Robot dumping area recognition optimization method and system based on hierarchical connectivity maximization

Abstract

The application belongs to the technical field of robot dumping recognition, and provides a robot dumping area recognition optimization method and system based on hierarchical connected maximization, and the scheme is as follows: in the rough recognition stage, according to the operation characteristics of a mobile operation robot, the target container is recognized and positioned through a vision module when far away from a target area, and a target container information acquisition position required in fine recognition is generated according to the positioning; in the fine recognition stage, the robot first acquires information of the target container and a source container and generates corresponding connectable areas of the target container and the source container, and then a connected maximization method is used to recognize and optimize the robot dumping area. The autonomous recognition problem of the mobile operation robot dumping area is solved, the method can be applied to most target containers and source containers, and the generalization ability of the robot dumping operation skill is enhanced.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of robot pouring recognition, and particularly relates to a robot pouring area recognition optimization method and system based on hierarchical connectivity maximization. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute the prior art.

[0003] The pouring skill of a robot is an operation skill with very wide application, and has many direct application scenarios whether in home service or laboratory operation. The pouring operation has a complex dynamic process due to the diversity of the operation objects (involving liquids, particles or powders, etc.), which brings great challenges to the robot operation skill learning. How to learn accurate and widely applicable pouring operation skills has become an important topic in the field of robot operation skill learning.

[0004] The existing research on robot pouring operation skill learning mainly focuses on the control and learning of fixed robot arm pouring actions, and a fixed pouring area is often given to the robot, which makes it difficult to realize the autonomous selection and optimization of the pouring area by the robot, so that the robot is difficult to autonomously complete the whole process of the pouring task. SUMMARY

[0005] In order to solve at least one technical problem in the above background art, the present application provides a robot pouring area recognition optimization method and system based on hierarchical connectivity maximization, which can realize the recognition and optimization of the pouring area of a mobile working robot, can be widely adapted to different pouring scenes, can be easily extended to the pouring operation skill learning process of a fixed robot arm, and provides support for the learning and realization of accurate and widely applicable pouring operation skills of a robot.

[0006] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0007] The first aspect of the present application provides a robot pouring area recognition optimization method based on hierarchical connectivity maximization, comprising the following steps:

[0008] Preliminarily positioning a target container based on an environmental information image to obtain a target container space area;

[0009] Generating a trajectory planning and a path planning of a mobile base of a mobile robot according to the target container space area, and controlling the robot arm to move to an information collection position for fine recognition according to the above planning to obtain a pouring area; wherein the fine recognition process specifically comprises:

[0010] Obtaining image information of the target container and a source container;

[0011] Based on the image information of the target container and the source container, corresponding connectable areas are generated;

[0012] A pour point and a pour direction of the source container in the plane that make the intersection of the connectable areas of the source container and the target container maximum are searched by using a connectivity maximization method, a position of the pour point on the source container is further obtained, and a pour area is obtained by combining the position of the pour point.

[0013] The second aspect of the present application provides a robot pour area recognition optimization system based on hierarchical connectivity maximization, comprising:

[0014] A coarse recognition module is configured to preliminarily locate the target container based on the image of the environment information, to obtain a spatial area of the target container, and to generate a trajectory planning and a path planning of a mobile base of the mobile robot according to the spatial area of the target container;

[0015] A fine recognition module is configured to control the robotic arm to move to an information acquisition position for fine recognition according to the planning, to obtain a pour area; the fine recognition process specifically comprises:

[0016] Image information of the target container and the source container is acquired;

[0017] Based on the image information of the target container and the source container, corresponding connectable areas are generated;

[0018] A pour point and a pour direction of the source container in the plane that make the intersection of the connectable areas of the source container and the target container maximum are searched by using a connectivity maximization method, a position of the pour point on the source container is further obtained, and a pour area is obtained by combining the position of the pour point.

[0019] The third aspect of the present application provides a computer readable storage medium.

[0020] A computer readable storage medium has a computer program stored thereon, and the program is executed by a processor to implement the steps in the robot pour area recognition optimization method based on hierarchical connectivity maximization of the first aspect.

[0021] The fourth aspect of the present application provides a computer device.

[0022] A computer device comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps in the robot pour area recognition optimization method based on hierarchical connectivity maximization of the first aspect when executing the program.

[0023] Compared with the prior art, the present application has the following beneficial effects:

[0024] 1. The application proposes a connected maximization method to determine the optimal dumping area of the robot and proposes a corresponding simplified calculation method, solving the problem of autonomous recognition of the dumping area of the mobile working robot.

[0025] 2. The recognition method proposed in the application has strong generalization function and can be applied to most target containers and source containers, enhancing the generalization ability of the robot dumping operation skill.

[0026] 3. The application proposes a hierarchical connected maximization dumping area recognition framework for mobile working robot dumping area recognition, which can be applied to mobile working robots and can also be migrated to fixed mechanical arms. It has a wide range of applications.

[0027] The advantages of the additional aspects of the application will be partially given in the following description, partially will become obvious from the following description, or will be known by the practice of the application. BRIEF DESCRIPTION OF DRAWINGS

[0028] The drawings accompanying the specification of the application are used to provide further understanding of the application, the illustrative embodiments of the application and the description thereof are used to explain the application, and do not constitute improper limitation of the application.

[0029] Figure 1 is a mobile working robot system and dumping scene provided by the embodiment of the application;

[0030] Figure 2 is a hierarchical connected maximization robot dumping area recognition and optimization method flowchart provided by the embodiment of the application;

[0031] Figure 3 is a connectable area model diagram of the target container provided by the embodiment of the application;

[0032] Figure 4 is a connectable area model diagram of the source container provided by the embodiment of the application;

[0033] Figure 5 is a projection plane, rectangular segmentation and weight distribution diagram of the connectable area of the source container provided by the embodiment of the application;

[0034] Figure 6 is a projection plane and opening estimation containing diagram of the connectable area of the source container provided by the embodiment of the application;

[0035] Figure 7 is a simplified search process execution diagram provided by the embodiment of the application. DETAILED DESCRIPTION

[0036] The application will be further described below in combination with the drawings and embodiments.

[0037] It should be noted that the following detailed description is exemplary in nature and is intended to provide further description of the application. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0038] It is also to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments in accordance with the present application. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, steps, operations, elements, components, and / or groups thereof, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and / or groups thereof.

[0039] In order to enable the mobile work robot to master the skills of autonomous recognition and optimization of dumping area and realize autonomous completion of the complete dumping task process, the present application proposes a dumping area recognition and optimization method based on hierarchical connectivity maximization, which realizes autonomous recognition and optimization of the dumping area by the mobile work robot.

[0040] For the proposed method of the present application, a mobile work robot system and a dumping task scene as shown in Figure 1 are built.

[0041] The mobile work robot system includes a mobile base module (which includes an omni-directional wheel motion control subsystem, a radar positioning subsystem and other subsystems to realize the motion control and obstacle avoidance functions of the mobile robot), a mechanical arm module (which includes a mechanical arm submodule and a gripper submodule for holding a container, for realizing the clamping of a source container and the realization of camera actions and dumping actions), a vision module (which includes an RGB-D camera, a camera support and the like, for the recognition of a source container, a target container and a work environment); the dumping scene also includes a source container (a container for dumping material, which contains the material to be dumped in the initial state, and the opening is an axisymmetric pattern), a target container (a container for receiving material) and a matching work environment.

[0042] The goal of the robot dumping operation skill is to accurately and widely dump material from a source container into a target container. Accurate dumping requires the robot to minimize material overflow throughout the dumping operation, and widely adaptable dumping requires the robot to adapt to various source containers, target containers and dumping scenes.

[0043] In order to realize accurate and widely applicable completion of dumping operation of a mobile working robot, the application provides a dumping area identification and optimization method based on hierarchical connectivity maximization. The core point of the method is to adopt hierarchical processing on the dumping area identification and optimization process according to the working characteristics of the mobile working robot, and a source container connectivity area model capable of realizing prediction of a fluid flow area in the source container is provided. Finally, the identification of the optimal dumping area is realized through connectivity maximization.

[0044] Embodiment one

[0045] As shown in Figure 2 , the embodiment provides a dumping area identification and optimization method for a robot based on hierarchical connectivity maximization, which comprises the following steps:

[0046] First stage: rough identification stage

[0047] The rough identification stage identifies and locates the target container through a vision module when the target area is far away according to the working characteristics of the mobile working robot, and generates the information acquisition position of the target container required in the fine identification stage, comprising the following steps:

[0048] Step 1: obtaining environment information (image I work ) by using a vision module.

[0049] Step 2: extracting image I work information to identify and locate the target container, and obtaining the target container area Ω receiver .

[0050] Step 3: generating the information acquisition position P camera of the vision module in the fine identification stage according to the spatial area Ω receiver of the target container.

[0051] In step 2, the following contents are included:

[0052] A YOLO algorithm trained by a container data set is used to identify the target container, and the area of the target container in the image is obtained and the identification frame is marked;

[0053] According to the position of the identified target container in the image, the image is subjected to grayscale processing and binary processing to obtain the target container area centroid coordinates (x c , y c ), and further obtain the center coordinates (x u , y u ) of the upper edge of the target container, and then cooperate with the target distance obtained by the depth camera to calculate the world coordinates (x w , y w , z w ) of the target container;

[0054] Wherein, step 3 includes the following contents:

[0055] Increase h on the Z-axis of the target container world coordinates camera Get the information collection position P of the fine recognition stage vision module camera w , yw w +z camera +h camera ).hThe value of h is different according to the camera model and resolution, and the distance that can obtain the best shooting effect is set.

[0056] Wherein, h camera is the distance that needs to be kept from the object when the camera takes pictures to obtain information, which needs to comprehensively consider the reachable range of the mechanical arm and the camera shooting parameters, for example, the distance test range of RealSense camera is 0.1-10m, so the camera distance from the measurement object should be kept at a distance of more than 0.1m, and cannot be close.

[0057] Get the information collection position P camera after the coarse recognition stage, use the Rapidly-Expan ding Random Tree (RRT) algorithm to plan the motion trajectory of the mobile chassis, and control the mobile chassis to move to the vicinity of the target position according to the motion trajectory, and then use the Inverse Kinematics method to generate the joint parameters of the mechanical arm according to the information collection position P camera generated in the coarse recognition stage, and move the mechanical arm vision module to the information collection position.

[0058] Second stage: fine recognition stage

[0059] In the fine recognition stage, the robot first obtains the information of the target container and the source container and generates the corresponding communicable area, and then uses the connected maximum method to identify and optimize the robot pouring area.

[0060] The specific method is as follows:

[0061] Step 1: use the vision module to obtain the image information of the target container (image I receiver ).

[0062] Step 2: extract the image I receiver information to obtain the edge track τ receiver and the center O receiver of the opening of the target container, and generate the communicable area Ω receiver of the source container according to the communicable area model.

[0063] Step 3: use the vision module to obtain the image information of the source container (image I container ).​

[0064] Step 4: Extracting image I container information, obtaining source container opening edge information τ container , long axis length l of the opening l and short axis length l s , pouring point m of the opening, generating the communicable region Ω of the source container according to the communicable region model container .

[0065] Step 5: identifying and optimizing the robot pouring region using the connectivity maximization method, and combining the pouring point position to obtain the pouring region Ω pour .

[0066] Wherein, step 2 includes the following contents:

[0067] (1) respectively denoising, graying, and binarizing the image I receiver , using the contour detection method and the feature moment method in OpenCV to obtain the opening edge track τ receiver and the center O receiver of the target container.

[0068] (2) according to the opening edge track τ receiver of the target container, generating the region Ω receiver with the track surrounding the area as the base and the parameter h receiver as the height, which is the communicable region of the target container, as shown in Figure 3 .

[0069] Wherein, h receiver may be defined as infinity, or can be defined according to the following formula:

[0070] h receiver = l l / 2 + h container

[0071] Wherein, h container is the model parameter of the source container communicable region. h receiver is the length of the upward projection of the opening plane of the target container.

[0072] Step 4 includes the following steps:

[0073] (1) respectively denoising, graying, and binarizing the image I container , using the contour detection method and the feature moment method in OpenCV to obtain the opening edge track τ container and the center O receiver of the target container, and calculating the long axis length l l and the short axis length l sThe intersection of the vertical line of the centroid on the short axis of the circumscribed rectangle of the contour is taken as the pouring point m of the source container.

[0074] (2) This step will generate the connectable region of the source container, according to the calculated long axis distance l l of the opening region s , the area of the liquid flowing out of the source container is simplified as a right triangle with an angle of α and a height of l l / 2+h container , and a triangular prism with a height of l s , where the angle α can be set to different values according to the state of the fluid and the pouring angular velocity of the source container. The higher the viscosity of the liquid, the smaller the value of α, and the greater the pouring angular velocity of the source container, the greater the value of α. When the fluid properties are unknown, α can be temporarily set to 45°. h container is the distance from the pouring point m of the source container to the opening surface of the target container, and is an adjustable parameter greater than or equal to 0. h container can be set to 1 cm, that is, the pouring point of the source container is 1 cm away from the upper surface of the target container. In addition, let the projection of the pouring point m on the target opening surface be m', and construct a unit vector in the direction of the cup opening with m' as the starting point. The connectable region Ω container of the source container constructed as described above is shown in Figure 4 .

[0075] The formula of the connectable region of the source container is shown in equation 1:

[0076] Ω container = l s ×(l l / 2+h container ) 2 tanα (1)

[0077] The advantages and rationality of using this method to set the connectable region are:

[0078] Fluid flows out of an opening, and using precise fluid mechanics principles to predict the trajectory of the fluid and the target area is a complex calculation process, and the generalization ability of the results of fluid mechanics calculation is poor due to the different properties of different fluids. By using the simplified model of the parameterized (α, l l , l s , h) triangular prism, the fluid flow area can be well covered, and it has wide applicability and can easily identify and optimize the pouring area. In addition, Figure 4In the case of the source container angle of 90° and the circular opening of the circular container, the change of the pouring angle in the actual pouring process can be adapted by adjusting the model parameter a. The opening shape and size of the source container can also be generalized within the axisymmetric graph range, and the model is suitable for a wide range of applications.

[0079] Step 5 includes the following steps:

[0080] The maximum value of the intersection region of the target container and the source container is calculated and represented as follows:

[0081]

[0082]

[0083] The purpose of the maximum value operation is to search for the position and direction of the source container and the target container that can be connected to the maximum intersection region of the plane Plane receiver , and further obtain the position of the pouring point m on the source container.

[0084] The further optimization problem can be simplified using the following calculation method, specifically:

[0085] (1) Project the source container connectable region Ω container onto the plane Plane receiver , and set the projection region as F, and the projection region shape as a rectangle.

[0086] (2) Divide the projection plane F of the source container connectable region into a×b rectangles, as shown in Figure 5 , and calculate the weight w ij of the rectangle according to the parameters l l / 2+h container of the connectable region.

[0087] The weight calculation formula is as follows:

[0088] w ij =i×(l l / 2+h container ) / a

[0089] (3) In Plane receiver , search for the sum of weights of the projection plane of the source container connectable region that is contained in the target container opening trajectory τ receiver (Actually, if the weight rectangle is contained in an area of more than 1 / 2, it is considered to be fully contained), and the corresponding pouring point m of the maximum weight value is the pouring point and pouring direction of the source container, as shown in . Figure 6

[0090] ​​(4) Further, the search process can be simplified according to the existing pouring knowledge, and the target container opening centroid O receiver is taken as the starting point, the interval angle θ is radially outward, and the target container opening trajectory τ receiver intersects, and the intersection is taken as the starting point, and the is searched along the direction of the centroid, as shown in Figure 7

[0091] (5) The final target container opening trajectory τ receiver The region composed of the pouring point m with the largest weight sum in the target container opening trajectory τ pour .

[0092] Embodiment Two

[0093] The embodiment provides a robot pouring area recognition and optimization system based on hierarchical connectivity maximization, which comprises the following steps:

[0094] A rough recognition module is used to preliminarily locate a target container based on environmental information images to obtain a target container space region; and a trajectory planning and path planning of a mobile base of a mobile robot are generated according to the target container space region;

[0095] A fine recognition module is used to control a mechanical arm to move to an information acquisition position to perform fine recognition according to the planning, so as to obtain a pouring area; wherein the fine recognition process specifically comprises the following steps:

[0096] Image information of the target container and a source container is acquired;

[0097] Corresponding connectable regions are generated based on the image information of the target container and the source container;

[0098] A connectivity maximization method is used to search for a pouring point and a pouring direction of the source container in a plane such that the intersection of the connectable regions of the source container and the target container is the largest, a position of the pouring point on the source container is further obtained, and a pouring region is obtained by combination according to the position of the pouring point.

[0099] Embodiment Three

[0100] The embodiment provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to realize the steps in the robot pouring area recognition and optimization method based on hierarchical connectivity maximization in the embodiment one.

[0101] Embodiment Four

[0102] ​The embodiment provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and the processor implements the steps in the robot dumping area recognition optimization method based on hierarchical connectivity maximization as described in the embodiment one when executing the program.

[0103] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage, etc.) containing computer-usable program code.

[0104] The present application is described with reference to flowcharts and / or block diagrams of the method, device (system), and computer program product according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for carrying out the functions specified in the flowcharts and / or block diagrams.

[0105] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce a product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for carrying out the functions specified in the flowcharts and / or block diagrams.

[0106] These computer program instructions can also be loaded into a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 means for carrying out the functions specified in the flowcharts and / or block diagrams.

[0107] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by a computer program instructing relevant hardware, and the program can be stored in a computer readable storage medium. When the program is executed, the program can include the processes of the above-mentioned embodiment methods. The storage medium can be a magnetic disc, an optical disc, a read-only memory (ROM), a random access memory (RAM), or the like.

[0108] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for robot pour area recognition optimization based on hierarchical connectivity maximization, characterized in that, The method comprises the following steps: Preliminary positioning of the target container based on environmental information images to obtain a target container spatial region; Generating trajectory planning and path planning of the moving base of the mobile robot according to the spatial region of the target container, and controlling the movement of the mechanical arm to an information acquisition position for fine identification according to the planning to obtain a pouring region; wherein the fine identification process specifically comprises: Obtaining image information of the target container and the source container; Generating corresponding connectable regions based on the image information of the target container and the source container; Using a connectivity maximization method to search for a pouring point and a pouring direction of the source container in a plane such that the intersection of the connectable regions of the source container and the target container is maximum, further obtaining the position of the pouring point on the source container, and combining the positions of the pouring points to obtain the pouring region; The generating of the connectable region of the source container based on the image information of the source container comprises: Performing noise reduction, grayscale and binarization processing on the image information of the source container, obtaining the opening edge trajectory and the centroid of the target container by using a contour detection method and a feature moment method, and calculating the long axis distance and the short axis distance of the opening contour according to the circumscribed rectangle of the contour, and calculating the vertical intersection point of the centroid on the short axis of the circumscribed rectangle as the pouring point of the source container; According to the calculated long axis distance of the opening area and the short axis distance , the area of the liquid flowing out of the source container is simplified as a right triangle with the angle of α, the base area of , and the height of , wherein the angle of α can be set to different values according to the state of the liquid and the pouring angle speed of the source container, wherein is the distance from the pouring point m of the source container to the opening surface of the target container; the source container connectable area is calculated and generated in combination with the above parameters and the formula of the source container connectable area. The using of the connectivity maximization method to search for the pouring point and the pouring direction of the source container in the plane such that the intersection of the connectable regions of the source container and the target container is maximum comprises: Projecting the connectable region of the source container to a search plane to obtain a projection plane; Equally dividing the projection plane of the connectable region of the source container into multiple rectangles, and calculating the weighting values of the rectangles according to the parameters of the connectable region; The projection plane of the source container communicable area searched in the search plane is contained in the sum of weights of the target container opening trajectory, and the maximum value of the weights corresponds to The corresponding pouring point m is the pouring point and pouring direction of the source container, wherein, The projection of the pouring point m on the target container opening surface is A unit vector is constructed along the target container opening direction with as the starting point.

2. The hierarchical connectivity maximization based robot pour region identification optimization method of claim 1, wherein, The preliminary positioning of the target container based on the environmental information images to obtain the target container spatial region comprises: Identifying the target container by using a trained YOLO algorithm to obtain the region of the target container in the image and mark the identification frame; Performing grayscale processing and binarization processing on the image according to the position of the identified target container in the image to obtain the centroid coordinates of the target container region and further obtain the center coordinates of the upper edge of the target container, and then calculating the world coordinates of the target container by using the target distance obtained by the depth camera.

3. The hierarchical connectivity maximization based robot pour region identification optimization method of claim 1, wherein, The generating of the trajectory planning and path planning of the moving base of the mobile robot according to the spatial region of the target container, and the controlling of the movement of the mechanical arm to the information acquisition position according to the planning comprises: increasing on the Z axis of the target container world coordinates obtaining the information collection position of the fine recognition stage, refers to the distance required to be kept from the photographed object when the camera takes pictures to obtain information After obtaining the information acquisition position, planning the motion trajectory of the moving base by using a rapidly-exploring random tree algorithm, and controlling the moving base to move to the vicinity of the target position according to the motion trajectory, and then generating the joint parameters of the mechanical arm by using an inverse kinematics method according to the information acquisition position generated in the coarse identification stage, and controlling the movement of the visual module of the mechanical arm to the information acquisition position.

4. The hierarchical connectivity maximization based robot pour region identification optimization method of claim 1, wherein, The generating of the connectable region of the target container based on the image information of the target container comprises: Performing noise reduction, grayscale and binarization processing on the image of the target container, and obtaining the opening edge trajectory and the centroid of the target container by using a contour detection method and a feature moment method; According to the opening edge trajectory of the target container, the area surrounded by the trajectory is taken as the base, and the parameter is high, and the generated area is the communicable area of the target container, wherein, is the upward projection length of the opening plane of the target container.

5. The hierarchical connectivity maximization based robot pour region identification optimization method of claim 1, wherein, Further obtaining the position of the pouring point on the source container, and combining the positions of the pouring points to obtain the pouring region comprises: With the target container opening centroid as the starting point, interval angle Outwardly doing ray and target container opening trajectory intersection, and this as the starting point, Along the ray to the centroid direction search, the final target container opening trajectory contains the maximum weight of the region composed of the m group of pouring points, which is the identified and optimized pouring area.

6. A system for identifying and optimizing dump areas based on hierarchical connectivity maximization, characterized in that, The method for robot pouring area recognition optimization based on hierarchical connectivity maximization according to any one of claims 1-5, comprising: a coarse recognition module for preliminarily positioning a target container based on an environmental information image to obtain a target container space region; and generating a trajectory plan and a path plan of a mobile base of a mobile robot according to the target container space region; a fine recognition module for controlling a mechanical arm to move to an information acquisition position for fine recognition according to the plan to obtain a pouring area; wherein the fine recognition process specifically comprises: acquiring image information of the target container and a source container; generating corresponding connectable regions based on the image information of the target container and the source container; using a connectivity maximization method to search for a pouring point and a pouring direction of the source container in a plane such that the intersection of the connectable regions of the source container and the target container is the largest, further obtaining the position of the pouring point on the source container, and combining the pouring point position to obtain the pouring area.

7. A computer-readable storage medium having stored thereon a computer program, characterized in that, The program is executed by the processor to implement the steps in the method for robot pouring area recognition optimization based on hierarchical connectivity maximization according to any one of claims 1-5.

8. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps in the method for robot pouring area recognition optimization based on hierarchical connectivity maximization according to any one of claims 1-5.

Images