Automatic eggplant picking robot and target detection method
通过简化机械臂结构和结合目标检测模块,实现了茄子自动采摘机器人在大棚内高效、精准的茄子采摘,解决了现有技术中机械臂结构复杂导致的操作效率低的问题。
Patent Information
- Application Number
- CN202411109295.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-13
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-08-13
AI Technical Summary
The existing eggplant picking robots have low operating efficiency in the limited space in the greenhouse, and the robotic arm structure is complex, making it difficult to adapt to the operation between the branches of lush plants, resulting in poor picking results.
An eggplant automatic picking robot was designed, adopting a simplified robotic arm structure, including a quadrilateral connecting rod mechanism and synchronous belt transmission, combined with a walking device and a target detection module, and high-precision positioning and identification are achieved through a depth camera and YOLO target detection model to achieve high-precision positioning and identification. Unmanned picking.
It improves the operation flexibility and control simplicity of the robotic arm, realizes high-precision eggplant picking, reduces driving torque requirements, improves picking efficiency and accuracy, and adapts to limited space operations.
Smart Images

Figure CN118901406A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of agricultural picking, and in particular to an automatic eggplant picking robot and a target detection method. Background Art
[0002] Eggplants are mainly grown in greenhouses with a relatively high degree of standardization. The spacing between each ridge of eggplants in the greenhouse is 70-80cm, and the spacing between each plant is 30-40cm. At present, the planting, care, and picking of eggplants are mainly done manually, but the roots of mature eggplants are tough, and manual shearing is laborious. It is necessary to bend over or squat during the operation, which is labor-intensive and inefficient. Automated eggplant picking is an advanced agricultural technology that aims to use automated equipment to improve the efficiency and accuracy of eggplant picking. The automated eggplant picking system includes a visual recognition system and a robotic system, which can achieve efficient, accurate, and reliable operations, significantly improving agricultural production efficiency and product quality.
[0003] CN221178502U discloses a round eggplant harvesting device, comprising a base, a collecting basket and a fixed seat are fixed on the surface of the base, a flip plate is rotatably connected to the side of the collecting basket, a moving wheel is rotatably connected to the lower end of the base, a shock absorbing structure is rotatably connected between the base and the moving wheel, a rotating steering gear is fixed inside the fixed seat, a positioning plate is transmission-connected to the upper end of the rotating steering gear, a variable frame is rotatably connected to the upper end of the positioning plate, a rotating steering gear is rotationally connected to the upper end of the variable frame, a rotating steering gear is rotatably connected to the upper end of the variable frame, a rotating steering gear is arranged inside the variable frame and the rotating frame, a driving motor is connected to the upper end of the rotating frame, and a harvesting saw blade is transmission-connected to the surface of the driving motor.
[0004] CN109618670B discloses an intelligent robot for picking and sorting fruits and vegetables, comprising a walking and lifting mechanism, a terminal picking mechanism, a fruit receiving and conveying mechanism, a conveying and sorting mechanism and an induction control system, wherein: the walking and lifting mechanism comprises a mobile base and a lifting platform installed on the mobile base; the terminal picking mechanism is installed on the lifting platform, comprising a multi-section mechanical arm, a front-end mechanical claw, and a shearing mechanism installed on the mechanical claw, which moves to the position of fruits and vegetables through the multi-section mechanical arm, grabs fruits and vegetables with the mechanical claw, and cuts off the fruit stems with the shearing mechanism; the fruit receiving and conveying mechanism is installed on the lifting platform, comprising a fruit receiving net bag and a fruit conveying pipe connected to the bottom of the net bag, wherein the diameter of the fruit conveying pipe is larger than the diameter of one fruit or vegetable and smaller than the diameter of two fruits and vegetables, and the fruits and vegetables picked by the mechanical claw The fruits and vegetables are sent into the receiving net bag and fall one by one from the fruit delivery pipe; the conveying and sorting mechanism is installed on the surface of the lifting platform, which includes a slope section, a conveying section and a sorting section in sequence. The slope section is located at the starting end of the conveying section, including an inclined bottom plate and baffles on both sides, and the starting position of the slope section is located directly below the fruit delivery pipe. The conveying section includes an active roller, a driven roller and a conveyor belt. The sorting section is located at the end of the conveying section, including a bracket, a steering valve installed on the top of the bracket, and a sorting plate installed on the steering valve. A plurality of fruit and vegetable frames are placed side by side on the mobile base facing the end of the sorting plate. The fruits and vegetables dropped from the fruit delivery pipe roll down to the conveying section through the slope section and are sent to the sorting section. The sorting section selects the corresponding fruit and vegetable frame for unloading according to the size and maturity of the fruits and vegetables; the induction control system includes a controller, Multiple infrared sensors, gravity sensors and visual sensors connected to the controller are installed side by side on the baffles on both sides of the slope section to sense the rolling distance of fruits and vegetables. The detected data is sent to the controller, and the controller calculates the fruit diameter based on the distance data. The specific diameter calculation method is that each pair of infrared sensors calculates a fruit diameter value D=La*t / 2-b*t / 2, where L is the distance between the two side plates, a is the light speed of the infrared sensor emitted on one side baffle, b is the light speed of the infrared sensor emitted on the other side baffle, and t is the time taken for the light emission to touch the fruit and reflect back to the receiver. Then, the multiple infrared measurement data are averaged to obtain the diameter of the fruit. The gravity sensor is installed on the conveyor belt below the table It senses the gravity of fruits and vegetables above the conveyor belt and transmits the weight value back to the controller. The controller calculates the fruit maturity R=G / C based on the weight value and the diameter of the fruit, where G represents the weight of the fruit and C represents the volume of the fruit, which is calculated by the diameter. The fuzzy PID analysis system is then used to calculate the maturity coefficient of each fruit and compare it with the maturity critical value preset in the controller. If it exceeds the critical value, it means it is over-mature and a fruit and vegetable box should be placed separately. If it does not exceed the critical value, its circumference size is compared with the pre-stored size in the controller. If it is larger than the pre-stored size, it is placed in a fruit and vegetable box. If it is smaller than the pre-stored size, it is placed in another fruit and vegetable box. The visual sensor is installed on the mechanical claw to obtain the position of the fruit.
[0005] CN108401682B discloses an end effector of a spherical fruit picking robot, comprising a cylinder, a connecting rod, an actuator body and a cutting mechanism; wherein the actuator body comprises a cylindrical shell, a connecting plate, a push rod, a push plate and a second supporting ring; the cutting mechanism comprises a cutting rod assembly; one end of the connecting rod is connected to the piston of the cylinder, and the other end is connected to the push rod; the bottom surface of one side of the cylindrical shell is open, and the bottom surface of the other side is connected to the connecting plate; a first through hole is provided in the center of the connecting plate, and the push rod is not connected to the piston. One end of the connecting rod passes through the first through hole, enters the interior of the cylindrical shell, and is connected to the push plate; a second circular groove is provided on the outer side of the cylindrical shell, the second circular groove is coaxial with the cylindrical shell, and a second supporting circular ring is provided in the second circular groove; the cutting rod assembly comprises a first cutting rod, a first fixing block and a blade; one end of the first cutting rod is sequentially connected to the first fixing block and the blade in a direction perpendicular to the rod body, and the other end is a free end; the side surface of the cylindrical shell is axially opened from the bottom surface of the opening to the side connected to the connecting plate cutting notch; the free end of the first cutting rod is placed in the cutting notch, and the blade is placed in the direction of the central axis of the cylindrical shell; the free end of the first cutting rod is connected to one end of the first connecting block, and the other end of the first connecting block is connected to the push plate through the first connecting piece; the first connecting piece and the first connecting block are connected in a hinged manner; a second through hole is provided on the first connecting block, and the second supporting ring passes through the second through hole; the number of the cutting notches is 2, the lengths of the two cutting notches are equal and are symmetrically arranged relative to the central axis of the cylindrical shell; the number of the cutting rod assemblies is 2, the free ends of the first cutting rods of the two cutting rod assemblies are respectively placed in the two cutting notches, and the structures, sizes and connection methods of the two cutting rod assemblies are the same; the cutting rod assembly also includes a second cutting rod, which is arranged between the first cutting rod and the first fixing block, and the two ends of the second cutting rod are respectively connected to the first cutting rod and the first fixing block; the width of the second cutting rod is greater than the width of the first cutting rod.
[0006] Due to the regular arrangement of eggplant plants in the greenhouse, robots can be used to replace manual labor for daily eggplant care, picking and other operations. The current picking robots used for spherical fruits have complex structures, cumbersome control processes, and complex walking structures and mechanical arm structures that are difficult to adapt to the limited space in the greenhouse, resulting in poor picking results. Summary of the invention
[0007] Long-term practice has found that harvesting robots must fully consider the working environment during operation. The structure of the robotic arm directly affects the efficiency of subsequent operations. Since eggplant grows between lush plant branches, usually only increasing the number of robotic arms is used to increase the degree of freedom, which often leads to the complexity of the entire mechanical structure, and then to complex control algorithms, and cannot effectively adapt to technical problems such as picking operations in limited spaces.
[0008] In view of this, the present invention aims to propose an automatic eggplant picking robot comprising: a picking actuator, a walking device, and a target detection module;
[0009] The picking actuator includes a driving motor assembly, a mechanical arm assembly, and an end effector assembly; one end of the mechanical arm assembly is connected to the driving motor assembly, and the other end of the mechanical arm assembly is connected to the end effector assembly; the mechanical arm assembly includes a first arm, a second arm, a third arm, and a fourth arm, and the driving motor assembly includes a first driving motor and a second driving motor; one end of the first arm is connected to the power output shaft of the first driving motor, and the other end of the first arm is connected to the second arm and the third arm respectively through a first shaft; one end of the fourth arm is connected to the end effector assembly, and the other end of the fourth arm is connected to the second arm and the third arm through a second shaft; the power output shaft of the second driving motor is fixedly provided with a first synchronous wheel, the first shaft is provided with a second synchronous wheel and a third synchronous wheel, the second shaft is provided with a fourth synchronous wheel, the first synchronous wheel and the second synchronous wheel can be powered by a synchronous belt, and the third synchronous wheel and the fourth synchronous wheel are powered by a synchronous belt;
[0010] The walking device includes a chassis mechanism and a wheel hub motor, the picking actuator is connected to the chassis mechanism through a screw assembly, the chassis mechanism is connected to the wheel hub motor, and the wheel hub motor is used to drive the wheel;
[0011] The target detection module is fixedly arranged at one end of the robotic arm assembly connected to the end effector assembly; the target detection module is used to identify ripe eggplant fruits.
[0012] In one embodiment, a first reduction mechanism is provided between the power output shaft of the second drive motor and the first synchronous wheel, and a second reduction mechanism is provided between the power output shaft of the first drive motor and the first arm;
[0013] The first speed reduction mechanism and the second speed reduction mechanism are respectively detachably connected to the bracket, and the bracket is fixedly connected to the screw nut.
[0014] In one embodiment, the first arm can be connected to the output shaft of the second reduction mechanism through a first connecting seat, a bearing seat is provided between the second synchronous wheel and the third synchronous wheel, and the first arm can be connected to the first shaft through the bearing seat; the second arm and the third arm can be connected to the two ends of the first shaft through the second connecting seat and the third connecting seat respectively; the second arm and the third arm can be connected to the second shaft through the bearing seats respectively.
[0015] In one embodiment, the fourth arm can be connected to the second shaft via a second connecting seat.
[0016] In one embodiment, the second synchronous wheel and the third synchronous wheel have the same size and the same number of teeth.
[0017] In one embodiment, the end effector assembly includes an electric cutting assembly, and the electric cutting assembly is detachably connected to the fourth arm.
[0018] In one embodiment, the chassis mechanism includes a chassis bracket, a steering wheel and at least two wheels, the hub motor is coaxially arranged in the wheel, and the chassis bracket is a plurality of connecting rods connected end to end to form an internal space capable of accommodating an energy storage battery; the chassis bracket is connected to the steering wheel, and the chassis bracket is connected to the wheel.
[0019] In one embodiment, the target detection module includes a depth camera and a host computer module, the depth camera is used to collect image data and can detect the target; the host computer module is connected to the depth camera to receive the image data transmitted by the depth camera.
[0020] The present invention also discloses a target detection method for executing the above-mentioned automatic eggplant picking robot, the target detection method comprising:
[0021] Step S1, training an eggplant target detection model based on YOLO, and deploying the trained eggplant target detection model to a depth camera;
[0022] Step S2, using a depth camera to collect image data, calculating the relative position between the target eggplant and the depth camera, and calculating the relative position between the target eggplant and the electric cutting component;
[0023] Step S3, calculating the required movement distances of the screw assembly and the first arm, the second arm, the third arm, and the fourth arm in the mechanical arm assembly, and controlling the first drive motor, the second drive motor, and the screw assembly to perform operations.
[0024] The present invention provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the target detection method of the present application as described above.
[0025] The automatic eggplant picking robot disclosed in the present invention is provided with a picking actuator, a walking device, and a target detection module; a driving motor assembly and an end effector assembly are connected through a mechanical arm assembly, and a quadrilateral linkage mechanism composed of a first arm, a second arm, a third arm, and a fourth arm is combined with a synchronous belt transmission, and the first driving motor drives the first arm to adjust the mechanical arm angle, and the second driving motor transmits power to the second arm, the third arm, and the fourth arm through the first synchronous wheel, the second synchronous wheel, the third synchronous wheel, and the fourth synchronous wheel. The mechanical movement and angle adjustment, the structure is simple and easy to operate and control, the mechanical arm assembly adopts a parallelogram linkage mechanism to make the mechanical arm structure compact, the pitch dynamic performance is high, the end angle operation is flexible, the degree of freedom is separated, and the control solution is simple. Assisted by synchronous belt transmission, the driving mass distribution is concentrated at the root, saving the required driving torque, and under the interaction of the two, the end effector assembly in the automatic eggplant picking robot can be quickly positioned to a suitable position in space. The working space is adjusted by the walking device, and the target detection module quickly converts the spatial position of the target into the control instructions of the first drive motor and the second drive motor, which can realize the detection and high-precision positioning functions of the target eggplant and realize high-precision operation. The present invention also discloses a target detection method for the above-mentioned automatic eggplant picking robot, which makes the entire robot more intelligent, can quickly adjust the working space, perform highly reliable and accurate identification and positioning measurement of the target eggplant rhizome, and guide the robot to automatically cut the eggplant fruit stem based on machine vision to realize unmanned picking operation.
[0026] Other features and advantages of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:
[0028] Figure 1 This is a schematic structural diagram of an automatic eggplant picking robot according to an embodiment of the present invention;
[0029] Figure 2 This is a schematic diagram of the structure of a mechanical arm assembly of an automatic eggplant picking robot according to an embodiment of the present invention;
[0030] Figure 3 This is a schematic diagram of the chassis structure of an automatic eggplant picking robot according to an embodiment of the present invention;
[0031] Figure 4 This is a schematic structural diagram of a screw rod assembly of an automatic eggplant picking robot according to an embodiment of the present invention;
[0032] Figure 5 The present invention is a schematic diagram of the connection structure between the yaw motor and the bracket of the automatic eggplant picking robot according to an embodiment of the present invention.
[0033] Marking Description
[0034] 1 drive motor assembly 2 robotic arm assembly
[0035] 3 End effector assembly 4 Chassis mechanism
[0036] 5 wheel hub motor 11 first drive motor
[0037] 12 Second drive motor 13 Yaw motor
[0038] 14 screw drive motor 15 support seat
[0039] 16 screw 17 screw support rod
[0040] 21 first arm 22 second arm
[0041] 23 third arm 24 fourth arm
[0042] 31 target detection module 32 electric cutting component
[0043] 41 first axis 42 second axis
[0044] 51 first synchronous wheel 52 second synchronous wheel
[0045] 53 third synchronous wheel 54 fourth synchronous wheel
[0046] 6 Screw rod assembly 61 first connecting seat
[0047] 62 bearing seat 63 fourth connecting seat
[0048] 64 second connection seat 65 third connection seat
[0049] 71 first reduction mechanism 72 second reduction mechanism
[0050] 73 third reduction mechanism 81 bracket
[0051] 82 screw nut 91 wheel
[0052] 92 steering wheel 93 chassis bracket DETAILED DESCRIPTION
[0053] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.
[0054] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.
[0055] It should be noted that the terms "first", "second", "third", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so as to describe the embodiments of the present invention described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0056] In the prior art, the harvesting robot must fully consider the working environment during the operation process. The structure of the mechanical arm directly affects the later operation efficiency. Since eggplant grows between lush plant branches, usually only increasing the number of mechanical arms to increase the degree of freedom often leads to complex entire mechanical structure, which in turn leads to complex control algorithms and cannot effectively adapt to harvesting operations in limited spaces. The present invention provides an automatic eggplant harvesting robot, such as Figure 1-3 As shown, the automatic eggplant picking robot comprises:
[0057] The automatic eggplant picking robot comprises: a picking actuator, a walking device, and a target detection module 31;
[0058] The picking actuator includes a drive motor assembly 1, a mechanical arm assembly 2, and an end effector assembly 3; one end of the mechanical arm assembly 2 is connected to the drive motor assembly 1, and the other end of the mechanical arm assembly 2 is connected to the end effector assembly 3; the mechanical arm assembly 2 includes a first arm 21, a second arm 22, a third arm 23, and a fourth arm 24, and the drive motor assembly 1 includes a first drive motor 11 and a second drive motor 12; one end of the first arm 21 is connected to the power output shaft of the first drive motor 11, and the other end of the first arm 21 is connected to the second arm 22, the fourth arm 24 through a first shaft 41. The third arm 23 is connected; one end of the fourth arm 24 is connected to the end effector assembly 3, and the other end of the fourth arm 24 is connected to the second arm 22 and the third arm 23 through a second shaft 42 (42); the power output shaft of the second drive motor 12 is fixedly provided with a first synchronous wheel, the first shaft 41 is provided with a second synchronous wheel 52 and a third synchronous wheel 53, the second shaft 42 is provided with a fourth synchronous wheel 54, the first synchronous wheel 51 and the second synchronous wheel 52 can be powered by a synchronous belt, and the third synchronous wheel 53 and the fourth synchronous wheel 54 are powered by a synchronous belt;
[0059] The walking device includes a chassis mechanism 4 and a wheel hub motor 5, the picking actuator is connected to the chassis mechanism 4 through a screw assembly 6, the chassis mechanism 4 is connected to the wheel hub motor 5, and the wheel hub motor 5 is used to drive the wheel 91;
[0060] The target detection module 31 is fixedly disposed at one end of the robotic arm assembly 2 connected to the end effector assembly 3; the target detection module 31 is used to identify ripe eggplant fruits.
[0061] The automatic eggplant picking robot disclosed in the present invention is provided with a picking actuator, a walking device, and a target detection module 31; the driving motor assembly 1 and the end effector assembly 3 are connected through the mechanical arm assembly 2, and the quadrilateral linkage mechanism composed of the first arm 21, the second arm 22, the third arm 23 and the fourth arm 24 is combined with the synchronous belt transmission, the first driving motor 11 drives the first arm 21 to adjust the mechanical arm angle, and the second driving motor 12 transmits the power to the second arm 22, the third arm 23, and the fourth arm 24 through the first synchronous wheel 51, the second synchronous wheel 52, the third synchronous wheel 53, and the fourth synchronous wheel 54. The mechanical movement and angle adjustment, the structure is simple and easy to operate and control, the mechanical arm assembly 2 adopts a parallelogram linkage mechanism to make the mechanical arm structure compact, the pitch dynamic performance is high, the end angle operation is flexible, the degree of freedom is separated, and the control solution is simple. Assisted by the synchronous belt transmission, the driving mass distribution is concentrated at the root, saving the required driving torque, and under the interaction of the two, the end effector assembly 3 in the automatic eggplant picking robot can be quickly positioned to a suitable position in space. By adjusting the working space through the walking device, the target detection module 31 quickly converts the spatial position of the target into control instructions for the first drive motor 11 and the second drive motor 12, which can realize the detection and high-precision positioning functions of the target eggplant, and realize highly reliable and accurate identification and positioning measurement of the target eggplant rhizome. Based on machine vision, the robot is guided to automatically cut the eggplant stem, realizing unmanned picking operations.
[0062] In order to reduce the output speed of the drive motor and increase the torque, improve the accuracy and stability of the torque output, realize more precise operation of the robot arm, and improve the high accuracy and high stability of the entire system. In a preferred embodiment, the power output shaft of the second drive motor 12 and the first synchronous wheel 51 are provided with a first reduction mechanism 71, and the power output shaft of the first drive motor 11 and the first arm 21 are provided with a second reduction mechanism 72;
[0063] The first reduction mechanism 71 and the second reduction mechanism 72 are detachably connected to the bracket 81, respectively. In order to further stabilize the connection between the entire picking actuator and the robot body, the housings of the first drive motor 11 and the second drive motor 12 are fixedly connected to the bracket 81. In another more preferred embodiment of the present invention, in order to improve the accuracy and stability of torque output and reduce the vibration of power output, the first reduction mechanism 71 and the second reduction mechanism 72 include a harmonic reducer, and the power output shaft of the second drive motor 12 and the power output shaft of the first drive motor 11 are coaxially connected to the second reduction mechanism 72 and the first reduction mechanism 71, respectively.
[0064] In order to better realize the vertical and horizontal movement of the picking actuator, such as Figure 4-5In a preferred embodiment, the screw 16 is driven to rotate by the screw drive motor 14, and the screw 16 is vertically fixedly installed on the chassis mechanism 4 through the screw support rod 17. The screw 16 rotates to realize the linear motion of the screw nut 82 and the support seat 15 in the direction of the vertical horizontal plane. The bracket 81 is fixedly connected to the power output shaft of the yaw motor 13, and the bracket 81 can rotate a certain angle following the rotation of the yaw motor 13 to realize the rotation of the entire picking actuator in the horizontal direction. The screw nut 82 is fixed to the support seat 15. In a more preferred case, the screw nut 82 and the support seat 15 are integrally processed. The housing of the yaw motor 13 is fixed to the support seat 15 by threads, and the support seat 15 is provided with a through hole for extending the power output shaft of the yaw motor 13 so as to be connected to the bracket 81. In another more preferred embodiment, in order to achieve a more accurate rotation angle, a third reduction mechanism 73 is coaxially fixedly arranged between the bracket 81 and the yaw motor 13.
[0065] In the process of picking the target eggplant, due to the interweaving of plant stems and leaves, it is a complex automated operation task, in order to achieve flexible movement and precise control of the robot arm at each joint. In a preferred embodiment, the first arm 21 can be connected to the output shaft of the second reduction mechanism 72 through the first connecting seat 61, and a bearing seat 62 is provided between the second synchronous wheel 52 and the third synchronous wheel 53. The first arm 21 can be connected to the first shaft 41 through the bearing seat 62; the second arm 22 and the third arm 23 can be connected to the two ends of the first shaft 41 through the second connecting seat 64 and the third connecting seat 65 respectively; the second arm 22 and the third arm 23 can be connected to the second shaft 42 through the bearing seat 62 respectively. In order to better realize mechanical transmission, a bearing is provided in the bearing seat 62, for example, a bearing is provided between the bearing seat 62 and the first shaft 41, which can realize rotation, but limit radial and axial relative movement. In order to rotate with high precision and better limit, in a more preferred case, the bearing includes a sliding bearing. The first connecting seat 61 is connected to the output shaft of the second reduction mechanism 72 by interference fit, and the second connecting seat 64 and the third connecting seat 65 are respectively connected to both ends of the first shaft 41 by interference fit.
[0066] In order to better transmit power from the second shaft 42 to the fourth arm 24 and thus realize the picking operation, in a preferred embodiment, the fourth arm 24 can be connected to the second shaft 42 via a fourth connecting seat 63 .
[0067] In order to keep the fourth arm 24 parallel to the first arm 21 in direction, in a preferred embodiment, the second synchronous wheel 52 and the third synchronous wheel 53 have the same size and the same number of teeth.
[0068] In order to better cut off the stems of the target eggplants and achieve cutting and picking after rapid positioning, in a preferred embodiment, the end effector assembly 3 includes an electric cutting assembly 32, and the electric cutting assembly 32 is detachably connected to the fourth arm 24.
[0069] In order to better realize the power supply of the robot, make the overall structure more stable and compact, and make the chassis more stable during walking, in a preferred embodiment, the chassis mechanism 4 includes a chassis bracket 93, a steering wheel 92 and at least two wheels 91, the wheel 91 is coaxially provided with the hub motor 5, the chassis bracket 93 is a plurality of connecting rods connected end to end to form an internal space capable of accommodating an energy storage battery; the chassis bracket 93 is connected to the steering wheel 92, and the chassis bracket 93 is connected to the wheel 91.
[0070] In order to better locate the target, in a preferred embodiment, the target detection module 31 includes a depth camera and a host computer module, the depth camera is used to collect image data and detect the target; the host computer module is connected to the depth camera to receive the image data transmitted by the depth camera. The depth camera includes a binocular camera.
[0071] The present invention also provides a target detection method for executing the above-mentioned automatic eggplant picking robot, the target detection method comprising:
[0072] Step S1, training an eggplant object detection model based on YOLO (You Only Look Once), and deploying the trained eggplant object detection model to a depth camera;
[0073] Step S2, using a depth camera to collect image data, calculating the relative position between the target eggplant and the depth camera, and calculating the relative position between the target eggplant and the electric cutting component 32;
[0074] Step S3, calculating the required movement distances of the screw assembly 6, the first arm 21, the second arm 22, the third arm 23, and the fourth arm 24 in the robotic arm assembly 2, and controlling the first drive motor 11, the second drive motor 12 and the screw assembly 6 to perform operations.
[0075] The present invention also discloses a target detection method for the above-mentioned automatic eggplant picking robot. The target detection method makes the entire robot more intelligent, can quickly adjust the working space, perform highly reliable and accurate identification and positioning measurement of the target eggplant rhizomes, and guide the robot to automatically cut the eggplant stalks based on machine vision, thereby realizing unmanned picking operations.
[0076] The YOLO framework stands out for its remarkable balance between speed and accuracy, and can quickly and reliably identify objects in images. In a more preferred embodiment of the present invention, the YOLO framework includes the YOLO v8 version. Among them, the target eggplant is located in the target detection method, and the main process of positioning includes:
[0077]
[0078] L is the projection of the line connecting the target eggplant and the depth camera on the horizontal plane, X is the distance from the origin in the horizontal coordinate of the depth camera coordinate system, Y is the distance from the origin in the vertical coordinate of the depth camera coordinate system, and Z is the depth value of the target eggplant measured by the depth camera.
[0079]
[0080] Among them, θ 1 The angle that the yaw motor needs to rotate.
[0081]
[0082] Among them, θ 2 is the rotation angle required by the second drive motor 12, θ 3 is the angle between the axis of the second arm 22 and the horizontal plane, L 1 is the length of the second arm 22 .
[0083] H=(Y+L 1 sinθ 3 -L 1 sin(θ 3 +θ 2 )
[0084] Where H is the distance the screw needs to move.
[0085] The present invention provides a machine-readable storage medium, on which instructions are stored, and the instructions are used to enable a machine to execute the target detection method of the present application as described above.
[0086] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.
[0087] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0088] In the several embodiments provided by the present invention, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.
[0089] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0090] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0091] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a mobile terminal, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk and other media that can store program codes.
[0092] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. An automatic eggplant picking robot, characterized in that: The automatic eggplant picking robot comprises: a picking actuator, a walking device, and a target detection module; The picking actuator comprises a driving motor assembly (1), a mechanical arm assembly (2), and an end effector assembly (3); one end of the mechanical arm assembly (2) is connected to the driving motor assembly (1), and the other end of the mechanical arm assembly (2) is connected to the end effector assembly (3); the mechanical arm assembly (2) comprises a first arm (21), a second arm (22), a third arm (23), and a fourth arm (24); the driving motor assembly (1) comprises a first driving motor (11) and a second driving motor (12); one end of the first arm (21) is connected to a power output shaft of the first driving motor (11), and the other end of the first arm (21) is connected to the second arm (22) via a first shaft (41). , the third arm (23); one end of the fourth arm (24) is connected to the end effector assembly (3), and the other end of the fourth arm (24) is connected to the second arm (22) and the third arm (23) through a second shaft (42); the power output shaft of the second drive motor (12) is fixedly provided with a first synchronous wheel (51), the first shaft (41) is provided with a second synchronous wheel (52) and a third synchronous wheel (53), the second shaft (42) is provided with a fourth synchronous wheel (54), the first synchronous wheel (51) and the second synchronous wheel (52) can be driven by a synchronous belt, and the third synchronous wheel (53) and the fourth synchronous wheel (54) are driven by a synchronous belt; The walking device comprises a chassis mechanism (4) and a wheel hub motor (5); the picking actuator is connected to the chassis mechanism (4) via a screw rod assembly (6); the chassis mechanism (4) is connected to the wheel hub motor (5); and the wheel hub motor (5) is used to drive a wheel (91); The target detection module (31) is fixedly arranged at one end of the mechanical arm component (2) connected to the end effector component (3); the target detection module (31) is used to identify ripe eggplant fruits.
2. The automatic eggplant picking robot according to claim 1, characterized in that: A first reduction mechanism (71) is provided between the power output shaft of the second drive motor (12) and the first synchronous wheel (51), and a second reduction mechanism (72) is provided between the power output shaft of the first drive motor (11) and the first arm (21); The first speed reduction mechanism (71) and the second speed reduction mechanism (72) are respectively detachably connected to a bracket (81), and the bracket (81) is fixedly connected to a screw nut (82).
3. The automatic eggplant picking robot according to claim 2, characterized in that: The first arm (21) can be connected to the output shaft of the second reduction mechanism (72) through a first connecting seat (61); a bearing seat (62) is provided between the second synchronous wheel (52) and the third synchronous wheel (53); the first arm (21) can be connected to the first shaft (41) through the bearing seat (62); the second arm (22) and the third arm (23) can be connected to the two ends of the first shaft (41) through a second connecting seat (64) and a third connecting seat (65) respectively; the second arm (22) and the third arm (23) can be connected to the second shaft (42) through the bearing seat (62) respectively.
4. The automatic eggplant picking robot according to claim 3, characterized in that: The fourth arm (24) can be connected to the second shaft (42) via a fourth connecting seat (63).
5. The automatic eggplant picking robot according to claim 1, characterized in that: The second synchronous wheel (52) and the third synchronous wheel (53) have the same size and the same number of teeth.
6. The automatic eggplant picking robot according to claim 1, characterized in that: The end effector assembly (3) comprises an electric cutting assembly (32), and the electric cutting assembly (32) is detachably connected to the fourth arm (24).
7. The automatic eggplant picking robot according to any one of claims 1 to 6, characterized in that: The chassis mechanism (4) comprises a chassis bracket (93), a steering wheel (92) and at least two wheels (91); the wheel (91) is coaxially provided with the wheel hub motor (5); the chassis bracket (93) is a plurality of connecting rods connected end to end to form an internal space capable of accommodating an energy storage battery; the chassis bracket (93) is connected to the steering wheel (92); and the chassis bracket (93) is connected to the wheel (91).
8. The automatic eggplant picking robot according to any one of claims 1 to 6, characterized in that: The target detection module (31) comprises a depth camera and a host computer module, wherein the depth camera is used to collect image data and detect targets; and the host computer module is connected to the depth camera and is used to receive image data transmitted by the depth camera.
9. A target detection method for executing the automatic eggplant picking robot according to any one of claims 1 to 8, characterized in that: The target detection method comprises: Step S1, training an eggplant target detection model based on YOLO, and deploying the trained eggplant target detection model to a depth camera; Step S2, using a depth camera to collect image data, calculating the relative position between the target eggplant and the depth camera, and calculating the relative position between the target eggplant and the electric cutting component (32); Step S3, calculating the required movement distances of the screw assembly (6), the first arm (21), the second arm (22), the third arm (23), and the fourth arm (24) in the mechanical arm assembly (2), and controlling the first drive motor (11), the second drive motor (12) and the screw assembly (6) to perform operations.
10. A machine-readable storage medium, characterized in that: The machine-readable storage medium stores instructions, which are used to enable the machine to execute the target detection method according to claim 9 of the present invention.
Citation Information
Patent Citations
Fruit picking robot manipulator
CN102806565A
Fruit and vegetable picking robot
CN109079744A
Multi-manipulator truss tomato picking robot and picking method thereof
CN110249785A
Ridge-crossing double-arm cooperative strawberry harvesting robot, control system and picking method
CN114679963A
Intelligent eggplant picking mechanical device
CN116830901A
