Lightweight three-degree-of-freedom fruit picking robot and picking method
By designing a lightweight three-degree-of-freedom fruit picking robot, using a binocular camera and a three-degree-of-freedom robotic arm to achieve accurate detection and picking of solanaceous vegetables, the problem of low recognition and judgment accuracy of picking robots in the existing technology has been solved, and high efficiency has been achieved. , precise vegetable picking, improving yield and quality.
Patent Information
- Application Number
- CN202411710581.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-27
- Publication Date
- 2025-01-24
AI Technical Summary
In the prior art, the identification and judgment accuracy of the picking robot of the eggplant fruit vegetable is low, and the accurate identification of vegetables cannot be achieved, resulting in low picking efficiency and easy to cause vegetable damage or mis-picking and leakage problems, which seriously restricts the picking robot in the eggplant fruit Applications in the field of vegetables.
A lightweight three-degree-of-freedom fruit picking robot is designed, which uses a drive chassis, a three-degree-of-freedom mechanical arm, an end actuator and a detection and positioning system. It uses binocular cameras and central control equipment to realize the detection and positioning of solanaceous vegetables. Using The three-degree-of-freedom robotic arm and end effector enable precise picking and collection of solanaceous vegetables.
It achieves high-precision detection and positioning of solanaceous vegetables, ensuring good picking consistency and high picking efficiency, avoiding vegetable damage and wrong picking and missed picking, and improving vegetable yield and quality.
Smart Images
Figure CN119347835A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent picking, and in particular to a lightweight three-degree-of-freedom fruit picking robot and a picking method. Background Art
[0002] Solanaceae vegetables such as tomatoes, eggplants, and peppers (including sweet peppers) are not only delicious, but also have many benefits for human health. For example, eggplants have the effects of lowering cholesterol and protecting cardiovascular system, tomatoes play a role in anti-oxidation and reducing cardiovascular disease, and peppers are rich in vitamin C and antioxidants, which help to enhance the immune system.
[0003] In recent years, with the continuous advancement of urbanization, a large number of rural laborers have poured into cities, resulting in a serious shortage of laborers for picking during the harvest season, which has increased agricultural labor costs, made it impossible to pick vegetables in time, and caused a decrease in vegetable production and quality. At the same time, most traditional artificial materials rely on the human eye to locate the position of mature eggplant vegetables. First, the experience requirements of the pickers are high (the pickers need to judge whether the eggplant vegetables are ripe or not), and the human influence factors are large. Second, during the long-term picking process, visual fatigue is prone to occur, which in turn affects the picking positioning accuracy and judgment accuracy of eggplant vegetables, further reducing the picking efficiency. Third, manual judgment is prone to problems such as wrong picking, missed picking, and mis-picking, thereby causing damage to eggplant vegetables. In addition, in order to cope with the problem of labor shortage, there is also a way of picking vegetables by picking robots in the prior art, but the existing picking robots have low recognition and judgment accuracy and cannot achieve accurate identification of vegetables, which not only greatly affects the picking efficiency, but also easily causes the picking mechanical arm to damage the vegetables or cause problems such as wrong picking and missed picking, which seriously restricts the application of picking robots in the field of eggplant vegetables. Summary of the invention
[0004] In view of the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a lightweight three-degree-of-freedom fruit picking robot, which can automatically detect and locate Solanaceae vegetables, thereby realizing precise clamping, shearing and collection of Solanaceae vegetables, with high positioning accuracy, good picking consistency and high picking efficiency, effectively solving the current problem of shortage of agricultural picking labor and improving vegetable yield and quality.
[0005] Another object of the present invention is to provide a picking method for the above-mentioned picking robot.
[0006] The purpose of the present invention is achieved through the following technical solutions: A lightweight three-degree-of-freedom fruit picking robot and a picking method, comprising a driving chassis, a three-degree-of-freedom mechanical arm, an end actuator and a detection and positioning system, wherein the driving chassis comprises a bearing base and a walking mechanism, and the walking mechanism is arranged at the bottom of the bearing base; the three-degree-of-freedom mechanical arm is arranged at the end face of the bearing base, and the end actuator is arranged at the end of the three-degree-of-freedom mechanical arm (i.e., the end away from the bearing base); the end actuator comprises a shell, a driving component, two clamping components, two groups of clamping and transporting components, two cutting wheels and a collecting box, the driving component is arranged inside the shell, and an opening is arranged at the end of the shell away from the driving component, and two clamping components are symmetrically distributed at the opening, the driving component and the clamping component are connected through the clamping and transporting component, the cutting wheel is arranged at the opening of the shell and is controlled by the driving component, and the collecting box is fixedly arranged at the bottom of the shell, and a notch is provided on its end face corresponding to the clamping component; the detection and positioning system comprises a binocular camera and a central control device, the binocular camera is installed on the bearing base and is relatively stationary with the three-degree-of-freedom mechanical arm, and the central control device is arranged on the bearing chassis.
[0007] Based on further optimization of the above scheme, the three-degree-of-freedom robotic arm includes a fixed bracket, a rotating disk, a horizontal guide rail, a vertical guide rail and a cantilever. The fixed bracket is fixedly arranged on the end face of the supporting base and the rotating disk is rotatably arranged on the end face of the fixed bracket. The end face of the rotating disk is fixedly arranged with a horizontal guide rail through a positioning bracket and a vertical guide rail is slidably arranged on the horizontal guide rail through a first sliding seat. The cantilever is slidably arranged on the vertical guide rail through a second sliding seat and the end of the cantilever away from the second sliding seat is fixedly connected to the side surface of the shell away from the opening.
[0008] Based on further optimization of the above scheme, the driving assembly includes a DC reduction motor, a boss gear, two driving shafts, a driving gear and a driven gear. The DC reduction motor is fixedly arranged on the end face of the shell close to the cantilever, and the output shaft of the DC reduction motor passes through the end face of the shell and is rotatably connected to the bottom surface of the inner cavity of the shell; the boss gear is fixedly sleeved on the outer wall of the output shaft of the DC reduction motor located in the inner cavity of the shell; the two driving shafts are respectively rotatably arranged in the inner cavity of the shell on the side of the boss gear away from the cantilever, and the two driving shafts are symmetrically arranged about the center line of the shell, and the outer walls of the two driving shafts are respectively fixedly sleeved on the driving gear and the driven gear, and the driving gear and the driven gear are meshed with each other, and the driving gear and the boss gear are meshed with each other.
[0009] Based on the further optimization of the above scheme, the clamping assembly includes an upper clamping block, a lower clamping block, a limiting boss and a fastening screw. The upper clamping block and the lower clamping block are combined to form a clamping claw with an internal hollow, and the clamping claw is clamped at the opening end of the corresponding shell. The clamping claws of the two groups of clamping assemblies are symmetrically arranged, and the limiting boss is fixedly arranged at the opening end of the shell corresponding to the clamping claw. The fastening screw spirally penetrates the limiting boss and is connected to the corresponding clamping claw; the clamping and transporting assembly includes a driving wheel, a driven shaft, a driven wheel, an intermediate shaft, an intermediate wheel and the same A step belt, a driving wheel is fixedly sleeved on the outer wall of the corresponding driving shaft and is located on the upper side of the corresponding driving gear or driven gear, a driven shaft is rotatably arranged in the inner cavity of one end of the clamping claw away from the limiting boss, and the outer wall of the driven shaft is fixedly sleeved on the driven wheel corresponding to the driving wheel, an intermediate shaft is rotatably arranged in the inner cavity of the clamping claw between the driven shaft and the limiting boss, and the outer wall of the intermediate shaft is fixedly sleeved on the intermediate wheel corresponding to the driven wheel, the outer diameter of the intermediate wheel is larger than the outer diameter of the driven wheel, the driving wheel and the corresponding driven wheel and the intermediate wheel are connected through a synchronous belt, and synchronous serrations are evenly distributed on the outer ring of the synchronous belt.
[0010] Based on further optimization of the above scheme, the cutting wheels are fixedly sleeved on the outer walls of the two driving shafts and located between the driving wheel and the driving gear or the driven gear. The outer diameter of the cutting wheel is larger than the outer diameter of the driving gear or the driven gear and the two cutting wheels are staggered.
[0011] A picking method for a lightweight three-degree-of-freedom fruit picking robot, comprising: Step 1: The walking mechanism drives the entire picking robot to move to the position to be picked, and the binocular camera is started to collect images in the planting field in real time; Step 2: The central control device detects the solanaceous vegetables according to the images collected by the binocular camera, and obtains the picking points and three-dimensional point clouds of the solanaceous vegetables in the camera coordinate system; Step 3: The central control device controls the three-degree-of-freedom robotic arm according to the picking point of the eggplant and solanaceous vegetables, so that the end actuator reaches the corresponding picking point; Step 4: Control the terminal actuator to pick and collect the nightshade vegetables; Step 5: All mechanisms are reset and the picking robot moves to the next point to carry out the next picking and collection of Solanaceae vegetables.
[0012] Based on the further optimization of the above scheme, the specific method of "the central control device detects solanaceous vegetables according to the images collected by the binocular camera" in step 2 is: First, convert the color space of the image acquired by the binocular camera: for V (Right now Value ):
[0013] forS (Right now Saturation ):
[0014] for H (Right now Hue ):
[0015] At the same time, if H<0, then H=H+360° ; Then, preset HSV The mask is created by the threshold range. If the image pixel after the color space conversion is within the threshold range, it is determined to meet the color condition and the pixel mask value is 255. If the image pixel after the color space conversion is not within the threshold range, it is determined to not meet the color condition and the pixel mask value is 0, and a mask image is generated. Finally, a morphological closing operation is performed on the target area obtained from the mask image, that is, the target area is first expanded and then eroded to obtain the outer contour of the Solanaceae vegetables.
[0016] Based on the further optimization of the above scheme, the specific method of "obtaining the picking points and three-dimensional point cloud of solanaceous vegetables in the camera coordinate system" in step 2 is: The light structure projection technology is used to emit infrared light with patterns, and the reflected infrared light is sent to the RealSense Vision D4 processor, which processes the depth value. Z ; Then, the Solanaceae vegetables are selected by the outer contour of the Solanaceae vegetables, and the Solanaceae vegetable boundary box and the coordinates of the upper left corner vertex corresponding to the boundary box are obtained. (x,y) , width (i.e. the distance from the left to the right of the bounding box) w , height (i.e. the distance from the top to the bottom of the bounding box) h , get the coordinates of the picking point through the bounding box (u,v) :
[0017] Record the pixel coordinates of all Solanaceae vegetable picking points in the image (u i ,v i ) , i=1,2,…,n ; Finally, the pixel coordinates of the picking points of solanaceous vegetables are converted into three-dimensional space coordinates （X C ,Y C ,Z C ） :
[0018] Where: (u 0 ,v 0 ) Indicates the pixel coordinates corresponding to the origin of the coordinate system; dx , dy Respectively represent the size of each pixel; fx , fy Respectively represent the focal length of the camera on the x-axis and y-axis; Z C Indicates the depth information of the corresponding picking point in the camera coordinate system, through the depth value Z get; The three-dimensional point cloud of all Solanaceae vegetables was obtained by transforming the pixel coordinates of all the recorded picking points.
[0019] Based on the further optimization of the above scheme, the step three is specifically as follows: According to the Cartesian coordinate system, a coordinate system is established with the center of the rotating disk as the base point. The three-degree-of-freedom robot arm is divided into the X-axis, i.e., the rotation joint, the Y-axis, i.e., the vertical translation joint, and the Z-axis, i.e., the lateral translation joint. The offset of the picking position point A of the end actuator relative to the base point is obtained, i.e., the distance of point A along the positive direction of the X-axis is R , along the positive direction of the Z axis is L ; Since the calculation of the joint angle and the horizontal arm extension are not affected by the plumb bob direction, the picking position point of the spatial coordinate system is A With picking point A’ 、 Right now （X C ,Y C ,Z C ） Projected on the XZ plane, obtain the rotation joint angles in different situations θ Horizontal arm extension △L : when X C ＞0、X C -R＞0 hour:
[0020] when X C ≥0、X C -R＜0 hour:
[0021] when X C＜0、X C -R＜0 hour:
[0022] The three-degree-of-freedom robotic arm achieves picking control through a linkage control method, specifically: first, the rotation joint and the vertical translation joint are controlled to operate, and when the end actuator reaches the height of the picking point, the lateral translation joint is started to extend so that the end actuator reaches the picking point.
[0023] Based on further optimization of the above scheme, when the three-degree-of-freedom robotic arm is reset in step five, the lateral translation joint is firstly started to be reset, and then the rotation joint and the vertical translation joint are started to be reset at the same time.
[0024] The following are the technical effects of the solution of the present invention: The present application uses an end actuator composed of a shell, a driving component, two clamping components, two groups of clamping and transporting components, two cutting wheels and a collecting box. The clamping component is used to achieve the segmentation of the target Solanaceae vegetables from the adjacent Solanaceae vegetables, and the clamping and transporting component is used to achieve the transportation and pulling of the Solanaceae vegetable picking part, so that the picking part is cut off by the two cutting wheels at the end during the transportation and movement, and the Solanaceae vegetables are collected in the collecting box. Through the picking method of the structure of the present application, firstly, the adjacent Solanaceae vegetables are avoided from being damaged or falling due to repeated clamping, moving, twisting and other actions of the end mechanism during the picking process; secondly, the effective cutting of Solanaceae vegetables is ensured, and the slender and soft branches of Solanaceae vegetables (i.e. Solanaceae vegetables) are avoided from being easily separated from the adjacent cluster branches; thirdly, the separation of the target branches and the main trunk can be achieved only by the transmission force, so as to avoid the problems of damage or displacement of the main trunk caused by strong pulling, thereby ensuring the subsequent positioning and picking accuracy and ensuring the smoothness of continuous picking. The three-degree-of-freedom robotic arm consisting of a fixed bracket, a rotating disk, a horizontal guide rail, a vertical guide rail and a cantilever can achieve smooth rotation and movement of the end actuator, effectively avoiding problems such as inaccurate control caused by shaking during the robotic arm controlling the end actuator, and damage to surrounding clusters of eggplants and solanaceous vegetables.
[0025] In addition, the present application can effectively avoid interference from environmental factors and adjacent clusters and accurately obtain the outline of the target Solanaceae vegetables through the conversion of color space in combination with threshold setting and morphological operations; and utilize methods such as obtaining picking points and converting coordinate systems, which is not only suitable for the three-degree-of-freedom robotic arm specific to the present application and ensures picking accuracy, but also effectively completes precise control to avoid problems such as wrong picking, missed picking, and over-picking, thereby achieving efficient picking of Solanaceae vegetables with good picking consistency, effectively improving the quality of picked Solanaceae vegetables and reducing the breakage rate during the picking process. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] Figure 1 Schematic diagram of the overall structure of the picking robot in an embodiment of the present invention.
[0027] Figure 2 Schematic diagram of the structure of the three-degree-of-freedom mechanical arm of the picking robot in an embodiment of the present invention.
[0028] Figure 3 Schematic diagram of the structure of the end effector of the picking robot in an embodiment of the present invention.
[0029] Figure 4 It is a bottom view of the end effector of the picking robot in the embodiment of the present invention.
[0030] Figure 5 Schematic diagram of the interior of the end effector of the picking robot in an embodiment of the present invention.
[0031] Figure 6 It is a schematic structural diagram of the clamping and transporting assembly of the end effector of the picking robot in an embodiment of the present invention.
[0032] Figure 7 This is a simplified model diagram of the three-degree-of-freedom mechanical arm of the picking robot in an embodiment of the present invention when picking.
[0033] Among them, 11, bearing base; 12, walking mechanism; 21, fixed bracket; 22, rotating disk; 220, positioning bracket; 23, horizontal guide rail; 230, first sliding seat; 24, vertical guide rail; 240, second sliding seat; 25, cantilever; 31, housing; 321, DC reduction motor; 322, boss gear; 323, driving shaft; 324, driving gear; 325, driven gear; 331, upper clamping block; 332, lower clamping block; 333, limiting boss; 334, fastening screw; 341, driving wheel; 342, driven shaft; 343, driven wheel; 344, intermediate shaft; 345, intermediate wheel; 346, synchronous belt; 3460, synchronous sawtooth; 35, cutting wheel; 36, collection box. DETAILED DESCRIPTION
[0034] The technical solutions in the embodiments of the present invention will be described clearly and completely 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, rather than all the embodiments.
[0035] Embodiment 1: A lightweight three-degree-of-freedom fruit picking robot comprises a driving chassis, a three-degree-of-freedom mechanical arm, an end actuator and a detection and positioning system. The driving chassis comprises a bearing base 11 and a walking mechanism 12. The walking mechanism 12 is arranged at the bottom of the bearing base 11 (such as Figure 1As shown, in this embodiment, the walking mechanism 12 is composed of four walking wheels and a walking bracket, and the four walking wheels are arranged at the bottom of the supporting base 11 through the walking bracket).
[0036] The three-degree-of-freedom mechanical arm is arranged on the end surface of the bearing base 11, and the three-degree-of-freedom mechanical arm includes a fixed bracket 21, a rotating disk 22, a horizontal guide rail 23, a vertical guide rail 24 and a cantilever 25. The fixed bracket 21 is fixedly arranged on the end surface of the bearing base 11 and the rotating disk 22 is rotatably arranged on the end surface of the fixed bracket 21 (such as Figure 1 As shown, the end surface of the rotating disk 22 is fixedly provided with a horizontal guide rail 23 by a positioning bracket 220, and a vertical guide rail 24 is slidably provided on the horizontal guide rail 23 by a first sliding seat 230 (as shown in FIG. Figure 2 As shown in the figure, the cantilever 25 is slidably arranged on the vertical guide rail 24 through the second sliding seat 240, and the end of the cantilever 25 away from the second sliding seat 240 is fixedly connected to the side of the housing 31 away from the opening (that is, the connection between the three-degree-of-freedom robot arm and the end actuator is realized). In addition, the rotation of the rotating disk 22 is controlled by the motor arranged on the fixed bracket 21, and the first sliding seat 230 realizes horizontal sliding on the horizontal guide rail 23 through the screw and motor arranged on the horizontal guide rail 23, thereby completing the horizontal sliding of the vertical guide rail 24, the cantilever 25 and the end actuator; the second sliding seat 240 realizes up and down lifting on the vertical guide rail 24 through the screw and motor arranged on the vertical guide rail 24, thereby completing the vertical lifting of the cantilever 25 and the end actuator. The end actuator includes a housing 31, a driving assembly, two clamping assemblies, two sets of clamping and transporting assemblies, two cutting wheels 35 and a collection box 36. The driving assembly is arranged inside the housing 31 and an opening is arranged at one end of the housing away from the driving assembly. The driving assembly includes a DC reduction motor 321, a boss gear 322, two driving shafts 323, a driving gear 324 and a driven gear 325. The DC reduction motor 321 is fixedly arranged on an end surface of the housing 31 close to the cantilever 25 (such as Figure 3 The output shaft of the DC reduction motor 321 passes through the end surface of the housing 31 and is rotatably connected to the bottom surface of the inner cavity of the housing 31; the boss gear 322 is fixedly sleeved on the outer wall of the output shaft of the DC reduction motor 321 located in the inner cavity of the housing 31; the two driving shafts 323 are rotatably arranged on the side of the boss gear 322 away from the cantilever 25 (i.e. Figure 5 The two driving shafts 323 are symmetrically arranged about the center line of the housing 31, and the outer walls of the two driving shafts 323 are respectively fixedly sleeved with a driving gear 324 and a driven gear 325, and the driving gear 324 and the driven gear 325 are meshed with each other (the outer diameters of the driving gear 324 and the driven gear 325 are the same), and the driving gear 324 and the boss gear 322 are meshed with each other (as shown in the upper left side of the figure). Figure 5 as shown).
[0037] Two clamping components are symmetrically distributed at the opening, and the clamping components include an upper clamping block 331, a lower clamping block 332, a limiting boss 333 and a fastening screw 334. The upper clamping block 331 and the lower clamping block 332 are combined to form a clamping claw with an internal hollow (combined Figure 3 , Figure 4 , Figure 5 As shown) and the clamping claws are clamped at the opening end of the corresponding housing 31, the clamping claws of the two sets of clamping assemblies are symmetrically arranged, the opening end of the housing 31 is fixedly provided with a limiting boss 333 corresponding to the clamping claws, and the fastening screw 334 spirally penetrates the limiting boss 333 and is connected to the corresponding clamping claw (as shown Figure 5 The driving assembly and the clamping assembly are connected by a clamping and transporting assembly, which includes a driving wheel 341, a driven shaft 342, a driven wheel 343, an intermediate shaft 344, an intermediate wheel 345 and a synchronous belt 346 (as shown); Figure 6 As shown in the figure, the driving wheel 341 is fixedly sleeved on the outer wall of the corresponding driving shaft 323 and is located on the upper side of the corresponding driving gear 324 or the driven gear 325 (at the same time, the outer diameter of the driving wheel 341 is smaller than the outer diameter of the driving gear 324 or the driven gear 325), the driven shaft 342 is rotatably arranged in the inner cavity of the end of the clamping claw away from the limiting boss 333, and the outer wall of the driven shaft 342 corresponds to the driving wheel 341 and is fixedly sleeved on the driven wheel 343, the intermediate shaft 344 is rotatably arranged in the inner cavity of the clamping claw between the driven shaft 342 and the limiting boss 333, and the outer wall of the intermediate shaft 344 corresponds to the driven wheel 343 and is fixedly sleeved on the intermediate wheel 345, the outer diameter of the intermediate wheel 345 is larger than the outer diameter of the driven wheel 343, the driving wheel 341 is connected to the corresponding driven wheel 343 and the intermediate wheel 345 by a synchronous belt 346, and the outer ring of the synchronous belt 346 is evenly distributed with synchronous saw teeth 3460 (as shown in the figure). Figure 3 , Figure 4 , Figure 5 As shown). The cutting wheel 35 is arranged at the opening of the housing 31 and is controlled by the driving assembly. Specifically, the cutting wheels 35 are respectively fixedly sleeved on the outer walls of the two driving shafts 323 and are located between the driving wheel 341 and the driving gear 324 or the driven gear 325. The outer diameter of the cutting wheel 35 is larger than the outer diameter of the driving gear 324 or the driven gear 325 and the two cutting wheels 35 are staggered. The collecting box 36 is fixedly arranged at the bottom of the housing 31 and a notch is provided on its end surface corresponding to the clamping assembly (as shown in FIG. Figure 4 as shown).
[0038] The detection and positioning system includes a binocular camera and a central control device. The binocular camera is installed on the supporting base 31 and is relatively stationary with the three-degree-of-freedom robot arm (such as Figure 1 As shown, in this embodiment, the binocular camera is installed at the top end of the vertical guide rail 24), and the central control equipment is set on the supporting chassis 31.
[0039] Embodiment 2: As another preferred embodiment of the present invention, in order to achieve continuous picking and improve picking efficiency, based on the scheme of Example 1, a collection net bag or a connecting pipe can be set at the bottom of the collection box 36, and the connecting pipe is connected to the loading box set on the end surface of the supporting base 11 (a sponge layer is evenly laid inside the loading box to avoid damage to the nightshade vegetables), thereby realizing the collection of the collected nightshade vegetables. At the same time, if a connecting pipe and a loading box are set, soft buffer blocks (such as rubber blocks or silicone blocks, etc.) are evenly arranged at the connection between the connecting pipe and the loading box to cushion the falling eggplant and fruit vegetables and prevent them from being damaged due to excessive gravitational acceleration.
[0040] Embodiment 3: A picking method using a lightweight three-degree-of-freedom fruit picking robot as described in Example 1 or Example 2 comprises: Step 1: The walking mechanism drives the entire picking robot to move to the position to be picked, and the binocular camera is started to collect images in the planting field in real time; Step 2: The central control device detects the solanaceous vegetables based on the images collected by the binocular camera, specifically: First, convert the color space of the image acquired by the binocular camera: for V (Right now Value ):
[0041] for S (Right now Saturation ):
[0042] for H (Right now Hue ):
[0043] At the same time, if H<0, then H=H+360° ; Make sure the output value is within the valid range, i.e. 0≤V≤1、0≤S≤1、0≤H≤360 °; Then, preset HSV The mask is created by the threshold range. If the image pixel after the color space conversion is within the threshold range, it is determined to meet the color condition and the pixel mask value is 255. If the image pixel after the color space conversion is not within the threshold range, it is determined to not meet the color condition and the pixel mask value is 0, and a mask image is generated. Finally, a morphological closing operation is performed on the target area obtained from the mask image, that is, the target area is first expanded and then eroded to obtain the outer contour of the Solanaceae vegetables.
[0044] And obtain the picking points and three-dimensional point clouds of solanaceous vegetables in the camera coordinate system, specifically: The light structure projection technology is used to emit infrared light with patterns, and the reflected infrared light is sent to the RealSense Vision D4 processor, which processes the depth value. Z ; Then, the Solanaceae vegetables are selected by the outer contour of the Solanaceae vegetables, and the Solanaceae vegetable boundary box and the coordinates of the upper left corner vertex corresponding to the boundary box are obtained. (x,y) , width (i.e. the distance from the left to the right of the bounding box) w , height (i.e. the distance from the top to the bottom of the bounding box) h , get the coordinates of the picking point through the bounding box (u,v) :
[0045] Record the pixel coordinates of all Solanaceae vegetable picking points in the image (u i ,v i ) , i=1,2,…,n ; Finally, the pixel coordinates of the picking points of solanaceous vegetables are converted into three-dimensional space coordinates （X C ,Y C ,Z C ） :
[0046] Where: (u 0 ,v 0 ) Indicates the pixel coordinates corresponding to the origin of the coordinate system; dx , dy Respectively represent the size of each pixel; fx , fy Respectively represent the focal length of the camera on the x-axis and y-axis; Z C Indicates the depth information of the corresponding picking point in the camera coordinate system, through the depth value Z get; The three-dimensional point cloud of all Solanaceae vegetables was obtained by transforming the pixel coordinates of all the recorded picking points.
[0047] Step 3: The central control device controls the three-degree-of-freedom robotic arm according to the picking point of eggplant and tomato vegetables (the central control device can use STM32F103ZE as the main control chip) to make the end actuator reach the corresponding picking point. The specific control method is: According to the Cartesian coordinate system, a coordinate system is established with the center of the rotating disk as the base point. The three-degree-of-freedom robot arm is divided into the X-axis, i.e., the rotation joint, the Y-axis, i.e., the vertical translation joint, and the Z-axis, i.e., the lateral translation joint. The offset of the picking position point A of the end actuator relative to the base point is obtained, i.e., the distance of point A along the positive direction of the X-axis is R , along the positive direction of the Z axis is L ; Since the calculation of the joint angle and the horizontal arm extension are not affected by the plumb bob direction, the picking position point of the spatial coordinate system is A With picking point A’ 、 Right now （X C ,Y C ,Z C ） Projected on the XZ plane, obtain the rotation joint angles in different situations θ Horizontal arm extension △L : when X C ＞0、X C -R＞0 hour:
[0048] when X C ≥0、X C -R＜0 hour:
[0049] when X C ＜0、X C -R＜0 hour:
[0050] The three-degree-of-freedom robotic arm achieves picking control through a linkage control method, specifically: first, the rotation joint and the vertical translation joint are controlled to operate, and when the end actuator reaches the height of the picking point, the lateral translation joint is started to extend so that the end actuator reaches the picking point.
[0051] Step 4: Control the terminal actuator to pick and collect the eggplant and fruit vegetables, specifically: start the DC reduction motor 321 to drive the boss gear 322 to rotate, the boss gear 322 drives the two driving shafts 323 to rotate simultaneously and in the opposite direction through the driving gear 324 and the driven gear 325, the two driving shafts 323 realize the simultaneous and opposite rotation of the two driving wheels 341, and then drive the two synchronous belts 346 to rotate simultaneously and in the opposite direction through the driven shaft 342, the driven wheel 343, the intermediate shaft 344, and the intermediate wheel 345. During the rotation of 46, the synchronous sawtooth 3460 clamps the picking point of the branch of the Solanaceae vegetable at the end of the clamping claw (i.e., the end away from the DC reduction motor 321) and synchronously pulls it toward the end close to the DC reduction motor 321; at the same time, the two driving shafts 323 synchronously realize the simultaneous and reverse rotation of the two cutting wheels 35. When the synchronous sawtooth 3460 pulls the picking point of the branch to the outer ring of the cutting wheel 35, the cutting wheel 35 cuts off the branch, and the Solanaceae vegetable after the branch is cut off falls into the collection box 36, completing the picking and collection.
[0052] Step 5: All mechanisms are reset and the picking robot moves to the next point. When the three-degree-of-freedom robotic arm is reset, the lateral translation joint is reset first, and then the rotation joint and the vertical translation joint are reset at the same time; and the next picking and collection of Solanaceae vegetables is carried out.
Claims
1. A lightweight three-degree-of-freedom fruit picking robot, characterized by: It includes a driving chassis, a three-degree-of-freedom robotic arm, an end actuator and a detection and positioning system. The driving chassis includes a bearing base and a walking mechanism, and the walking mechanism is arranged at the bottom of the bearing base; the three-degree-of-freedom robotic arm is arranged on the end face of the bearing base and the end actuator is arranged at the end of the three-degree-of-freedom robotic arm; the end actuator includes a shell, a driving component, two clamping components, two groups of clamping and transporting components, two cutting wheels and a collection box. The driving component is arranged inside the shell and an opening is arranged at one end of the shell away from the driving component, and two clamping components are symmetrically distributed at the opening, and the driving component and the clamping component are connected through the clamping and transporting component, the cutting wheel is arranged at the opening of the shell and is controlled by the driving component, and the collection box is fixedly arranged at the bottom of the shell and a slot is opened on its end face corresponding to the clamping component; the detection and positioning system includes a binocular camera and a central control device, the binocular camera is installed on the bearing base and is relatively stationary with the three-degree-of-freedom robotic arm, and the central control device is arranged on the bearing chassis.
2. A lightweight three-degree-of-freedom fruit picking robot and picking method according to claim 1, characterized in that: The three-degree-of-freedom robotic arm includes a fixed bracket, a rotating disk, a horizontal guide rail, a vertical guide rail and a cantilever. The fixed bracket is fixedly arranged on the end face of the supporting base and the rotating disk is rotatably arranged on the end face of the fixed bracket. The end face of the rotating disk is fixedly arranged with a horizontal guide rail through a positioning bracket and a vertical guide rail is slidably arranged on the horizontal guide rail through a first sliding seat. The cantilever is slidably arranged on the vertical guide rail through a second sliding seat and one end of the cantilever away from the second sliding seat is fixedly connected to a side surface of the shell away from the opening.
3. A lightweight three-degree-of-freedom fruit picking robot according to claim 1 or 2, characterized in that: The driving assembly includes a DC reduction motor, a boss gear, two driving shafts, a driving gear and a driven gear. The DC reduction motor is fixedly arranged on an end face of a shell close to the cantilever, and the output shaft of the DC reduction motor penetrates the end face of the shell and is rotatably connected to the bottom face of the inner cavity of the shell; the boss gear is fixedly sleeved on the outer wall of the output shaft of the DC reduction motor located in the inner cavity of the shell; the two driving shafts are respectively rotatably arranged in the inner cavity of the shell on the side of the boss gear away from the cantilever, and the two driving shafts are symmetrically arranged about the center line of the shell, and the outer walls of the two driving shafts are respectively fixedly sleeved on the driving gear and the driven gear, and the driving gear and the driven gear are meshed with each other, and the driving gear and the boss gear are meshed with each other.
4. A lightweight three-degree-of-freedom fruit picking robot according to claim 3, characterized in that: The clamping assembly includes an upper clamping block, a lower clamping block, a limiting boss and a fastening screw. The upper clamping block and the lower clamping block are combined to form a clamping claw with an internal hollow, and the clamping claw is clamped at the opening end of the corresponding shell. The clamping claws of the two groups of clamping assemblies are symmetrically arranged. The opening end of the shell corresponds to the clamping claw and is fixed with a limiting boss. The fastening screw spirally penetrates the limiting boss and is connected to the corresponding clamping claw. The clamping and transporting assembly includes a driving wheel, a driven shaft, a driven wheel, an intermediate shaft, an intermediate wheel and a synchronous belt. The driving wheel is fixed The driven shaft is rotatably arranged in the inner cavity of one end of the clamping claw away from the limiting boss, and the outer wall of the driven shaft is fixedly sleeved with the driven wheel corresponding to the driving wheel; an intermediate shaft is rotatably arranged in the inner cavity of the clamping claw between the driven shaft and the limiting boss, and the outer wall of the intermediate shaft is fixedly sleeved with the intermediate wheel corresponding to the driven wheel; the outer diameter of the intermediate wheel is larger than the outer diameter of the driven wheel; the driving wheel and the corresponding driven wheel and the intermediate wheel are connected through a synchronous belt, and synchronous serrations are evenly distributed on the outer ring of the synchronous belt.
5. A picking method using a lightweight three-degree-of-freedom fruit picking robot as claimed in any one of claims 1 to 4, characterized in that: include: Step 1: The walking mechanism drives the entire picking robot to move to the position to be picked, and the binocular camera is started to collect images in the planting field in real time; Step 2: The central control device detects the solanaceous vegetables according to the images collected by the binocular camera, and obtains the picking points and three-dimensional point clouds of the solanaceous vegetables in the camera coordinate system; Step 3: The central control device controls the three-degree-of-freedom robotic arm according to the picking point of the eggplant and solanaceous vegetables, so that the end actuator reaches the corresponding picking point; Step 4: Control the terminal actuator to pick and collect the nightshade vegetables; Step 5: All mechanisms are reset and the picking robot moves to the next point to carry out the next picking and collection of Solanaceae vegetables.
6. A picking method for a lightweight three-degree-of-freedom fruit picking robot according to claim 5, characterized in that: The specific method of "the central control device detects Solanaceae vegetables according to the images collected by the binocular camera" in step 2 is: First, convert the color space of the image acquired by the binocular camera: for V : for S : for H : At the same time, if H<0, then H=H+360° ; Then, preset HSV The mask is created by the threshold range. If the image pixel after the color space conversion is within the threshold range, it is determined to meet the color condition and the pixel mask value is 255. If the image pixel after the color space conversion is not within the threshold range, it is determined to not meet the color condition and the pixel mask value is 0, and a mask image is generated. Finally, a morphological closing operation is performed on the target area obtained from the mask image, that is, the target area is first expanded and then eroded to obtain the outer contour of the Solanaceae vegetables.
7. The picking method of a lightweight three-degree-of-freedom fruit picking robot according to claim 5, characterized in that: The specific method of "obtaining the picking points and three-dimensional point cloud of solanaceous vegetables in the camera coordinate system" in step 2 is: The light structure projection technology is used to emit infrared light with patterns, and the reflected infrared light is sent to the RealSense Vision D4 processor, which processes the depth value. Z ; Then, the Solanaceae vegetables are selected by the outer contour of the Solanaceae vegetables, and the Solanaceae vegetable boundary box and the coordinates of the upper left corner vertex corresponding to the boundary box are obtained. (x,y) ,width w ,high h , get the coordinates of the picking point through the bounding box (u,v) : Record the pixel coordinates of all Solanaceae vegetable picking points in the image (u i ,v i ) , i=1,2,…,n ; Finally, the pixel coordinates of the picking points of solanaceous vegetables are converted into three-dimensional space coordinates （X C ,Y C ,Z C ） : Where: (u 0 ,v 0 ) Indicates the pixel coordinates corresponding to the origin of the coordinate system; dx , dy Respectively represent the size of each pixel; fx , fy Respectively represent the focal length of the camera on the x-axis and y-axis; Z C Indicates the depth information of the corresponding picking point in the camera coordinate system, through the depth value Z get; The three-dimensional point cloud of all Solanaceae vegetables was obtained by transforming the pixel coordinates of all the recorded picking points.
8. The picking method of a lightweight three-degree-of-freedom fruit picking robot according to claim 7, characterized in that: The step three is specifically as follows: According to the Cartesian coordinate system, a coordinate system is established with the center of the rotating disk as the base point. The three-degree-of-freedom robot arm is divided into the X-axis, i.e., the rotation joint, the Y-axis, i.e., the vertical translation joint, and the Z-axis, i.e., the lateral translation joint. The offset of the picking position point A of the end actuator relative to the base point is obtained, i.e., the distance of point A along the positive direction of the X-axis is R , along the positive direction of the Z axis is L ; Since the calculation of the joint angle and the horizontal arm extension are not affected by the plumb bob direction, the picking position point of the spatial coordinate system is A With picking point A’ 、 Right now （X C ,Y C ,Z C ） Projected on the XZ plane, obtain the rotation joint angles in different situations θ Horizontal arm extension △L : when X C ＞0、X C -R＞0 hour: when X C ≥0、X C -R＜0 hour: when X C ＜0、X C -R＜0 hour: The three-degree-of-freedom robotic arm achieves picking control through a linkage control method, specifically: first, the rotation joint and the vertical translation joint are controlled to operate, and when the end actuator reaches the height of the picking point, the lateral translation joint is started to extend so that the end actuator reaches the picking point.