Pineapple multi-arm row-controlled harvesting robot and picking method thereof
By designing a pineapple multi-arm row harvesting robot, the automatic picking of pineapple fruits is achieved using binocular vision depth cameras and picking robots, the problem of low picking efficiency in the existing technology is solved and the degree of automation and picking efficiency is improved.
Patent Information
- Application Number
- CN202510318964.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-17
- Publication Date
- 2025-06-03
AI Technical Summary
The existing pineapple picking equipment has low automation, resulting in low picking efficiency and cannot meet the development needs of the modern pineapple industry.
A pineapple multi-arm pair harvesting robot is designed, including a chassis, a pineapple fruit collection frame and multiple picking robots. The position of the pineapple fruit is identified through a binocular visual depth camera, and the picking robot arm controls the picking end effector for picking and collecting.
It realizes automatic picking of pineapple fruits, improves picking efficiency, has a high degree of automation, adapts to pineapple plants of different heights and row spacing, reduces damage to the plants, and extends the sustainable harvesting cycle of the pineapple garden.
Smart Images

Figure CN120077848A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of agricultural machinery, and in particular to a pineapple multi-arm row harvesting robot and a picking method thereof. Background Art
[0002] Pineapple is one of the special tropical fruits. It is mainly grown in tropical and subtropical regions. It is widely planted and is an important pillar industry in tropical and subtropical regions. At present, pineapple picking still relies on manual harvesting. Because the outer skin of pineapples has obvious grid-like textures, and the upper crown buds and leaf edges have many thorns and other physical characteristics, manual picking requires wearing protective gear to avoid scratches. There are problems such as difficulty in picking and low operating efficiency, which cannot meet the development needs of the modern pineapple industry.
[0003] In order to solve the above problems, there are many existing pineapple picking equipments. Most of the existing pineapple picking equipments are hand-held portable and semi-automatic picking equipments, which still need manual assistance and have low automation, resulting in that the speed of pineapple picking operation is still relatively low. Summary of the invention
[0004] The purpose of the present invention is to overcome the above-mentioned problems and provide a pineapple multi-arm row harvesting robot, which can automatically pick pineapple fruits with a high degree of automation and can greatly improve the picking efficiency of pineapple fruits.
[0005] Another object of the present invention is to provide a pineapple picking method using a multi-arm row-by-row harvesting robot.
[0006] The purpose of the present invention is achieved through the following technical solutions:
[0007] A pineapple multi-arm row-by-row harvesting robot comprises a chassis, a pineapple fruit collecting frame arranged on the chassis, and a plurality of picking manipulators; wherein the plurality of picking manipulators are arranged at the front end of the chassis; during picking, each picking manipulator is arranged in one-to-one correspondence with each row of pineapple plants; the picking manipulator comprises a picking manipulator arm, a picking end effector arranged at the end of the picking manipulator arm, and a binocular vision depth camera.
[0008] The working principle of the above-mentioned pineapple multi-arm row harvesting robot is:
[0009] The position of pineapple fruits can be identified by the binocular vision depth camera, and the picking robot arm controls the movement of the picking end effector. The picking end effector will pick the pineapple fruits and put them into the pineapple fruit collection frame for collection. Each picking robot is set one by one corresponding to each row of pineapple plants. Multiple picking robots can realize the picking of pineapple fruits on multiple rows of pineapple plants, and the picking efficiency is high.
[0010] A preferred embodiment of the present invention, wherein the chassis is a gantry chassis, including a gantry and a crawler mechanism provided at the lower end of the gantry. A height adjustment mechanism for adjusting the height of the gantry is provided between the crawler mechanism and the gantry; the picking manipulator is provided on the gantry. By providing the gantry, it is convenient for the harvesting robot to travel in the pineapple field. The gantry can avoid pineapple plants. The height of the gantry can be adjusted through the height adjustment mechanism, and then the height of the picking manipulator can be adjusted, so that the harvesting robot can adapt to pineapple plants of different heights.
[0011] Preferably, a robotic arm mounting platform is provided on the gantry, and the picking robotic arm is provided on the robotic arm mounting platform. By providing the robotic arm mounting platform, it is convenient for the installation of the picking manipulator.
[0012] Preferably, the gantry includes a first frame body and a second frame body. A plug rod is provided on the first frame body, and a slot is provided on the second frame body. The plug rod is connected with the slot in a matching manner, and a locking mechanism is provided between the plug rod and the slot. The crawler mechanism includes two sets of crawler assemblies. One set of crawler assemblies is provided at the lower end of the first frame body, and the other set of crawler assemblies is provided at the lower end of the second frame body. By providing the plug rod and the slot, the distance between the first frame body and the second frame body can be adjusted, and then the distance between the two sets of crawler assemblies can be adjusted, so that the harvesting robot can adapt to pineapple plants with different row spacings. Through the locking mechanism, the first frame body and the second frame body can be locked, so that the first frame body and the second frame body are fixedly connected to ensure the stability of picking.
[0013] Preferably, the picking end effector includes a mounting body, a plurality of finger mechanisms provided on the mounting body, and a picking driving mechanism for driving each finger mechanism to move; wherein,
[0014] The finger mechanism includes a proximal finger and a distal finger. One end of the proximal finger is hinged to the mounting body, and the other end of the proximal finger is hinged to one end of the distal finger; torsion springs and pulleys are provided between the proximal finger and the mounting body and between the proximal finger and the distal finger;
[0015] The picking driving mechanism includes a first driving mechanism for driving the proximal fingers to move and a second driving mechanism for driving the distal fingers to move; both the first driving mechanism and the second driving mechanism include a servo motor, a steering wheel and a driving rope arranged on the mounting body; the steering wheel is arranged on the servo motor, one end of the driving rope of the first driving mechanism is connected to the steering wheel, and the other end is connected to the proximal finger after passing through a pulley; one end of the driving rope of the second driving mechanism is connected to the steering wheel, and the other end is connected to the distal finger after passing through a pulley. In the above structure, during picking, the servo motor of the first driving mechanism drives the steering wheel thereon to rotate, drives the driving rope to move, tightens the driving rope, and then drives the proximal fingers to move closer inward. In this process, the elastic force of the torsion spring is overcome. When the proximal fingers move closer inward, the distal fingers will be driven to move synchronously; the servo motor of the second driving mechanism drives the steering wheel thereon to rotate, drives the driving rope to move, tightens the driving rope, and then drives the distal fingers to move closer inward. In this process, the elastic force of the torsion spring is overcome. When the distal fingers move closer inward, multiple finger mechanisms synchronously envelope the pineapple fruit, and the picking robotic arm rotates to apply a breaking torque to complete the breaking of the pineapple fruit.
[0016] Preferably, the harvesting robot further includes a main control module and a pressure detection module provided on each picking manipulator. The main control module is connected to the pressure detection module. The pressure detection module includes a plurality of two-point distributed thin-film pressure sensors. The two-point distributed thin-film pressure sensors are arranged in one-to-one correspondence with the finger mechanisms. Each two-point distributed thin-film pressure sensor includes a thin-film pressure sensor provided in the inner flexible silica gel of the proximal finger and the inner flexible silica gel of the distal finger. The process of clamping and picking the pineapple fruit is as follows: First, move the picking end effector close to the pineapple fruit to be picked. When the installation body contacts the surface of the pineapple fruit, stop moving. At this time, the center of the pineapple fruit contacts the palm position on the installation body. Then, the servo motor of the first driving mechanism drives the output shaft thereon to rotate. The output shaft drives the steering wheel connected thereto to rotate. The driving rope connected to the steering wheel and the proximal finger drives the proximal finger to move closer to the inside under the action of tension, and the distal finger will also move synchronously until the proximal finger clamps the pineapple fruit. The thin-film pressure sensor in the inner flexible silica gel of the proximal finger receives the force signal. After transmitting the force signal to the main control module, the main control module controls the output shaft on the servo motor of the first driving mechanism to stop rotating. Subsequently, the servo motor of the second driving mechanism drives the output shaft thereon to rotate. The output shaft drives the steering wheel connected thereto to rotate. The driving rope connected to the steering wheel and the distal finger drives the distal finger to move closer to the inside under the action of tension until the distal finger clamps the pineapple fruit. The thin-film pressure sensor in the inner flexible silica gel of the distal finger receives the force signal. After transmitting the force signal to the main control module, the main control module controls the output shaft on the servo motor of the second driving mechanism to stop rotating. Thus, the clamping action is completed. The inner surfaces of the proximal finger and the distal finger are provided with flexible silica gel, which can increase the contact area with the pineapple fruit, increase friction and reduce mechanical damage to the pineapple fruit. Then, rotate the picking manipulator arm and simulate the stem-breaking picking method of manual picking to break the pineapple fruit off the stem. Finally, the picking end effector is controlled by the picking manipulator arm to transport the pineapple fruit to a designated position (pineapple fruit collection box). The output shafts on the servo motors of the first driving mechanism and the second driving mechanism drive the steering wheels to rotate in the reverse direction. Under the elastic force of the torsion springs between the proximal finger and the installation body and between the proximal finger and the distal finger, the finger mechanism opens and resets to the initial position, and the pineapple fruit falls to the collection point (pineapple fruit collection box). Thus, the pineapple picking work is completed.
[0017] Preferably, the proximal finger and the distal finger are connected by a pivot pin, and the proximal finger and the mounting body are also connected by a pivot pin. The torsion spring and the pulley are coaxially arranged with the pivot pin. The number of torsion springs between the proximal finger and the distal finger and between the proximal finger and the mounting body is two. Installation grooves for accommodating the torsion springs are provided on both sides of the pulley. Wiring hole grooves are provided on both sides of the proximal finger, and the circuit of the thin-film pressure sensor is led out through the wiring hole grooves. The wiring hole grooves are provided for the circuit of the thin-film pressure sensor to be orderly led out from the inside of the picking end effector, avoiding loosening or winding of the circuit when the picking end effector works. The installation grooves are provided to facilitate the installation of the torsion springs and make the structure very compact.
[0018] Preferably, a camera bracket is provided on the mounting body, the binocular vision depth camera is installed on the camera bracket, and the end of the picking robotic arm is connected to the camera bracket. By providing the camera bracket, the connection between the picking end effector and the picking robotic arm can be realized, and at the same time, the binocular vision depth camera can be installed.
[0019] Preferably, the harvesting robot further includes a power module. The main control module includes a single-chip microcomputer chip. The thin-film pressure sensor is connected to the single-chip microcomputer chip through a Bluetooth module. The servo is connected to the single-chip microcomputer chip. The power module includes a 24V switching power supply. The switching power supply converts 220V alternating current into 24V direct current to supply power to the single-chip microcomputer chip. The single-chip microcomputer chip converts 24V direct current into 5V direct current to supply power to the servo.
[0020] A picking method of a pineapple multi-arm parallel-row harvesting robot. The number of picking manipulators is two. The picking method includes the following steps:
[0021] S1. Use the YOLOv8 object detection algorithm to identify pineapple fruits on the RGB color image, and synchronously obtain the corresponding depth information through the binocular vision depth camera. Combine point cloud computing to obtain the position coordinates of the pineapple fruits in the three-dimensional space. These position coordinates are the coordinates in the camera coordinate system. Then, through the hand-eye calibration method, coordinate transformation is performed to convert the three-dimensional coordinates of the pineapple fruits from the camera coordinate system to the robotic arm base coordinate system. The robotic arm base coordinate system is the coordinate system of the base of the picking robotic arm.
[0022] S2. Layout optimization of the double picking robotic arms: The double picking robotic arms are installed on the same side. By establishing an optimization model for the working space of the two arms and taking the maximum picking coverage rate and the minimum mutual interference of the picking robotic arms as the optimization objectives, the particle swarm optimization algorithm is used to solve the optimal installation position, angle, and forward step distance of the picking robotic arms.
[0023] Motion Planning and Obstacle Avoidance: A path optimization algorithm based on the artificial potential field method and the rapidly-exploring random tree is used for generating obstacle avoidance trajectories and local path correction respectively. Combining with the dynamic window algorithm and the real-time motion state of the picking robotic arm, the motion path and smoothness of the picking end effector are adjusted to ensure flexible adjustment under different pineapple plant distributions;
[0024] Parallel Picking Strategy: The genetic algorithm is used to optimize the task assignment, enabling the left and right picking robotic arms to pick alternately in time sequence. Combining with time scheduling optimization control, the picking time windows of the left and right picking robotic arms are adjusted;
[0025] S3. According to the three-dimensional coordinates of the pineapple fruit in the base coordinate system of the robotic arm, the picking robotic arm approaches the target pineapple fruit along the optimized trajectory and controls the picking end effector to perform the picking operation; the picking end effector performs a clamping action. After complete clamping, the rotation of the picking robotic arm applies a breaking moment to separate the fruit stalk from the pineapple fruit;
[0026] S4. The picking robotic arm places the picked pineapple fruit into the pineapple fruit collection box on the chassis;
[0027] S5. After placing the pineapple fruit, each picking robotic arm and the picking end effector are reset respectively to prepare for the next picking task;
[0028] S6. Repeat steps S2 - S5 to pick the pineapple fruits in the current planting row of the pineapple plants until all the pineapple fruits in the current planting row of the pineapple plants are picked.
[0029] The present invention has the following beneficial effects compared with the prior art:
[0030] 1. In the multi-arm row-by-row harvesting robot for pineapples of the present invention, each picking manipulator is set corresponding to each row of pineapple plants, and multiple picking manipulators can pick the pineapple fruits on multiple rows of pineapple plants, which can greatly improve the picking efficiency of pineapple fruits.
[0031] 2. In the multi-arm row-by-row harvesting robot for pineapples of the present invention, the position of the pineapple fruit can be recognized by the binocular vision depth camera to realize the automatic recognition of the pineapples to be picked in the current row. By controlling the picking end effector to move through the picking robotic arm, the picking end effector is automatically controlled to move to the target point to be picked, and the picking end effector will pick the pineapple fruit. Each picking robotic arm corresponds to picking one row of pineapples, with good coherence and high efficiency. After picking the pineapple fruit, it is put into the pineapple fruit collection box for collection, with a high degree of automation, further improving the picking efficiency of pineapple fruits.
[0032] 3. In the pineapple multi-arm row-by-row harvesting robot of the present invention, the proximal fingers and the distal fingers are arranged in a profiling arc shape, which can better adapt to the shape of the pineapple fruit, reduce the mechanical pressure concentration during the grasping process, improve the stability and adaptability of grasping, and is especially suitable for the picking of targets with complex surfaces.
[0033] 4. In the pineapple multi-arm row-by-row harvesting robot of the present invention, a driving rope is used for driving, replacing the traditional rigid connecting rod or hydraulic drive, significantly reducing the overall weight of the picking manipulator. At the same time, the rope drive component is simple and has a low maintenance cost, and can achieve efficient picking while ensuring sufficient driving force.
[0034] 5. In the pineapple multi-arm row-by-row harvesting robot of the present invention, the picking end effector adopts a full drive mode, controls each joint through independent servos and driving ropes, adapts to the size of the pineapple fruit, realizes precise control of the movement of the finger mechanism, and applies a rotational torque through the sixth axis of the picking robotic arm to simulate the manual picking method. After clamping the pineapple fruit, a certain torsional moment is applied along the fruit stem, so that the fruit stem gradually breaks at the stress point, realizing the lossless separation of the pineapple fruit from the pineapple plant; this method does not require an additional cutting device, can reduce the fruit stalk residue, improve the integrity of the pineapple fruit, ensure that the quality of the picked pineapple meets the market standards, and at the same time reduce the damage to the plant during the picking process and extend the sustainable harvesting cycle of the pineapple orchard.
[0035] 6. In the pineapple multi-arm row-by-row harvesting robot of the present invention, multiple picking robotic arms are installed in parallel on the same side, enabling the harvesting robot to efficiently pick on one side of the pineapple planting row, and through a task allocation strategy based on the genetic algorithm, reasonably allocate picking tasks, avoid interference between the double picking robotic arms, and at the same time ensure staggered operation to improve the operation efficiency. Compared with the single picking robotic arm scheme, the double picking robotic arms can efficiently cover the target area, reduce the time for the single picking robotic arm to move back and forth, and increase the number of pickings per unit time. Description of the Drawings
[0036] Figure 1 It is a three-dimensional structural schematic diagram of the first specific embodiment of a pineapple multi-arm row-by-row harvesting robot of the present invention during operation.
[0037] Figure 2 It is a three-dimensional structural schematic diagram of a pineapple multi-arm row-by-row harvesting robot of the present invention.
[0038] Figure 3 It is a front view of a pineapple multi-arm row-by-row harvesting robot of the present invention.
[0039] Figure 4 It is a three-dimensional structural schematic diagram of the picking manipulator of the present invention.
[0040] Figure 5Schematic diagram of the installation of the picking end effector and the binocular vision depth camera in the present invention.
[0041] Figure 6 Schematic three-dimensional structure diagram of the picking end effector in the present invention.
[0042] Figure 7 Top view of the picking end effector in the present invention.
[0043] Figure 8 Schematic three-dimensional structure diagram of the finger mechanism in the present invention.
[0044] Figure 9 Schematic diagram of the installation of the torsion spring and the pulley in the present invention.
[0045] Figure 10 Schematic three-dimensional structure diagram of the chassis of the second specific embodiment of the pineapple multi-arm in-row harvesting robot in the present invention.
[0046] Figure 11 Front view of the chassis of the second specific embodiment of the pineapple multi-arm in-row harvesting robot in the present invention. Specific embodiments
[0047] In order to enable those skilled in the art to well understand the technical solution of the present invention, the present invention will be further described below in conjunction with embodiments and drawings, but the embodiments of the present invention are not limited thereto.
[0048] Embodiment 1
[0049] See Figures 1-4 , this embodiment discloses a pineapple multi-arm in-row harvesting robot, including a chassis 1, a pineapple fruit collection box 2 arranged on the chassis 1, and a plurality of picking manipulators; wherein, a plurality of picking manipulators are arranged at the front end of the chassis 1; during picking, each picking manipulator is arranged corresponding to each row of pineapple plants; the picking manipulator includes a picking robotic arm 3, a picking end effector 4 arranged at the end of the picking robotic arm 3, and a binocular vision depth camera 5. Multi-arm in-row means that a plurality of picking robotic arms 3 are arranged corresponding to multiple rows of pineapple plants.
[0050] See Figures 1-4 , the binocular vision depth camera 5 constitutes a vision system. The pineapple fruit collection box 2 is arranged behind the relative position of the picking manipulator, and the picking robotic arm 3 is a six-degree-of-freedom picking robotic arm 3 with strong flexibility.
[0051] See Figures 1-4, the chassis 1 is a gantry chassis, including a gantry and a crawler mechanism arranged at the lower end of the gantry, and the picking manipulator is arranged on the gantry. By setting the gantry, it is convenient for the harvesting robot to travel in the pineapple field, and the gantry can avoid pineapple plants. A height adjustment mechanism for adjusting the height of the gantry is arranged between the crawler mechanism and the gantry.
[0052] See Figures 1-4 , a robotic arm mounting platform 7 is arranged on the gantry, and the picking robotic arm 3 is arranged on the robotic arm mounting platform 7. By setting the robotic arm mounting platform 7, it is convenient for the installation of the picking manipulator. The picking robotic arm 3 is arranged on a plane perpendicular to the advancing direction of the chassis 1 on the robotic arm mounting platform 7. The picking robotic arm 3 is a six-degree-of-freedom picking robotic arm, and the base of the six-degree-of-freedom picking robotic arm is fixedly connected to the robotic arm mounting platform 7 through threaded fasteners.
[0053] See Figures 1-9 , the picking end effector 4 includes a mounting body 41, a plurality of finger mechanisms 42 arranged on the mounting body 41, and a picking drive mechanism 43 for driving each finger mechanism 42 to move, that is, the number of the picking drive mechanisms 43 is the same as that of the finger mechanisms 42, and the picking drive mechanisms 43 are connected to the finger mechanisms 42 in a one-to-one correspondence.
[0054] See Figures 1-9 , the finger mechanism 42 includes a proximal finger 421 and a distal finger 422. One end of the proximal finger 421 is hinged to the mounting body 41, and the other end of the proximal finger 421 is hinged to one end of the distal finger 422; a torsion spring 423 and a pulley 424 are arranged between the proximal finger 421 and the mounting body 41 and between the proximal finger 421 and the distal finger 422. The mounting body 41 has a hollow hexagonal columnar structure. The number of finger mechanisms 42 is three, and the number of picking drive mechanisms 43 is three. The three finger mechanisms 42 are evenly distributed in an equilateral triangle manner, that is, evenly distributed along the circumferential direction, and are installed on the top edge of the mounting body 41; the picking drive mechanisms 43 are installed at the outer wall near the top of the hollow hexagonal columnar structure. Three groups of mounting holes are arranged at the top of the mounting body 41. The three groups of mounting holes are evenly distributed along the three sides of the hexagon, and the included angle between adjacent two groups of mounting holes is 120°, forming an equilateral triangle distribution structure. The internal of the mounting holes is equipped with a threaded structure for installing the finger mechanisms 42; three wiring holes are arranged on the side wall of the mounting body 41; the wiring holes are evenly arranged along the three sides of the hexagon respectively, and one wiring hole is arranged at the central position of each side. The wiring holes are all processed with rounded corners.
[0055] See Figures 1-9, the picking driving mechanism 43 includes a first driving mechanism for driving the movement of the proximal finger 421 and a second driving mechanism for driving the movement of the distal finger 422; both the first driving mechanism and the second driving mechanism include a servo 431, a servo disc 432 and a driving rope arranged on the mounting body 41; the servo disc 432 is arranged on the servo 431, one end of the driving rope of the first driving mechanism is connected to the servo disc 432, and the other end is connected to the proximal finger 421 after passing through the pulley 424; one end of the driving rope of the second driving mechanism is connected to the servo disc 432, and the other end is connected to the distal finger 422 after passing through the pulley 424. In the above structure, during picking, the servo 431 of the first driving mechanism drives the servo disc 432 thereon to rotate, drives the driving rope to move, tightens the driving rope, and further drives the proximal finger 421 to move inward. In this process, the elastic force of the torsion spring 423 will be overcome. When the proximal finger 421 moves inward, it will drive the distal finger 422 to move synchronously; the servo 431 of the second driving mechanism drives the servo disc 432 thereon to rotate, drives the driving rope to move, tightens the driving rope, and further drives the distal finger 422 to move inward. In this process, the elastic force of the torsion spring 423 will be overcome. When the distal finger 422 moves inward, the plurality of finger mechanisms 42 synchronously envelope the pineapple fruit 9, and the picking robotic arm 3 rotates to apply a breaking moment to complete the breaking of the pineapple fruit 9.
[0056] See Figures 1-9 , the servo 431 is installed on the side wall of the mounting body 41 through a servo bracket 433; at three adjacent sides below the top edge of the mounting body 41, each servo bracket 433 corresponds to the mounting hole position of a finger mechanism 42 and is evenly distributed in an equilateral triangle manner; eight screw holes are opened inside the mounting groove of the servo bracket 433 for installing the servo 431 of the first driving mechanism and the servo 431 of the second driving mechanism; the position of the servo bracket 433 is set at a certain vertical distance from the top mounting hole of the mounting body 41 to ensure that the axes of the output shafts of the servo 431 of the first driving mechanism and the servo 431 of the second driving mechanism are in the same horizontal plane as the axis of the pulley 424 guiding area at the top of the mounting body 41, so that the driving ropes pulled out from the output shafts of the servo 431 of the first driving mechanism and the servo 431 of the second driving mechanism can enter the pulley 424 guiding area horizontally without generating vertical offsets.
[0057] See Figures 1-9 , the servo 431 of the first driving mechanism and the servo 431 of the second driving mechanism are fixedly installed in the mounting groove of the servo bracket 433 through the bottom mounting holes with screws; the output shafts of the servo 431 of the first driving mechanism and the servo 431 of the second driving mechanism are connected to the servo disc 432 through a toothed circumferential card slot. The servo disc 432 is coaxial with the output shafts of the respective servos 431 and is at the same horizontal plane as the pulley 424 at the top of the mounting body 41.
[0058] See Figures 1-9 , the harvesting robot further includes a main control module and a pressure detection module disposed on each picking manipulator. The main control module is connected to the pressure detection module. The pressure detection module includes a plurality of two-point distributed thin-film pressure sensors. The two-point distributed thin-film pressure sensors are arranged in one-to-one correspondence with the finger mechanisms 42. Each two-point distributed thin-film pressure sensor includes a thin-film pressure sensor disposed in the inner flexible silica gel of the proximal finger 421 and the inner flexible silica gel of the distal finger 422. The process of clamping and picking the pineapple fruit 9 is as follows: First, move the picking end effector 4 close to the pineapple fruit 9 to be picked. When the mounting body 41 contacts the epidermis of the pineapple fruit, stop moving. At this time, the center of the pineapple fruit contacts the palm position on the mounting body 41. Then, the servo motor 431 of the first driving mechanism drives the output shaft thereon to rotate. The output shaft drives the steering wheel 432 connected thereto to rotate. The driving rope connected to the steering wheel 432 and the proximal finger 421 drives the proximal finger 421 to move inward under the action of tension. The distal finger 422 will also move synchronously until the proximal finger 421 clamps the pineapple fruit. The thin-film pressure sensor in the inner flexible silica gel of the proximal finger 421 receives the force signal. After transmitting the force signal to the main control module, the main control module controls the output shaft on the servo motor 431 of the first driving mechanism to stop rotating. Subsequently, the servo motor 431 of the second driving mechanism drives the output shaft thereon to rotate. The output shaft drives the steering wheel 432 connected thereto to rotate. The driving rope connected to the steering wheel 432 and the distal finger 422 drives the distal finger 422 to move inward under the action of tension until the distal finger 422 clamps the pineapple fruit. The thin-film pressure sensor in the inner flexible silica gel of the distal finger 422 receives the force signal. After transmitting the force signal to the main control module, the main control module controls the output shaft on the servo motor 431 of the second driving mechanism to stop rotating. Thus, the clamping action is completed. The inner surfaces of the proximal finger 421 and the distal finger 422 are provided with flexible silica gel, which can increase the contact area with the pineapple fruit, increase friction while reducing mechanical damage to the pineapple fruit. Then, rotate the picking robotic arm 3 and simulate the stem-breaking picking method of manual picking to break the pineapple fruit off the stem. Finally, the picking end effector 4 is controlled by the picking robotic arm 3 to transport the pineapple fruit to the designated position (the pineapple fruit collection box 2). The output shafts on the servo motors 431 of the first driving mechanism and the second driving mechanism drive the steering wheel 432 to rotate in the reverse direction. Under the elastic force of the torsion springs 423 between the proximal finger 421 and the mounting body 41 and between the proximal finger 421 and the distal finger 422, the finger mechanism 42 opens and resets to the initial position, and the pineapple fruit falls to the collection point (the pineapple fruit collection box 2). Thus, the pineapple picking work is completed.
[0059] See Figures 1-9, both between the proximal finger 421 and the distal finger 422 and between the proximal finger 421 and the mounting body 41 are connected by a pin, and the torsion spring 423 and the pulley 424 are coaxially arranged with the pin; the number of torsion springs 423 between the proximal finger 421 and the distal finger 422 and the number of torsion springs 423 between the proximal finger 421 and the mounting body 41 are both two. Installation grooves for accommodating the torsion spring 423 are provided on both sides of the pulley 424, and two opening grooves 4232 for the two ends of the torsion spring 423 to extend out are provided on the side of the accommodating groove 4231; wiring hole grooves 425 are provided on both sides of the proximal finger 421, and the circuit of the thin-film pressure sensor is led out from the wiring hole grooves 425. The wiring hole grooves 425 are provided to orderly lead out the circuit of the thin-film pressure sensor from the inside of the picking end effector 4, avoiding the loosening or winding of the circuit when the picking end effector 4 works; the installation grooves are provided to facilitate the installation of the torsion spring 423 and at the same time make the structure very compact.
[0060] See Figures 1-9 , the proximal finger 421 and the distal finger 422 are in an arc-shaped plate structure according to the arc characteristics of the surface of the pineapple fruit. Circular connection seats are provided at both ends of the proximal finger 421; a circular connection seat is also provided at one end of the distal finger 422, and a through shaft hole is opened on the connection seat; the pin is a metal cylinder, and the pin is matched with the shaft hole; one end of the pin is inserted and penetrated through the shaft hole of the connection seat by pushing, and both ends of the pin are fixed by a buckle or a threaded cap, and the surface of the pin is smooth; the torsion spring 423 is annular.
[0061] See Figures 1-9 , a camera bracket 8 is provided on the mounting body 41, the binocular vision depth camera 5 is installed on the camera bracket 8, the end of the picking robotic arm 3 is fixedly connected to the camera bracket 8 through a threaded fastener, and the picking end effector 4 is fixedly connected to the camera bracket 8 through the same set of threaded fasteners. By providing the camera bracket 8, the connection between the picking end effector 4 and the picking robotic arm 3 can be realized, and at the same time, the binocular vision depth camera 5 can be installed.
[0062] See Figures 1-9 , the harvesting robot further includes a power module. The main control module includes a single-chip microcomputer chip. The thin-film pressure sensor is connected to the single-chip microcomputer chip through a Bluetooth module, and the single-chip microcomputer chip is an STM32 control main board; the two-point distributed pressure sensor is used to detect whether the proximal finger 421 and the distal finger 422 clamp the pineapple fruit; the servo 431 is connected to the single-chip microcomputer chip. The power module includes a 24V switching power supply. The switching power supply converts 220V alternating current into 24V direct current to supply power to the single-chip microcomputer chip, and the single-chip microcomputer chip converts 24V direct current into 5V direct current to supply power to the servo 431.
[0063] See Figures 1-9 , the working principle of the above pineapple multi-arm row-by-row harvesting robot is as follows:
[0064] The position of the pineapple fruit can be recognized by the binocular vision depth camera 5. The picking robotic arm 3 controls the movement of the picking end effector 4, and the picking end effector 4 picks the pineapple fruit. After picking the pineapple fruit, it is placed in the pineapple fruit collection box 2 for collection. Each picking manipulator is arranged corresponding to each row of pineapple plants, and multiple picking manipulators can realize the picking of pineapple fruits on multiple rows of pineapple plants, with high picking efficiency. The pineapple fruit is recognized and positioned through the vision system to obtain the three-dimensional coordinate position of the pineapple fruit, and the position signal is sent to the single-chip microcomputer chip; the picking manipulator approaches the pineapple fruit. When the mounting body 41 touches the pineapple fruit, the output shaft of the servo motor 431 of the first driving mechanism drives the steering wheel 432 to rotate, driving the drive rope to tighten, thereby driving the proximal finger 421 to clamp towards the pineapple fruit. The output shaft of the servo motor 431 of the second driving mechanism drives the steering wheel 432 to rotate, driving the drive rope to tighten, thereby driving the distal finger 422 to clamp towards the fruit. The finger mechanism 42 adapts to the size of the pineapple fruit and clamps the fruit; the thin-film pressure sensors under the flexible silica gel on the inner sides of the proximal finger 421 and the distal finger 422 receive the pressure signal, which is fed back to the single-chip microcomputer chip, and the single-chip microcomputer chip then sends a signal to rotate the sixth axis of the picking robotic arm 3, driving the picking manipulator that clamps the pineapple fruit to rotate, applying a breaking moment to break the connection between the lower end of the stem of the pineapple fruit and the calyx; after the pineapple fruit is separated from the stem, it is moved to the pineapple collection point. The servo motors 431 of the first driving mechanism and the servo motors 431 of the second driving mechanism are reset, the drive ropes are relaxed, and the torsion spring 423 applies a reverse moment to the joint, causing the finger mechanism 42 to reset, and the pineapple fruit is placed into the pineapple fruit collection box 2. The joint refers to the hinged position between the proximal finger 421 and the distal finger 422 and the hinged position between the proximal finger 421 and the mounting body 41.
[0065] Embodiment 2
[0066] See Figures 1-9, other structures in this embodiment are the same as those in Embodiment 1. The difference is that when the pineapple multi-arm row-by-row harvesting robot in this embodiment is working, the picking manipulator approaches from the front of the pineapple fruit. When the installation main body 41 (also called the palm) contacts the epidermis of the pineapple fruit, the servos 431 of the first driving mechanism and the second driving mechanism respectively drive the proximal fingers 421 and the distal fingers 422 to move, and clamp the pineapple fruit 9. The thin-film pressure sensor monitors whether the grasping force reaches the set safety threshold. If the grasping force is insufficient and does not reach the set safety threshold, the force is continuously adjusted and the grasping state is maintained, that is, the servos 431 of the first driving mechanism and the second driving mechanism continue to drive; once the grasping force reaches the safety threshold, the servos 431 of the first driving mechanism and the second driving mechanism lock the clamping position to ensure that the pineapple fruit is firmly grasped. Subsequently, the sixth axis of the picking robotic arm 3 rotates to apply a breaking moment to drive the picking end effector 4 to perform pulling and twisting actions, and remove the pineapple fruit from the pineapple plant.
[0067] A personal computer is used as the upper computer, which is responsible for sending start and stop commands and receiving the status feedback of the system. The picking robotic arm 3 serves as the execution main body, receives the instructions from the upper computer and controls the movement of the picking end effector 4, and communicates with the main control module of the picking end effector 4 through the RS232 serial port at the same time. The main control module undertakes the core control task, receives the grasping force data from the thin-film pressure sensor, and adjusts the grasping force according to the data to prevent damage to the pineapple fruit. At the same time, control instructions are sent to the servo motor of the picking robotic arm 3 through the serial port to make the picking end effector 4 perform specific actions. The clamping and twisting actions of the picking end effector 4 are driven by the servos 431. Combining appropriate force control and real-time feedback, the purpose of stably picking the pineapple fruit is achieved. Through the information interaction between each module of the overall system, precise control of the picking process is realized, and finally the reliability and adaptability of mechanized picking are improved.
[0068] Embodiment 3
[0069] See Figures 10-11 , other structures in this embodiment are the same as those in Embodiment 1. The difference is that a height adjustment mechanism 6 for adjusting the height of the gantry is provided between the crawler mechanism and the gantry; the picking manipulator is arranged on the gantry. By setting the gantry, it is convenient for the harvesting robot to travel in the pineapple field. The gantry can avoid the pineapple plants. The height of the gantry can be adjusted through the height adjustment mechanism 6, and then the height of the picking manipulator can be adjusted, so that the harvesting robot can adapt to pineapple plants of different heights.
[0070] See Figures 10-11, the gantry includes a first frame 11 and a second frame 12. An insertion rod is provided on the first frame 11, and a slot is provided on the second frame 12. The insertion rod is connected to the slot in a mating manner, and a locking mechanism is provided between the insertion rod and the slot. The crawler mechanism includes two sets of crawler assemblies. One set of crawler assemblies is arranged at the lower end of the first frame 11, and the other set of crawler assemblies is arranged at the lower end of the second frame 12. By providing the insertion rod and the slot, the distance between the first frame 11 and the second frame 12 can be adjusted, and then the distance between the two sets of crawler assemblies can be adjusted, so that the harvesting robot can adapt to pineapple plants with different row spacings. Through the locking mechanism, the first frame 11 and the second frame 12 can be locked, so that the first frame 11 and the second frame 12 are fixedly connected to ensure the stability of picking. In another implementation manner of this embodiment, the base of the six-degree-of-freedom picking manipulator 3 is slidably connected to the manipulator mounting platform 7, and the picking manipulator 3 is fixed to the manipulator mounting platform 7 through a locking mechanism. The purpose is to facilitate the adjustment of the distance between the picking manipulators 3 and further improve flexibility. Two manipulator mounting platforms 7 can be provided, which are respectively arranged on the first frame 11 and the second frame 12. Two picking manipulators are provided, and the manipulator mounting platform 7 and the picking manipulator 3 are installed in one-to-one correspondence.
[0071] See Figures 10-11 , the crawler assembly includes a crawler frame 13 and a crawler 14 provided on the crawler frame 13. The height adjustment mechanism 6 includes hydraulic cylinder bodies provided between the first frame 11 and the crawler frame 13 and between the second frame 12 and the crawler frame 13. Through the telescopic movement of the hydraulic cylinder bodies, the height of the gantry is adjusted. A guide sleeve 15 is provided on the crawler frame 13, and the first frame 11 and the second frame 12 are slidably connected to the guide sleeve 15. The purpose is to play a guiding role. The total number of guide sleeves 15 is 6.
[0072] Embodiment 4
[0073] See Figures 1-9 , this embodiment discloses a picking method for a pineapple multi-arm in-row harvesting robot. The number of picking manipulators is two, and the picking method includes the following steps:
[0074] S1. Use the YOLOv8 object detection algorithm to identify pineapple fruits on the RGB color image, and synchronously obtain the corresponding depth information through the binocular vision depth camera 5. Combine point cloud computing to obtain the position coordinates of the pineapple fruits in the three-dimensional space. The position coordinates are the coordinates (X C , Y C , Z C) After that, coordinate transformation is performed through the hand-eye calibration method to convert the position coordinates of the pineapple fruit in the three-dimensional space from the camera coordinate system to the robotic arm base coordinate system, where the robotic arm base coordinate system is the coordinate system of the base of the picking robotic arm 3;
[0075] S2. Layout optimization of the dual picking robotic arms 3: The dual picking robotic arms 3 are installed on the same side. Based on the same-side installation structure, by establishing an optimization model for the working space of the two arms and taking the maximum picking coverage rate and the minimum interference between the picking robotic arms 3 as the optimization objectives, the particle swarm optimization algorithm (PSO) is used to solve the optimal installation position, angle, and forward step of the picking robotic arm 3, so as to optimize the picking efficiency of the dual picking robotic arms 3 in a narrow working space;
[0076] Motion planning and obstacle avoidance: Based on the path optimization algorithms of the artificial potential field method (APF) and the rapidly-exploring random tree (RRT), they are respectively used for generating obstacle avoidance trajectories and local path correction. Combining the dynamic window algorithm (DWA) and the real-time motion state of the picking robotic arm 3, the motion path and smoothness of the picking end effector 4 are adjusted to ensure flexible adjustment under different distributions of pineapple plants;
[0077] Parallel picking strategy: The genetic algorithm (GA) is used to optimize the task allocation, enabling the left and right picking robotic arms 3 to pick alternately in time sequence to reduce conflicts. Combining time scheduling optimization control, the picking time windows of the left and right picking robotic arms 3 are adjusted to avoid collisions in adjacent areas;
[0078] S3. According to the three-dimensional coordinates of the pineapple fruit in the robotic arm base coordinate system, the picking robotic arm 3 approaches the target pineapple fruit along the optimized trajectory and controls the picking end effector 4 to perform the picking operation; the picking end effector 4 performs the clamping action. After complete clamping, the sixth axis of the picking robotic arm 3 rotates to apply a breaking moment to complete the separation of the fruit stem from the pineapple fruit;
[0079] S4. The picking robotic arm 3 drives the picking end effector 4 to move and places the picked pineapple fruit into the pineapple fruit collection box 2 on the chassis 1;
[0080] S5. After placing the pineapple fruit, each picking robotic arm 3 and the picking end effector 4 return to their original positions respectively to prepare for the next picking task;
[0081] S6. Repeat steps S2 - S5 to pick the pineapple fruits in the current planting row of the pineapple plant until all the pineapple fruits in the current planting row of the pineapple plant are picked.
[0082] In step S1, the specific steps of obtaining the position coordinates of the pineapple fruit in the three-dimensional space and performing coordinate transformation through the hand-eye calibration method to convert the position coordinates of the pineapple fruit in the three-dimensional space from the camera coordinate system to the robotic arm base coordinate system are as follows:
[0083] S11. Set the optical center of the binocular vision depth camera 5 as the origin, and establish a camera coordinate system (X C , Y C , Z C ), where the X-axis is horizontal to the right (i.e., the X-axis is perpendicular to the optical axis and to the right), the Y-axis is vertical downward (i.e., the Y-axis is perpendicular to the optical axis and downward), and the Z-axis is along the optical axis forward (depth). Let the internal parameter matrix of the binocular vision depth camera 5 be:
[0084]
[0085] Where: f x , f y are the focal lengths (unit: pixel), c x , c y are the coordinates of the principal point (optical center) (unit: pixel). Using the camera projection model, the relationship between the pixel coordinates (u, v) and the camera coordinate system (X C , Y C , Z C ) is:
[0086]
[0087] Through the camera projection model, convert the pixel coordinates (u, v) into three-dimensional coordinates in the camera coordinate system:
[0088]
[0089] Among them, after obtaining the Z C value from the binocular vision depth camera 5, the X C , Y C can be obtained:
[0090] S12. Complete the conversion from the camera coordinate system to the base coordinate system of the picking manipulator 3 through the fixed transformation matrices E T C and B T E :
[0091]
[0092] First, through the E T C transformation, convert the three-dimensional coordinates (position coordinates in three-dimensional space) of the pineapple fruit in the camera coordinate system (X C , Y C , Z C ) to the end coordinate system of the picking manipulator 3 (X E , Y E , Z E), so that the picking robotic arm 3 can accurately identify the target position; then use B T E to further convert the coordinates (X E , Y E , Z E ) of the end of the picking robotic arm to the coordinates (X B , Y B , Z B ) of the robotic arm base; finally, obtain the three-dimensional coordinates of the pineapple fruit in the robotic arm base coordinate system; the end point E of the picking robotic arm 3 is the place where the picking end effector 4 is connected; the robotic arm base point B is the place where the whole picking robotic arm 3 is fixed on the robotic arm mounting platform 7;
[0093] Among them, the fixed transformation matrix between the camera coordinate system and the robotic arm end coordinate system E E T C is composed of the rotation matrix and the translation vector E P C :
[0094]
[0095] Among them, is the rotation matrix from the camera coordinate system to the robotic arm end coordinate system:
[0096]
[0097] The translation vector E P C represents the position of the optical center of the binocular vision depth camera 5 in the robotic arm end coordinate system:
[0098]
[0099] The transformation matrix from the robotic arm end coordinate system E to the robotic arm base coordinate system B is:
[0100]
[0101] Among them, is the rotation matrix from the robotic arm end coordinate system to the robotic arm base coordinate system, B P E is the position vector of the robotic arm end coordinate system in the robotic arm base coordinate system.
[0102] Furthermore, step S2 further includes the following steps:
[0103] In the parallel picking strategy of the dual picking robotic arm 3, if simultaneous picking is inevitable, optimize the trajectory so that the paths of the two picking robotic arms 3 do not cross. Use a trajectory optimization algorithm to minimize the trajectory crossing error and optimize the motion path of the picking robotic arm 3. The trajectory optimization algorithm is as follows:
[0104]
[0105] And an additional constraint can be added: |||P L -P R ‖-d safe |<∈.
[0106] Where, J min represents the objective function of trajectory optimization, minimizing the trajectory crossing error of the picking robotic arm 3; P L (left picking robotic arm 3) and P R (right picking robotic arm 3) are the path planning points of the left and right picking robotic arms 3 at time i; d safe is the safety distance when the left and right picking robotic arms 3 pick; ∈ is the acceptable error range to ensure that the basic safety requirements are not violated during trajectory optimization.
[0107] The above is a preferred embodiment of the present invention. However, the embodiments of the present invention are not limited by the above content. Any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.
Claims
1. A pineapple multi-arm row harvesting robot, characterized in that: It includes a chassis, a pineapple fruit collection frame arranged on the chassis, and multiple picking manipulators; wherein the multiple picking manipulators are arranged at the front end of the chassis; when picking, each picking manipulator is arranged in one-to-one correspondence with each row of pineapple plants; the picking manipulator includes a picking manipulator arm, a picking end effector arranged at the end of the picking manipulator arm, and a binocular vision depth camera.
2. A pineapple multi-arm row harvesting robot according to claim 1, characterized in that: The chassis is a gantry chassis, including a gantry and a crawler mechanism arranged at the lower end of the gantry, and a height adjustment mechanism for adjusting the height of the gantry is provided between the crawler mechanism and the gantry; the picking manipulator is arranged on the gantry.
3. A pineapple multi-arm row harvesting robot according to claim 1, characterized in that: A mechanical arm mounting platform is provided on the gantry, and the picking mechanical arm is arranged on the mechanical arm mounting platform.
4. A pineapple multi-arm row harvesting robot according to claim 2, characterized in that: The gantry includes a first frame and a second frame, the first frame is provided with an insertion rod, the second frame is provided with a slot, the insertion rod is cooperatively connected with the slot, a locking mechanism is provided between the insertion rod and the slot, and the track mechanism includes two groups of track assemblies, one group of track assemblies is arranged at the lower end of the first frame, and the other group of track assemblies is arranged at the lower end of the second frame.
5. The pineapple multi-arm row harvesting robot according to claim 1, characterized in that: The picking end effector comprises a mounting body, a plurality of finger mechanisms arranged on the mounting body, and a picking drive mechanism for driving each finger mechanism to move; wherein, The finger mechanism comprises a proximal finger and a distal finger, one end of the proximal finger is hinged to the mounting body, and the other end of the proximal finger is hinged to one end of the distal finger; a torsion spring and a pulley are provided between the proximal finger and the mounting body and between the proximal finger and the distal finger; The picking drive mechanism includes a first drive mechanism for driving the movement of the proximal fingers and a second drive mechanism for driving the movement of the distal fingers; the first drive mechanism and the second drive mechanism both include a servo, a steering wheel and a driving rope arranged on a mounting body; the steering wheel is arranged on the servo, one end of the driving rope of the first drive mechanism is connected to the steering wheel, and the other end is connected to the proximal fingers after passing through a pulley; one end of the driving rope of the second drive mechanism is connected to the steering wheel, and the other end is connected to the distal fingers after passing through a pulley.
6. A pineapple multi-arm row harvesting robot according to claim 5, characterized in that: The harvesting robot also includes a main control module and a pressure detection module arranged on each picking manipulator, the main control module is connected to the pressure detection module, the pressure detection module includes a plurality of two-point distributed thin film pressure sensors, the two-point distributed thin film pressure sensors are arranged in a one-to-one correspondence with the finger mechanism, each two-point distributed thin film pressure sensor includes a thin film pressure sensor arranged in the inner flexible silicone of the proximal finger and the inner flexible silicone of the distal finger.
7. The pineapple multi-arm row harvesting robot according to claim 5, characterized in that: The proximal finger and the distal finger, as well as the proximal finger and the mounting body are connected via an axle pin, and the torsion spring and the pulley are coaxially arranged with the axle pin; there are two torsion springs between the proximal finger and the distal finger, as well as two torsion springs between the proximal finger and the mounting body, and mounting grooves for accommodating the torsion springs are provided on both sides of the pulley; wiring holes are provided on both sides of the proximal finger, and the lines of the thin film pressure sensor are led out through the wiring holes.
8. The pineapple multi-arm row-by-row harvesting robot according to claim 5, characterized in that: A camera bracket is provided on the installation body, the binocular vision depth camera is installed on the camera bracket, and the end of the picking mechanical arm is connected to the camera bracket.
9. The pineapple multi-arm row-by-row harvesting robot according to claim 6, characterized in that: The harvesting robot also includes a power module, the main control module includes a single-chip microcomputer chip, the thin film pressure sensor is connected to the single-chip microcomputer chip via a Bluetooth module, the servo is connected to the single-chip microcomputer chip, the power module includes a 24V switching power supply, the switching power supply converts 220V AC power into 24V DC power to power the single-chip microcomputer chip, and the single-chip microcomputer chip converts 24V DC power into 5V DC power to power the servo.
10. A method for picking pineapples using a multi-arm row-by-row harvesting robot as claimed in any one of claims 1 to 9, characterized in that: The number of the picking manipulators is two, and the picking method comprises the following steps: S1. Use the YOLOv8 target detection algorithm to identify pineapple fruits on the RGB color image, and synchronously obtain the corresponding depth information through the binocular vision depth camera, and combine point cloud computing to obtain the position coordinates of the pineapple fruit in the three-dimensional space, which are the coordinates of the camera coordinate system; then, use the hand-eye calibration method to perform coordinate conversion, so that the three-dimensional coordinates of the pineapple fruit are converted from the camera coordinate system to the robot arm base coordinate system, and the robot arm base coordinate system is the coordinate system of the base of the picking robot arm; S2. Optimization of the layout of dual picking robotic arms: The dual picking robotic arms are installed on the same side. By establishing a dual-arm workspace optimization model and taking the maximum picking coverage and minimum mutual interference of the picking robotic arms as the optimization goals, the particle swarm optimization algorithm is used to solve the optimal installation position, angle and forward step of the picking robotic arms. Motion planning and obstacle avoidance: The path optimization algorithm based on the artificial potential field method and the rapidly expanding random tree is used for obstacle avoidance trajectory generation and local path correction respectively. The dynamic window algorithm and the real-time motion state of the picking robot arm are combined to adjust the motion path and smoothness of the picking end effector to ensure flexible adjustment under different pineapple plant distribution conditions; Parallel picking strategy: Genetic algorithm is used to optimize task allocation, so that the left and right picking robot arms can pick fruits alternately in time sequence. Combined with time scheduling optimization control, the picking time windows of the left and right picking robot arms are adjusted. S3, according to the three-dimensional coordinates of the pineapple fruit in the coordinate system of the robot arm base, the picking robot arm approaches the target pineapple fruit according to the optimized trajectory, and controls the picking end effector to perform the picking operation; the picking end effector performs the clamping action, and after the complete clamping, the rotation of the picking robot arm applies the breaking torque to complete the separation of the fruit stem and the pineapple fruit; S4, the picking robot arm puts the picked pineapple fruits into the pineapple fruit collection frame on the chassis; S5, after placing the pineapple fruit, each picking robot arm and the picking end effector are reset to prepare for the next picking task; S6, repeat steps S2-S5 to pick the pineapple fruits in the current pineapple plant planting row until all pineapple fruits in the current pineapple plant planting row are picked.
Citation Information
Patent Citations
Pineapple picking machine and picking method thereof
CN110744570A
Pineapple picking robot based on binocular vision
CN114872007A
Pineapple picking robot and picking method thereof
CN116210445A
Bionic self-adaptive rope-driven pear fruit picking manipulator
CN116277122A
Cited By
Multi-thread path planning method and device for flexible cable type picking mechanical arm
CN121515223A