Vision-tactile fusion active avoidance tomato picking hand and method

Abstract

The application provides a tomato picking hand with active avoidance of visual and tactile fusion, which comprises a perception component, a posture adjustment component, a shearing execution component, a control component, a sleeve frame, a fixing frame, a main body support and a driving part; the visual information feedback obtained by the perception component enables the picking hand to have two degrees of freedom for adjusting the posture; the posture adjustment component adjusts the posture of the picking hand according to the visual information feedback, reduces the accuracy requirement for the initial picking pose, guarantees the adaptability to tomatoes growing in different environments and improves the picking success rate; the control component executes corresponding active avoidance actions through the tactile feedback data obtained by the perception component, so that the damage to the fruits and their growing environment is avoided; and the perception component provides real-time information in the whole picking process, thereby improving the reliability of the picking hand.

Description

Technical Field

[0001] This invention relates to the field of agricultural harvesting equipment technology, specifically to a visual-tactile fusion method and technique for actively avoiding tomato harvesters. Background Technology

[0002] With the continuous improvement of agricultural modernization, facility-based and large-scale planting methods are gradually expanding, especially for agricultural products such as tomatoes. However, currently, greenhouse tomatoes are still largely harvested manually, one by one, which has problems such as low efficiency and high labor intensity. Therefore, it is necessary to introduce robotic automated harvesting into the harvesting of agricultural products such as tomatoes.

[0003] Patent document CN119427409A discloses an intelligent harvester based on an improved YOLOv5 model, including a base and a drive controller. The base and drive controller are connected by a linkage mechanism. A harvesting turntable is provided at the other end of the drive controller. A tray is provided on the side of the harvesting turntable away from the drive controller. Multiple mechanical claws are provided on the circumference of the tray. Each mechanical claw is slidably mounted on the slide rail of the harvesting turntable via a moving block assembly on its lower side. The mechanical claws are rotatably connected to the moving block assembly. All mechanical claws and the tray together form a harvesting space. A camera is installed on the drive controller, and the improved YOLOv5 model is deployed in the camera. During harvesting, the drive controller identifies and locates the fruit to be harvested through the camera, then controls the linkage mechanism to rotate and move closer to the fruit. Subsequently, it controls the rotation of the mechanical claws on the moving block assembly and the sliding of the moving block assembly on the harvesting turntable to expand the harvesting space to cover the fruit and then shrink the harvesting space to clamp the fruit. Finally, it controls the harvesting turntable to rotate to complete the harvesting. However, the harvesting method described in this patent document still relies on a camera to locate the target tomato, which makes it difficult to guarantee the reliability of obtaining the tomato's pose from visual information, and the end effector has no other degrees of freedom to adapt to the complex tomato growing environment.

[0004] A Chinese patent with publication number CN207040242U discloses a machine vision-based intelligent tomato harvesting device, including a base. A weighing sensor is installed at the bottom of the inner cavity of the storage box. An alarm device is installed on the right side of the storage box. Vertical bars are installed at the top and bottom of the horizontal bar of the folded support rod. Lights are installed on the top left side of the top vertical bar and the bottom left side of the bottom vertical bar. A laser rangefinder and a color recognition module are respectively installed at the center left side of the top vertical bar and the center left side of the bottom vertical bar. The device uses a camera, laser rangefinder, and color recognition module to collect tomato information, improving the accuracy of identifying ripe tomatoes. Furthermore, when the weight of the tomatoes in the storage box reaches a set value, the alarm device sounds, serving as a reminder and preventing the tomato box from overflowing due to overfilling. Remote monitoring of the harvesting process is achieved through a wireless communication module.

[0005] Current tomato harvesting robots mostly rely on depth cameras to acquire the spatial pose of the target tomato for localization, and then the robotic arm moves the harvesting actuator to the target location for picking. However, the actual tomato growing environment is complex, with intertwined stems and leaves, and the fruit often being obscured. Furthermore, the light conditions for tomatoes in different locations often vary. Obtaining accurate visual information is difficult, the reliability of visual information is low, and it is difficult to achieve good harvesting results with only single-modal information. In unstructured environments, higher demands are placed on the dexterity of the harvesting actuator. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this invention is to provide a method for actively avoiding tomato pickers by integrating visual and tactile fusion.

[0007] According to the present invention, a visual-tactile fusion active avoidance tomato picker includes: a sensing component, a posture adjustment component, a cutting execution component, a control component, a frame, a fixing frame, a main support component, and a driving component;

[0008] The fixing frame and the sleeve frame are respectively arranged on adjacent sides of the main support member; The sensing component is used to acquire visual information and contact force information of the target fruit bunch in two modalities; The posture adjustment component is used to drive the frame to adjust the first posture and the second posture based on the visual information data of the target fruit bunch; The control component is used to drive the frame to move based on the contact force information of the target fruit bunch; The driver is connected to the cutting execution component, which is installed inside the frame. The driver is used to drive the cutting execution component to cut the target fruit string.

[0009] Preferably, the attitude adjustment component includes a first attitude adjustment unit and a second attitude adjustment unit; The first posture adjustment part is disposed between the main support member and the fixed frame. The first posture adjustment part is used to drive the main support member to rotate around the first rotation axis. The extension direction of the first rotation axis is the orientation of the mounting surface of the main support member on the fixed frame, thereby adjusting the first posture. The second posture adjustment part is disposed between the main support member and the sleeve frame. The second posture adjustment part is used to drive the sleeve frame to rotate around the second rotation axis. The extension direction of the second rotation axis is the orientation of the mounting surface of the main support member on which the sleeve frame is mounted, thereby adjusting the second posture.

[0010] Preferably, the first attitude adjustment part is a motor, which is mounted on the fixed frame, and the output end of the motor is connected to the main support member; The second attitude adjustment unit includes a stepper motor, an input gear, an output gear, a bearing, and a pressure cover; The stepper motor is fixed on the main support member, the input gear is connected to the output end of the stepper motor, and the output gear meshes with the input gear. The output gear is mounted on the bearing, and the bearing is connected to the cover; the cover is used to support the bearing, and the bearing is used to support the output gear. The output gear is fixedly connected to the sleeve frame, and the output gear rotates synchronously with the sleeve frame.

[0011] Preferably, the pressure cap is provided with a bearing support surface and an output gear through hole, the bearing support surface is used to fix the bearing, and the output gear through hole is used to place the output gear; The output gear is provided with a drive component through hole, a bearing mating surface, and a sleeve frame mating surface; the drive component through hole is used to allow the drive end of the drive component to pass through, the bearing mating surface is fixedly connected to the bearing, and the sleeve frame mating surface is fixedly connected to the sleeve frame.

[0012] Preferably, the sensing components include a depth camera and a contact sensor; The depth camera is mounted on the fixed frame and is used to acquire visual information data of the target fruit bunch; The contact sensor is installed around the inner side of the frame and is used to acquire contact force information of the target fruit bunch.

[0013] Preferably, the driving component is a cylinder, and the shearing actuator is mounted on the piston rod end of the cylinder, the cylinder being used to drive the shearing actuator to move linearly.

[0014] Preferably, the shearing actuator includes a guide frame, an upper pressure plate, a blade, and a lower support block; The upper pressure plate and the lower support block are installed on one side of the guide frame, the blade is installed between the upper pressure plate and the lower support block, and the other side of the guide frame is connected to the drive component.

[0015] Preferably, the lower support block is provided with a gathering plate, which is used to push the target fruit bunch toward the middle of the blade.

[0016] Preferably, the lower support block is provided with a clamping contact area, which is used to cooperate with one side of the frame to clamp the fruit stem of the target fruit bunch.

[0017] According to the present invention, a visual-tactile fusion active avoidance tomato picking method is provided, employing the visual-tactile fusion active avoidance tomato picking hand described in any one of the above-mentioned methods, comprising: The perception component acquires visual information data about the target fruit string; The posture adjustment component drives the frame to adjust the first posture and the second posture based on the visual information data obtained by the sensing component; The control component controls the picking hand to move the frame directly below the fruit bunch, and the control component controls the picking hand to move vertically upward; During the vertical upward movement of the picking hand, the sensing component acquires real-time data on the contact force information of the target fruit bunch; When the data of the contact force information of the target fruit bunch reaches the set contact force threshold, the control component controls the picking hand to move the frame a fixed distance to the opposite side of the contact surface that has reached the set contact force threshold. The frame moves to the position of the stem of the target fruit bunch, and the target fruit bunch is completely inside the frame. The driving component drives the cutting execution component to move and cut the target fruit bunch.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention utilizes visual information feedback acquired by a sensing component, along with a posture adjustment component that grants the harvester two degrees of freedom to adjust their posture. The posture adjustment component adjusts the harvester's posture based on visual feedback, reducing the precision requirements for the initial harvesting position, ensuring adaptability to tomatoes in different growing environments, and increasing the harvesting success rate. The control component executes corresponding active avoidance actions based on tactile feedback data acquired by the sensing component, preventing damage to the fruit and its growing environment. The sensing component provides real-time information throughout the harvesting process, improving the reliability of the harvester. Attached Figure Description

[0019] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 This is a schematic diagram of the first angle structure of the present invention, which mainly embodies the visual-tactile fusion to actively avoid tomato pickers. Figure 2 This is a schematic diagram of the second angle structure of the present invention, which mainly embodies the visual-tactile fusion to actively avoid the tomato picker. Figure 3 This is a schematic diagram of the top structure of the invention, which mainly embodies the visual-tactile fusion and actively avoids tomato pickers. Figure 4 This is a schematic diagram illustrating the AA cross-sectional structure of the present invention; Figure 5This is a schematic diagram illustrating the structure of the pressure cap, which is the main feature of this invention. Figure 6 This is a schematic diagram illustrating the structure of the shearing execution component, which is the main feature of this invention. Figure 7 This is a schematic diagram illustrating the structure of the output gear, which is the main feature of this invention. Figure 8 This is a schematic diagram illustrating the structure of the present invention, which mainly embodies the visual-tactile fusion of the active avoidance mechanism for tomato pickers adjusting their first posture during cutting. Figure 9 This is a schematic diagram illustrating the structure of the present invention, which mainly embodies the visual-tactile fusion of the active avoidance mechanism for tomato pickers to adjust their second posture when performing cutting. Figure 10 This is a partial structural diagram illustrating the active avoidance mechanism of a tomato picker during shearing, which is mainly based on visual-tactile fusion. Figure 11 This is a schematic diagram illustrating the structure of the tomato picker during the cutting process, which mainly embodies the fusion of visual and tactile senses.

[0020] The diagram shows: 1. Fixture; 2. Depth camera; 3. Motor; 4. Main support component; 5. Stepper motor; 6. Drive component; 601. Piston rod; 7. Input gear; 8. Output gear; 801. Drive component through hole; 802. Bearing mating surface; 803. Sleeve frame mating surface; 9. Bearing; 10. Pressure cap; 1001. Bearing support surface; 1002. Output gear through hole; 11. Guide frame; 12. Upper pressure plate; 13. Blade; 14. Lower support block; 1401. Clamping contact area; 1402. Gathering plate; 15. Contact sensor; 16. Sleeve frame; 17. Target fruit bunch. Detailed Implementation

[0021] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0022] like Figure 1 As shown, a visual-tactile fusion active avoidance tomato picker provided by the present invention includes: a sensing component, a posture adjustment component, a cutting execution component, a control component, a frame 16, a fixing frame 1, a main support component 4, and a driving component 6. The fixing frame 1 and the sleeve frame 16 are respectively set on the adjacent sides of the main support member 4; The perception component is used to acquire visual and contact force data of the target fruit bunch; these are used to locate the target fruit bunch and actively avoid it.

[0023] The posture adjustment component is used to drive the frame 16 to adjust the first posture and the second posture based on the visual information data of the target fruit bunch. Specifically, the posture adjustment component drives the main support 4 to rotate together with the frame 16 relative to the fixed frame 1, thereby adjusting the first posture; the posture adjustment component drives the frame 16 to rotate relative to the main support 4, thereby adjusting the second posture.

[0024] The control component is used to control the movement of the picking hand and to control the movement of the frame 16 based on the contact force information of the target fruit bunch.

[0025] The driving component 6 is installed on the main support component 4. The driving end of the driving component 6 passes through the sleeve frame 16 and is connected to the shearing execution component. The shearing execution component is installed inside the sleeve frame 16. The driving component is used to drive the shearing execution component to move and shear the target fruit bunch.

[0026] The sensing component acquires visual information and contact force information of the target fruit bunch in two modalities. Based on the visual information of the target fruit bunch acquired by the sensing component, the posture adjustment component drives the frame 16 to adjust the first posture and the second posture. The adjusted frame 16 adapts to the growth posture of the target fruit bunch and minimizes the contact between the fruit, fruit stem and frame 16.

[0027] The control component controls the movement of the picking hand, causing the target fruit bunch to enter the frame 16. The sensing component acquires the contact force information of the target fruit bunch in real time. When the acquired contact force information of the target fruit bunch reaches the set contact force threshold, the control component controls the picking hand to move the frame 16 a fixed distance to the opposite side of the contact surface of the target fruit bunch, performing an active avoidance action.

[0028] After the avoidance is completed, the target fruit bunch is in a suitable cutting position within the frame 16. The drive component 6 drives the cutting execution component to move and cut the target fruit bunch.

[0029] This invention designs a visual-tactile fusion active avoidance mechanism for tomato harvesters. Through visual information feedback acquired by a sensing component, a posture adjustment component provides the harvester with two degrees of freedom to adjust their posture. The posture adjustment component adjusts the harvester's posture based on visual feedback, reducing the precision requirements of the initial harvesting position, ensuring adaptability to tomatoes in different growing environments, and improving the harvesting success rate. The control component executes corresponding active avoidance actions based on tactile feedback data acquired by the sensing component, preventing damage to the fruit and its growing environment. The sensing component provides real-time information throughout the harvesting process, improving the reliability of the harvester.

[0030] In some feasible implementations, the attitude adjustment component includes a first attitude adjustment unit and a second attitude adjustment unit; The first posture adjustment unit is located between the main support member 4 and the fixed frame 1. The first rotation axis is a center line of the main support member 4. The extension direction of the first rotation axis is the orientation of the mounting surface of the fixed frame 1 on which the main support member 4 is mounted. The first posture adjustment unit is used to drive the main support member 4 to rotate around the first rotation axis and adjust the first posture.

[0031] The second attitude adjustment unit is located between the main support member 4 and the sleeve frame 16. The other center line of the main support member 4 is the second rotation axis, and the extension direction of the second rotation axis is the orientation of the mounting surface of the main support member mounting the sleeve frame 16. The second attitude adjustment unit is used to drive the sleeve frame 16 to rotate around the second attitude adjustment unit to adjust the second attitude.

[0032] The first and second rotation axes are perpendicular to each other, thus enabling the frame 16 to rotate in two degrees of freedom.

[0033] In some feasible implementations, the first attitude adjustment unit is a motor 3, which is mounted on the fixed frame 1. The output end of the motor 3 is connected to the main support member 4, and the motor 3 is used to drive the main support member 4 and the output shaft of the motor 3 to rotate, thereby realizing the first attitude adjustment. Specifically, the motor 3 is mounted on the side of the fixed frame 1 and embedded in the main support member 4. The main support member 4 has a space to accommodate the motor 3, so that the main support member 4 can rotate synchronously without interference under the drive of the motor 3. Specifically, the motor 3 is fixed to the fixed frame 1 by bolts.

[0034] The second attitude adjustment unit includes a stepper motor 5, an input gear 7, an output gear 8, a bearing 9, and a cover 10.

[0035] The stepper motor 5 is fixed to the main support member 4. The input gear 7 is connected to the output end of the stepper motor 5, and the output gear 8 meshes with the input gear 7. The stepper motor 5 drives the input gear 7 to rotate, and the rotation of the input gear 7 causes the output gear 8, which meshes with the input gear 7, to rotate together. Specifically, the stepper motor 5 is fixed to the main support member 4 by bolts. Specifically, the input gear 7 and the output shaft of the stepper motor 5 are fitted together as a single unit.

[0036] The output gear 7 is mounted on the bearing 9 and is fixedly fitted on the inner ring of the bearing 9. The output gear 7 and the inner ring of the bearing 9 rotate synchronously. The bearing 9 is mounted on the cover 10. The cover 10 is used to support the bearing 9, and the bearing 9 is used to support the output gear 7. Specifically, the cover 10 is fixed to the main support member 4 by bolts.

[0037] The output gear 8 is fixedly connected to the sleeve 16. The output gear 8 and the sleeve 16 rotate synchronously to realize the second posture adjustment of the sleeve 16.

[0038] In some feasible embodiments, the cover is provided with a bearing support surface 1001 and an output gear through hole 1002. The bearing support surface 1001 is used to fix the bearing 9, and the output gear through hole 1002 is used to install the output gear 8.

[0039] The output gear 8 is provided with a drive component through hole 801, a bearing mating surface 802, and a sleeve mating surface 803; the drive end of the drive component extends and retracts through the drive component through hole 801, and the drive component through hole 801 allows the drive end of the drive component to pass through without interference.

[0040] The bearing mating surface 802 is fixedly connected to the bearing 9, thereby achieving a fixed connection between the bearing 9 and the output gear 8. The bearing 9 is fixed to the bearing mating surface 802, thus enabling the bearing 9 to support the output gear 7.

[0041] The sleeve frame mating surface 803 is fixedly connected to the sleeve frame 16, thereby achieving a fixed connection between the sleeve frame 16 and the output end gear 8.

[0042] The output gear 8 extends outward through the output gear through hole 1002 to the sleeve engagement surface 803. The sleeve 16 engages with the sleeve engagement surface 803, thereby achieving a fixed connection between the output gear 8 and the sleeve 16. Consequently, the output gear 8 and the sleeve 16 rotate synchronously.

[0043] In some feasible implementations, the sensing components include a depth camera 2 and a contact sensor 15.

[0044] Depth camera 2 is mounted on the mounting bracket 1. Depth camera 2 is used to acquire visual information data of the target fruit bunch; specifically, depth camera 2 is fixed to the mounting bracket 1 by bolts.

[0045] Multiple contact sensors 15 are mounted on the inner periphery of the non-installed shearing actuator of the frame 16. The number and model of the sensors on the front, left and right sides of the frame 16 are the same. The contact sensors 15 are used to acquire data on the contact force information of the target fruit bunch.

[0046] Specifically, there are 6 contact sensors 15, with two on the front, two on the left, and two on the right of the frame 16. The contact sensors 15 on the same side are grouped together, resulting in a total of three groups of sensors. This enables contact sensing within the frame 16 and executes corresponding active avoidance actions based on the feedback data from the contact sensors 15, thus preventing damage to the target fruit bunch and its growing environment.

[0047] When any contact sensor 15 in a set of sensors detects that the contact force has reached a set threshold, it is considered that a touch has occurred on that side, and the control component then controls the harvester to move a fixed distance to the opposite side. Specifically, when a collision is detected by a set of sensors on the left side, the control component controls the harvester to move to the right; when a collision is detected by a set of sensors on the front side, the control component controls the harvester to move to the rear.

[0048] By using visual and tactile feedback, the harvester's posture and harvesting actions are adjusted in real time, ensuring the adaptability of tomatoes to different growing environments and improving the harvesting success rate.

[0049] In some feasible implementations, the driving component 6 is a cylinder, and the shearing actuator is mounted on the piston rod 601 end of the cylinder. The movement of the piston rod 601 drives the shearing actuator to move linearly, completing the shearing and clamping actions. Driven by the cylinder, the shearing and clamping actions can be completed quickly, improving harvesting efficiency. Specifically, the cylinder is fixedly connected to the main support component 4 by bolts.

[0050] In some feasible implementations, the shearing execution assembly includes a guide frame 11, an upper pressure plate 12, a blade 13, and a lower support block 14; the upper pressure plate 12 and the lower support block 14 are mounted on one side of the guide frame 11, the blade 13 is mounted between the upper pressure plate 12 and the lower support block 14, and the other side of the guide frame 11 is connected to the drive member 6.

[0051] Specifically, the lower support block 14 is fixedly connected to the guide frame 11, the blade 13 is connected to the lower support block 14, and the upper pressure plate 12 is connected to the lower support block 14. The guide frame 11, the upper pressure plate 12, the blade 13, and the lower support block 14 are fixedly connected as a whole by bolts. The shearing execution assembly is a whole with no moving parts. The drive component 6 extends to drive the shearing execution assembly to move linearly. When the guide frame 11 drives the upper pressure plate 12, the blade 13, and the lower support block 14 to move together to one side of the sleeve frame 16 and engage with the sleeve frame 16, the shearing and clamping action is completed.

[0052] In some feasible implementations, a gathering plate 1402 is provided on the lower support block 14. The gathering plate 1402 is used to push the target fruit bunch towards the middle position of the blade 13 to avoid the fruit of the target fruit bunch from colliding with the picker and to ensure the cutting effect.

[0053] In some feasible implementations, the lower support block 14 is provided with a clamping contact area 1401, which is used to cooperate with a contact sensor on one side of the frame 16 to clamp the fruit stem of the target fruit bunch.

[0054] In some feasible implementations, the clamping contact area 1401 provided on the lower support block 14 is also provided at the corresponding position of the contact sensor 15 on the front side of the sleeve frame 16. The two clamping contact areas are used to cooperate in clamping the fruit stems of the target fruit bunch.

[0055] This invention also provides a visual-tactile fusion active avoidance tomato picking method, employing any of the above-mentioned visual-tactile fusion active avoidance tomato picking hands, including: The depth camera 2 acquires visual information data of the target fruit string 17, the motor 3 drives the main support 4 to rotate around the output shaft of the motor 3, and the frame 16 rotates together with the main support 4 to achieve the first posture adjustment of the frame 16.

[0056] Stepper motor 5 drives the input gear 7 to rotate. The rotation of the input gear 7 drives the output gear 8, which meshes with the input gear 7, to rotate together. The output gear 8 drives the sleeve 16 to rotate synchronously, thereby realizing the second posture adjustment of the sleeve 16.

[0057] At this time, the harvester adapts to the growth posture of the target fruit bunch and minimizes the contact between the frame 16 and the fruit and stem when it is inserted into the target fruit bunch.

[0058] The control component controls the movement of the harvesting arm, which moves the frame 16 directly below the target fruit bunch 17, and then moves vertically upward at a certain speed, fitting the target fruit bunch 17 into the frame 16. The harvesting arm then continues to move the frame 16 vertically upward. Specifically, the control component is a robotic arm that controls the movement of the harvesting arm.

[0059] As the frame 16 moves vertically upward, the contact sensor 15 installed inside the frame 16 detects contact in real time. When the contact force information obtained by the contact sensor 15 on one side reaches the set contact force threshold, the control component controls the picking hand to actively move a fixed distance to the opposite side to perform an active avoidance action.

[0060] In one feasible implementation, the contact sensor 15 is provided with a contact force threshold and a maximum number of adjustments. That is, when the number of times to avoid to the right or left exceeds 2 at this height position, the side with the smaller contact force will no longer actively avoid it. At this time, the target fruit string 17 is close to the contact sensor 15 position on the side with the smaller contact force.

[0061] When the contact force between the left-side and front-side contact sensors both exceeds a set threshold, the control component controls the sleeve frame 16 to move a fixed distance to the left front of the sleeve frame, with a reference distance value of 10mm. Adjustments are stopped after more than two attempts to avoid damage to the main vine.

[0062] When the contact force between the right-side and front-side contact sensors both exceeds a set threshold, the control component controls the sleeve frame 16 to move a fixed distance to the right front of the sleeve frame, with a reference distance of 10mm. Adjustments are stopped after more than two attempts to avoid damage to the main vine.

[0063] The frame 16 moves to the fruit stem position, at which point the target fruit bunch 17 is fully inside the frame 16. The cylinder drives the shearing actuator on the piston rod 601 to move inside the frame 16. The blade 13 cuts the fruit stem, and the lower support block 14 and the contact sensor 15 located on the front side of the frame 16 squeeze and cooperate to complete the clamping of the target fruit bunch.

[0064] In the description of this application, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.

[0065] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A visual-tactile fusion method for actively avoiding tomato pickers, characterized in that, include: Sensing component, attitude adjustment component, shearing execution component, control component, frame (16), fixing frame (1), main support component (4), driving component (6); The fixing frame (1) and the sleeve frame (16) are respectively arranged on adjacent sides of the main support member (4); The sensing component is used to acquire visual information and contact force information of the target fruit bunch (17) in two modalities; The posture adjustment component is used to drive the frame (16) to adjust the first posture and the second posture based on the visual information data of the target fruit bunch (17); The control component is used to drive the frame (16) to move based on the contact force information of the target fruit bunch (17); The driving component (6) is connected to the cutting execution component, which is installed inside the frame (16). The driving component (6) is used to drive the cutting execution component to cut the target fruit string (17).

2. The visual-tactile fusion active avoidance tomato picker as described in claim 1, characterized in that, The attitude adjustment component includes a first attitude adjustment unit and a second attitude adjustment unit; The first posture adjustment part is disposed between the main support member (4) and the fixed frame (1). The first posture adjustment part is used to drive the main support member (4) to rotate around the first rotation axis. The extension direction of the first rotation axis is the orientation of the mounting surface of the fixed frame (1) on which the main support member (4) is mounted, and to adjust the first posture. The second posture adjustment part is disposed between the main support member (4) and the sleeve frame (16). The second posture adjustment part is used to drive the sleeve frame (16) to rotate around the second rotation axis. The extension direction of the second rotation axis is the orientation of the mounting surface of the main support member (4) on which the sleeve frame (16) is mounted, and to adjust the second posture.

3. The visual-tactile fusion active avoidance tomato picker as described in claim 2, characterized in that, The first attitude adjustment part is a motor (3), the motor (3) is mounted on the fixed frame (1), and the output end of the motor (3) is connected to the main support member (4); The second attitude adjustment unit includes a stepper motor (5), an input gear (7), an output gear (8), a bearing (9), and a cover (10). The stepper motor (5) is fixed on the main support member (4), the input end gear (7) is connected to the output end of the stepper motor (5), and the output end gear (8) meshes with the input end gear (7); The output gear (8) is mounted on the bearing (9), and the bearing (9) is connected to the cover (10); the cover (10) is used to support the bearing (9), and the bearing (9) is used to support the output gear (8). The output gear (8) is fixedly connected to the sleeve (16), and the output gear (8) and the sleeve (16) rotate synchronously.

4. The visual-tactile fusion active avoidance tomato picker as described in claim 3, characterized in that, The pressure cap is provided with a bearing support surface (1001) and an output gear through hole (1002). The bearing support surface (1001) is used to fix the bearing (9) and the output gear through hole (1002) is used to place the output gear (8). The output gear (8) is provided with a drive component through hole (801), a bearing mating surface (802), and a sleeve frame mating surface (803); the drive component through hole (801) is used to allow the drive end of the drive component (6) to pass through, the bearing mating surface (802) is fixedly connected to the bearing (9), and the sleeve frame mating surface (803) is fixedly connected to the sleeve frame (16).

5. The visual-tactile fusion active avoidance tomato picker as described in claim 1, characterized in that, The sensing components include a depth camera (2) and a contact sensor (15); The depth camera (2) is mounted on the fixed frame (1) and is used to acquire visual information data of the target fruit bunch (17); The contact sensor (15) is installed around the inner side of the frame (16) and is used to acquire data on the contact force information of the target fruit bunch (17).

6. The visual-tactile fusion active avoidance tomato picker as described in claim 1, characterized in that, The driving component (6) is a cylinder, and the shearing execution assembly is installed at the piston rod (601) end of the cylinder. The cylinder is used to drive the shearing execution assembly to move linearly.

7. The visual-tactile fusion active avoidance tomato picker as described in claim 1, characterized in that, The shearing execution assembly includes a guide frame (11), an upper pressure plate (12), a blade (13), and a lower support block (14). The upper pressure plate (12) and the lower support block (14) are installed on one side of the guide frame (11), the blade (13) is installed between the upper pressure plate (12) and the lower support block (14), and the other side of the guide frame (11) is connected to the drive member (6).

8. The visual-tactile fusion active avoidance tomato picker as described in claim 7, characterized in that, A gathering plate (1402) is provided on the lower support block (14), which is used to push the target fruit bunch (17) towards the middle position of the blade (13).

9. The visual-tactile fusion active avoidance tomato picker as described in claim 7, characterized in that, The lower support block (14) is provided with a clamping contact area (1401), which is used to cooperate with one side of the frame (16) to clamp the fruit stem of the target fruit bunch (17).

10. A visual-tactile fusion method for actively avoiding obstacles during tomato picking, characterized in that, The active avoidance of tomato pickers using visual-tactile fusion as described in any one of claims 1 to 9 includes: The perception component acquires visual information data of the target fruit string (17); The posture adjustment component drives the frame (16) to adjust the first posture and the second posture based on the visual information data obtained by the perception component; The control component controls the picking hand to move the frame (16) directly below the fruit bunch, and the control component controls the picking hand to move vertically upward; During the vertical upward movement of the picking hand, the sensing component acquires the contact force information of the target fruit bunch (17) in real time; When the data of the contact force information of the target fruit bunch (17) reaches the set contact force threshold, the control component controls the picking hand to move the frame (16) a fixed distance to the opposite side of the contact surface that has reached the set contact force threshold. The frame (16) moves to the stem position of the target fruit bunch (17), and the drive (6) drives the cutting execution component to move to cut the target fruit bunch (17).

Images