioned steps can be understood as a cropped version of the original image, containing a 3D label. This region image may contain information that affects the recognition of the 3D label. Therefore, denoising processing is used to eliminate noise caused by the surface material of the 3D label or bright spots caused by lighting in the region image. Currently, denoising processing can employ any existing image denoising algorithm, such as a smoothing filter algorithm; this application embodiment is not limited to this. Then, based on the denoised region image, the pose of the autonomous mobile device is further determined. Eliminating noise and bright spots helps determine the pose based on the region image, improving the accuracy of pose determination for the autonomous mobile device.

[0040] Optionally, the denoising process for the region image to update the region image may include: performing shape recognition on the region image to determine the graphic region in the region image that matches the 3D label; and performing highlight filling on the graphic region to update the region image.

[0041] Shape recognition involves identifying pixel regions in the region image that resemble the shape of a 3D label, and determining the position of the pixel portion corresponding to the 3D label in the region image (i.e., the graphic region that matches the 3D label). Highlight filling is then applied to this graphic region to resolve issues such as overexposure of highlights and bright spots caused by light shining onto the 3D label.

[0042] For example, such as Figure 1B As shown, the 3D label consists of two spheres. Figure 1B The image shown is a processed region image of the original image acquired by the autonomous mobile device. It can be seen that before highlight filling, the sphere exhibits overexposure in the region image, which is detrimental to sphere recognition. For example, the center of the circle in the region image can be determined first, and then a circle with a preset radius can be determined based on this center. The pixels within this circle can be filled, and the filling color can be set according to the actual situation. The method for determining the center can employ any existing circle center determination algorithm, and the preset radius can be set by relevant technicians based on practical experience or extensive experimentation; this application does not limit this. The above implementation reduces the impact of overexposure or light spots on subsequent image processing and pose determination by filling pixels in the region image that have the same shape as the 3D label, thus helping to improve the accuracy of pose determination for the autonomous mobile device.

[0043] The technical solution of this application embodiment identifies the position and pixel information of the 3D label in the original image acquired by the autonomous mobile device, and determines the pose of the autonomous mobile device based on the recognition of the 3D label. This solves the problems of high requirements and poor recognition accuracy of planar label recognition, and improves the positioning accuracy of the autonomous mobile device in the label recognition scenario.

[0044] Example 2

[0045] Figure 2A This is a schematic diagram of a 3D tag provided in Embodiment 2 of this application. This embodiment refines the 3D tag design based on the foregoing technical solutions, improving the positioning accuracy of autonomous mobile devices. The 3D tag specifically includes:

[0046] 3D labels may include positioning devices and fixing devices; the positioning devices may be installed in the label application area by means of the fixing devices.

[0047] The positioning device assists the autonomous mobile device in identifying calibration points. Specifically, the positioning device is the label content of the 3D tag, and should be a three-dimensional object that can be recognized by the autonomous mobile device based on the image, such as a sphere or cylinder. Due to the special properties of a sphere, its planar image obtained from any direction is always a circle; therefore, a predetermined number of spheres are preferred as the positioning device. A fixing device is used to fix the positioning device, facilitating the acquisition of the 3D tag image by the autonomous mobile device; for example, an acrylic sheet can be used. Of course, the size, material, shape, and installation method of the positioning device are not limited in this embodiment.

[0048] like Figure 2A As shown, the positioning device consists of multiple spheres 210 mounted on a fixing device 220 according to a preset arrangement. This preset arrangement can be set by technical personnel based on actual conditions. In practice, the materials used for the three-dimensional positioning device and the fixing device can be matte to reduce glare and other issues caused by light. Additionally, an infrared reflective coating can be added to the surface to filter out ambient light and other noise, in conjunction with an infrared camera.

[0049] Optionally, the depth to which the positioning device is embedded in the fixing device shall not exceed a preset depth. It should be noted that, in order to obtain more easily identifiable image information when acquiring images of the 3D tag, the depth to which the positioning device is embedded in the fixing device should be appropriately adjusted to ensure that the exposed portion of the positioning device can be accurately identified by the image processing algorithm. Furthermore, the preset depth is determined based on the length of the positioning device along a fixed direction.

[0050] Taking a 3D spherical label as an example, such as Figure 2B As shown, the sphere 210 is embedded in the fixing device 220, and the direction in which the sphere 210 is embedded in the fixing device 220 can be represented as the direction from A to B along the auxiliary line AB. The preset embedding depth can be preset according to specific circumstances. In one specific embodiment, because the portion of the sphere 210 embedded in the fixing device 220 may cause occlusion (non-transparent fixing device) or optical distortion (transparent fixing device) when acquiring images, it affects the recognition of the sphere's shape. Therefore, in order to enable recognition even when the relative angle between the autonomous mobile device and the 3D label is 45°, the preset depth is set to be no more than 30% of the sphere's radius.

[0051] Furthermore, inlaying is not the only way to fix a positioning device to a fixing device. For example... Figure 2CAs shown, a hole can also be made in the fixing device, and the preset depth can be controlled by the size of the hole. A thin rope is then passed through the hole to fix the sphere to the fixing device, with most of the sphere exposed. This assists the autonomous mobile device in acquiring images of the 3D tag, recognizing the 3D tag, and determining its own pose. The above-described fixing methods are for illustrative purposes only and should not be construed as limiting the fixing methods of the 3D tag in the embodiments of this application.

[0052] Example 3

[0053] Figure 3 The flowchart of a pose determination method provided in Embodiment 3 of this application is shown in the figure. This embodiment of the application further refines the pose determination operation based on the foregoing embodiments and technical solutions to improve the positioning accuracy. This method can be executed by a pose determination device, which can be implemented in software and / or hardware and specifically configured in an autonomous mobile device.

[0054] In this embodiment, the 3D label consists of at least two spheres arranged in a preset pattern. The preset arrangement can be determined based on the experience of those skilled in the art or through extensive experimentation; this embodiment does not limit the number of spheres or the preset arrangement of all spheres.

[0055] refer to Figure 3 The pose determination method shown includes the following steps:

[0056] S310. Acquire the original image, including the stereo tag, collected by the autonomous mobile device.

[0057] S320. Determine the region image of the area where the 3D label is located in the original image.

[0058] S330. Based on the region image, determine the centers of at least two spheres in the region image.

[0059] It is understood that the region image includes pixel information corresponding to at least two spheres in the stereo label. Using a preset image processing method, the centers of at least two spheres can be determined based on the pixel information in the images corresponding to those spheres. The centers of each sphere can be used to assist the autonomous mobile device in label recognition and its own pose determination. Of course, the preset image processing method can be any combination of one or more image processing algorithms in the prior art, and this application embodiment does not limit this.

[0060] In one optional implementation, determining the centers of at least two spheres in the region image based on the region image may include: determining the positions and radii of at least two spheres in the region image based on the region image; determining the virtual centers of at least two spheres based on the sphere positions and a preset arrangement; and determining the centers of the spheres based on the sphere radii, the virtual centers, and a preset circle fitting algorithm.

[0061] The region image is processed accordingly to obtain the positions and radii of at least two spheres on the 3D label. These positions can be the positions of the spheres relative to the region image, the relative positions between the spheres, or the positions of the spheres relative to the autonomous mobile device. Since these different position parameters can be transformed into each other through spatial coordinate transformation, this embodiment of the application does not limit the specific parameters. For example, the Hough ring detection method can be used to process the region image to calculate the positions and radii of the spheres in the image.

[0062] A preset arrangement can represent the adjacency relationship of each sphere in the 3D label. Based on the position of each sphere and the adjacency relationship between them, information from non-label spheres in the region image is removed. It is understandable that due to lighting or other external factors, the image acquired by the autonomous mobile device may contain pixel information resembling the shape of the 3D label spheres. Based on the adjacency relationship between the spheres, pixel information of spheres whose shapes do not conform to the relationship is excluded, thus obtaining the correct information of each sphere in the image. A preset image processing algorithm determines the virtual center of each sphere in the image. It is understandable that because the position and radius of the spheres obtained by Hough ring detection have certain errors, the obtained virtual center will also have corresponding errors.

[0063] Based on the virtual sphere center and radius, a preset circle fitting algorithm is used to reduce the errors generated in the aforementioned steps, thereby obtaining a high-precision sphere center at the sub-pixel level. The preset circle fitting algorithm can be any existing circle fitting algorithm, such as a combination of nonlinear optimization algorithms like least squares, Gauss-Newton iteration, and RANSAC.

[0064] In the above embodiments, by obtaining the sphere positions and preset arrangements, the pixel information of non-labeled spheres is filtered out, improving the sphere recognition accuracy and helping to improve the accuracy of subsequent pose determination of autonomous mobile devices. Simultaneously, through circle fitting, the center of the sphere is further determined with higher precision, which helps to improve the accuracy of pose determination of autonomous mobile devices.

[0065] S340. Determine the pose of the autonomous mobile device based on the center of the sphere.

[0066] Based on the position of the center of each sphere relative to the autonomous mobile device, the three-dimensional position and orientation of the autonomous mobile device relative to the stereo label are calculated using a preset three-dimensional pose estimation algorithm (such as the SolvePnP algorithm).

[0067] The technical method of this application embodiment determines the center of each sphere in the region image based on the region image, and then determines the pose of the autonomous mobile device based on each sphere center. Identifying the center of each sphere in a 3D label can effectively solve the errors caused by perspective distortion and other factors when identifying planar labels in the prior art, thereby improving the accuracy of pose determination for the autonomous mobile device.

[0068] Example 4

[0069] Figure 4 This is a structural diagram of the pose determination device provided in Embodiment 4 of this application. This embodiment is applicable to autonomous positioning using stereo tags. The device can be implemented in software and / or hardware and can be configured in an autonomous mobile device. Figure 4 As shown, the pose determination device 400 may include: an original image acquisition module 410, a region image determination module 420, and a pose determination module 430, wherein,

[0070] The original image acquisition module 410 is used to acquire original images, including stereo tags, collected by the autonomous mobile device;

[0071] The region image determination module 420 is used to determine the region image of the area where the stereo label is located in the original image;

[0072] The pose determination module 430 is used to determine the pose of the autonomous mobile device based on the positioning image.

[0073] The technical solution of this application embodiment identifies the position and pixel information of the 3D label in the original image acquired by the autonomous mobile device, and determines the pose of the autonomous mobile device based on the recognition of the 3D label. This solves the problems of high requirements and poor recognition accuracy of planar label recognition, and improves the positioning accuracy of the autonomous mobile device in the label recognition scenario.

[0074] In one alternative embodiment, the 3D label may include a positioning device and a fixing device; the positioning device may be installed in the label affixing area by the fixing device.

[0075] In one alternative embodiment, the depth to which the positioning device is embedded in the fixing device is no greater than a preset depth.

[0076] In one alternative implementation, the preset depth is determined based on the length of the positioning device along a fixed direction.

[0077] In one alternative implementation, the 3D label consists of at least two spheres arranged in a preset manner.

[0078] In one alternative embodiment, the pose determination module 430 may include:

[0079] The sphere center determination unit is used to determine the centers of at least two spheres in the region image based on the region image.

[0080] The pose determination unit is used to determine the pose of the autonomous mobile device based on the center of the ball.

[0081] In one alternative implementation, the sphere center determination unit may include:

[0082] The position radius determination sub-unit is used to determine the position and radius of at least two spheres in the region image based on the region image.

[0083] The virtual sphere center determination subunit is used to determine the virtual sphere centers of at least two spheres based on their positions and a preset arrangement.

[0084] The sphere center determination sub-unit is used to determine the sphere center based on the sphere radius, the virtual sphere center, and a preset circle fitting algorithm.

[0085] In one alternative embodiment, the pose determination module 430 may include:

[0086] The image update unit is used to denoise the region image in order to update the region image;

[0087] The pose localization unit is used to determine the pose of the autonomous mobile device based on the updated region image.

[0088] In one alternative implementation, the image updating unit may include:

[0089] The shape recognition subunit is used to perform shape recognition on the region image and determine the graphic region in the region image that matches the 3D label;

[0090] The highlight fill subunit is used to perform highlight fill processing on the graphic area to update the area image.

[0091] The pose determination device provided in this application can execute the pose determination method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects for executing each pose determination method.

[0092] Example 5

[0093] Figure 5 This is a structural diagram of an autonomous mobile device provided in Embodiment 5 of this application. Figure 5A block diagram is shown of an exemplary autonomous mobile device 512 suitable for implementing embodiments of this application. Figure 5 The autonomous mobile device 512 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0094] like Figure 5 As shown, the autonomous mobile device 512 is presented in the form of a general-purpose computing device. The components of the autonomous mobile device 512 may include, but are not limited to: one or more processors or processing units 516, system memory 528, and bus 518 connecting different system components (including system memory 528 and processing unit 516).

[0095] Bus 518 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0096] The autonomous mobile device 512 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the autonomous mobile device 512, including volatile and non-volatile media, removable and non-removable media.

[0097] System memory 528 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 530 and / or cache memory 532. Autonomous mobile device 512 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 534 may be used to read and write non-removable, non-volatile magnetic media (…). Figure 5 Not shown; commonly referred to as a "hard drive"). Although Figure 5 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 518 via one or more data media interfaces. Memory 528 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0098] A program / utility 540 having a set (at least one) of program modules 542 may be stored, for example, in memory 528. Such program modules 542 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 542 typically perform the functions and / or methods described in the embodiments of this application.

[0099] The autonomous mobile device 512 can also communicate with one or more external devices 514 (e.g., keyboard, pointing device, display 524, etc.), one or more devices that enable a user to interact with the autonomous mobile device 512, and / or any device that enables the autonomous mobile device 512 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 522. Furthermore, the autonomous mobile device 512 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 520. As shown, network adapter 520 communicates with other modules of the autonomous mobile device 512 via bus 518. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with the autonomous mobile device 512, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0100] The processing unit 516 executes various functional applications and data processing by running at least one of the multiple programs stored in the system memory 528, such as implementing the pose determination method provided in the embodiments of this application.

[0101] Example 6

[0102] This application also provides a computer-readable storage medium storing a computer program (or computer-executable instructions) thereon. When the program is executed by a processor, it is used to perform the pose determination method provided in this application: acquiring an original image including a stereo tag collected by an autonomous mobile device; determining a region image of the area where the stereo tag is located in the original image; and determining the pose of the autonomous mobile device based on the region image.

[0103] The computer storage medium in this application embodiment can be any combination of one or more computer-readable media. A computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0104] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0105] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0106] Computer program code for performing the operations of the embodiments of this application can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0107] Note that the above are merely preferred embodiments and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of this application, the scope of which is determined by the scope of the appended claims.

Claims

1. A pose determination method, characterized in that, include: Acquire raw images, including stereo tags, from autonomous mobile devices; A region image is determined in the original image where the 3D label is located; wherein, the 3D label is composed of at least two spheres arranged in a preset manner, the preset arrangement being used to characterize the adjacency relationship of each sphere in the 3D label; based on the position of each sphere and the adjacency relationship between each sphere, information represented by non-label spheres in the region image is removed; The pose of the autonomous mobile device is determined based on the region image; The step of determining the pose of the autonomous mobile device based on the region image includes: Based on the region image, determine the center of the at least two spheres in the region image; based on the center of the spheres, determine the pose of the autonomous mobile device.

2. The method according to claim 1, characterized in that, The 3D label includes a positioning device and a fixing device; the positioning device is installed in the label affixing area via the fixing device.

3. The method according to claim 2, characterized in that, The depth to which the positioning device is embedded in the fixing device is no greater than a preset depth.

4. The method according to claim 3, characterized in that, The preset depth is determined based on the length of the positioning device along a fixed direction.

5. The method according to claim 1, characterized in that, Determining the centers of the at least two spheres in the region image based on the region image includes: Based on the region image, determine the sphere positions and sphere radii of the at least two spheres in the region image; Based on the positions of the spheres and the preset arrangement, determine the virtual centers of the at least two spheres; The center of the sphere is determined based on the radius of the sphere, the virtual center of the sphere, and a preset circle fitting algorithm.

6. The method according to any one of claims 1-4, characterized in that, Determining the pose of the autonomous mobile device based on the region image further includes: The image of the region is denoised in order to update the image of the region; The pose of the autonomous mobile device is determined based on the updated region image.

7. The method according to claim 6, characterized in that, The step of denoising the region image to update the region image includes: Shape recognition is performed on the region image to determine the graphic region in the region image that matches the 3D label; The graphic region is filled with highlights to update the region image.

8. A pose determination device, characterized in that, include: The raw image acquisition module is used to acquire raw images, including stereo tags, collected by the autonomous mobile device; A region image determination module is used to determine the region image of the area where the 3D label is located in the original image; wherein, the 3D label is composed of at least two spheres arranged in a preset manner, the preset arrangement is used to characterize the adjacency relationship of each sphere in the 3D label; based on the position of each sphere and the adjacency relationship between each sphere, the information represented by non-label spheres in the region image is removed; The pose determination module is used to determine the pose of the autonomous mobile device based on the region image. The pose determination module includes: A sphere center determination unit is used to determine the center of the at least two spheres in the region image based on the region image. The pose determination unit is used at the center of the sphere to determine the pose of the autonomous mobile device.

9. An autonomous mobile device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement a pose determination method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements a pose determination method as described in any one of claims 1-7.

Images