Mechanical arm sorting device and sorting method based on visual identification

Abstract

The invention discloses a mechanical arm sorting device and method based on visual recognition, and belongs to the technical field of sorting equipment, the mechanical arm sorting device comprises a conveying belt arranged in the first direction, a recognition assembly and a carrying assembly, the recognition assembly and the carrying assembly are arranged above the conveying belt, and the recognition assembly is used for recognizing the front view posture and the placement angle of a workpiece conveyed in the first direction; the carrying assembly comprises a first mechanical arm and a second mechanical arm, the first mechanical arm is used for carrying and clamping circular structural parts, and the second mechanical arm is used for carrying and clamping non-circular structural parts; and the control assembly is used for sending the front view posture and the placement angle, obtained by the recognition assembly, of the workpiece to the carrying assembly and controlling the carrying assembly to adjust the clamping force and the clamping angle so as to grab and carry the workpiece. The corresponding mechanical arms are selected through the control assembly, the clamping distance and the clamping force of the two mechanical arms are adjusted through the obtained front view postures of the workpieces, the clamping angles of the two mechanical arms are adjusted, and accurate clamping of different workpieces is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent sorting equipment, and in particular relates to a mechanical arm sorting device and a sorting method based on visual recognition. Background Art

[0002] Intelligent manufacturing is a human-machine integrated intelligent system composed of intelligent machines and human experts. It performs intelligent activities in the manufacturing process, such as analysis, thrust, judgment, conception and decision-making, to achieve semi-automatic or even fully automated manufacturing.

[0003] In the field of manual logistics, there are many intelligent sorting robots, which recognize objects through QR codes, grab objects through robots and store them at set workstations, so as to realize the rapid transfer and storage of items. Conventional sorting robots are often used for grabbing and storing specific workpieces or products, such as square boxes of set sizes or workpieces of fixed sizes. However, in the actual processing and assembly process, there are often diversified assembly requirements, that is, the objects that need to be grabbed often have different shapes and sizes, and conventional manipulators cannot complete the grabbing of products of different shapes. Summary of the invention

[0004] In response to one or more of the above-mentioned defects or improvement needs of the prior art, the present invention provides a robotic arm sorting device based on visual recognition, so as to solve the problem that the existing sorting device cannot complete the grasping of products of different shapes.

[0005] To achieve the above object, the present invention provides a robotic arm sorting device based on visual recognition, which comprises:

[0006] a conveyor belt arranged along a first direction;

[0007] An identification component and a transport component are disposed above the conveyor belt, and the identification component and the transport component are spaced apart along a first direction;

[0008] The recognition component is used to recognize the front view posture and placement angle of the workpiece transported along the first direction;

[0009] The transport assembly comprises a first mechanical arm and a second mechanical arm, wherein the first mechanical arm is used for transporting and clamping circular structural parts, and the second mechanical arm is used for transporting and clamping non-circular structural parts;

[0010] And a control component, which is communicated with the identification component and the conveying component respectively, and is used to send the formal posture and placement angle of the workpiece obtained by the identification component to the conveying component, and control the conveying component to adjust the clamping force and clamping angle to grab and convey the workpiece.

[0011] As a further improvement of the present invention, the first mechanical arm comprises a first mounting base, a first connecting arm is provided on the first mounting base, and a universal movable joint is provided at one end of the first connecting arm away from the first mounting base;

[0012] The universal movable joint is provided with clamping claws, and the clamping claws include three clamping claws arranged at 120 degrees to each other, and the three clamping claws form a circular clamping gap.

[0013] As a further improvement of the present invention, a first steering arm is further provided between the universal movable joint and the clamping claw, and the clamping claw comprises at least a first clamping claw and a second clamping claw, the first clamping claw and the second clamping claw are both arranged on the first steering arm, and the clamping sizes of the first clamping claw and the second clamping claw are different.

[0014] As a further improvement of the present invention, the second robotic arm includes a second mounting base, a second connecting arm is provided on the second mounting base, the second mounting base and the second connecting arm are rotatably arranged, a third clamping claw and a fourth clamping claw are respectively provided on the second connecting arm, and the clamping gaps of the third clamping claw and the fourth clamping claw are different.

[0015] As a further improvement of the present invention, the third clamping claw and the fourth clamping claw each include two clamping claws arranged in pairs, and a proximity switch bracket is provided at one end of the two clamping claws facing away from each other.

[0016] The present application also includes a sorting method based on visual recognition, which performs sorting and recognition by using the mechanical arm sorting device based on visual recognition, and includes the following steps:

[0017] Obtain workpiece requirements, place the set model workpiece on the conveyor belt, and the conveyor belt will transport it;

[0018] The recognition component obtains the front view posture and the placement angle of the workpiece, and the recognition component transmits the front view posture and the swing angle of the workpiece to the control component;

[0019] The control component selects the robot arm according to the workpiece signal, controls the robot arm to move to the workpiece displacement path and adjusts the robot arm's grasping angle, and the robot arm grasps the workpiece and places the workpiece at a set workstation.

[0020] As a further improvement of the present invention, the control component selects the robot arm according to the workpiece model and specifically includes the following steps:

[0021] The control component obtains the workpiece image information transmitted by the recognition component, and determines the type of the workpiece according to the workpiece image information;

[0022] When the workpiece is a circular component, the control component controls the first robot arm to grab the workpiece;

[0023] When the workpiece is a rectangular or L-shaped component, the control component controls the second robot arm to grab the workpiece.

[0024] As a further improvement of the present invention, the control component further comprises the following steps of selecting a robot arm according to the workpiece model:

[0025] The control component obtains the workpiece size according to the workpiece image information;

[0026] When the workpiece is a circular component, the grasping size of the first clamping claw and the second clamping claw of the first mechanical arm is obtained, the first clamping claw or the second clamping claw is selected according to the size of the workpiece, and the first mechanical arm is rotated so that the first clamping claw or the second clamping claw faces the direction of the workpiece;

[0027] When the workpiece is a rectangular or L-shaped component, the grasping size of the third clamping claw and the fourth clamping claw of the second robotic arm is obtained, the third clamping claw or the fourth clamping claw is selected according to the size of the workpiece, and the second robotic arm is rotated so that the third clamping claw or the fourth clamping claw faces the direction of the workpiece.

[0028] As a further improvement of the present invention, the control component controls the robot arm to move to the workpiece displacement path, specifically comprising the following steps:

[0029] The visual component divides the conveyor belt into N transmission areas (N ≥ 2) along the width direction;

[0030] The visual component obtains the position of the workpiece on the conveyor belt;

[0031] The control component obtains the area where the workpiece is located according to the workpiece image information, and the control component controls the robot arm to move to the corresponding transmission area.

[0032] As a further improvement of the present invention, the control component controls the mechanical arm to adjust the grasping angle specifically including the following steps:

[0033] The control component obtains the workpiece image information transmitted by the recognition component, determines the workpiece type according to the workpiece image information, and retrieves the six-view information of the workpiece according to the workpiece type;

[0034] The control component compares the workpiece front view posture with the workpiece six-view information to obtain the workpiece shape and size under the workpiece front view posture. The control component controls the robot arm to adjust the clamping gap so that the clamping size of the robot arm matches the size of the workpiece.

[0035] The control component compares the workpiece's front-view posture with the workpiece's six-view information to obtain a deflection angle of the workpiece relative to the first direction in the front-view posture. The control component controls the robotic arm to rotate a corresponding angle so that the robotic arm's clamping angle is the same as the workpiece's deflection angle.

[0036] The above-mentioned improved technical features can be combined with each other as long as they do not conflict with each other.

[0037] In general, compared with the prior art, the above technical solution conceived by the present invention has the following beneficial effects:

[0038] (1) The robot arm sorting device based on visual recognition of the present invention obtains the front view posture and placement angle of the workpiece through the recognition component, and selects the corresponding robot arm structure through the control component to achieve clamping of circular or other non-circular structural parts, and adjusts the clamping distance and clamping force of the two robot arms and the clamping angle of the two robot arms according to the front view posture of the workpiece, so as to achieve accurate clamping of different workpieces. It avoids the problem that conventional robot arms cannot clamp various special-shaped workpieces, and can avoid the problem of unstable clamping or rigid contact with the workpiece caused by excessive or insufficient clamping force when the sorting device clamps workpieces of different sizes or shapes, thereby ensuring that the workpiece will not be damaged due to clamping.

[0039] (2) The robot arm sorting device based on visual recognition of the present invention realizes the grasping of gears of different sizes and rectangular parts and L-shaped structural parts by arranging a first clamping claw and a second clamping claw on the first robot arm, and arranging a third clamping claw and a fourth clamping claw on the second robot arm, thereby greatly improving the grasping efficiency of workpieces.

[0040] (3) The visual recognition-based sorting method of the present invention divides the transmission area of ​​the conveyor belt into partitions and calls the robot arm to the set position, thereby improving the efficiency of grasping the workpiece; secondly, by selecting clamping claws of different types and sizes according to the type of workpiece, accurate grasping of the workpiece is achieved, ensuring stable grasping of the workpiece while avoiding damage to the workpiece; at the same time, the present application accurately identifies the placement direction and angle of the workpiece through the recognition component, and calls the six-view information of the workpiece to adjust the clamping gap of the clamping claw accordingly, while ensuring that the clamping claw and the workpiece are accurately matched, avoiding the problem of the clamping claw over-contracting and clamping the workpiece, thereby greatly improving the sorting efficiency of the workpiece and avoiding clamping the workpiece. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 is a schematic diagram of the overall structure of a robotic arm sorting device based on visual recognition in an embodiment of the present invention;

[0042] Figure 2 Schematic diagram of the connection structure between the first mechanical arm and the main arm in an embodiment of the present invention;

[0043] Figure 3 It is a schematic diagram of the overall structure of the second mechanical arm in an embodiment of the present invention.

[0044] In all the drawings, the same reference numerals represent the same technical features, specifically:

[0045] 1. Conveyor belt; 2. Identification component; 3. Handling component;

[0046] 301, first mechanical arm; 302, second mechanical arm; 303, main arm; 304, first mounting base; 305, first connecting arm; 306, universal movable joint; 307, first clamping claw; 308, second clamping claw; 309, second mounting base; 310, second connecting arm; 311, third clamping claw; 312, fourth clamping claw; 313, proximity switch bracket. DETAILED DESCRIPTION

[0047] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0048] In the description of the present invention, it should be understood that, unless otherwise specified, terms such as “center”, “longitudinal”, “lateral”, “length”, “width”, “thickness”, “up”, “down”, “front”, “back”, “left”, “right”, “vertical”, “horizontal”, “top”, “bottom”, “inside”, “outside”, “clockwise”, “counterclockwise”, “axial”, “radial”, “circumferential”, etc., indicating orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention.

[0049] In addition, unless otherwise specified, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0050] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0051] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0052] Example:

[0053] See also Figures 1 to 3 The visual recognition-based robotic arm sorting device in the preferred embodiment of the present invention includes a conveyor belt 1 arranged along a first direction, an identification component 2 and a conveying component 3 arranged above the conveyor belt 1, and the identification component 2 and the conveying component 3 are arranged in sequence along the first direction; wherein the identification component 2 is used to identify the front view posture and placement angle of the workpiece transported along the first direction; the conveying component 3 is used to carry the workpiece, and includes a first robotic arm 301 and a second robotic arm 302, wherein the first robotic arm 301 is used to carry and clamp circular structural parts, and the second robotic arm 302 is used to carry and clamp non-circular structural parts with at least two clamping surfaces; and a control component, which is communicated with the identification component 2 and the conveying component 3 respectively, and is used to send the front view posture and placement angle of the workpiece obtained by the identification component 2 to the conveying component 3, and control the conveying component 3 to adjust the clamping force and clamping angle to grab and carry the workpiece.

[0054] Specifically, conventional material sorting devices are usually used for logistics handling, which usually have boxes or cushioning foam on the outside of the items, and the items are usually packaged in a rectangular structure, so that as long as the conventional robot arm grabs the goods, it will not damage the goods even if it applies a certain degree of pressure to the goods. Relatively speaking, this application is mainly used for the transportation of parts for large equipment, vehicles or instruments and equipment, etc., so as to facilitate installation by operators or intelligent robots. In the actual grasping process, it is mainly used for the grasping and installation of gears, mechanical parts, etc., which are mainly metal structural parts or hard plastic structural parts, which themselves are basically special-shaped structural parts according to the shape and size of the workpiece, and do not have elasticity. Therefore, conventional sorting manipulators cannot be used for grasping workpieces. On the one hand, they cannot adapt to the grasping of various special-shaped structural parts, and on the other hand, they cannot control the grasping force, which can easily cause damage to the workpiece.

[0055] The robot arm sorting device based on visual recognition in the present application obtains the front view posture and placement angle of the workpiece respectively through the recognition component 2, and selects the corresponding robot arm structure through the control component to achieve clamping of circular or other non-circular structural parts, and adjusts the clamping distance and clamping force of the two robot arms and the clamping angle of the two robot arms according to the obtained front view posture of the workpiece, so as to achieve accurate clamping of different workpieces. It avoids the problem that conventional robot arms cannot clamp various special-shaped workpieces, and can avoid the problem of unstable clamping or rigid contact with the workpiece caused by excessive or insufficient clamping force when the sorting device clamps workpieces of different sizes or shapes, thereby ensuring that the workpiece will not be damaged due to clamping.

[0056] Specifically, the first direction in the present application refers to the arrangement direction of the conveyor belt 1, that is, the transmission direction of the conveyor belt 1. The front view posture of the workpiece in the present application refers to the posture of the workpiece placed on the conveyor belt 1 facing the identification component 2, that is, the identification component 2 is mainly used to identify the shape and posture of the workpiece facing vertically upward. Specifically, when the workpiece is a gear or other circular structure, the front view posture of the workpiece is usually a circular surface, and it needs to be grasped by the first robot arm 301. When the workpiece is a rectangular or L-shaped structure, the front view posture of the workpiece is a rectangular or L-shaped, and it needs to be grasped by the second robot arm 302.

[0057] Furthermore, as a preferred embodiment of the present invention, the first robotic arm 301 and the second robotic arm 302 in the present application are both installed through a mounting frame, which is a rectangular hollow aluminum alloy frame, and a fixed pad is provided at the mounting mating surface between the mounting frame and the first robotic arm 301 or the second robotic arm 302, each end of the fixed pad is connected to the frame of the mounting frame, and the first robotic arm 301 or the second robotic arm 302 is installed on the fixed pad.

[0058] Further preferably, the first mechanical arm 301 and the second mechanical arm 302 are mainly used for moving in a plane and vertically grasping, and their range of motion is limited. Therefore, a main arm 303 is also provided between the fixed pad and the first mechanical arm 301 or the second mechanical arm 302, and the main arm 303 is connected to the first mechanical arm 301 or the second mechanical arm 302 through a spherical joint; the other end of the main arm 303 is installed on the fixed pad. The main arm 303 is hollowed out inside and is composed of multiple groups of arms with angles of 120° to each other, so as to ensure the bearing capacity of the main arm 303 and reduce the weight of the main arm 303 to improve the movement stability of the handling assembly 3. At the same time, the main arm 303 is connected to the servo motor through a flange, and the servo motor is arranged on the fixed pad to drive the main arm 303 to move to adjust the position of the first mechanical arm 301 or the second mechanical arm 302.

[0059] Furthermore, as a preferred embodiment of the present invention, the first robotic arm 301 in the present application includes a first mounting base 304, which is connected to the main arm 303, and a first connecting arm 305 is provided on the first mounting base 304. A universal movable joint 306 is provided at one end of the first connecting arm 305 facing away from the first mounting base 304, and a clamping claw is provided on the universal movable joint 306. The clamping claw includes three clamping claws arranged at 120° to each other, and the three clamping claws form a circular clamping gap for grabbing gears.

[0060] Further preferably, a first steering arm is further provided between the universal movable joint 306 and the clamping claw of the present application, and the clamping claw includes at least a first clamping claw 307 and a second clamping claw 308, and the first clamping claw 307 and the second clamping claw 308 are both provided on the first steering arm, and the first clamping claw 307 and the second clamping claw 308 have the same structure, and the clamping sizes of the first clamping claw 307 and the second clamping claw 308 are different. In the actual grasping process, due to the large size difference of gears of different models and sizes, it is difficult to complete the clamping of gears of all models and sizes with a single clamping claw, so the present application corresponds to setting the first clamping claw 307 and the second clamping claw 308 on the first mechanical arm 301, and when it is necessary to grasp the gear of the corresponding model and size, the first steering arm is driven to rotate through the universal movable joint 306, so that the first clamping claw 307 or the second clamping claw 308 faces the gear to complete the grasping of the gear.

[0061] Further preferably, a rack is provided on the inner side of the clamping jaw of the present application, and the rack is used to mesh with the surface of the gear to increase the stability of the clamping jaw gripping the gear. Preferably, the first clamping jaw 307 and the second clamping jaw 308 in the present application are arranged in different directions of the first steering arm, so that when the first clamping jaw 307 or the second clamping jaw 308 is used respectively, the two will not interfere with each other.

[0062] Further preferably, the second mechanical arm 302 in the present application includes a second mounting base 309, which is also connected to the main arm 303. A second connecting arm 310 is provided on the second mounting base 309, which is rotatably arranged between the second mounting base 309 and the second connecting arm 310, and a third clamping claw 311 and a fourth clamping claw 312 are respectively provided on the second connecting arm 310, and the third clamping claw 311 and the fourth clamping claw 312 have the same structure, and the clamping gaps of the third clamping claw 311 and the fourth clamping claw 312 are different. The third clamping claw 311 and the fourth clamping claw 312 are used to clamp rectangular or L-shaped or other shaped workpieces. As with the case of gears, the sizes of different workpieces also vary greatly. Therefore, the present application provides a third clamping claw 311 and a fourth clamping claw 312 with different clamping gaps to clamp workpieces of different sizes.

[0063] Further preferably, the third clamping claw 311 and the fourth clamping claw 312 in the present application both include a pair of clamping claws, and the two clamping claws can be used to clamp the workpiece. At the same time, a proximity switch bracket 313 is provided at the ends of the two clamping claws facing away from each other, and the proximity switch bracket 313 is used for the clamping claws to abut against the side of the workpiece after clamping the surface of the workpiece, so as to achieve stable clamping of the workpiece.

[0064] Furthermore, as an optional embodiment of the present invention, the present application may also provide a third robot arm, which is mainly used to grab and carry the workpieces that are boxed in sets, so as to transport the workpieces required for assembly in sets to the set processing stations.

[0065] Furthermore, with respect to the robot arm sorting device based on visual recognition in the present application, the present application also includes a sorting method based on visual recognition, which includes the following steps:

[0066] Obtain workpiece demand, place the set model workpiece on conveyor belt 1, and conveyor belt 1 will transport it;

[0067] The recognition component 2 obtains the front view posture and the placement angle of the workpiece, and the recognition component 2 transmits the front view posture and the swing angle of the workpiece to the control component;

[0068] The control component selects the robot arm according to the workpiece signal, controls the robot arm to move to the workpiece displacement path and adjusts the robot arm's grasping angle, and the robot arm grasps the workpiece and places the workpiece at a set workstation.

[0069] The core of the sorting method based on visual recognition in this application is to obtain the model and state of the workpiece through the control component, select the robot arm accordingly, and adjust the clamping gap size and grasping angle of the robot arm to avoid damage to the workpiece while achieving accurate grasping of the workpiece.

[0070] Further, as a preferred embodiment of the present invention, the control component in the present application selects the robot arm according to the workpiece model and specifically includes the following steps:

[0071] The control component obtains the workpiece image information transmitted by the recognition component 2, and determines the type of the workpiece according to the workpiece image information;

[0072] When the workpiece is a circular component, the control component controls the first mechanical arm 301 to grab the workpiece;

[0073] When the workpiece is a rectangular or L-shaped member, the control component controls the second robot arm 302 to grab the workpiece.

[0074] Specifically, when the workpiece is a gear structure, the first mechanical arm 301 selects the first clamping claw 307 or the second clamping claw 308 of the corresponding size, moves the first mechanical arm 301 to the top of the workpiece displacement path, and obtains the transmission rate of the workpiece, and the first mechanical arm 301 drops and clamps the gear. When the workpiece is a rectangular or L-shaped component, the workpiece size and clamping center of gravity under the standard state are retrieved to obtain the workpiece clamping point, and then the second mechanical arm 302 is moved to the top of the workpiece displacement path, and the second mechanical arm 302 drops and clamps the workpiece.

[0075] Further preferably, the control component selects the robot arm according to the workpiece model and further comprises the following steps:

[0076] The control component obtains the workpiece size according to the workpiece image information;

[0077] When the workpiece is a circular component, the grasping size of the first clamping claw 307 and the second clamping claw 308 of the first mechanical arm 301 is obtained, the first clamping claw 307 or the second clamping claw 308 is selected according to the size of the workpiece, and the first mechanical arm 301 is rotated so that the first clamping claw 307 or the second clamping claw 308 faces the direction of the workpiece;

[0078] When the workpiece is a rectangular or L-shaped component, the grasping size of the third clamping claw 311 and the fourth clamping claw 312 of the second robotic arm 302 is obtained, and the third clamping claw 311 or the fourth clamping claw 312 is selected according to the size of the workpiece. The second robotic arm 302 rotates and makes the third clamping claw 311 or the fourth clamping claw 312 face the direction of the workpiece.

[0079] Further, as a preferred embodiment of the present invention, the control component in the present application controls the robot arm to move to the workpiece displacement path specifically including the following steps:

[0080] The visual component divides the conveyor belt 1 into N transmission areas (N≥2) along the width direction;

[0081] The visual component obtains the position of the workpiece on conveyor belt 1;

[0082] The control component obtains the area where the workpiece is located according to the workpiece image information, and the control component controls the robot arm to move to the corresponding transmission area.

[0083] In the actual clamping process, since the robot arm is set beside the conveyor belt 1, in order to ensure the high-speed movement and clamping of the robot arm, a clamping area is usually defined on the conveyor belt 1, and then the clamping area is divided into multiple transmission areas, so as to facilitate the high-speed and accurate movement of the robot arm to the set clamping position to facilitate the clamping of the workpiece.

[0084] Further preferably, as a preferred embodiment of the present invention, the control component in the present application controls the mechanical arm to adjust the gripping angle specifically includes the following steps:

[0085] The control component obtains the workpiece image information transmitted by the recognition component 2, determines the workpiece type according to the workpiece image information, and retrieves the workpiece six-view information according to the workpiece type;

[0086] The control component compares the workpiece front view posture with the workpiece six-view information to obtain the workpiece shape and size under the workpiece front view posture. The control component controls the robot arm to adjust the clamping gap so that the clamping size of the robot arm matches the size of the workpiece.

[0087] The control component compares the workpiece's front-view posture with the workpiece's six-view information to obtain a deflection angle of the workpiece relative to the first direction in the front-view posture. The control component controls the robotic arm to rotate a corresponding angle so that the robotic arm's clamping angle is the same as the workpiece's deflection angle.

[0088] Since the workpiece in this application is not a box structure in traditional logistics, it may be an irregular structure. Therefore, the placement posture of the workpiece on the conveyor belt 1 will affect the gripping of the workpiece by the robot arm. Based on this, this application sets the six-view information of the workpiece and obtains the inclination state of the workpiece, so as to obtain the best gripping position of the workpiece under any placement posture, and adjusts the position and gripping angle of the robot arm accordingly to achieve accurate gripping of the workpiece.

[0089] Specifically, the matching of the clamping size of the robot arm and the workpiece size here means that the gap between the clamping claws of the robot arm is 1 to 10 mm larger than the workpiece size. When the clamping claws of the robot arm fall on both sides of the workpiece, the workpiece can be clamped by slightly adjusting the clamping force of the clamping claws and avoiding damage to the workpiece caused by excessive clamping force.

[0090] The visual recognition-based sorting method of the present invention divides the transmission area of ​​the conveyor belt 1 into partitions, and calls the robot arm to the set position, thereby improving the efficiency of grasping the workpiece; secondly, by selecting clamping claws of different types and sizes according to the type of workpiece, accurate grasping of the workpiece is achieved, ensuring stable grasping of the workpiece while avoiding damage to the workpiece; at the same time, the application accurately identifies the placement direction and angle of the workpiece through the identification component 2, and calls up the six-view information of the workpiece to adjust the clamping gap of the clamping claw accordingly, while ensuring that the clamping claw is accurately matched with the workpiece, avoiding the problem of the clamping claw over-contracting and clamping the workpiece, thereby greatly improving the sorting efficiency of the workpiece and avoiding clamping the workpiece.

[0091] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A robotic arm sorting device based on visual recognition, characterized in that: include: a conveyor belt arranged along a first direction; An identification component and a transport component are disposed above the conveyor belt, and the identification component and the transport component are spaced apart along a first direction; The recognition component is used to recognize the front view posture and placement angle of the workpiece transported along the first direction; The transport assembly comprises a first mechanical arm and a second mechanical arm, wherein the first mechanical arm is used for transporting and clamping circular structural parts, and the second mechanical arm is used for transporting and clamping non-circular structural parts; And a control component, which is communicated with the identification component and the conveying component respectively, and is used to send the formal posture and placement angle of the workpiece obtained by the identification component to the conveying component, and control the conveying component to adjust the clamping force and clamping angle to grab and convey the workpiece.

2. The robot arm sorting device based on visual recognition according to claim 1 is characterized in that: The first mechanical arm comprises a first mounting base, a first connecting arm is disposed on the first mounting base, and a universal movable joint is disposed at one end of the first connecting arm away from the first mounting base; The universal movable joint is provided with clamping claws, and the clamping claws include three clamping claws arranged at 120 degrees to each other, and the three clamping claws form a circular clamping gap.

3. The robot arm sorting device based on visual recognition according to claim 2 is characterized in that: A first steering arm is also provided between the universal movable joint and the clamping claw. The clamping claw comprises at least a first clamping claw and a second clamping claw. The first clamping claw and the second clamping claw are both provided on the first steering arm, and the clamping sizes of the first clamping claw and the second clamping claw are different.

4. The robot arm sorting device based on visual recognition according to claim 1, characterized in that: The second robotic arm includes a second mounting base, a second connecting arm is provided on the second mounting base, the second mounting base and the second connecting arm are rotatably arranged, a third clamping claw and a fourth clamping claw are respectively provided on the second connecting arm, and the clamping gaps of the third clamping claw and the fourth clamping claw are different.

5. The robot arm sorting device based on visual recognition according to claim 4 is characterized in that: The third clamping claw and the fourth clamping claw each include two clamping claws arranged in pairs, and a proximity switch bracket is provided at one end of the two clamping claws facing away from each other.

6. A sorting method based on visual recognition, which performs sorting and recognition by using a mechanical arm sorting device based on visual recognition as described in any one of claims 1 to 5, characterized in that: The steps include: Obtain workpiece requirements, place the set model workpiece on the conveyor belt, and the conveyor belt will transport it; The recognition component obtains the front view posture and the placement angle of the workpiece, and the recognition component transmits the front view posture and the swing angle of the workpiece to the control component; The control component selects the robot arm according to the workpiece signal, controls the robot arm to move to the workpiece displacement path and adjusts the robot arm's grasping angle, and the robot arm grasps the workpiece and places the workpiece at a set workstation.

7. The visual recognition-based sorting method according to claim 6, characterized in that: The control component selects the robot arm according to the workpiece model and specifically includes the following steps: The control component obtains the workpiece image information transmitted by the recognition component, and determines the type of the workpiece according to the workpiece image information; When the workpiece is a circular component, the control component controls the first robot arm to grab the workpiece; When the workpiece is a rectangular or L-shaped component, the control component controls the second robot arm to grab the workpiece.

8. The visual recognition-based sorting method according to claim 7, characterized in that: The control component further comprises the following steps of selecting a robot arm according to the workpiece model: The control component obtains the workpiece size according to the workpiece image information; When the workpiece is a circular component, the grasping size of the first clamping claw and the second clamping claw of the first mechanical arm is obtained, the first clamping claw or the second clamping claw is selected according to the size of the workpiece, and the first mechanical arm is rotated so that the first clamping claw or the second clamping claw faces the direction of the workpiece; When the workpiece is a rectangular or L-shaped component, the grasping size of the third clamping claw and the fourth clamping claw of the second robotic arm is obtained, the third clamping claw or the fourth clamping claw is selected according to the size of the workpiece, and the second robotic arm is rotated so that the third clamping claw or the fourth clamping claw faces the direction of the workpiece.

9. The visual recognition-based sorting method according to claim 6, characterized in that: The control component controls the robot arm to move to the workpiece displacement path, specifically comprising the following steps: The visual component divides the conveyor belt into N transmission areas (N ≥ 2) along the width direction; The visual component obtains the position of the workpiece on the conveyor belt; The control component obtains the area where the workpiece is located according to the workpiece image information, and the control component controls the robot arm to move to the corresponding transmission area.

10. The visual recognition-based sorting method according to claim 6, characterized in that: The control component controls the robot arm to adjust the gripping angle, specifically comprising the following steps: The control component obtains the workpiece image information transmitted by the recognition component, determines the workpiece type according to the workpiece image information, and retrieves the six-view information of the workpiece according to the workpiece type; The control component compares the workpiece front view posture with the workpiece six-view information to obtain the workpiece shape and size under the workpiece front view posture. The control component controls the robot arm to adjust the clamping gap so that the clamping size of the robot arm matches the size of the workpiece. The control component compares the workpiece's front-view posture with the workpiece's six-view information to obtain a deflection angle of the workpiece relative to the first direction in the front-view posture. The control component controls the robotic arm to rotate a corresponding angle so that the robotic arm's clamping angle is the same as the workpiece's deflection angle.

Images