Apple picking and follow-up collecting device and method with global self-adaptive parallel mechanical arms

By using a fully adaptive parallel robotic arm and a follow-up collection device, the problems of large motion inertia, slow operation speed, and limited workspace of orchard picking robots in orchards with different tree ages and varieties have been solved, achieving efficient and low-damage fruit picking and collection, and improving the level of intelligent management of orchards.

CN121970610APending Publication Date: 2026-05-05SHANDONG AGRICULTURAL UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHANDONG AGRICULTURAL UNIVERSITY
Filing Date
2026-03-19
Publication Date
2026-05-05

AI Technical Summary

Technical Problem

Existing orchard harvesting robots suffer from problems such as large motion inertia, slow operating speed, limited working space, many blind spots, and high fruit damage rate. They are particularly difficult to achieve efficient and low-damage automated harvesting in orchards with different tree ages and varieties.

Method used

Employing a fully adaptive parallel robotic arm, combined with a longitudinal speed-increasing module and a lateral telescopic module, the system acquires the three-dimensional coordinates and depth information of the fruit through a vision control system. This enables flexible movement of the robotic arm and the follow-up collection of the fruit, allowing for layered positioning and graded harvesting. The system also incorporates a follow-up conveyor belt for synchronous fruit collection.

Benefits of technology

It enables full-area harvesting of orchards with different tree ages and varieties, improving harvesting efficiency and accuracy, reducing fruit damage rate, and is highly adaptable and widely applicable. It also reduces labor costs and promotes intelligent management of orchards.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121970610A_ABST
    Figure CN121970610A_ABST
Patent Text Reader

Abstract

The invention discloses a global self-adaptive parallel mechanical arm apple picking and follow-up collecting device and method.The device comprises a central controller, a longitudinal multi-speed lifting module, a transverse telescopic module, a parallel mechanical arm and a visual control system, and the longitudinal multi-speed lifting module, the transverse telescopic module and the parallel mechanical arm are electrically connected with the central controller; the visual control system is used for obtaining three-dimensional coordinates and depth information of fruits, and the central controller controls the longitudinal speed-multiplying lifting module to drive the picking operation platform to move to a plurality of height stations in sequence. Meanwhile, after the transverse telescopic module is controlled to transversely move the picking operation platform according to three-dimensional coordinates and depth information obtained by the visual control system at each height station, the central controller controls the parallel mechanical arm to pick the fruits, and the picked fruits are collected in a follow-up mode through the follow-up fruit collecting module. According to the device, full-area covering picking of fruits at different heights and different depth positions of tree crowns is achieved, and the environmental suitability of the device is greatly improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of intelligent orchard management machinery technology, specifically to a fully adaptive parallel robotic arm apple picking and follow-up collection device and method. Background Technology

[0002] Apple harvesting is a crucial part of orchard management, directly impacting orchard production efficiency, fruit quality, and farmers' economic benefits. With the increasing scale and intensification of orchard operations, traditional manual harvesting methods are no longer sufficient to meet production demands. Manual harvesting is not only labor-intensive and has seen rising labor costs annually, but it also suffers from low efficiency and is susceptible to damage due to operator skill and fatigue, resulting in high fruit damage rates and economic losses. Therefore, developing automated apple harvesting equipment to achieve efficient and low-damage apple harvesting has become an urgent need for intelligent orchard management.

[0003] Orchards of different ages and varieties exhibit significant differences in canopy height and fruit distribution depth. Even within the same orchard, the fruit on the same tree can be distributed at different heights, such as the bottom, middle, and top layers, and the depth of the fruit from the edge of the canopy varies. This places extremely high demands on the spatial adaptability and operational flexibility of automated harvesting devices.

[0004] Currently, most orchard harvesting robots use either serial robotic arms or parallel robotic arms with fixed bases. Serial robotic arms, composed of multiple joints connected in series, offer a certain range of motion but suffer from inherent problems such as high inertia and slow movement. This results in slow response times and long harvesting cycles, severely limiting overall harvesting efficiency and failing to meet the high-efficiency harvesting needs of large-scale orchards. Parallel robotic arms, with their compact structure, high motion precision, and fast response, offer advantages in high-speed operations. However, existing parallel robotic arms mostly use fixed base designs, severely limiting their workspace and preventing flexible adjustments based on fruit depth. This leads to numerous blind spots in complex environments with varying fruit growth depths, making it difficult to effectively harvest fruits inside the canopy or deep within the tree, resulting in poor applicability.

[0005] Furthermore, most existing apple-picking robots use a fixed chassis and container design, meaning the container's position remains constant. After each harvest, the robotic arm must retract a long distance from the harvesting position back to the top of the container to unload the fruit. This process not only consumes a significant amount of work time, further reducing harvesting efficiency, but also makes the fruit susceptible to damage from shaking and collisions during the long retraction, thus affecting its commercial value.

[0006] Therefore, developing a fully adaptive, highly efficient, and low-damage parallel robotic arm for apple picking and follow-up collection is of significant practical importance and application value. Summary of the Invention

[0007] In order to solve the above-mentioned technical problems, this application proposes the following technical solution: In a first aspect, embodiments of this application provide a fully adaptive parallel robotic arm apple picking and follow-up collection device, comprising: a central controller and a longitudinal speed-multiplying lifting module, a lateral telescopic module, a parallel robotic arm, and a vision control system disposed on the parallel robotic arm, all electrically connected to the central controller. The vision control system is used to acquire the three-dimensional coordinates and depth information of the fruit. The central controller controls the longitudinal speed-multiplying lifting module to drive the picking platform to move sequentially to multiple height positions. Simultaneously, based on the three-dimensional coordinates and depth information acquired by the vision control system at each height position, the central controller controls the lateral telescopic module to move the picking platform laterally. After the movement is completed, the central controller controls the parallel robotic arm to pick the fruit and the follow-up fruit collection module to collect the picked fruit.

[0008] In one possible implementation, the longitudinal lifting module includes a longitudinal drive cylinder fixedly mounted on the side of the mobile frame column and a vertical guide rail fixedly mounted on the mobile frame. The push rod of the longitudinal drive cylinder is provided with a double-speed sprocket set. One end of the chain is mounted on the column of the mobile frame, and the other end of the chain passes around the double-speed sprocket set and is connected to the harvesting platform. The harvesting platform is slidably mounted on the vertical guide rail.

[0009] In one possible implementation, the harvesting platform includes a horizontal frame and a lateral telescopic module. A horizontal linear guide rail is fixedly mounted on the horizontal frame. The lateral telescopic module includes a lateral drive cylinder and a slide table. The cylinder body of the lateral drive cylinder is fixedly connected to the slide table base. The horizontal frame is slidably connected to the slide table. The movable end of the push rod of the lateral drive cylinder is connected to a parallel robotic arm base fixedly mounted on the horizontal frame. A parallel robotic arm and a rigid connecting arm are mounted on the parallel robotic arm base. The rigid connecting arm is fixedly connected to a follow-up fruit collection module.

[0010] In one possible implementation, the follow-up fruit collection module includes a follow-up conveyor belt, which is connected to a conveyor belt drive motor. Anti-drop baffles are provided at both ends of the follow-up conveyor belt, and spacer baffles are evenly spaced on the surface of the follow-up conveyor belt. A fruit receiving box is provided at the receiving end of the follow-up conveyor belt, and the frame of the fruit receiving box is fixedly connected to the first end of a rigid connecting arm. The second end of the rigid connecting arm is fixedly connected to the base of the parallel robotic arm. The output end of the follow-up conveyor belt is slidably connected to the moving frame via a linear slide mechanism. The output end of the follow-up conveyor belt is located above the main collection channel, and a collection box is provided below the output end of the main collection channel.

[0011] In one possible implementation, the end effector of the parallel robotic arm is provided with an end effector and a depth camera, wherein the end effector is a flexible gripping structure robotic arm.

[0012] In one possible implementation, the mobile frame is a gantry frame structure, and a mobile chassis is configured at the bottom of the mobile frame.

[0013] In one possible implementation, the vision control system includes a depth camera and multiple position sensors, which are electrically connected to the input of a central controller. The output of the central controller is electrically connected to a lateral drive cylinder, a longitudinal drive cylinder, and a parallel robotic arm driver, which is electrically connected to a parallel robotic arm.

[0014] Secondly, this application provides a method for apple picking and follow-up collection using a fully adaptive parallel robotic arm, including: controlling the longitudinal speed-doubled lifting module to be placed at the initial target position through a central controller, while simultaneously controlling the lateral telescopic module to retract to the initial position, thus completing the initialization operation; After initialization, the central controller controls the longitudinal speed-increasing module to drive the harvesting platform to the bottom, middle, and top height positions sequentially. The calculation formula is as follows: in, For the first The target height for each workstation. To hold a high position in a low-level job Interlayer spacing The vertical reachable radius of a single height station of a parallel robotic arm. This refers to the interlayer coverage overlap rate; The vision control system at each height station acquires the three-dimensional coordinates and depth information of each layer of fruit. The depth of the fruit from the parallel robotic arm base is calculated based on the acquired 3D coordinates and depth information, using the following formula: in, For the first The depth of each fruit relative to the base of the parallel robotic arm. For the first The orientation coordinates of each fruit in the base coordinate system The coordinates of the base depth reference plane; The harvesting area is determined based on the calculated depth information, and harvesting operations are carried out based on the determined area. After harvesting, the parallel robotic arm releases the fruit to the follow-up conveyor belt, then resets or proceeds to harvest the next fruit. The follow-up conveyor belt transports the released fruit to the main collection channel, and then the main collection channel transports it to the collection box. After all fruits at the current height station have been harvested, control the longitudinal double-speed lifting module to move to the next height station until the harvesting of all fruits at the three height stations is completed.

[0015] In one possible implementation, determining the harvesting area based on the calculated depth information and performing harvesting operations based on the determined area includes: If the calculated depth is greater than 0 and less than or equal to the first preset threshold, it is determined to be a direct picking area, and the central controller controls the parallel robotic arm to perform the picking operation. If the calculated depth is greater than the first preset threshold and less than or equal to the second preset threshold, it is determined to be a compensation picking area, and the compensation distance is calculated using the following formula: in, For the first The lateral scaling compensation distance corresponding to each fruit The gap distance is defined as the distance where the depth of the fruit exceeds the reachability of the parallel robotic arm itself. This is the maximum compensation stroke for the lateral telescopic module. This is a limiting function used to... Limited to the range Inside; The horizontal telescopic module drives the parallel robotic arm to move forward based on the calculated compensation distance and then controls the movement of the parallel robotic arm to perform the picking operation. During the movement, the rigid connecting arm drives the follower conveyor belt to slide horizontally synchronously along the linear slide mechanism, so that the fruit receiving end of the follower conveyor belt is always located below the working space of the parallel robotic arm.

[0016] Compared with the prior art, the beneficial effects of this application are as follows: This application utilizes a central controller to coordinate the longitudinal speed-increasing module and the lateral telescopic module, enabling the harvesting platform to move smoothly and quickly to three height positions: bottom, middle, and top. This adapts to the differences in canopy height among fruit trees of different ages and varieties. Simultaneously, based on the fruit depth information obtained from the vision control system, the lateral position of the parallel robotic arm can be flexibly adjusted. This effectively compensates for the shortcomings of traditional fixed-base parallel robotic arms, such as limited working space and blind spots. Combined with the high speed and high precision of the parallel robotic arm itself, it achieves full-area harvesting of fruits at different heights and depths of the canopy, significantly improving the environmental adaptability of the device.

[0017] This application achieves synchronous picking and collection through a follow-up fruit collection module. The follow-up conveyor belt is fixed to the base of the parallel robotic arm through a rigid connecting arm and can move laterally synchronously with the parallel robotic arm, so that the fruit receiving end is always located below the working space of the parallel robotic arm. After the robotic arm finishes picking, it can directly release the fruit onto the follow-up conveyor belt without the need for long-distance retraction to unload the fruit. This shortens the single picking cycle, improves the overall operation efficiency, and avoids damage to the fruit caused by long-distance shaking and collision, thus protecting the commercial value of the fruit.

[0018] This application achieves automated and standardized apple harvesting through a process of initialization, layered positioning, depth determination, graded picking, and follow-up collection. It adopts differentiated picking strategies for fruits at different depths and rationally divides the direct picking area and the compensation picking area, further improving the accuracy and efficiency of picking. At the same time, this method can be adapted to the fruit distribution characteristics of orchards with different tree ages and varieties, without the need for large-scale adjustments to the device. It has strong versatility, can effectively reduce orchard labor costs, and promote the intelligent and large-scale management upgrade of orchards. Attached Figure Description

[0019] Figure 1 A schematic diagram of the overall structure of a fully adaptive parallel robotic arm apple picking and follow-up collection device provided in this application embodiment; Figure 2 Side view of the fully adaptive parallel robotic arm apple picking and follow-up collection device provided in the embodiments of this application; Figure 3 This is a flowchart illustrating the global adaptive parallel robotic arm apple picking and follow-up collection method provided in an embodiment of this application.

[0020] Figure 1 The symbols in the middle are: 1-moving frame, 2-longitudinal drive cylinder, 3-double speed sprocket set, 4-picking platform, 5-lateral drive cylinder, 6-slide table, 7-parallel robotic arm, 8-follower conveyor belt, 9-rigid connecting arm, 10-linear slide mechanism, 11-main collection channel, 12-depth camera, 13-central controller, 14-end manipulator, 15-fruit receiving box. Detailed Implementation

[0021] The present solution will now be described in conjunction with the accompanying drawings and specific embodiments.

[0022] Figure 1 A schematic diagram of a globally adaptive parallel robotic arm for apple picking and follow-up collection, provided in an embodiment of this application, is shown below. Figure 1This embodiment of an all-domain adaptive parallel robotic arm apple picking and follow-up collection device includes: a central controller 13 and a longitudinal speed-multiplying lifting module, a lateral telescopic module, a parallel robotic arm 7, and a vision control system mounted on the parallel robotic arm 7, all electrically connected to the central controller 13. The vision control system is used to acquire the three-dimensional coordinates and depth information of the fruit. The central controller 13 controls the longitudinal speed-multiplying lifting module to drive the picking platform 4 to move sequentially to multiple height positions. At the same time, based on the three-dimensional coordinates and depth information acquired by the vision control system at each height position, it controls the lateral telescopic module to move the picking platform 4 laterally. After the movement is completed, the central controller 13 controls the parallel robotic arm 7 to pick the fruit and the follow-up fruit collection module to collect the picked fruit.

[0023] See Figure 2 In this embodiment, the longitudinal lifting module includes a longitudinal drive cylinder 2 fixedly mounted on the side of the column of the mobile frame 1 and a vertical guide rail fixedly mounted on the mobile frame 1. The mobile frame 1 is a gantry frame structure, and a mobile chassis is configured at the bottom of the mobile frame 1. A double-speed sprocket set 3 is provided at the top of the push rod of the longitudinal drive cylinder 2. One end of the chain is mounted on the column of the mobile frame, and the other end of the chain passes around the double-speed sprocket set and is connected to the harvesting platform. The harvesting platform 4 is slidably mounted on the vertical guide rail.

[0024] In this embodiment, the harvesting platform 4 includes a horizontal frame and a lateral telescopic module. A horizontal linear guide rail is fixedly installed on the horizontal frame. The lateral telescopic module includes a lateral drive cylinder 5 and a slide table 6. The cylinder body of the lateral drive cylinder 5 is fixedly connected to the slide table seat of the slide table 6. The horizontal frame is slidably connected to the slide table 6. The movable end of the push rod of the lateral drive cylinder 5 is connected to the parallel robotic arm base fixedly installed on the horizontal frame. The parallel robotic arm base is provided with a parallel robotic arm 7 and a rigid connecting arm 9. The rigid connecting arm 9 is fixedly connected to the follow-up fruit collection module. The end of the parallel robotic arm 7 is provided with an end effector and a depth camera 12. The end effector is a flexible gripping structure robotic arm.

[0025] The follow-up fruit collection module includes a follow-up conveyor belt 8, which is connected to a conveyor belt drive motor. Anti-drop baffles are provided at both ends of the follow-up conveyor belt 8, and spacer baffles are provided at equal intervals on the surface of the follow-up conveyor belt 8. A fruit receiving box 15 is provided at the receiving end of the follow-up conveyor belt. The frame of the fruit receiving box 15 is fixedly connected to the first end of the rigid connecting arm 9, and the second end of the rigid connecting arm 9 is fixedly connected to the base of the parallel robotic arm. The output end of the follow-up conveyor belt 8 is slidably connected to the mobile frame 1 through a linear slide mechanism 10. The output end of the follow-up conveyor belt 8 is located above the main collection channel 11, and a collection box is provided below the output end of the main collection channel 11.

[0026] The vision control system includes a depth camera 12 and multiple position sensors. The depth camera 12 and multiple position sensors are electrically connected to the input terminals of the central controller 13. The output terminals of the central controller 13 are electrically connected to the lateral drive cylinder 5, the longitudinal drive cylinder 2 and the parallel robotic arm driver. The parallel robotic arm driver is electrically connected to the parallel robotic arm 7.

[0027] In this embodiment, the central controller 13 is equipped with a coordinate calculation module, which is used to determine whether to activate the horizontal telescopic module for depth compensation based on the depth coordinates of the fruit.

[0028] Corresponding to the global adaptive parallel robotic arm apple picking and follow-up collection device provided in the above embodiments, this application also provides a global adaptive parallel robotic arm apple picking and follow-up collection method.

[0029] See Figure 2 This application provides a method for global adaptive parallel robotic arm apple picking and follow-up collection, comprising: To facilitate the positioning and depth determination of the fruit target, a parallel robotic arm base coordinate system is established. Its origin Located at the reference center position of the parallel robotic arm base, The YB axis points in the depth direction of the tree canopy, with the inward direction towards the tree being positive. The ZB axis represents the lateral extension direction of the telescopic module, with forward extension being positive. Establish the depth camera coordinate system. The depth camera outputs the center point of the fruit at... The coordinates below are The target point coordinates are transformed to the base coordinate system using camera extrinsic parameters, satisfying the following: in, Let be the coordinates of the center point of the fruit under ΣB. This is the rotation matrix from the camera coordinate system to the base coordinate system; This is the translation vector from the camera coordinate system to the base coordinate system.

[0030] S101, through the central controller, controls the longitudinal double-speed lifting module to the lowest position, and simultaneously controls the lateral telescopic module to retract to the initial position, completing the initialization operation. Initialization can be determined according to the following zero-position relationship: in: The current vertical height of the harvesting platform is obtained from the longitudinal position sensor. This is the initial height of the longitudinal speed-increasing module; This represents the current extension amount of the lateral telescopic module. This represents the initial extension amount of the lateral telescopic module.

[0031] To improve positioning accuracy, this embodiment also corrects the zero offset of the position sensor: the calculation formula is as follows: in: The original sensor reading; This is the zero bias correction value obtained during initialization.

[0032] S102, after the initialization operation is completed, the central controller controls the longitudinal speed lifting module to drive the picking platform to move sequentially to the bottom, middle and top height positions, and acquires the three-dimensional coordinates and depth information of each layer of fruit through the vision control system at each height position.

[0033] In this embodiment, the three height stations can meet the following requirements: in, For the first The target height for each workstation. This refers to the height of the bottom workstation. The interlayer spacing is used to achieve overlapping coverage between layers and reduce harvesting blind spots. The optimal interlayer spacing should satisfy the following conditions: in, The vertical reachable radius of a single height station of a parallel robotic arm. To achieve the interlayer coverage overlap rate, η=0.5 in this embodiment, ensuring 50% overlap coverage between adjacent height stations. The parallel robotic arm has a depth working range of 0-400mm, a vertical and horizontal working range of ±400mm, a lateral telescopic module compensation stroke of 400mm, and a total longitudinal lifting module stroke of 800mm. At each height station, the depth camera acquires the first... The fruit in the camera coordinate system coordinates below And its coordinates in the base coordinate system are obtained through extrinsic parameter transformation. The following coordinates: in, For the first The center point of each fruit is at The coordinates below, For the first The center point of each fruit is at The coordinates below.

[0034] S103, calculate the depth of the fruit from the parallel robotic arm base based on the acquired three-dimensional coordinates and depth information.

[0035] In this embodiment, depth is defined as the distance at which the target point is located. The formula for calculating the projected distance in the direction is: in, For the first The depth of each fruit relative to the base of the parallel robotic arm. For the first The orientation coordinates of each fruit in the base coordinate system The coordinates are the reference plane coordinates for the depth of the base, and can be 0 or the reference value obtained from the calibration.

[0036] S104: Determine the picking area based on the calculated depth information, and then perform the picking operation based on the determined area.

[0037] In this embodiment, the upper limit of the effective depth of operation of the parallel robotic arm without moving the lateral telescopic module is denoted as: The maximum compensation stroke of the lateral telescopic module is denoted as Therefore, set the partition threshold accordingly: in: For direct picking area threshold, To compensate for the upper limit threshold of the picking area, the rules for determining the picking area are as follows: in, For the first The depth of each fruit relative to the base of the parallel robotic arm is used to calculate the compensation distance when the area is determined to be a compensation picking zone. The compensation distance satisfies the following conditions: in, For the first The lateral scaling compensation distance corresponding to each fruit The gap distance is defined as the distance where the depth of the fruit exceeds the reachability of the parallel robotic arm itself. This is the maximum compensation stroke for the lateral telescopic module. This is a limiting function used to... Limited to the range The target extension amount of the lateral telescopic module is: ,in: The extension amount of the horizontal telescopic module to the control target of the k-th fruit; This represents the initial lateral extension. To compensate for the distance.

[0038] When a picking area is identified as a direct harvesting area, the central controller controls the parallel robotic arm to directly perform the harvesting action; when a picking area is identified as a compensation harvesting area, the central controller first controls the lateral telescopic module to... After moving forward to complete depth compensation, the parallel robotic arm is then controlled to perform the picking action.

[0039] S105 After harvesting, the parallel robotic arm releases the fruit to the follow-up conveyor belt, then resets or proceeds to harvest the next fruit. The follow-up conveyor belt transports the released fruit to the main collection channel, and then the main collection channel transports it to the collection box.

[0040] In this embodiment, regardless of whether the follower conveyor belt is in an extended or retracted state, its output end is always located within the vertical projection range of the main collection channel, realizing continuous relay transportation of the fruit.

[0041] S106, after all fruits at the current height station have been harvested, control the longitudinal double-speed lifting module to move to the next height station until the harvesting of all fruits at the three-level height station is completed.

[0042] In this embodiment, the target quantity to be harvested at the current workstation is used. As a criterion for workstation completion: ,in, Let represent the number of fruits to be picked after being identified and screened by the vision control system at the i-th height station. This is the target height for the next workstation. When the following conditions are met: in, This is the current height workstation number; The target quantity to be harvested at the top-level workstation.

[0043] In this embodiment, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0044] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0045] The above description is merely a specific embodiment of this application. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the protection scope of this application. The protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A globally adaptive parallel robotic arm for apple picking and follow-up collection, characterized in that, include: The system includes a central controller, a longitudinal speed-increasing module, a lateral telescopic module, a parallel robotic arm, and a vision control system mounted on the parallel robotic arm. The vision control system acquires the three-dimensional coordinates and depth information of the fruit. The central controller controls the longitudinal speed-increasing module to drive the harvesting platform to move sequentially to multiple height positions. Simultaneously, based on the three-dimensional coordinates and depth information acquired by the vision control system at each height position, it controls the lateral telescopic module to move the harvesting platform laterally. After the movement is completed, the central controller controls the parallel robotic arm to harvest the fruit and uses the follow-up fruit collection module to collect the harvested fruit.

2. The global adaptive parallel robotic arm apple picking and follow-up collection device according to claim 1, characterized in that, The longitudinal lifting module includes a longitudinal drive cylinder fixedly mounted on the side of the mobile frame column and a vertical guide rail fixedly mounted on the mobile frame. The push rod of the longitudinal drive cylinder is provided with a double-speed sprocket set. One end of the chain is mounted on the column of the mobile frame, and the other end of the chain passes around the double-speed sprocket set and is connected to the harvesting platform. The harvesting platform is slidably mounted on the vertical guide rail.

3. The global adaptive parallel robotic arm apple picking and follow-up collection device according to claim 2, characterized in that, The harvesting platform includes a horizontal frame and a lateral telescopic module. A horizontal linear guide rail is fixedly installed on the horizontal frame. The lateral telescopic module includes a lateral drive cylinder and a slide table. The cylinder body of the lateral drive cylinder is fixedly connected to the slide table base of the slide table. The horizontal frame is slidably connected to the slide table. The movable end of the push rod of the lateral drive cylinder is connected to a parallel robotic arm base fixedly installed on the horizontal frame. A parallel robotic arm and a rigid connecting arm are installed on the parallel robotic arm base. The rigid connecting arm is fixedly connected to a follow-up fruit collection module.

4. The global adaptive parallel robotic arm apple picking and follow-up collection device according to claim 3, characterized in that, The following fruit collection module includes a following conveyor belt, which is connected to a conveyor belt drive motor. Anti-drop baffles are provided at both ends of the following conveyor belt, and spacer baffles are provided at equal intervals on the surface of the following conveyor belt. A fruit receiving box is provided at the receiving end of the following conveyor belt. The frame of the fruit receiving box is fixedly connected to the first end of the rigid connecting arm, and the second end of the rigid connecting arm is fixedly connected to the base of the parallel robotic arm. The output end of the following conveyor belt is slidably connected to the moving frame through a linear slide mechanism. The output end of the following conveyor belt is located above the main collection channel, and a collection box is provided below the output end of the main collection channel.

5. The global adaptive parallel robotic arm apple picking and follow-up collection device according to claim 4, characterized in that, The parallel robotic arm is equipped with an end effector and a depth camera at its end, and the end effector is a flexible gripping structure robotic arm.

6. The global adaptive parallel robotic arm apple picking and follow-up collection device according to claim 2, characterized in that, The mobile frame is a gantry frame structure, and a mobile chassis is configured at the bottom of the mobile frame.

7. The global adaptive parallel robotic arm apple picking and follow-up collection device according to claim 1, characterized in that, The vision control system includes a depth camera and multiple position sensors. The depth camera and multiple position sensors are electrically connected to the input terminal of the central controller. The output terminal of the central controller is electrically connected to the lateral drive cylinder, the longitudinal drive cylinder and the parallel robotic arm driver. The parallel robotic arm driver is electrically connected to the parallel robotic arm.

8. A method for apple picking and follow-up collection using a fully adaptive parallel robotic arm, employing the fully adaptive parallel robotic arm apple picking and follow-up collection device as described in any one of claims 1-7, characterized in that, include: The central controller controls the longitudinal speed-increasing module to be placed at the initial target position, and at the same time controls the lateral telescopic module to retract to the initial position, thus completing the initialization operation. After initialization, the central controller controls the longitudinal speed-increasing module to drive the harvesting platform to the bottom, middle, and top height positions sequentially. The calculation formula is as follows: in, For the first The target height for each workstation. To hold a high position in a low-level job Interlayer spacing The vertical reachable radius of a single height station of a parallel robotic arm. This refers to the interlayer coverage overlap rate; The vision control system at each height station acquires the three-dimensional coordinates and depth information of each layer of fruit. The depth of the fruit from the parallel robotic arm base is calculated based on the acquired 3D coordinates and depth information, using the following formula: in, For the first The depth of each fruit relative to the base of the parallel robotic arm. For the first The orientation coordinates of each fruit in the base coordinate system The coordinates of the base depth reference plane; The harvesting area is determined based on the calculated depth information, and harvesting operations are carried out based on the determined area. After harvesting, the parallel robotic arm releases the fruit to the follow-up conveyor belt, then resets or proceeds to harvest the next fruit. The follow-up conveyor belt transports the released fruit to the main collection channel, and then the main collection channel transports it to the collection box. After all fruits at the current height station have been harvested, control the longitudinal double-speed lifting module to move to the next height station until the harvesting of all fruits at the three height stations is completed.

9. The method for apple picking and follow-up collection by a fully adaptive parallel robotic arm according to claim 8, characterized in that, The process of determining the harvesting area based on the calculated depth information and performing harvesting operations based on the determined area includes: If the calculated depth is greater than 0 and less than or equal to the first preset threshold, it is determined to be a direct picking area, and the central controller controls the parallel robotic arm to perform the picking operation. If the calculated depth is greater than the first preset threshold and less than or equal to the second preset threshold, it is determined to be a compensation picking area, and the compensation distance is calculated using the following formula: in, For the first The lateral scaling compensation distance corresponding to each fruit The gap distance is defined as the distance where the depth of the fruit exceeds the reachability of the parallel robotic arm itself. This is the maximum compensation stroke for the lateral telescopic module. This is a limiting function used to... Limited to the range Inside; The horizontal telescopic module drives the parallel robotic arm to move forward based on the calculated compensation distance and then controls the movement of the parallel robotic arm to perform the picking operation. During the movement, the rigid connecting arm drives the follower conveyor belt to slide horizontally synchronously along the linear slide mechanism, so that the fruit receiving end of the follower conveyor belt is always located below the working space of the parallel robotic arm.