Domestic waste classification treatment method, device and system

Abstract

The present invention relates to the field of garbage sorting, specifically to a method, device and system for sorting and processing domestic garbage, which visually identifies the distribution characteristic information of domestic garbage in a target area, divides the target area into several garbage community sub-areas and determines their grabbing parameters; based on the distribution position information of all garbage community sub-areas and the distribution position information of the manipulator group around the target area, determines the matching relationship between the manipulator and the garbage community sub-area and combines the grabbing parameters to determine the grabbing action path of the manipulator in the garbage community sub-area, and realizes the manipulator's special grabbing operation on the garbage community sub-area; based on the spatial morphological characteristic information of the domestic garbage that the manipulator is about to grab, adjusts the grabbing action state of the manipulator. The present invention can control all manipulators to sort garbage by region and category, quickly and accurately pick out garbage of the same category in a sorting operation of a manipulator, and realize efficient sorting and processing of domestic garbage.

Description

Technical Field

[0001] The present invention relates to the field of garbage sorting, and in particular to a method, device and system for sorting and processing domestic garbage. Background Art

[0002] Garbage classification and recycling is an important part of realizing the recycling of garbage resources. The daily average amount of urban domestic waste is huge, and urban domestic waste includes different types of recyclable resources such as glass bottles, plastic bottles and waste paper. At present, the recycling rate of urban domestic waste is low, mainly due to factors such as the complicated classification and sorting of domestic waste and the heavy workload.

[0003] Specifically, the existing domestic waste is sorted and recycled using a semi-automatic method, that is, the domestic waste is moved by a conveyor belt, and at the same time, the sorting personnel pick out the recyclable waste visible to the naked eye near the conveyor belt. The operating efficiency of the above-mentioned domestic waste sorting and recycling method mainly depends on the sorting personnel's visual recognition accuracy and sorting operation proficiency. When the sorting personnel are faced with a large amount of domestic waste stacked on each other, it will not be possible to accurately and quickly screen out all recyclable waste, which seriously affects the efficiency of domestic waste sorting and processing. Therefore, how to synchronously control multiple manipulators to carry out partitioned classification and processing of domestic waste, and ensure that each manipulator can accurately sort out domestic waste belonging to the same category, is of great significance to improving the automation and efficiency of domestic waste sorting and processing. Summary of the invention

[0004] In order to accurately control all the manipulators to sort the garbage by region and category during the synchronous sorting process of domestic garbage by using multiple manipulators, quickly and accurately pick out the garbage of the same or similar category in one sorting operation of one manipulator, and realize the efficient sorting process of domestic garbage, the present invention provides a method for sorting domestic garbage, which comprises the following steps:

[0005] Visually identify the target area to obtain the domestic waste distribution characteristic information of the target area; divide the target area into a number of garbage cluster sub-areas based on the domestic waste distribution characteristic information; determine the grabbing parameters of the garbage cluster sub-areas based on the relative position information of the domestic waste in the garbage cluster sub-areas;

[0006] Based on the distribution position information of all garbage cluster sub-areas in the target area and the distribution position information of the robot group around the target area, determine the matching relationship of all robots under the robot group to implement sorting and grasping of all garbage cluster sub-areas; based on the matching relationship and the grasping parameters, determine the grasping action path of each robot in its matching garbage cluster sub-area;

[0007] Visual recognition is performed on the real-scene space corresponding to the grasping action path of the manipulator to obtain spatial morphological feature information of the domestic waste that the manipulator is about to grasp; based on the spatial morphological feature information, the grasping action state of the manipulator on the domestic waste is adjusted, thereby controlling the manipulator to grasp and transfer the domestic waste to the corresponding sorting and recycling area.

[0008] Preferably, the target area is visually identified to obtain the domestic waste distribution characteristic information of the target area; based on the domestic waste distribution characteristic information, the target area is divided into a plurality of garbage community sub-areas; based on the relative position information of the domestic waste in the garbage community sub-areas, the sorting and grabbing parameters of the garbage community sub-areas are determined, specifically:

[0009] Perform visible light photography and infrared light photography on the target area to obtain visible light images and infrared light images of the global scope of the target area; identify the visible light image to obtain the domestic waste distribution location information of the target area; perform reflectance spectrum recognition on the infrared light image to obtain the material type information of all domestic waste in the target area; divide the target area into a number of garbage community sub-areas based on the domestic waste distribution location information and the material type information; wherein all domestic waste in each garbage community sub-area have the same or similar material type attributes, and the actual distance between any two domestic wastes in each garbage community sub-area does not exceed a preset distance threshold;

[0010] Based on the boundary information of the garbage community sub-area, a three-dimensional image of the garbage community sub-area is acquired and converted from the visible light image; object contours are recognized in the three-dimensional image to obtain contour feature information of domestic garbage in the garbage community sub-area; based on the domestic garbage contour features, the mutual stacking state of all domestic garbage in the garbage community sub-area is determined, so as to determine the grabbing order corresponding to grabbing all domestic garbage in the garbage community sub-area to the same sorting and recycling area, and use this as the grabbing parameter of the garbage community sub-area.

[0011] Preferably, based on the distribution position information of all garbage cluster sub-areas in the target area and the distribution position information of the robot group around the target area, the matching relationship of all robots under the robot group to sort and grasp all garbage cluster sub-areas is determined; based on the matching relationship and the grasping parameters, the grasping action path of each robot in its matching garbage cluster sub-area is determined, specifically:

[0012] Based on the boundary distribution position information of all garbage cluster sub-areas in the target area, determine the sorting and grasping space ranges corresponding to all garbage cluster sub-areas; based on the distribution position coordinates of all manipulators under the manipulator group around the target area and the activity space range of each manipulator, determine the grasping action coverage range of each manipulator in the target area;

[0013] Based on the sorting and grasping space range of all garbage cluster sub-areas and the grasping action coverage range of all manipulators, determine the initial mapping relationship of all manipulators implementing sorting and grasping on all garbage cluster sub-areas; based on the initial mapping relationship, predict collision events for all manipulators implementing sorting and grasping operations synchronously, determine the location information of collision events for the manipulators implementing sorting and grasping operations synchronously, and change and adjust the initial mapping relationship to obtain the matching relationship of all manipulators implementing sorting and grasping on all garbage cluster sub-areas;

[0014] Based on the matching relationship and the grabbing order included in the grabbing parameters, the grabbing action path corresponding to the manipulator grabbing all domestic garbage to the same sorting and recycling area in its matching garbage cluster sub-area is determined.

[0015] Preferably, the real scene space corresponding to the grasping action path of the manipulator is visually identified to obtain spatial morphological feature information of the domestic waste that the manipulator is about to grasp; based on the spatial morphological feature information, the grasping action state of the manipulator on the domestic waste is adjusted, so as to control the manipulator to grasp and transfer the domestic waste to the corresponding sorting and recycling area, specifically:

[0016] The real scene space corresponding to the grasping action path of the manipulator is photographed in three dimensions to obtain surface contour posture information and surface morphological feature information of the domestic garbage to be grasped by the manipulator; based on the surface contour posture information and the surface morphological feature information, surface sub-region distribution information of the domestic garbage to be grasped suitable for grasping by the manipulator is determined, and this is used as the spatial morphological feature information of the domestic garbage to be grasped;

[0017] Based on the surface sub-area distribution information, all suitable grasping contact position points of the manipulator on the surface of the domestic waste are determined; based on the allowable variation range of the posture angle of the grasping contact component of the manipulator, at least a part of the suitable grasping contact position points are selected as the position points where the manipulator applies the grasping action on the domestic waste; and based on the at least a part of the suitable grasping contact position points, the grasping action posture of the grasping contact component of the manipulator is adjusted, so as to control the manipulator to grasp and transfer the domestic waste to the corresponding sorting and recycling area.

[0018] In another aspect, the present invention provides a device for sorting and processing domestic waste, the device comprising:

[0019] computer readable storage medium;

[0020] processor;

[0021] Several visual recognition cameras;

[0022] A robot cluster, including several robots that can work independently;

[0023] The computer readable storage medium stores a computer program, and when the computer program is executed by the processor, the computer program controls the operation of the plurality of visual recognition cameras and the manipulator cluster; wherein,

[0024] The processor controls one of the visual recognition cameras to perform visual recognition on the target area to obtain the domestic waste distribution characteristic information of the target area; based on the domestic waste distribution characteristic information, the target area is divided into a plurality of garbage cluster sub-areas;

[0025] The processor determines the grabbing parameters of the garbage cluster sub-area based on the relative position information of the domestic garbage in the garbage cluster sub-area;

[0026] The processor determines the matching relationship of all manipulators under the manipulator group for sorting and grasping all garbage cluster sub-regions based on the distribution position information of all garbage cluster sub-regions in the target region and the distribution position information of the manipulator group around the target region; determines the grasping action path of each manipulator in its matching garbage cluster sub-region based on the matching relationship and the grasping parameters;

[0027] The processor controls another visual recognition camera to perform visual recognition on the real scene space corresponding to the grasping action path of the manipulator to obtain spatial morphological feature information of the domestic waste that the manipulator is about to grasp;

[0028] The processor adjusts the grabbing action state of the manipulator on the domestic waste based on the spatial morphological feature information, thereby controlling the manipulator to grab and transfer the domestic waste to a corresponding sorting and recycling area.

[0029] In another aspect, the present invention provides a system for sorting and processing domestic waste, the system comprising the following modules:

[0030] A first visual recognition module is used to perform visual recognition on a target area to obtain domestic waste distribution characteristic information of the target area; based on the domestic waste distribution characteristic information, the target area is divided into a plurality of waste cluster sub-areas;

[0031] A grabbing parameter determination module, used for determining the grabbing parameters of the garbage cluster sub-area based on the relative position information of the domestic garbage in the garbage cluster sub-area;

[0032] A sorting and grasping matching relationship determination module is used to determine the matching relationship of all manipulators under the manipulator group to implement sorting and grasping on all garbage cluster sub-areas based on the distribution position information of all garbage cluster sub-areas in the target area and the distribution position information of the manipulator group around the target area;

[0033] A grasping action path module, used for determining the grasping action path of each manipulator in its matched garbage cluster sub-area based on the matching relationship and the grasping parameters;

[0034] The second visual recognition module is used to visually recognize the real scene space corresponding to the grasping action path of the manipulator to obtain the spatial morphological feature information of the domestic waste that the manipulator is about to grasp;

[0035] The grabbing action adjustment module is used to adjust the grabbing action state of the manipulator on the domestic waste based on the spatial morphological feature information, thereby controlling the manipulator to grab and transfer the domestic waste to a corresponding sorting and recycling area.

[0036] Preferably, the first visual recognition module visually recognizes the target area to obtain the domestic waste distribution characteristic information of the target area; based on the domestic waste distribution characteristic information, the target area is divided into a number of waste community sub-areas, specifically:

[0037] Perform visible light photography and infrared light photography on the target area to obtain visible light images and infrared light images of the global scope of the target area; identify the visible light image to obtain the domestic waste distribution location information of the target area; perform reflectance spectrum recognition on the infrared light image to obtain the material type information of all domestic waste in the target area; divide the target area into a number of garbage community sub-areas based on the domestic waste distribution location information and the material type information; wherein all domestic waste in each garbage community sub-area have the same or similar material type attributes, and the actual distance between any two domestic wastes in each garbage community sub-area does not exceed a preset distance threshold;

[0038] The grabbing parameter determination module determines the grabbing parameters of the garbage cluster sub-area based on the relative position information of the domestic garbage in the garbage cluster sub-area, specifically:

[0039] Based on the boundary information of the garbage community sub-area, a three-dimensional image of the garbage community sub-area is acquired and converted from the visible light image; object contours are recognized in the three-dimensional image to obtain contour feature information of domestic garbage in the garbage community sub-area; based on the domestic garbage contour features, the mutual stacking state of all domestic garbage in the garbage community sub-area is determined, so as to determine the grabbing order corresponding to grabbing all domestic garbage in the garbage community sub-area to the same sorting and recycling area, and use this as the grabbing parameter of the garbage community sub-area.

[0040] Preferably, the sorting and grasping matching relationship determination module determines the matching relationship of all manipulators under the manipulator group to implement sorting and grasping on all garbage cluster sub-areas based on the distribution position information of all garbage cluster sub-areas in the target area and the distribution position information of the manipulator group around the target area, specifically:

[0041] Based on the boundary distribution position information of all garbage cluster sub-areas in the target area, determine the sorting and grasping space ranges corresponding to all garbage cluster sub-areas; based on the distribution position coordinates of all manipulators under the manipulator group around the target area and the activity space range of each manipulator, determine the grasping action coverage range of each manipulator in the target area;

[0042] Based on the sorting and grasping space range of all garbage community sub-areas and the grasping action coverage range of all manipulators, the initial mapping relationship of all manipulators implementing sorting and grasping on all garbage community sub-areas is determined; based on the initial mapping relationship, collision events are predicted for the sorting and grasping operations implemented synchronously by all manipulators, and the position information of the collision events when the manipulators implement the sorting and grasping operations synchronously is determined, so as to change and adjust the initial mapping relationship and obtain the matching relationship of all manipulators implementing sorting and grasping on all garbage community sub-areas.

[0043] Preferably, the grasping action path module determines the grasping action path of each manipulator in its matching garbage cluster sub-area based on the matching relationship and the grasping parameters, specifically:

[0044] Based on the matching relationship and the grabbing order included in the grabbing parameters, the grabbing action path corresponding to the manipulator grabbing all domestic garbage to the same sorting and recycling area in its matching garbage cluster sub-area is determined.

[0045] Preferably, the second visual recognition module performs visual recognition on the real scene space corresponding to the grasping action path of the manipulator to obtain the spatial morphological feature information of the domestic waste that the manipulator is about to grasp, specifically:

[0046] The real scene space corresponding to the grasping action path of the manipulator is photographed in three dimensions to obtain surface contour posture information and surface morphological feature information of the domestic garbage to be grasped by the manipulator; based on the surface contour posture information and the surface morphological feature information, surface sub-region distribution information of the domestic garbage to be grasped suitable for grasping by the manipulator is determined, and this is used as the spatial morphological feature information of the domestic garbage to be grasped;

[0047] The grabbing action adjustment module adjusts the grabbing action state of the manipulator on the domestic waste based on the spatial morphological feature information, thereby controlling the manipulator to grab and transfer the domestic waste to the corresponding sorting and recycling area, specifically:

[0048] Based on the surface sub-area distribution information, all suitable grasping contact position points of the manipulator on the surface of the domestic waste are determined; based on the allowable variation range of the posture angle of the grasping contact component of the manipulator, at least a part of the suitable grasping contact position points are selected as the position points where the manipulator applies the grasping action on the domestic waste; and based on the at least a part of the suitable grasping contact position points, the grasping action posture of the grasping contact component of the manipulator is adjusted, so as to control the manipulator to grasp and transfer the domestic waste to the corresponding sorting and recycling area.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] Visual recognition is performed on the target area to obtain the distribution characteristic information of domestic waste in the target area; based on the distribution characteristic information of domestic waste, the target area is divided into several garbage cluster sub-areas; based on the relative position information of domestic waste in the garbage cluster sub-areas, the grabbing parameters of the garbage cluster sub-areas are determined. The target area may be, but is not limited to, a local surface area of ​​the conveyor belt, wherein marking lines may be provided around the local surface area to facilitate accurate definition of the spatial scope of the target area during the visual recognition process. By setting multiple visual recognition cameras above the target area, images of different properties of the global scope of the target area are captured, and the distribution status of all domestic waste in the target area is characterized from the two aspects of spatial position and material type to obtain the distribution characteristic information of domestic waste. Based on the distribution characteristic information of domestic waste, a part of domestic waste with relatively concentrated distribution positions in the target area and the same or similar material types is divided into the same waste community, and the local range occupied by the waste community in the target area is determined as a waste community sub-area. In this way, the target area is divided into several waste community sub-areas from the spatial position level, which is convenient for the subsequent control of a manipulator to grab and sort all domestic waste in a waste community sub-area, and realize the classification and selection of domestic waste with the same or similar materials. In addition, there are multiple domestic wastes distributed in the same waste community sub-area and these domestic wastes may be stacked on each other. If the manipulator grabs the domestic waste in the same waste community sub-area at will, it may cause the stacked domestic waste to flip over, change the original spatial distribution state of the domestic waste in the waste community sub-area, and cannot guarantee that the manipulator can accurately grasp each domestic waste according to the originally determined distribution of the domestic waste position. For this reason, based on the relative position information of the domestic waste in the waste community sub-area, the grabbing parameters such as the grabbing order of the waste community sub-area are determined to provide reliable guidance for the manipulator to perform garbage grabbing actions in the waste community sub-area.

[0051] Based on the distribution location information of all garbage cluster sub-areas in the target area and the distribution location information of the robot group around the target area, the matching relationship of all robots under the robot group for sorting and grasping all garbage cluster sub-areas is determined; based on the matching relationship and grasping parameters, the grasping action path of each robot in its matching garbage cluster sub-area is determined. In order to achieve independent and efficient grasping of domestic garbage in all garbage cluster sub-areas, a robot cluster is set up around the target area, where the robot cluster contains several robots, and each robot is controlled to grasp garbage specifically in only one garbage cluster sub-area, so that all robots in the robot group can synchronously and independently grasp garbage in different garbage cluster sub-areas, and achieve synchronous and efficient sorting of all garbage cluster sub-areas in the target area. All the robots under the robot cluster are distributed in different positions around the target area, so the relative position relationship between each robot and each garbage community sub-area in the target area is not the same, and the grabbing action coverage range of each robot is limited. For garbage community sub-areas that are far away from the robot, the robot will not be able to completely cover and grab all domestic garbage in the corresponding garbage community sub-areas. In order to ensure that each garbage community sub-area can be assigned a suitable manipulator to grab and sort all the domestic garbage within it, based on the distribution position information of all garbage community sub-areas in the target area and the distribution position information of the manipulator group around the target area, the actual coverage range of the grabbing action of each manipulator under the manipulator group on each garbage community sub-area and the potential collision between all manipulators when they simultaneously grab and sort garbage, thereby determining the matching relationship of all manipulators under the manipulator group to implement sorting and grabbing on all garbage community sub-areas, where the matching relationship may be but is not limited to the corresponding relationship of each manipulator under the manipulator group to implement sorting and grabbing on one of the garbage community sub-areas. Subsequently, each manipulator will implement sorting and grabbing on a corresponding garbage community sub-area according to the matching relationship, but will not implement sorting and grabbing on other garbage community sub-areas. At the same time, it can also avoid collisions between different manipulators when they are simultaneously implementing sorting and grabbing on the corresponding garbage community sub-areas, thereby reducing work interference between different manipulators. In addition, the robot needs to grab all the domestic garbage in the matching garbage community sub-area one by one and transfer them to the same sorting and recycling area. In order to enable the robot to quickly transfer the grabbed domestic garbage to the sorting and recycling area every time it grabs domestic garbage from the garbage community sub-area, the grabbing action path of each robot in its matching garbage community sub-area is determined based on the matching relationship and grabbing parameters, thereby determining the shortest grabbing action path to grab and transfer all domestic garbage to the sorting and recycling area, thereby improving the speed of garbage sorting and transfer.

[0052] Visual recognition is performed on the real-life space corresponding to the grasping action path of the manipulator to obtain the spatial morphological feature information of the domestic garbage that the manipulator is about to grasp; based on the spatial morphological feature information, the grasping action state of the manipulator on the domestic garbage is adjusted, so as to control the manipulator to grasp and transfer the domestic garbage to the corresponding sorting and recycling area. Different domestic garbage in the garbage community sub-area has different spatial morphological features such as shape and size, placement posture orientation and surface roughness distribution. If the manipulator always grasps all domestic garbage with the same gesture, it may not be able to grasp all domestic garbage accurately and firmly, and it is easy for domestic garbage to slip from the manipulator, increasing the grasping time cost of the manipulator. To this end, the real-life space corresponding to the grasping action path of the manipulator is visually recognized to obtain the spatial morphological feature information of the domestic garbage that the manipulator is about to grasp, so as to adjust the grasping action state of the manipulator on the domestic garbage, and ensure that the manipulator can accurately adjust the grasping gesture according to the appearance of different domestic garbage, so as to stably grasp and transfer the domestic garbage to the corresponding sorting and recycling area. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0054] Figure 1 It is a flow chart of the method for classifying and treating domestic waste provided by the present invention.

[0055] Figure 2 It is a structural diagram of the domestic waste classification and processing device provided by the present invention.

[0056] Figure 3 It is a schematic diagram of the grabbing action of the mechanical arm on domestic waste in the domestic waste classification and processing device provided by the present invention.

[0057] Figure 4 It is a structural diagram of the domestic waste classification and processing system provided by the present invention.

[0058] Description of the accompanying drawings: 1. Target area; 2. Garbage community sub-area; 3. Processor; 4. Visual recognition camera; 5. Robot arm; 6. Domestic garbage; 7. Suitable grasping contact position point. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only the parts related to the present invention rather than all structures are shown in the accompanying drawings. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0060] The terms "including" and "having" and any variations thereof in the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.

[0061] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0062] See also Figure 1 As shown, Figure 2 Referring to the above, the present invention provides a method for sorting and treating domestic waste, the method comprising the following steps:

[0063] S100, visually identify the target area 1 to obtain the domestic waste distribution characteristic information of the target area 1; based on the domestic waste distribution characteristic information, divide the target area 1 into a number of garbage community sub-areas 2; based on the relative position information of the domestic waste 6 in the garbage community sub-areas 2, determine the grabbing parameters of the garbage community sub-areas 2.

[0064] Furthermore, based on the distribution characteristic information of domestic waste, the target area 1 is divided into a number of garbage cluster sub-areas 2; based on the relative position information of domestic waste 6 in the garbage cluster sub-areas 2, the sorting and grabbing parameters of the garbage cluster sub-areas 2 are determined, specifically:

[0065] The target area 1 is photographed with visible light and infrared light to obtain a visible light image and an infrared light image of the global scope of the target area 1; the visible light image is identified to obtain the distribution position information of the domestic garbage 6 in the target area 1; the infrared light image is identified with a reflection spectrum to obtain the material type information of all the domestic garbage 6 in the target area 1; based on the distribution position information and material type information of the domestic garbage 6, the target area 1 is divided into a number of garbage community sub-areas 2; wherein all the domestic garbage 6 in each garbage community sub-area 2 have the same or similar material type attributes, and the actual distance between any two domestic garbage 6 in each garbage community sub-area 2 does not exceed a preset distance threshold;

[0066] Based on the boundary information of the garbage community sub-area 2, a three-dimensional image of the garbage community sub-area 2 is acquired and converted from the visible light image; object contours are recognized in the three-dimensional image to obtain contour feature information of domestic garbage in the garbage community sub-area 2; based on the contour features of domestic garbage, the mutual stacking state of all domestic garbage 6 in the garbage community sub-area 2 is determined, thereby determining the grabbing order corresponding to grabbing all domestic garbage 6 in the garbage community sub-area 2 to the same sorting and recycling area, which is used as the grabbing parameter of the garbage community sub-area 2.

[0067] Domestic waste classification and treatment can be implemented on the conveyor belt. In order to carry out targeted classification and treatment of domestic waste transported on the conveyor belt, the surface of the conveyor belt can be divided into several target areas with equal areas on the conveyor belt surface. Each target area corresponds to a local surface area of ​​the conveyor belt. In order to improve the visual recognition resolution of the target area, a marking line of a specific color can be set around the local surface area to facilitate accurate positioning of the spatial range of the target area during the visual recognition process.

[0068] A plurality of visual recognition cameras (i.e. Figure 24 in the figure), these visual recognition cameras may include but are not limited to visible light cameras and infrared light cameras. The above recognition cameras are used to shoot visible light and infrared light of the target area to obtain visible light images and infrared light images of the global scope of the target area, shoot different nature images of the global scope of the target area, characterize the distribution state of all domestic waste in the target area from the two aspects of spatial position and material type, and obtain domestic waste distribution characteristic information. Specifically, the visible light image reflects the distribution position of all domestic waste in the target area. The visible light image is identified to obtain the domestic waste distribution position information of the target area, so as to accurately locate all domestic waste in the target area. Domestic waste may include glass bottles, plastic bottles, waste paper and other items with different material types, and the absorption rate of different materials to infrared light is not the same, so the reflectivity of different materials to infrared light is also different accordingly. By performing reflection spectrum recognition on the infrared light image, the material type information of all domestic waste in the target area is obtained. By shooting the target area with visible light and infrared light, the spatial positioning and material type identification of all domestic waste in the target area can be achieved, and the location of each domestic waste in the target area and its material type are matched one by one, so as to achieve the distribution status characterization of all domestic waste in the target area. Based on the domestic waste distribution location information and material type information, the target area is divided into several garbage community sub-areas, so that a part of domestic waste with relatively concentrated distribution locations in the target area and the same or similar material types are divided into the same garbage community, and the local area occupied by the garbage community in the target area is determined as a garbage community sub-area. In this way, the target area is divided into several garbage community sub-areas from the spatial location level, which is convenient for the subsequent control of a manipulator to grab and sort all domestic waste in a garbage community sub-area, and realize the classification and selection of domestic waste with the same or similar materials.

[0069] In addition, there are multiple domestic garbage distributed in the same garbage community sub-area and these domestic garbage may be stacked on each other. If the manipulator randomly grabs the domestic garbage in the same garbage community sub-area, it may cause the stacked domestic garbage to flip over, change the original spatial distribution state of the domestic garbage in the garbage community sub-area, and cannot guarantee that the manipulator can accurately grasp each domestic garbage according to the originally determined domestic garbage position distribution. To this end, based on the boundary information of the garbage community sub-area, a three-dimensional image of the garbage community sub-area is obtained and converted from the visible light image, and the contour feature information of the domestic garbage in the garbage community sub-area is identified from the three-dimensional image, so as to obtain the mutual stacking state of all domestic garbage in the garbage community sub-area (such as the location of two domestic garbage in a stacking relationship, and / or the size of the stacking area of ​​two domestic garbage, etc.), so as to determine the grabbing order corresponding to grabbing all domestic garbage in the garbage community sub-area to the same sorting and recycling area, which is used as the subsequent grabbing parameter of the manipulator for all domestic garbage in the garbage community sub-area, and provides reliable guidance for the manipulator to perform garbage grabbing actions in the garbage community sub-area, and minimizes the interference of the manipulator to other domestic garbage during the process of grabbing domestic garbage.

[0070] S200, based on the distribution position information of all garbage community sub-areas 2 in the target area 1 and the distribution position information of the robot group around the target area 1, determine the matching relationship of all robots 5 under the robot group for sorting and grasping all garbage community sub-areas 2; based on the matching relationship and grasping parameters, determine the grasping action path of each robot 5 in its matching garbage community sub-area 2.

[0071] Furthermore, based on the distribution position information of all garbage cluster sub-areas 2 in the target area 1 and the distribution position information of the robot group around the target area 1, the matching relationship of all the robots 5 under the robot group to sort and grasp all the garbage cluster sub-areas 2 is determined; based on the matching relationship and grasping parameters, the grasping action path of each robot 5 in its matching garbage cluster sub-area 2 is determined, specifically:

[0072] Based on the boundary distribution position information of all garbage cluster sub-areas 2 in the target area 1, the sorting and grabbing space ranges corresponding to all garbage cluster sub-areas 2 are determined; based on the distribution position coordinates of all manipulators 5 under the manipulator group around the target area 1 and the activity space range of each manipulator 5, the grabbing action coverage range of each manipulator 5 in the target area 1 is determined;

[0073] Based on the sorting and grasping spatial range of all garbage cluster sub-areas 2 and the grasping action coverage range of all manipulators 5, determine the initial mapping relationship of all manipulators 5 implementing sorting and grasping on all garbage cluster sub-areas 2; based on the initial mapping relationship, predict collision events for all manipulators 5 implementing sorting and grasping operations simultaneously, determine the location information of collision events for the manipulators 5 implementing sorting and grasping operations simultaneously, and change and adjust the initial mapping relationship to obtain the matching relationship of all manipulators 5 implementing sorting and grasping on all garbage cluster sub-areas 2;

[0074] Based on the matching relationship and the grabbing sequence included in the grabbing parameters, the robot 5 is determined to grab all the domestic garbage 6 in its matching garbage cluster sub-area 2 to the corresponding grabbing action path of the same sorting and recycling area.

[0075] In order to achieve independent and efficient domestic waste grabbing in all garbage community sub-areas, a robot cluster is set up around the target area, where the robot cluster includes several robots, and each robot is controlled to only grab garbage from one garbage community sub-area. In this way, all robots in the robot group can grab garbage from different garbage community sub-areas at the same time, thereby achieving synchronous and efficient sorting of all garbage community sub-areas in the target area.

[0076] All the manipulators under the manipulator cluster are distributed in different positions around the target area, so that the relative position relationship between each manipulator and each garbage community sub-area in the target area is not the same, and the grabbing action coverage of each manipulator is limited. For garbage community sub-areas that are far away from the manipulator, the manipulator will not be able to completely cover and grab all the domestic garbage in the corresponding garbage community sub-area. In order to determine which garbage community sub-area the manipulator matches, it is necessary to determine the grabbing action coverage of the manipulator in the target area. Specifically, based on the boundary distribution position information of all garbage community sub-areas in the target area, the sorting and grabbing space range corresponding to each garbage community sub-area is determined; based on the distribution position coordinates of all the manipulators under the manipulator group around the target area and the activity space range of each manipulator, the grabbing action coverage of each manipulator in the target area is determined.

[0077] In order to ensure that each garbage community sub-area can be assigned a suitable robot to grab and sort all the domestic waste within it, based on the distribution position information of all garbage community sub-areas in the target area and the distribution position information of the robot group around the target area, the actual coverage range of the grabbing action of each robot under the robot group for each garbage community sub-area and the potential collision between all robots when they grab and sort garbage simultaneously are determined, thereby determining the matching relationship of all robots under the robot group for sorting and grabbing in all garbage community sub-areas. Specifically, based on the sorting and grasping space range of all garbage community sub-areas and the grasping action coverage range of all manipulators, the initial mapping relationship of all manipulators implementing sorting and grasping on all garbage community sub-areas is determined; based on the initial mapping relationship, collision event prediction is performed on the sorting and grasping operations implemented synchronously by all manipulators, and the location information of the collision events of the sorting and grasping operations implemented synchronously by the manipulators is determined, so as to change and adjust the initial mapping relationship, and obtain the matching relationship of all manipulators implementing sorting and grasping on all garbage community sub-areas; wherein the matching relationship may be but is not limited to the corresponding relationship of each manipulator under the manipulator group implementing sorting and grasping on one of the garbage community sub-areas, and subsequently each manipulator will implement sorting and grasping on the corresponding garbage community sub-area according to the matching relationship, but will not implement sorting and grasping on other garbage community sub-areas, and at the same time, it can also avoid collisions during the sorting and grasping process of different manipulators on the corresponding garbage community sub-areas at the same time, thereby reducing the work interference between different manipulators.

[0078] In addition, the robot needs to grab all the domestic waste in the matching garbage community sub-area one by one and transfer them to the same sorting and recycling area (such as a recycling box for collecting and storing the same type of domestic waste). In order to enable the robot to quickly transfer the grabbed domestic waste to the sorting and recycling area every time it grabs domestic waste from the garbage community sub-area, the grabbing action path of each robot in its matching garbage community sub-area is determined based on the matching relationship and grabbing order, thereby determining the shortest grabbing action path to grab and transfer all domestic waste to the sorting and recycling area, thereby improving the speed of garbage sorting and transfer.

[0079] S300, visually identify the real-life space corresponding to the grasping action path of the manipulator 5 to obtain the spatial morphological feature information of the domestic waste 6 that the manipulator 5 is about to grasp; based on the spatial morphological feature information, adjust the grasping action state of the manipulator 5 on the domestic waste 6, thereby controlling the manipulator 5 to grasp and transfer the domestic waste 6 to the corresponding sorting and recycling area.

[0080] Furthermore, the real scene space corresponding to the grasping action path of the manipulator 5 is visually recognized to obtain the spatial morphological feature information of the domestic garbage 6 that the manipulator 5 is about to grasp; based on the spatial morphological feature information, the grasping action state of the manipulator 5 on the domestic garbage 6 is adjusted, so as to control the manipulator 5 to grasp and transfer the domestic garbage 6 to the corresponding sorting and recycling area, specifically:

[0081] The real scene space corresponding to the grasping action path of the manipulator 5 is photographed in three dimensions to obtain the surface contour posture information and surface morphological feature information of the domestic garbage 6 that the manipulator 5 is about to grasp; based on the surface contour posture information and the surface morphological feature information, the surface sub-region distribution information suitable for grasping by the manipulator 5 is determined, and this is used as the spatial morphological feature information of the domestic garbage 6 that is about to be grasped;

[0082] Based on the surface sub-area distribution information, all suitable grasping contact position points 7 of the manipulator 5 on the surface of the domestic waste 6 are determined; based on the allowable range of variation of the posture angle of the grasping contact component of the manipulator 5, at least a part of the suitable grasping contact position points 7 are selected as the position points where the manipulator 5 applies the grasping action of the domestic waste 6; and based on at least a part of the suitable grasping contact position points 7, the grasping action posture of the grasping contact component of the manipulator 5 is adjusted, so as to control the manipulator 5 to grasp and transfer the domestic waste 6 to the corresponding sorting and recycling area.

[0083] Different domestic garbage in the garbage community sub-area has different spatial morphological characteristics such as shape, size, placement posture orientation and surface roughness distribution. If the manipulator always grabs all domestic garbage with the same gesture, it may not be able to accurately and firmly grab all domestic garbage, and it is easy for domestic garbage to slip from the manipulator, increasing the time cost of the manipulator's grabbing. For this reason, the camera built into the manipulator is used to take three-dimensional photos of the real-life space corresponding to the manipulator's grabbing action path to obtain the surface contour posture information and surface morphological feature information of the domestic garbage that the manipulator is about to grab; based on the surface contour posture information and surface morphological feature information, the surface sub-area distribution information suitable for the manipulator to grab the domestic garbage to be grabbed is determined, and the shape, size and roughness distribution of the global external surface of the domestic garbage are fully characterized through the surface sub-area distribution information.

[0084] Based on the surface sub-area distribution information, all suitable gripping contact position points of the manipulator on the surface of the domestic waste are determined; wherein the suitable gripping contact position points may be, but are not limited to, position points on the surface of the domestic waste where the actual curvature is less than or equal to a preset curvature threshold and the roughness is greater than or equal to a preset roughness threshold. Also based on the allowable range of variation of the attitude angle of the gripping contact component of the manipulator (such as the allowable range of variation of the gripping clamping arm of the manipulator in the pitch direction, yaw direction and roll direction), at least a portion of suitable gripping contact position points are selected as position points where the manipulator applies gripping action to the domestic waste (such as Figure 3 As shown in the figure, the gripping contact parts of the manipulator can directly and firmly grip the domestic waste at the selected gripping action application point by changing their own posture angle, effectively preventing the manipulator from slipping during the gripping of domestic waste, and ensuring that the manipulator can accurately adjust the gripping gesture according to the appearance of different domestic waste, thereby stably gripping and transferring the domestic waste to the corresponding sorting and recycling area.

[0085] See also Figure 2 As shown, the present invention provides a household waste classification and processing device, the device comprising:

[0086] computer readable storage medium;

[0087] Processor 3;

[0088] A number of visual recognition cameras 4;

[0089] A robot 5 cluster, including a number of robots 5 capable of working independently;

[0090] A computer program is stored on the computer-readable storage medium, and when the computer program is executed by the processor 3, the computer program controls the operation of a cluster of several visual recognition cameras 4 and manipulators 5; wherein,

[0091] The processor 3 controls one of the visual recognition cameras 4 to perform visual recognition on the target area 1 to obtain the domestic waste distribution characteristic information of the target area 1; based on the domestic waste distribution characteristic information, the target area 1 is divided into a plurality of waste community sub-areas 2;

[0092] The processor 3 determines the grabbing parameters of the garbage cluster sub-area 2 based on the relative position information of the domestic garbage 6 in the garbage cluster sub-area 2;

[0093] The processor 3 determines the matching relationship of all the manipulators 5 under the manipulator group to sort and grasp all the garbage cluster sub-areas 2 based on the distribution position information of all the garbage cluster sub-areas 2 in the target area 1 and the distribution position information of the manipulator group around the target area 1; based on the matching relationship and grasping parameters, determines the grasping action path of each manipulator 5 in its matching garbage cluster sub-area 2;

[0094] The processor 3 controls another visual recognition camera 4 to perform visual recognition on the real scene space corresponding to the grasping action path of the manipulator 5, and obtains the spatial morphological feature information of the domestic garbage 6 that the manipulator 5 is about to grasp;

[0095] The processor 3 adjusts the grabbing action state of the robot 5 on the domestic waste 6 based on the spatial morphological feature information, thereby controlling the robot 5 to grab and transfer the domestic waste 6 to the corresponding sorting and recycling area.

[0096] See also Figure 4 As shown, the present invention provides a system for sorting and processing domestic waste, which includes the following modules:

[0097] The first visual recognition module is used to visually recognize the target area 1 to obtain the domestic waste distribution characteristic information of the target area 1; based on the domestic waste distribution characteristic information, the target area 1 is divided into a plurality of waste community sub-areas 2;

[0098] A grabbing parameter determination module, used to determine the grabbing parameters of the garbage cluster sub-area 2 based on the relative position information of the domestic garbage 6 in the garbage cluster sub-area 2;

[0099] A sorting and grasping matching relationship determination module is used to determine the matching relationship of all manipulators 5 under the manipulator group to implement sorting and grasping on all garbage cluster sub-areas 2 based on the distribution position information of all garbage cluster sub-areas 2 in the target area 1 and the distribution position information of the manipulator group around the target area 1;

[0100] A grasping action path module, used to determine the grasping action path of each manipulator 5 in its matching garbage cluster sub-area 2 based on the matching relationship and grasping parameters;

[0101] The second visual recognition module is used to visually recognize the real scene space corresponding to the grasping action path of the manipulator 5 to obtain the spatial morphological feature information of the domestic garbage 6 that the manipulator 5 is about to grasp;

[0102] The grabbing action adjustment module is used to adjust the grabbing action state of the manipulator 5 on the domestic waste 6 based on the spatial morphological feature information, thereby controlling the manipulator 5 to grab and transfer the domestic waste 6 to the corresponding sorting and recycling area.

[0103] Furthermore, the first visual recognition module visually recognizes the target area 1 to obtain the domestic waste distribution characteristic information of the target area 1; based on the domestic waste distribution characteristic information, the target area 1 is divided into a number of waste community sub-areas 2, specifically:

[0104] The target area 1 is photographed with visible light and infrared light to obtain a visible light image and an infrared light image of the global scope of the target area 1; the visible light image is identified to obtain the distribution position information of the domestic garbage 6 in the target area 1; the infrared light image is identified with a reflection spectrum to obtain the material type information of all the domestic garbage 6 in the target area 1; based on the distribution position information and material type information of the domestic garbage 6, the target area 1 is divided into a number of garbage community sub-areas 2; wherein all the domestic garbage 6 in each garbage community sub-area 2 have the same or similar material type attributes, and the actual distance between any two domestic garbage 6 in each garbage community sub-area 2 does not exceed a preset distance threshold;

[0105] The grabbing parameter determination module determines the grabbing parameters of the garbage cluster sub-area 2 based on the relative position information of the domestic garbage 6 in the garbage cluster sub-area 2, specifically:

[0106] Based on the boundary information of the garbage community sub-area 2, a three-dimensional image of the garbage community sub-area 2 is acquired and converted from the visible light image; object contours are recognized in the three-dimensional image to obtain contour feature information of domestic garbage in the garbage community sub-area 2; based on the contour features of domestic garbage, the mutual stacking state of all domestic garbage 6 in the garbage community sub-area 2 is determined, thereby determining the grabbing order corresponding to grabbing all domestic garbage 6 in the garbage community sub-area 2 to the same sorting and recycling area, which is used as the grabbing parameter of the garbage community sub-area 2.

[0107] Furthermore, the sorting and grasping matching relationship determination module determines the matching relationship of all the manipulators 5 under the manipulator group to implement sorting and grasping on all the garbage cluster sub-areas 2 based on the distribution position information of all the garbage cluster sub-areas 2 in the target area 1 and the distribution position information of the manipulator group around the target area 1, specifically:

[0108] Based on the boundary distribution position information of all garbage cluster sub-areas 2 in the target area 1, the sorting and grabbing space ranges corresponding to all garbage cluster sub-areas 2 are determined; based on the distribution position coordinates of all manipulators 5 under the manipulator group around the target area 1 and the activity space range of each manipulator 5, the grabbing action coverage range of each manipulator 5 in the target area 1 is determined;

[0109] Based on the sorting and grasping space range of all garbage community sub-areas 2 and the grasping action coverage range of all manipulators 5, the initial mapping relationship of all manipulators 5 implementing sorting and grasping in all garbage community sub-areas 2 is determined; based on the initial mapping relationship, collision events are predicted for all manipulators 5 implementing sorting and grasping operations simultaneously, and the location information of the collision events of the manipulators 5 implementing sorting and grasping operations simultaneously is determined, so as to change and adjust the initial mapping relationship and obtain the matching relationship of all manipulators 5 implementing sorting and grasping in all garbage community sub-areas 2.

[0110] Furthermore, the grabbing action path module determines the grabbing action path of each manipulator 5 in its matching garbage cluster sub-area 2 based on the matching relationship and the grabbing parameters, specifically:

[0111] Based on the matching relationship and the grabbing sequence included in the grabbing parameters, the robot 5 is determined to grab all the domestic garbage 6 in its matching garbage cluster sub-area 2 to the corresponding grabbing action path of the same sorting and recycling area.

[0112] Furthermore, the second visual recognition module visually recognizes the real scene space corresponding to the grasping action path of the manipulator to obtain the spatial morphological feature information of the domestic waste that the manipulator is about to grasp, specifically:

[0113] The second visual recognition module visually recognizes the real scene space corresponding to the grasping action path of the manipulator 5, and obtains the spatial morphological feature information of the domestic garbage 6 that the manipulator 5 is about to grasp, specifically:

[0114] The real scene space corresponding to the grasping action path of the manipulator 5 is photographed in three dimensions to obtain the surface contour posture information and surface morphological feature information of the domestic garbage 6 that the manipulator 5 is about to grasp; based on the surface contour posture information and the surface morphological feature information, the surface sub-region distribution information suitable for grasping by the manipulator 5 is determined, and this is used as the spatial morphological feature information of the domestic garbage 6 that is about to be grasped;

[0115] The grabbing action adjustment module adjusts the grabbing action state of the manipulator 5 on the domestic garbage 6 based on the spatial morphological feature information, thereby controlling the manipulator 5 to grab and transfer the domestic garbage 6 to the corresponding sorting and recycling area, specifically:

[0116] Based on the surface sub-area distribution information, all suitable grasping contact position points 7 of the manipulator 5 on the surface of the domestic waste 6 are determined; based on the allowable range of variation of the posture angle of the grasping contact component of the manipulator 5, at least a part of the suitable grasping contact position points 7 are selected as the position points where the manipulator 5 applies the grasping action of the domestic waste 6; and based on at least a part of the suitable grasping contact position points 7, the grasping action posture of the grasping contact component of the manipulator 5 is adjusted, so as to control the manipulator 5 to grasp and transfer the domestic waste 6 to the corresponding sorting and recycling area.

[0117] The operation and effects of the domestic waste classification and treatment device and the domestic waste classification and treatment system of the present invention correspond to and are consistent with the above-mentioned domestic waste classification and treatment method, and the domestic waste classification and treatment device and the domestic waste classification and treatment system will not be repeated here.

[0118] Through the description of the above implementation methods, technicians in this field can clearly understand that each implementation method can be implemented by adding a necessary general hardware platform, and of course can also be implemented by combining hardware and software. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a computer product, and the present invention can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0119] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it, and other embodiments may also be used. Although the present invention has been described in detail with reference to the aforementioned embodiments, ordinary technicians in this field should understand that they can still modify the technical solutions recorded in the aforementioned embodiments, or replace some of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for classifying and treating domestic waste, characterized in that: The method comprises the following steps: Visually identify the target area to obtain the domestic waste distribution characteristic information of the target area; divide the target area into a number of garbage cluster sub-areas based on the domestic waste distribution characteristic information; determine the grabbing parameters of the garbage cluster sub-areas based on the relative position information of the domestic waste in the garbage cluster sub-areas, which are specifically: Perform visible light photography and infrared light photography on the target area to obtain visible light images and infrared light images of the global scope of the target area; identify the visible light image to obtain the domestic waste distribution location information of the target area; perform reflectance spectrum recognition on the infrared light image to obtain the material type information of all domestic waste in the target area; divide the target area into a number of garbage community sub-areas based on the domestic waste distribution location information and the material type information; wherein all domestic waste in each garbage community sub-area have the same or similar material type attributes, and the actual distance between any two domestic wastes in each garbage community sub-area does not exceed a preset distance threshold; Based on the boundary information of the garbage community sub-area, a three-dimensional image of the garbage community sub-area is obtained and converted from the visible light image; object contour recognition is performed on the three-dimensional image to obtain contour feature information of domestic garbage in the garbage community sub-area; based on the domestic garbage contour feature, the mutual stacking state of all domestic garbage in the garbage community sub-area is determined, so as to determine the grabbing order corresponding to grabbing all domestic garbage in the garbage community sub-area to the same sorting and recycling area, and use this as the grabbing parameter of the garbage community sub-area; Based on the distribution position information of all garbage cluster sub-areas in the target area and the distribution position information of the robot group around the target area, determine the matching relationship of all robots under the robot group to implement sorting and grasping of all garbage cluster sub-areas; based on the matching relationship and the grasping parameters, determine the grasping action path of each robot in its matching garbage cluster sub-area; Visual recognition is performed on the real-scene space corresponding to the grasping action path of the manipulator to obtain spatial morphological feature information of the domestic waste that the manipulator is about to grasp; based on the spatial morphological feature information, the grasping action state of the manipulator on the domestic waste is adjusted, thereby controlling the manipulator to grasp and transfer the domestic waste to the corresponding sorting and recycling area.

2. The method according to claim 1, characterized in that Based on the distribution position information of all garbage cluster sub-areas in the target area and the distribution position information of the robot group around the target area, the matching relationship of all robots under the robot group for sorting and grasping all garbage cluster sub-areas is determined; based on the matching relationship and the grasping parameters, the grasping action path of each robot in its matching garbage cluster sub-area is determined, specifically: Based on the boundary distribution position information of all garbage cluster sub-areas in the target area, determine the sorting and grasping space ranges corresponding to all garbage cluster sub-areas; based on the distribution position coordinates of all manipulators under the manipulator group around the target area and the activity space range of each manipulator, determine the grasping action coverage range of each manipulator in the target area; Based on the sorting and grasping space range of all garbage cluster sub-areas and the grasping action coverage range of all manipulators, determine the initial mapping relationship of all manipulators implementing sorting and grasping on all garbage cluster sub-areas; based on the initial mapping relationship, predict collision events for all manipulators implementing sorting and grasping operations synchronously, determine the location information of collision events for the manipulators implementing sorting and grasping operations synchronously, and change and adjust the initial mapping relationship to obtain the matching relationship of all manipulators implementing sorting and grasping on all garbage cluster sub-areas; Based on the matching relationship and the grabbing order included in the grabbing parameters, the grabbing action path corresponding to the manipulator grabbing all domestic garbage to the same sorting and recycling area in its matching garbage cluster sub-area is determined.

3. The method according to claim 1, characterized in that: The real scene space corresponding to the grasping action path of the manipulator is visually recognized to obtain spatial morphological feature information of the domestic waste that the manipulator is about to grasp; based on the spatial morphological feature information, the grasping action state of the manipulator on the domestic waste is adjusted, thereby controlling the manipulator to grasp and transfer the domestic waste to the corresponding sorting and recycling area, specifically: Performing three-dimensional photography on the real scene space corresponding to the grasping action path of the manipulator to obtain surface contour posture information and surface morphology feature information of the domestic garbage that the manipulator is about to grasp; Based on the surface contour posture information and the surface morphology feature information, determining the surface sub-region distribution information of the domestic waste to be grasped that is suitable for the manipulator to grasp, and using this as the spatial morphology feature information of the domestic waste to be grasped; Based on the surface sub-area distribution information, determining all suitable grasping contact position points of the manipulator on the surface of the domestic waste; Based on the allowable variation range of the posture angle of the gripping contact component of the manipulator, select at least a portion of suitable gripping contact position points as the position points where the manipulator applies the gripping action on the domestic waste; then based on the at least a portion of suitable gripping contact position points, adjust the gripping action posture of the gripping contact component of the manipulator, so as to control the manipulator to grab and transfer the domestic waste to the corresponding sorting and recycling area.

4. A household waste classification and processing device, characterized in that: The device comprises: computer readable storage medium; processor; Several visual recognition cameras; A robot cluster, including several robots that can work independently; The computer readable storage medium stores a computer program, and when the computer program is executed by the processor, the computer program controls the operation of the plurality of visual recognition cameras and the manipulator cluster; wherein, The processor controls one of the visual recognition cameras to perform visual recognition on the target area to obtain the domestic waste distribution characteristic information of the target area; based on the domestic waste distribution characteristic information, the target area is divided into a plurality of garbage community sub-areas, which are specifically: Perform visible light photography and infrared light photography on the target area to obtain visible light images and infrared light images of the global scope of the target area; identify the visible light image to obtain the domestic waste distribution location information of the target area; perform reflectance spectrum recognition on the infrared light image to obtain the material type information of all domestic waste in the target area; divide the target area into a number of garbage community sub-areas based on the domestic waste distribution location information and the material type information; wherein all domestic waste in each garbage community sub-area have the same or similar material type attributes, and the actual distance between any two domestic wastes in each garbage community sub-area does not exceed a preset distance threshold; The processor determines the grabbing parameters of the garbage cluster sub-area based on the relative position information of the domestic garbage in the garbage cluster sub-area, which are specifically: Based on the boundary information of the garbage community sub-area, a three-dimensional image of the garbage community sub-area is obtained and converted from the visible light image; object contour recognition is performed on the three-dimensional image to obtain contour feature information of domestic garbage in the garbage community sub-area; based on the domestic garbage contour feature, the mutual stacking state of all domestic garbage in the garbage community sub-area is determined, so as to determine the grabbing order corresponding to grabbing all domestic garbage in the garbage community sub-area to the same sorting and recycling area, and use this as the grabbing parameter of the garbage community sub-area; The processor determines the matching relationship of all manipulators under the manipulator group for sorting and grasping all garbage cluster sub-regions based on the distribution position information of all garbage cluster sub-regions in the target region and the distribution position information of the manipulator group around the target region; determines the grasping action path of each manipulator in its matching garbage cluster sub-region based on the matching relationship and the grasping parameters; The processor controls another visual recognition camera to perform visual recognition on the real scene space corresponding to the grasping action path of the manipulator to obtain spatial morphological feature information of the domestic waste that the manipulator is about to grasp; The processor adjusts the grabbing action state of the manipulator on the domestic waste based on the spatial morphological feature information, thereby controlling the manipulator to grab and transfer the domestic waste to a corresponding sorting and recycling area.

5. A system for sorting and treating domestic waste, characterized in that: The system includes the following modules: The first visual recognition module is used to visually recognize the target area and obtain the domestic waste distribution characteristic information of the target area; based on the domestic waste distribution characteristic information, the target area is divided into a plurality of garbage community sub-areas, which are specifically: Perform visible light photography and infrared light photography on the target area to obtain visible light images and infrared light images of the global scope of the target area; identify the visible light image to obtain the domestic waste distribution location information of the target area; perform reflectance spectrum recognition on the infrared light image to obtain the material type information of all domestic waste in the target area; divide the target area into a number of garbage community sub-areas based on the domestic waste distribution location information and the material type information; wherein all domestic waste in each garbage community sub-area have the same or similar material type attributes, and the actual distance between any two domestic wastes in each garbage community sub-area does not exceed a preset distance threshold; The grabbing parameter determination module is used to determine the grabbing parameters of the garbage cluster sub-area based on the relative position information of the domestic garbage in the garbage cluster sub-area, which is specifically: Based on the boundary information of the garbage community sub-area, a three-dimensional image of the garbage community sub-area is obtained and converted from the visible light image; object contour recognition is performed on the three-dimensional image to obtain contour feature information of domestic garbage in the garbage community sub-area; based on the domestic garbage contour feature, the mutual stacking state of all domestic garbage in the garbage community sub-area is determined, so as to determine the grabbing order corresponding to grabbing all domestic garbage in the garbage community sub-area to the same sorting and recycling area, and use this as the grabbing parameter of the garbage community sub-area; A sorting and grasping matching relationship determination module is used to determine the matching relationship of all manipulators under the manipulator group to implement sorting and grasping on all garbage cluster sub-areas based on the distribution position information of all garbage cluster sub-areas in the target area and the distribution position information of the manipulator group around the target area; A grasping action path module, used for determining the grasping action path of each manipulator in its matched garbage cluster sub-area based on the matching relationship and the grasping parameters; The second visual recognition module is used to visually recognize the real scene space corresponding to the grasping action path of the manipulator to obtain the spatial morphological feature information of the domestic waste that the manipulator is about to grasp; The grabbing action adjustment module is used to adjust the grabbing action state of the manipulator on the domestic waste based on the spatial morphological feature information, thereby controlling the manipulator to grab and transfer the domestic waste to a corresponding sorting and recycling area.

6. The system according to claim 5, characterized in that The sorting and grasping matching relationship determination module determines the matching relationship of all the manipulators under the manipulator group to sort and grasp all the garbage cluster sub-areas based on the distribution position information of all the garbage cluster sub-areas in the target area and the distribution position information of the manipulator group around the target area, specifically: Based on the boundary distribution position information of all garbage cluster sub-areas in the target area, determine the sorting and grasping space ranges corresponding to all garbage cluster sub-areas; based on the distribution position coordinates of all manipulators under the manipulator group around the target area and the activity space range of each manipulator, determine the grasping action coverage range of each manipulator in the target area; Based on the sorting and grasping space range of all garbage community sub-areas and the grasping action coverage range of all manipulators, the initial mapping relationship of all manipulators implementing sorting and grasping on all garbage community sub-areas is determined; based on the initial mapping relationship, collision events are predicted for the sorting and grasping operations implemented synchronously by all manipulators, and the position information of the collision events when the manipulators implement the sorting and grasping operations synchronously is determined, so as to change and adjust the initial mapping relationship and obtain the matching relationship of all manipulators implementing sorting and grasping on all garbage community sub-areas.

7. The system according to claim 6, characterized in that The grasping action path module determines the grasping action path of each manipulator in its matching garbage cluster sub-area based on the matching relationship and the grasping parameters, specifically: Based on the matching relationship and the grabbing order included in the grabbing parameters, the grabbing action path corresponding to the manipulator grabbing all domestic garbage to the same sorting and recycling area in its matching garbage cluster sub-area is determined.

8. The system according to claim 5, characterized in that The second visual recognition module visually recognizes the real scene space corresponding to the grasping action path of the manipulator to obtain the spatial morphological feature information of the domestic waste that the manipulator is about to grasp, specifically: Performing three-dimensional photography on the real scene space corresponding to the grasping action path of the manipulator to obtain surface contour posture information and surface morphology feature information of the domestic garbage that the manipulator is about to grasp; Based on the surface contour posture information and the surface morphology feature information, determining the surface sub-region distribution information of the domestic waste to be grasped that is suitable for the manipulator to grasp, and using this as the spatial morphology feature information of the domestic waste to be grasped; The grabbing action adjustment module adjusts the grabbing action state of the manipulator on the domestic waste based on the spatial morphological feature information, thereby controlling the manipulator to grab and transfer the domestic waste to the corresponding sorting and recycling area, specifically: Based on the surface sub-area distribution information, all suitable grasping contact position points of the manipulator on the surface of the domestic waste are determined; based on the allowable variation range of the posture angle of the grasping contact component of the manipulator, at least a part of the suitable grasping contact position points are selected as the position points where the manipulator applies the grasping action on the domestic waste; and based on the at least a part of the suitable grasping contact position points, the grasping action posture of the grasping contact component of the manipulator is adjusted, so as to control the manipulator to grasp and transfer the domestic waste to the corresponding sorting and recycling area.

Images