Method and device for generating robot sensing map and robot

Abstract

The invention relates to the technical field of robots, and discloses a method for generating a robot sensing map, which comprises the following steps: performing online mapping through a simultaneous positioning and mapping technology to obtain an initial sensing map under a robot coordinate system; according to the prior information of the robot, determining a target area needing to be complemented in the initial perception map and a space entity type corresponding to the target area; and complementing the target area according to the spatial entity type to obtain a target perception map. According to the method, an initial sensing map under a robot coordinate system is obtained by online mapping through a simultaneous positioning and mapping technology, a specific target area is complemented in combination with prior information, and the robot can obtain a more complete environment model by complementing missing data of a view limited area, so that the robot can more accurately perform navigation and obstacle avoidance, and the navigation efficiency of the robot is improved. And the accuracy of environment perception when the robot executes the task is improved. The invention further discloses a device for generating the robot perception map, the robot and a storage medium.

Description

Technical Field

[0001] The present application relates to the field of robotics technology, and for example, to a method and device for generating a robot perception map, and a robot. Background Art

[0002] At present, with the rapid development of automation technology, robots are increasingly used in industrial inspection, environmental monitoring, safety inspection and other fields. Especially in transportation fields such as railways and subways, robots are used to regularly inspect the bottom and sides of vehicles to ensure the safety and reliability of transportation systems. Therefore, when robots perform these tasks, how to accurately perceive and reconstruct a complete three-dimensional environment model under the condition of limited sensor field of view has become an urgent problem to be solved.

[0003] Related technologies disclose technical solutions for enabling robots to achieve online perception through a variety of sensors. For example, environmental information can be obtained through visual perception, distance perception, and environmental perception sensors. Visual perception technology obtains image data through cameras and uses image processing and computer vision algorithms for analysis; distance perception technology uses sensors such as laser radar to detect obstacles and build maps.

[0004] In the process of implementing the embodiments of the present application, it is found that there are at least the following problems in the related technology:

[0005] When using related technologies to perceive the task environment through sensors, the sensor's field of view is limited, making it impossible to obtain a complete three-dimensional model of the target. For example, when a laser radar is scanning, it is difficult to cover all areas due to its fixed scanning angle and range, especially for objects directly in front of the robot's direction of travel, and only a partial view can be obtained.

[0006] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0007] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical components or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.

[0008] The embodiments of the present application provide a method and device for generating a robot perception map, a robot, and a computer-readable storage medium, so as to improve the accuracy of the robot's perception of the environment when the robot perceives the task environment through sensors, and reduce the impact of the limited field of view of the sensor on the robot's execution of tasks.

[0009] In some embodiments, the method includes: obtaining an initial perception map in the robot coordinate system by online mapping through simultaneous positioning and mapping technology; determining, based on the robot's prior information, the target area that needs to be completed in the initial perception map and the spatial entity type corresponding to the target area; completing the target area according to the spatial entity type to obtain a target perception map.

[0010] Optionally, the target area is completed according to the spatial entity type to obtain a target perception map, including: using a point cloud processing algorithm to segment the point cloud data of the initial perception map to obtain initial point cloud data corresponding to the target area; fitting and / or denoising the initial point cloud data to obtain target geometric parameters corresponding to the spatial entity type; completing the initial point cloud data according to the target geometric parameters, and merging the completed target point cloud data with the point cloud data of the initial perception map to obtain a target perception map.

[0011] Optionally, the initial point cloud data is fitted and / or denoised to obtain target geometric parameters corresponding to the spatial entity type, including: fitting the initial point cloud data using a fitting algorithm to identify the geometric surface corresponding to the spatial entity type of the target area to obtain a fitting model; denoising the fitting model and obtaining the target geometric parameters of the denoised fitting model.

[0012] Optionally, the initial point cloud data is completed according to the target geometric parameters, including: generating a complete geometric model according to the target geometric parameters and parametric equations corresponding to the spatial entity type; aligning the complete geometric model with the point cloud data of the initial perception map, adjusting the position and orientation of the complete geometric model, and obtaining the target point cloud data.

[0013] Optionally, the completed target point cloud data is merged with the point cloud data of the initial perception map, including: obtaining the target position of the spatial entity of the target area; moving the complete geometric model corresponding to the target point cloud data to the target position; and cropping the merged point cloud data according to the size of the real object.

[0014] Optionally, an initial perception map in the robot coordinate system is obtained by online mapping through simultaneous positioning and mapping technology, including: online mapping through simultaneous positioning and mapping based on point cloud data of a lidar and / or inertial data of an inertial measurement unit.

[0015] Optionally, based on the robot's prior information, the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area are determined, including: obtaining prior information and sensor data of the current task scene; identifying the target area that cannot be covered by the sensor's field of view based on the prior information and sensor data; cropping the target area according to the corresponding algorithm, and identifying the type of spatial entity corresponding to the target area.

[0016] In some embodiments, the device includes: a processor and a memory storing program instructions, and the processor is configured to execute the above-mentioned method for generating a robot perception map when executing the above-mentioned program instructions.

[0017] In some embodiments, the robot includes: a robot body; and the above-mentioned device for generating a robot perception map is installed on the robot body.

[0018] In some embodiments, the computer-readable storage medium stores program instructions, and when the program instructions are run, the method for generating a robot perception map is executed.

[0019] The method and device for generating a robot perception map, the robot, and the computer-readable storage medium provided in the embodiments of the present application can achieve the following technical effects:

[0020] The initial perception map in the robot coordinate system is obtained by online mapping using simultaneous localization and mapping technology, and the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area are determined based on the robot's prior information. Finally, the target area is completed according to the spatial entity type to obtain the target perception map. The initial perception map in the robot coordinate system is obtained by online mapping using simultaneous localization and mapping technology, and specific target areas are completed in combination with prior information. By completing the missing data in the restricted field of view, the robot can obtain a more complete environmental model, which enables the robot to navigate and avoid obstacles more accurately, improving the accuracy of the robot's perception of the environment when performing tasks.

[0021] The above general description and the following description are exemplary and explanatory only and are not intended to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] One or more embodiments are exemplarily described by corresponding drawings, which do not limit the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements, and the drawings do not constitute a scale limitation, and wherein:

[0023] Figure 1 is a schematic diagram of a method for generating a robot perception map provided by an embodiment of the present application;

[0024] Figure 2 is a schematic diagram of another method for generating a robot perception map provided by an embodiment of the present application;

[0025] Figure 3 is a schematic diagram of another method for generating a robot perception map provided by an embodiment of the present application;

[0026] Figure 4is a schematic diagram of another method for generating a robot perception map provided by an embodiment of the present application;

[0027] Figure 5 is a schematic diagram of a device for generating a robot perception map provided in an embodiment of the present application;

[0028] Figure 6 It is a schematic diagram of a robot provided in an embodiment of the present application.

[0029] Reference numerals:

[0030] 800: device for chiller control; 801: processor; 802: memory; 803: communication interface; 804: bus; 900: robot. DETAILED DESCRIPTION

[0031] In order to be able to understand the features and technical contents of the embodiments of the present application in more detail, the implementation of the embodiments of the present application is described in detail below in conjunction with the accompanying drawings. The attached drawings are for reference only and are not used to limit the embodiments of the present application. In the following technical description, for the convenience of explanation, a full understanding of the disclosed embodiments is provided through multiple details. However, one or more embodiments can still be implemented without these details. In other cases, to simplify the drawings, well-known structures and devices can be simplified for display.

[0032] The terms "first", "second", etc. in the specification and claims of the embodiments of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the embodiments of the present application described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions.

[0033] Unless otherwise stated, the term "plurality" means two or more.

[0034] In the embodiment of the present application, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0035] The term "and / or" is a description of the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or, A and B.

[0036] The term "correspondence" may refer to an association relationship or a binding relationship. The correspondence between A and B means that there is an association relationship or a binding relationship between A and B.

[0037] The embodiment of the present application discloses a robot, including a perception module, a determination module and a completion module. The perception module is configured to obtain an initial perception map in the robot coordinate system by online mapping through simultaneous positioning and mapping technology. The determination module is configured to determine the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area based on the robot's prior information. The completion module is configured to complete the target area according to the spatial entity type to obtain a target perception map. The robot also includes a control module, which includes a processor, which is electrically connected to the above-mentioned electrical components and is used to control the above-mentioned electrical components to perform actions.

[0038] Figures 1 to 4 Schematic diagram of a method for generating a robot perception map provided in an embodiment of the present application. Any of the following methods can be executed in a robot, or in a server or terminal device that is connected to the robot for communication. In the embodiment of the present application, the robot is used as the execution subject to explain the solution.

[0039] Based on the above robot structure, such as Figure 1 As shown, the embodiment of the present application provides a method for generating a robot perception map, comprising:

[0040] S01, the robot obtains the initial perception map in the robot coordinate system through online mapping using simultaneous localization and mapping technology.

[0041] In this step, SLAM (Simultaneous Localization and Mapping) technology refers to the process in which a robot starts from an unknown position in an unknown environment, perceives the environment through its own sensors (such as lidar, camera, etc.) during movement, realizes its own positioning, and builds an incremental map at the same time. In SLAM, the robot can continuously update and optimize its own position estimation and map information by perceiving the features in the surrounding environment (such as points, lines, surfaces, etc.) and combining its own motion information, thereby helping the robot to achieve autonomous navigation, environmental perception, obstacle avoidance and other functions, and improving the robot's intelligence level and adaptability.

[0042] S02, the robot determines the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area based on the robot's prior information.

[0043] In this step, prior information refers to the information related to the environment or task that the robot knows before perception and mapping. This information can be used to guide and optimize the SLAM process and improve the accuracy and efficiency of perception. For example, prior information can include a general understanding of the environment, such as indoor or outdoor environment, corridor, room size; or the types of objects expected to be found in the environment, such as vehicles, pedestrians, pillars, walls; or the approximate location of known objects or features, such as parking signs are usually located next to the road; or the constraints on the robot's operation, such as safety boundaries, areas that are not allowed to enter; etc.

[0044] S03, the robot completes the target area according to the spatial entity type to obtain a target perception map.

[0045] In this step, the initial perception map can be realized by using a point cloud algorithm and geometric characteristics and / or spatial parameter expressions of spatial geometric bodies.

[0046] Different targets are processed in parallel, so that different types of geometric entity completion can be performed simultaneously. In addition, the same type of completion in the same area, such as multiple pillars, can be calculated in parallel, thereby improving the completion efficiency.

[0047] The method for generating a robot perception map provided in an embodiment of the present application is adopted, and an initial perception map in the robot coordinate system is obtained by online mapping through simultaneous positioning and mapping technology, and the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area are determined according to the robot's prior information, and finally the target area is completed according to the spatial entity type to obtain a target perception map. The initial perception map in the robot coordinate system is obtained by online mapping through simultaneous positioning and mapping technology, and the specific target area is completed in combination with prior information. By completing the missing data in the area with limited field of view, the robot can obtain a more complete environmental model, thereby enabling the robot to navigate and avoid obstacles more accurately, and improving the accuracy of the robot's perception of the environment when performing tasks.

[0048] Based on the above robot structure, such as Figure 2 As shown, the embodiment of the present application provides a method for generating a robot perception map, comprising:

[0049] S01, the robot obtains the initial perception map in the robot coordinate system through online mapping using simultaneous localization and mapping technology.

[0050] S02, the robot determines the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area based on the robot's prior information.

[0051] S21, the robot uses a point cloud processing algorithm to segment the point cloud data of the initial perception map to obtain initial point cloud data corresponding to the target area.

[0052] In this step, the missing spatial entity is a pillar. Taking the pillar as an example, this paper explains how to use the point cloud processing algorithm to segment the point cloud data of the initial perception map. For the cropped target area point cloud, the DBSCAN (DBSCANDensity-Based Spatial Clustering of Applications with Noise) algorithm is used in combination with the SDF (Statistical Outlier Removal) algorithm to segment the point cloud of the initial perception map into individual pillar monomers, namely the initial point cloud data.

[0053] S22, the robot performs fitting and / or denoising processing on the initial point cloud data to obtain target geometric parameters corresponding to the spatial entity type.

[0054] S23, the robot completes the initial point cloud data according to the target geometric parameters, and merges the completed target point cloud data with the point cloud data of the initial perception map to obtain the target perception map.

[0055] Using the method for generating a robot perception map provided in an embodiment of the present application, the robot uses a point cloud processing algorithm to segment the point cloud data of the initial perception map to obtain initial point cloud data corresponding to the target area, and fits and / or denoises the initial point cloud data to obtain target geometric parameters corresponding to the spatial entity type, and finally completes the initial point cloud data according to the target geometric parameters, and merges the completed target point cloud data with the point cloud data of the initial perception map to obtain a target perception map. By filling the blank areas in the original point cloud data with a completion algorithm, the robot can generate a more complete environmental model, namely a target perception map, when performing a task, so that when the sensor performance is limited, the completion technology can also provide sufficient environmental information and reduce the robot's dependence on high-end sensors. Therefore, through point cloud segmentation and completion, especially in areas where the sensor's field of view is limited, the robot can obtain more accurate environmental perception.

[0056] Optionally, the robot fits and / or denoises the initial point cloud data to obtain target geometric parameters corresponding to the spatial entity type, including: the robot uses a fitting algorithm to fit the initial point cloud data, identifies the geometric surface corresponding to the spatial entity type of the target area, and obtains a fitting model; the robot denoises the fitting model and obtains the target geometric parameters of the denoised fitting model.

[0057] Among them, in the above steps, the initial point cloud data is processed. After the column monomers are segmented by the point cloud processing algorithm, they should be two planes that are perpendicular to each other and form an "L" shape. Therefore, it is necessary to further use the point cloud plane for fitting. The three-dimensional space plane mathematical model can be expressed by its general equation: Ax+By+Cz+D=0; where N=[A,B,C] is the normal vector of the plane, D is the distance between the plane and the origin, and the plane parameter model can be expressed by a column vector M: M=[A,B,C,D]. The RANSAC (Random Sample Consensus) algorithm is used to calculate the parameters of the plane model. After one fitting, the distance from the point to the plane is calculated; assuming that there is an arbitrary point in space (x,y,z), which is expressed as: P=[x,y,z,1]; then the distance from the point to the plane is calculated as: dis=M T *P; when the distance dis from the point to the plane is less than the distance threshold σ, the point is considered to be a point on the plane, so a plane point set is obtained And the remaining point set, and then continue to fit the second plane in the remaining point set When the number of points of P1 and P2 is greater than the number threshold σ, and the angle between the normal vectors of the two planes is close to 90°, the robot determines that the correct L-shaped point cloud has been obtained:

[0058]

[0059] Next, the distance dis is used to further remove the points outside the plane as noise points to obtain a cleaner L-shaped point cloud, that is, the target geometric parameters corresponding to the spatial entity type are obtained.

[0060] In this way, the robot uses the fitting algorithm to fit the initial point cloud data, identifies the geometric surface corresponding to the spatial entity type of the target area, obtains the fitting model, denoises the fitting model and obtains the target geometric parameters of the denoised fitting model. Through the fitting algorithm, the geometric shape representing the surface of the actual object can be identified from the initial point cloud data, and the structural quality of the data is improved by denoising. The fitting model provides accurate geometric parameters, such as the normal vector and offset of the plane, the radius and height of the cylinder, etc., and then the completion algorithm can generate more accurate completion data through accurate fitting models and geometric parameters, thereby improving the integrity of the overall point cloud map.

[0061] Optionally, the robot completes the initial point cloud data according to the target geometric parameters, including: the robot generates a complete geometric model according to the target geometric parameters and parametric equations corresponding to the spatial entity type; the robot aligns the complete geometric model with the point cloud data of the initial perception map, adjusts the position and orientation of the complete geometric model, and obtains the target point cloud data.

[0062] Among them, the target point cloud data is the completed geometric model.

[0063] Among them, in the above steps, the robot can generate an artificially synthesized complete column point cloud, a left L-shaped point cloud and a right L-shaped point cloud, that is, a complete geometric model, according to the size parameters of the actual column, such as length, width and height. The height of the complete column point cloud (z-direction length) is several times the height of the actual column, and the left L-shaped point cloud and the right L-shaped point cloud are generated according to the actual column height. Calculate the AABB (AxisAlignedBoundingBox axis-aligned bounding box) of each column, approximate the geometric center of the column Pillar_center with the center of the AABB, and determine whether it is a left column or a right column based on the distance from the center of the column to the origin. Then, the artificially synthesized complete column point cloud is aligned to the real column through the point cloud alignment algorithm to obtain a three-dimensional rigid body transformation matrix (rotation and translation matrix, (R, T)) that can fit the complete column point cloud to the column to be completed. Among them, the purpose of point cloud alignment is to use the information of its overlapping parts to solve the optimal relative position relationship between the two point clouds, that is, the rotation and translation matrix. Suppose N P The point set to be registered P(source), N Q The optimal transformation matrix (R, T) from P to Q should minimize the sum of the distances between the corresponding points of the two point clouds:

[0064]

[0065] Commonly used point cloud registration algorithms can be divided into coarse registration algorithms and fine registration algorithms. Because the synthetic pillars are required to fit the real pillars as closely as possible, the fine registration algorithm ICP (Iterative Closest Point) is used to achieve this. ICP requires an initial position to prevent falling into the local optimum. The geometric center Pillar_center mentioned above can be used as the initial position of the artificially synthesized pillars. Taking the left pillar Pillar_Left_1 as an example, the artificially synthesized left L-shaped pillar (source) is first moved to the geometric center Pillar_center of the actual pillar (target), and the variant algorithm point-to-planeICP is used to further align the two point clouds. The registration results are applied to the synthesized complete pillar point cloud, so that the complete pillar point cloud fits the pillar to be completed. Then, the same process is performed on each pillar, and finally the artificially synthesized target point cloud data placed on the actual pillar is obtained.

[0066] In this way, the robot generates a complete geometric model based on the target geometric parameters and the parametric equations corresponding to the spatial entity type, thereby providing a complete three-dimensional representation of the environment, including areas outside the sensor's field of view. The robot aligns the complete geometric model with the point cloud data of the initial perception map, adjusts the position and orientation of the complete geometric model, and obtains the target point cloud data. Through the alignment process, the spatial position and orientation of the complete geometric model can be accurately adjusted, thereby ensuring the consistency of the model with the actual environment.

[0067] Optionally, the robot merges the completed target point cloud data with the point cloud data of the initial perception map, including: the robot obtains the target position of the spatial entity of the target area; the robot moves the complete geometric model corresponding to the target point cloud data to the target position; the robot crops the merged point cloud data according to the size of the real object.

[0068] Among them, in the above steps, the target point cloud data can be cropped according to the prior information such as the height of the actual column to obtain point cloud data consistent with the actual column. At this time, all the target point cloud data, that is, the completed column, is already in the robot's coordinate system. The completed column is merged with the initial perception map point cloud to obtain the completed target perception map.

[0069] In this way, the robot obtains the target position of the spatial entity in the target area and moves the complete geometric model corresponding to the target point cloud data to the target position. By moving the completed geometric model to the target position, the robot can accurately locate the exact position of the spatial entity in the environment. The merged point cloud data provides a seamless environmental mapping that includes areas within and outside the field of view of the robot sensor. Finally, the robot crops the merged point cloud data according to the size of the real object. The cropping process ensures that the merged point cloud data is consistent with the real object in size and shape, improving the accuracy of the data.

[0070] Based on the above robot structure, such as Figure 3 As shown, the embodiment of the present application provides a method for generating a robot perception map, comprising:

[0071] S31, the robot builds a map online by simultaneously positioning and building a map based on the point cloud data of the laser radar and / or the inertial data of the inertial measurement unit.

[0072] In this step, SLAM online mapping is performed through the non-repetitive scanning lidar and inertial measurement unit carried by the robot. At the same time, the laser inertial odometer information of real-time mapping is used to convert the online map point cloud to the current robot coordinate system to form three-dimensional perception information of the robot's surrounding environment, namely the initial perception map.

[0073] S02, the robot determines the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area based on the robot's prior information.

[0074] S21, the robot uses a point cloud processing algorithm to segment the point cloud data of the initial perception map to obtain initial point cloud data corresponding to the target area.

[0075] S22, the robot performs fitting and / or denoising processing on the initial point cloud data to obtain target geometric parameters corresponding to the spatial entity type.

[0076] S23, the robot completes the initial point cloud data according to the target geometric parameters, and merges the completed target point cloud data with the point cloud data of the initial perception map to obtain the target perception map.

[0077] Using the method for generating a robot perception map provided in an embodiment of the present application, the robot builds a map online through simultaneous positioning and mapping based on the point cloud data of the laser radar and / or the inertial data of the inertial measurement unit. The laser radar provides accurate distance measurement, while the inertial data of the inertial measurement unit provides real-time information about the robot's movement. Through the fusion of the point cloud data of the laser radar and the inertial data of the inertial measurement unit, the environmental map can be updated in real time through the simultaneous positioning and mapping algorithm. Even in indoor or urban canyon environments with poor GPS signals, the positioning accuracy of the robot can be improved.

[0078] Based on the above robot structure, such as Figure 4 As shown, the embodiment of the present application provides a method for generating a robot perception map, comprising:

[0079] S31, the robot builds a map online by simultaneously positioning and building a map based on the point cloud data of the laser radar and / or the inertial data of the inertial measurement unit.

[0080] S41, the robot obtains prior information and sensor data of the current task scene.

[0081] S42, the robot identifies the target area that cannot be covered by the sensor field of view based on the prior information and the sensor data.

[0082] In this step, for example, in a train inspection scenario, the missing spatial entity can be considered as a combination of different types of spatial geometric bodies such as planes, cubes, cylinders, etc., and the prior information of the bottom and / or side inspection scenes can be used to determine the target area that needs to be completed in the current robot coordinate system.

[0083] S43, the robot crops the target area according to the corresponding algorithm and identifies the type of spatial entity corresponding to the target area.

[0084] In this step, the missing spatial entity is a column. Taking the column as an example, how to determine the spatial entity type corresponding to the target area is explained. In the initial perception map, each column can be considered as a cuboid with a fixed length, width and height. Although the center of each column is at a different position of the robot, all the columns around the robot can be cropped out through a fixed cropping frame to identify the spatial entity type corresponding to the target area.

[0085] S21, the robot uses a point cloud processing algorithm to segment the point cloud data of the initial perception map to obtain initial point cloud data corresponding to the target area.

[0086] S22, the robot performs fitting and / or denoising processing on the initial point cloud data to obtain target geometric parameters corresponding to the spatial entity type.

[0087] S23, the robot completes the initial point cloud data according to the target geometric parameters, and merges the completed target point cloud data with the point cloud data of the initial perception map to obtain the target perception map.

[0088] Using the method for generating a robot perception map provided in an embodiment of the present application, the robot obtains prior information and sensor data of the current task scene. Through the prior information, the robot has an expected understanding of the structure of the task scene and can predict areas that the sensor may not cover. Therefore, based on the prior information and sensor data, the robot can accurately identify the target area outside the sensor's field of view. The prior information also provides the types of objects that may exist. Therefore, in combination with the sensor data, the robot can crop the target area according to the corresponding algorithm and identify the type of spatial entity corresponding to the target area, so that the robot can select a completion algorithm specific to that type, thereby improving the efficiency and accuracy of the completion.

[0089] In actual application, except for pillars, other spatial entity fields can be completed by at least one of the above embodiments. For planes, for plane holes caused by limited viewing angles, the characteristics of the sensor itself, etc., the plane fitting method in the above process can be used to fit the spatial mathematical model of the plane to be completed, and then use the mathematical model to generate a dense plane point cloud within a certain range to play a role in completion. For cylinders, the spatial cylinder can also use the fitting method to obtain the parameters of the cylinder to be completed, and use the parametric equation to regenerate a uniformly dense cylindrical surface point cloud for the cylinder to achieve the purpose of completion. Among them, the general equation for fitting the spatial cylinder is: (x-x0) 2 +(y-y0) 2 +(z-z0) 2 -[a(x-x0)+b(y-y0)+

[0090] c(z-z0)] 2=r 2 .

[0091] Combination Figure 5 As shown, an embodiment of the present application provides a device 800 for generating a robot perception map, including a processor 801 and a memory 802. Optionally, the device may also include a communication interface 803 and a bus 804. Among them, the processor 801, the communication interface 803, and the memory 802 can communicate with each other through the bus 804. The communication interface 803 can be used for information transmission. The processor 801 can call the logic instructions in the memory 802 to execute the method for generating a robot perception map of the above embodiment.

[0092] In addition, the logic instructions in the memory 802 described above may be implemented in the form of software functional units and when sold or used as independent products, may be stored in a computer-readable storage medium.

[0093] The memory 802 is a computer-readable storage medium that can be used to store software programs and computer executable programs, such as program instructions / modules corresponding to the method in the embodiment of the present application. The processor 801 executes the function application and data processing by running the program instructions / modules stored in the memory 802, that is, the method for generating a robot perception map in the above embodiment is implemented.

[0094] The memory 802 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function; the data storage area may store data created according to the use of the terminal device, etc. In addition, the memory 802 may include a high-speed random access memory and may also include a non-volatile memory.

[0095] Combination Figure 6 As shown, an embodiment of the present application provides a robot 900, including: a robot body, and the above-mentioned device 700 (800) for generating a robot perception map. The device 700 (800) for generating a robot perception map is installed on the robot body. The installation relationship described here is not limited to placement inside the robot, but also includes installation connections with other components of the robot, including but not limited to physical connections, electrical connections, or signal transmission connections. It can be understood by those skilled in the art that the device 700 (800) for generating a robot perception map can be adapted to a feasible robot body, thereby realizing other feasible embodiments.

[0096] An embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions are configured to execute the above-mentioned method for generating a robot perception map.

[0097] The technical solution of the embodiment of the present application can be embodied in the form of a software product, which is stored in a storage medium and includes one or more instructions for a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in the embodiment of the present application. The aforementioned storage medium may be a non-transient storage medium, including: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0098] The above description and the accompanying drawings sufficiently illustrate the embodiments of the present application to enable those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, process and other changes. The embodiments represent only possible variations. Unless explicitly required, individual components and functions are optional, and the order of operations may vary. Portions and features of some embodiments may be included in or replace portions and features of other embodiments. Moreover, the terms used in this application are only used to describe the embodiments and are not used to limit the claims. As used in the description of the embodiments and the claims, the singular form of "a" or "an" shall be used unless the context clearly indicates otherwise.

[0099] (a), "an" and "the" are intended to include plural forms as well. Similarly, the term "and / or" as used in this application is intended to include any and all possible combinations of one or more of the associated listed items. In addition, when used in this application, the term "including"

[0100] "comprise" and its variations "comprises" and / or comprising refer to the presence of stated features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups thereof. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method or device including the elements. In this article, each embodiment may focus on the differences from other embodiments, and the same and similar parts between the various embodiments may refer to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method part disclosed in the embodiments, then the relevant parts can refer to the description of the method part.

[0101] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software may depend on the specific application and design constraints of the technical solution. The technicians may use different methods to implement the described functions for each specific application, but such implementations should not be considered to exceed the scope of the embodiments of the present application. The technicians may clearly understand that, for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above may refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here.

[0102] In the embodiments disclosed herein, the disclosed methods and products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units can be only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to implement this embodiment. In addition, each functional unit in the embodiment of the present application may be integrated in a processing unit, or each unit may exist physically alone, or two or more units may be integrated in one unit.

[0103] The flowchart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to the embodiment of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of the code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two continuous boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. In the description corresponding to the flowchart and the block diagram in the accompanying drawings, the operations or steps corresponding to different boxes can also occur in a different order from the order disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two continuous operations or steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which can depend on the functions involved. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or actions, or may be implemented by a combination of dedicated hardware and computer instructions.

Claims

1. A method for generating a robot perception map, characterized in that: include: The initial perception map in the robot coordinate system is obtained through online mapping using simultaneous localization and mapping technology; According to the robot's prior information, determine the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area; The target area is completed according to the spatial entity type to obtain a target perception map.

2. The method according to claim 1, characterized in that The target area is completed according to the spatial entity type to obtain the target perception area, including: Use a point cloud processing algorithm to segment the point cloud data of the initial perception map to obtain initial point cloud data corresponding to the target area; Perform fitting and / or denoising processing on the initial point cloud data to obtain target geometric parameters corresponding to the spatial entity type; The initial point cloud data is completed according to the target geometric parameters, and the completed target point cloud data is merged with the point cloud data of the initial perception map to obtain the target perception map.

3. The method according to claim 2, characterized in that Perform fitting and / or denoising on the initial point cloud data to obtain the target geometric parameters corresponding to the spatial entity type, including: Use the fitting algorithm to fit the initial point cloud data, identify the geometric surface corresponding to the spatial entity type of the target area, and obtain the fitting model; The fitting model is denoised and the target geometric parameters of the denoised fitting model are obtained.

4. The method according to claim 2, characterized in that: The initial point cloud data is completed according to the target geometric parameters, including: Generate a complete geometric model based on target geometric parameters and parametric equations corresponding to the spatial entity type; The complete geometric model is registered with the point cloud data of the initial perception map, and the position and orientation of the complete geometric model are adjusted to obtain the target point cloud data.

5. The method according to claim 2, characterized in that: Merge the completed target point cloud data with the point cloud data of the initial perception map, including: Get the target position of the spatial entity in the target area; Move the complete geometric model corresponding to the target point cloud data to the target position; The merged point cloud data is cropped according to the size of the real object.

6. The method according to any one of claims 1 to 5, characterized in that: The initial perception map in the robot coordinate system is obtained through online mapping using simultaneous localization and mapping technology, including: Online mapping is achieved through simultaneous localization and mapping based on LiDAR point cloud data and / or inertial measurement unit inertial data.

7. The method according to any one of claims 1 to 5, characterized in that: Based on the robot's prior information, determine the target area that needs to be completed in the initial perception map and the type of spatial entity corresponding to the target area, including: Obtain prior information and sensor data of the current mission scenario; Based on prior information and sensor data, identify target areas that are not covered by the sensor's field of view; The target area is cropped according to the corresponding algorithm, and the spatial entity type corresponding to the target area is identified.

8. A device for generating a robot perception map, comprising a processor and a memory storing program instructions, characterized in that: The processor is configured to execute the method for generating a robot perception map as described in any one of claims 1 to 7 when running the program instructions.

9. A robot, characterized in that: include: Robot body; The device for generating a robot perception map as described in claim 8 is installed on the robot body.

10. A computer-readable storage medium storing program instructions, characterized in that: When the program instructions are executed, the computer is used to execute the method for generating a robot perception map as described in any one of claims 1 to 7.

Images