Intelligent three-dimensional stock bin automatic storing and taking device and method

Abstract

This invention discloses an intelligent three-dimensional silo automatic storage and retrieval device and method, which solves the technical defects of existing vacuum suction cup storage and retrieval devices, such as interlayer adhesion of boards, difficulty in single-sheet retrieval, and easy damage to boards. The device includes an intelligent three-dimensional silo, a robotic arm, and a transfer assembly. The robotic arm includes an execution end, a central suction cup group, a movable end suction cup group, a pneumatic module, a vision recognition module, and a control module. The method achieves automated storage and retrieval through material retrieval and storage operations. During the material retrieval process, the length of the board is first identified and the suction cup is positioned. After initial adsorption and end lifting, the deformation size of the board is detected. Combined with step adjustment and multi-parameter adaptive correction, interlayer adhesion is resolved, achieving accurate single-sheet retrieval. Therefore, this invention effectively solves the problem of board adhesion, reduces the risk of board damage, adapts to different specifications and rigidities of wood boards, and improves storage and retrieval efficiency and stability.

Description

Technical Field

[0001] This invention relates to the field of transportation and storage device processing technology, specifically to an intelligent three-dimensional silo automatic storage and retrieval device and method. Background Technology

[0002] Intelligent automated storage and retrieval systems (AS / RS) are core equipment for the automated storage, scheduling, and retrieval of wood-based panels. They fall under the category of transportation or storage devices, such as loading or unloading conveyors, workshop conveyor systems, or pneumatic pipeline conveyors. These systems can efficiently classify and store wood-based panels of different colors and sizes, and are widely used in automated production lines for furniture manufacturing, whole-house customization, and panel processing. To improve surface texture and performance, wood-based panels typically undergo surface treatments such as sanding, painting, and lamination before being stored, resulting in a smooth, dense surface with high flatness.

[0003] To address the needs of storing and retrieving this type of board material, existing technologies include various automated gripping solutions such as vacuum suction cup robotic arms, clamp-type, and magnetic gripping. Clamp-type devices rely on mechanical grippers to hold the edges of the board, which can easily cause crush damage to the corners and edges, and is not suitable for thin or easily deformable wood-based panels. Magnetic gripping devices are only suitable for composite boards with metal substrates and cannot effectively grip non-magnetic boards such as pure wood or laminated boards, limiting their applicability. Vacuum suction cup robotic arms, due to their high gripping efficiency and minimal damage to the board surface, are the mainstream solution for automated storage and retrieval of wood-based panels.

[0004] However, such devices have the following problems in actual operation: When smooth, surface-treated sheets are stacked, the air between layers is easily squeezed out, and adjacent sheets adhere tightly to form a negative pressure adsorption between layers, resulting in sheet adhesion; when the robotic arm grabs the upper sheet, the negative pressure between layers will simultaneously pull the lower sheet, causing an abnormally increased gripping resistance, making it impossible to achieve accurate single-sheet picking. If the gripping is forcibly lifted, it can easily cause sheet deformation, edge chipping, and surface scratches. In severe cases, it can also cause the vacuum suction cup to overload and fall off, and the robotic arm to deviate in positioning, affecting the stability of picking and the sheet yield. Moreover, the frequent adhesion failures can lead to disordered storage and retrieval cycles and shutdown of automated production lines. Therefore, the existing common suction cup robotic arms are difficult to meet the continuous production requirements of high density, high reliability, and high surface quality. Summary of the Invention

[0005] The purpose of this invention is to provide an intelligent three-dimensional silo automatic storage and retrieval device and method, which does not have at least one of the above-mentioned disadvantages.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an intelligent three-dimensional silo automatic storage and retrieval method, comprising the following steps: S1. Material handling: The intelligent three-dimensional material bin receives the material handling instruction, the robotic arm moves to the top of the stack of boards to be picked, the vision recognition module identifies the length of the top layer of boards and transmits it to the control module; S1.1 Suction Cup Positioning: The control module controls the end suction cup group to move to the standard suction position according to the length of the board. The distance between the standard position and the position of the middle suction cup group is the standard distance L. The standard distance L is positively correlated with the length of the board. S1.2 Initial Adsorption: The actuator drives the middle and end suction cup assemblies to move down synchronously to complete the initial adsorption; S1.3, End Lifting: The pneumatic module drives the end suction cup assembly to lift with standard air pressure P0, while the middle suction cup assembly is fixed, lifting only the two ends of the sheet metal; S1.4 Deformation Detection and Judgment: The visual recognition module identifies the maximum deformation size at both ends of the board, which is the upward movement distance of both ends of the board. The maximum deformation size is compared with the threshold A0: If the deformation size is ≥ threshold A0, the robotic arm lifts the entire board and transfers it to the preset position to complete the material picking; if the deformation size is < threshold A0, the end suction cup group is de-aired and reset, and the end suction cup group moves away from the middle suction cup group by a step distance Z. Each time the end suction cup group moves, the actions of S1.2-S1.3 are repeated until the deformation size is ≥ threshold A0, and then the entire board is lifted to complete the material picking. S2, Material Storage Operation: Move stacked boards to empty storage locations by transferring components.

[0007] Furthermore, in step S1.4, when the initial deformation dimension d0 is less than the threshold A0, and the deformation dimension d1 of the end suction cup assembly is still less than the threshold A after the first step movement, the control module calculates the deformation coefficient a = (d1-d0) / d0, and corrects the threshold A0 to A1 = b*A0*a0 / a through the deformation coefficient a, where a0 is the reference threshold of the deformation coefficient a, and b is the pre-input adjustment coefficient.

[0008] Furthermore, during the stepping process in step S1.4, the control module corrects the standard air pressure P0 in the pneumatic module to B1 times, and B1=c*A1 / A0, where c is the adjustment coefficient.

[0009] Furthermore, when the initial deformation dimension d0 is less than the threshold A0, and the deformation dimension d1 of the end suction cup assembly is still less than the threshold A after the first step movement, the control module corrects the subsequent step distance Z1=e*Z0*a0 / a according to the deformation coefficient a, where Z0 is the reference threshold for the step distance Z, and e is the adjustment coefficient.

[0010] Furthermore, when the initial deformation dimension d0 is less than the threshold A0, and the deformation dimension d1 of the end suction cup group is still less than the threshold A after the first step movement, the control module will control the total number of steps to be three, and during the third step, control the end suction cup group to step to the end position. The distance between the end position and the middle suction cup group is 1 / 5-2 / 5 of the length of the plate, and at this time the actual air pressure B1*P0 in the pneumatic module and the threshold is A1.

[0011] Furthermore, the intelligent three-dimensional silo has several compartments. When the robotic arm performs its first material removal from the stacked boards in the compartment, it executes steps S1.2-S1.3. After successfully removing the top board from the compartment, the control module records the compartment's position information, the final actual air pressure of the pneumatic module, the final distance between the end suction cup group and the middle suction cup group, and the threshold A1. The control module then creates a mapping table between the position information, the actual air pressure value, the final distance, and the threshold A1. The data for the compartment is only updated after all the stacked boards in the compartment have been removed.

[0012] Furthermore, in the mapping table, the final actual air pressure of the pneumatic module increases linearly as the height of the plate being removed decreases.

[0013] Furthermore, in step S1.1, the standard distance L is 1 / 10 to 1 / 5 of the length of the board.

[0014] This invention also discloses an intelligent automated storage and retrieval device for a three-dimensional silo, used to implement an automated storage and retrieval method for an intelligent automated storage and retrieval silo, comprising: The intelligent automated storage and retrieval system is equipped with several storage compartments for storing sheet metal. At least one set of robotic arms, which are used to grab and transfer sheet metal from the corresponding compartment; The robotic arm includes an actuator, a central suction cup assembly, at least two sets of end suction cup assemblies, a pneumatic module, a vision recognition module, and a control module. Transfer components are used to move stacked boards to corresponding storage locations; The control module is electrically connected to the pneumatic module, vision recognition module, robotic arm, and transfer assembly, respectively.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention uses a gripping method of fixing in the middle and lifting at the ends to deform the rigidity of the board itself, thereby breaking the negative pressure adsorption between the layers, and then the board can be lifted and transferred. In addition, during the process of breaking the negative pressure adsorption between the layers, the end suction cup group is combined with step adjustment and multi-parameter adaptive correction to further release the adhesion between the layers of the board, ensuring accurate picking of a single board at a time, thereby solving the defects of abnormal picking resistance and multiple pieces sticking together in the prior art. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the overall structure of the present invention; Figure 2 This is a schematic diagram of the processing technology of the present invention.

[0018] The components include: 1. Intelligent three-dimensional silo; 2. Sheet material; 3. Robotic arm; 4. Middle suction cup assembly; 5. End suction cup assembly. Detailed Implementation

[0019] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0020] Intelligent automated storage and retrieval systems (AS / RS) are core equipment for the automated storage, scheduling, and retrieval of wood-based panels. They fall under the category of transportation or storage devices, such as loading or unloading conveyors, workshop conveyor systems, or pneumatic pipeline conveyors. These systems can efficiently classify and store wood-based panels of different colors and sizes, and are widely used in automated production lines for furniture manufacturing, whole-house customization, and panel processing. To improve surface texture and performance, wood-based panels typically undergo surface treatments such as sanding, painting, and lamination before being stored, resulting in a smooth, dense surface with high flatness.

[0021] To address the needs of storing and retrieving this type of board material, existing technologies include various automated gripping solutions such as vacuum suction cup robotic arms, clamp-type, and magnetic gripping. Clamp-type devices rely on mechanical grippers to hold the edges of the board, which can easily cause crush damage to the corners and edges, and is not suitable for thin or easily deformable wood-based panels. Magnetic gripping devices are only suitable for composite boards with metal substrates and cannot effectively grip non-magnetic boards such as pure wood or laminated boards, limiting their applicability. Vacuum suction cup robotic arms, due to their high gripping efficiency and minimal damage to the board surface, are the mainstream solution for automated storage and retrieval of wood-based panels.

[0022] However, such devices have the following problems in actual operation: When smooth, surface-treated sheets are stacked, the air between layers is easily squeezed out, and adjacent sheets adhere tightly to form a negative pressure adsorption between layers, resulting in sheet adhesion; when the robotic arm grabs the upper sheet, the negative pressure between layers will simultaneously pull the lower sheet, causing an abnormally increased gripping resistance, making it impossible to achieve accurate single-sheet picking. If the gripping is forcibly lifted, it can easily cause sheet deformation, edge chipping, and surface scratches. In severe cases, it can also cause the vacuum suction cup to overload and fall off, and the robotic arm to deviate in positioning, affecting the stability of picking and the sheet yield. Moreover, the frequent adhesion failures can lead to disordered storage and retrieval cycles and shutdown of automated production lines. Therefore, the existing common suction cup robotic arms are difficult to meet the continuous production requirements of high density, high reliability, and high surface quality.

[0023] Based on the above issues, please refer to Figures 1-2 An intelligent automated storage and retrieval method for a three-dimensional silo includes the following steps: S1. Material Retrieval Preparation: The intelligent three-dimensional material hopper 1 receives the material retrieval command, and the robotic arm 3 moves to a preset position above the stacked board 2 in the hopper to be retrieved. This preset position is usually the middle area of ​​the upper part of the hopper. The command of the robotic arm 3 is determined by the identity information of the board 2 to be retrieved. Normally, before each stack of board 2 is placed into the hopper, the identity information of the board 2, such as material, thickness and size, is recorded. Therefore, it is only necessary to feed back the identity information of the board 2 to the robotic arm 3, and the robotic arm 3 will automatically move to the corresponding hopper. After moving to the preset position, the vision recognition module is activated to identify the length and dimension of the top stack of board 2 (to further verify whether the identity information is accurate). Then, the identified length and dimension signal is transmitted to the control module of the robotic arm 3.

[0024] S1.1 Suction Cup Positioning: The control module controls the end suction cup group 5 to move to the standard suction position according to the length of the board 2. The distance between the standard position and the position of the middle suction cup group 4 is the standard distance L. The standard distance L is positively correlated with the length of the board 2. Therefore, adapting the standard distance L to the length of the board 2 can ensure that the end suction cup group 5 can act on the board 2 at a reasonable position, laying the foundation for effective deformation of the subsequent end lifting. When the length of the board 2 increases, the standard distance L increases synchronously to ensure sufficient lifting torque. When the length of the board 2 decreases, the standard distance L decreases synchronously to avoid excessive deformation of the board 2 due to excessive distance, while ensuring the stability of the suction.

[0025] S1.2 Initial adsorption: The robotic arm 3 starts, driving the middle suction cup group 4 and the end suction cup group 5 to move downwards synchronously until both suction cup groups achieve initial adsorption with the plate 2; S1.3, End Lifting: Activate the pneumatic module. The pneumatic module drives the end suction cup group 5 to lift upward with the preset standard air pressure P0, while the middle suction cup group 4 remains fixed. Only the two ends of the plate 2 are lifted upward. S1.4 Deformation Detection and Judgment: The visual recognition module identifies the maximum deformation size at both ends of the board 2. The maximum deformation size is the distance that the two ends of the board 2 move upward, and the maximum deformation size is compared with the threshold A0. The threshold A0 is an important standard for judging whether the negative pressure between the layers of the board 2 has been effectively broken and whether the adhesion has been released. Without the threshold A0 as a reference, it is impossible to quantify the degree of deformation at the ends of the board 2, and it is impossible to distinguish between "slight deformation (negative pressure not broken, still adhesion)" and "effective deformation (negative pressure broken, no adhesion)". This may lead to the robotic arm 3 lifting the whole prematurely (still adhesion, causing damage to the board 2 and failure to pick up the material) or over-lifting (causing excessive deformation of the board 2 and overload of the suction cup). The specific judgment logic is as follows: If the deformation size is greater than or equal to the threshold A0, it indicates that the upward deformation at both ends of the plate 2 has reached the preset standard, the interlayer negative pressure has been effectively broken, and the adhesion between adjacent plates 2 has been released. At this time, the plate 2 only maintains stable adsorption with the middle and end suction cup groups 5, without any excess pulling resistance. After receiving the judgment signal, the control module sends a command to control the robotic arm 3 to lift the entire plate smoothly, ensuring that the plate 2 remains horizontal during the lifting process, avoiding deformation or detachment of the plate 2 due to uneven force. Then, the plate 2 is accurately transferred to the preset processing station or transfer area, and the plate 2 is released smoothly to complete this material handling operation. If the deformation size is less than the threshold A0, it indicates that the deformation at both ends of the board 2 is insufficient, the interlayer negative pressure has not been completely broken, and there may still be adhesion between the upper board 2 and the lower board 2. If the robotic arm 3 is forcibly lifted as a whole at this time, the pulling resistance generated by the interlayer negative pressure will cause the lower board 2 to be driven synchronously, which will lead to the deformation of the board 2, and even cause the vacuum suction cup to overload and fall off, or the robotic arm 3 to deviate in positioning. Based on this, the control module immediately sends an air cut-off command to control the end suction cup group 5 to stop suction and reset to the initial suction height. Then, the end suction cup group 5 is controlled to move smoothly along the direction away from the middle suction cup group 4 at the execution end with a preset step distance Z. After each step movement is completed, the actions of S1.2-S1.3 are repeated, and the deformation size at both ends of the board 2 is detected again by the visual recognition module. This cycle is repeated until the detected deformation size is greater than or equal to the threshold A0 and the adhesion is confirmed to be completely released. Then, the robotic arm 3 is controlled to lift as a whole and transfer the board 2 to the preset position to complete the material picking operation, ensuring the safety and accuracy of the material picking process. It should be noted that the greater the distance between the end suction cup group 5 and the middle suction cup group 4, the greater the lifting torque on both ends of the board 2. Therefore, under the same lifting air pressure, the ends of the board 2 are more likely to deform upwards, and the deformation size is more likely to reach the preset threshold A0. This is also the main reason why the control module controls the end suction cup group 5 to move away from the middle suction cup group 4 step by step. By gradually increasing the distance between the two, the lifting torque at the ends of the board 2 is increased, the deformation of the board 2 is gradually increased, and thus the interlayer adhesion is broken.

[0026] It is worth mentioning that the present invention does not use the method of "moving the end suction cup group 5 directly to both ends of the board 2 at the beginning" to perform the lifting step of S1.3. Instead, it adopts the form of "preset standard distance L positioning + step movement adjustment". The reason and advantage are as follows: First, the technical solution in the present invention can avoid excessive deformation and damage to the board 2. If the end suction cup group 5 is moved to both ends of the board 2 at the beginning, the distance between the end and the middle suction cup group 4 will reach the maximum, and the lifting torque will also reach the maximum. Under the same lifting air pressure, the two ends of the board 2 will instantly undergo large deformation. Especially for thin and easily deformable wooden boards 2, it is very easy to cause the ends of the board 2 to bend and crack, or even cause the suction cup to detach from the surface of the board 2, which will make it impossible to effectively pick up the material. Secondly, it can adapt to different sizes of boards 2. The size of wooden boards 2 varies. If the end suction cup group 5 is directly fixed to both ends of the board 2, the position of the end of the board 2 needs to be accurately identified each time. The positioning accuracy requirement is extremely high, and positioning deviation is easy to occur. However, the preset standard distance L (positively correlated with the length of the board 2) can quickly complete the initial positioning. Then, by stepping, the adhesion of different sizes of boards 2 can be adjusted to reduce the positioning difficulty and improve the positioning efficiency. Third, it enables the gradual breaking of adhesion. The severity of interlayer adhesion varies depending on the surface condition and weight of board 2. If the maximum lifting torque is used initially, it will waste energy if the adhesion is slight, and may not be able to break the negative pressure in one go if the adhesion is severe. However, the step-by-step adjustment can gradually increase the lifting torque according to the deformation detection results, accurately match the degree of adhesion, ensure effective breaking of adhesion, avoid waste of resources, and improve the stability of material handling operations.

[0027] Therefore, in this invention, the method of "preset standard distance L positioning + step movement adjustment" can fully avoid the above problems.

[0028] Due to differences in the material of wood-based panels 2 (such as solid wood, MDF, particleboard, etc.) and surface treatment processes (such as sanding and polishing, water-based paint spraying, film coating, UV curing, etc.), the smoothness and density of the surface of panels 2 will vary, resulting in different interlayer bonding strengths after different panels 2 are stacked. In the above-mentioned material handling methods, the debonding judgment threshold A0 is mostly a fixed preset value, which cannot be dynamically adjusted according to the actual characteristics of the board 2 to be handled. This easily leads to two types of debonding misjudgment problems: First, for board 2 that is easily deformed and has low adhesion strength (such as thin-film coated board), if the fixed threshold A0 is too high, board 2 may be completely debonded but still judged as not debonded, and repeated end suction cup adjustment and lifting operations will be performed, which not only reduces the material handling efficiency, but may also cause board 2 to deform due to excessive lifting; Second, for board 2 that is not easily deformed and has high adhesion strength (such as thick solid wood board), if the fixed threshold A0 is too low, board 2 may be judged as qualified before it is completely debonded. When the robotic arm 3 lifts the whole, it will pull the lower board 2, resulting in problems such as multiple pieces of material handling, scratches on board 2, and chipping of edges and corners, which will affect the material handling quality.

[0029] Based on this, in one embodiment of the present invention, when the initial deformation dimension d0 is less than the threshold A0, and the deformation dimension d1 is still less than the threshold A0 after the first step movement of the end suction cup group 5, it indicates that the current adhesion strength of the board 2 is higher than the initial expectation, and the fixed threshold A0 can no longer be adapted. At this time, the control module will automatically start the threshold adaptive correction process and first calculate the deformation coefficient a = (d1-d0) / d0. Wherein, a is the deformation coefficient, reflecting the degree of deformation of the board 2 and the interlayer adhesion strength; d0 is the maximum deformation dimension at both ends of the board 2 when the end is first lifted, and d1 is the maximum deformation dimension at both ends of the board 2 when the end suction cup group 5 is lifted again after the first step movement; the value of a is directly related to the characteristics of the board 2. The larger the value of a, the better the flexibility of the board 2, the easier it is to deform, and the lower the interlayer adhesion strength. A small adjustment of the end suction cup position can achieve obvious deformation; the smaller the value of a, the stronger the rigidity of the board 2, the less easily it is deformed, and the higher the interlayer adhesion strength. A larger adjustment of the suction cup position or an increase in the lifting force is required to achieve debonding. After calculating the deformation coefficient a, the control module corrects the initial preset threshold A0 to A1 using the formula A1=b*A0*a0 / a. Here, A1 is the corrected debonding judgment threshold, which is the optimal judgment standard adapted to the actual characteristics of the current board 2; A0 is the initial preset threshold, which can be preset based on the average characteristics of commonly used board 2 on the production line, serving as the basis for threshold correction; a0 is the reference threshold for the deformation coefficient a, a pre-set standard deformation coefficient corresponding to the deformation characteristics of the most commonly used board 2 on the production line, used to calibrate the deviation of the current deformation coefficient a; b is a pre-input adjustment coefficient, typically ranging from 0.8 to 1.2, mainly used to fine-tune the threshold correction range, which can be flexibly set according to the type of board 2 and production precision requirements of different production lines to avoid excessive deviation of the corrected threshold A1. Therefore, when the deformation coefficient 'a' increases, according to the formula A1=b*A0*a0 / a, an increase in 'a' leads to a larger denominator, and A1 will decrease inversely. Considering the physical meaning of the value of 'a' (an increase in 'a' indicates that board 2 has good flexibility, is easily deformed, and has low interlayer adhesion strength), a decrease in A1 at this time means a decrease in the fixed threshold. This avoids situations where board 2 is completely detached but is still judged as not detached, leading to repeated adjustments and lifting operations with the end suction cups. This not only reduces material handling efficiency but may also cause deformation of board 2 due to excessive lifting. When the deformation coefficient 'a' decreases, according to the formula A1=b*A0*a0 / a, the denominator decreases, and A1 increases inversely. Considering the physical meaning of 'a' (a decreases because it indicates that sheet 2 is rigid, not easily deformed, and has high interlayer adhesion strength), an increase in A1 at this time means an increase in the fixed threshold. This prevents sheet 2 from being judged as qualified before it is fully debonded, which would cause problems such as pulling on the lower sheet 2 during the overall lifting by the robotic arm 3, resulting in multiple sheet picking, scratches on sheet 2, and edge chipping, thus affecting the quality of material picking.

[0030] The change in adjustment coefficient b will simultaneously affect the correction effect of A1, further adapting to the change in the value of a: when a increases, if b increases appropriately, it can offset the decrease in A1 caused by the increase in a, avoiding the problem of thin-film coated board 2 not detaching during material handling due to A1 being too low. When a decreases, if b decreases appropriately, it can avoid the problem of thick solid wood board 2 being over-lifted due to A1 being too high.

[0031] Therefore, this invention achieves adaptive correction of the judgment threshold, avoiding the limitation that a fixed threshold cannot adapt to complex working conditions. It enables the deformation judgment standard to be dynamically adjusted according to the adhesion degree and material differences of the sheet 2, effectively avoiding the problems of ineffective step adjustment and low material picking efficiency caused by a fixed threshold, and improving the accuracy and efficiency of material picking.

[0032] The above adaptive correction of the judgment threshold solves the problem that the fixed threshold cannot adapt to different adhesion levels and different materials of the board 2; however, the lifting air pressure of the end suction cup group 5 was not adjusted simultaneously. Since the corrected threshold A1 is a personalized judgment standard adapted to the current adhesion and deformation characteristics of the board 2, its value may be higher or lower than the original fixed threshold A0: if A1 is higher than A0, it means that a is less than b*a0, and the current board 2 is rigid, not easily deformed, and has high interlayer adhesion strength; if the original standard air pressure P0 is still used for lifting at this time, the lifting power will be insufficient, causing the board 2 to... If the deformation size cannot reach the corrected threshold A1 in a short period of time, the interlayer adhesion is difficult to break quickly, which can easily prolong the material picking cycle and reduce the material picking efficiency. If A1 is lower than A0, it means that a is greater than b*a0, which means that the current board 2 is flexible, easy to deform, and has low interlayer adhesion strength. If the original standard air pressure P0 is used for lifting at this time, the excessive lifting power will cause the deformation size of board 2 after stepping to exceed the A1 value excessively. This will not only waste the energy of the pneumatic module, but also cause the end of board 2 to bend and crack due to excessive lifting, or even cause the vacuum suction cup to detach from the surface of board 2.

[0033] Based on the above problems, in one embodiment of the present invention, during the step adjustment process, the control module no longer uses a fixed standard air pressure P0, but combines the influencing factors: the ratio of the corrected threshold to the original threshold (A1 / A0) and the adjustment coefficient c to calculate the air pressure correction coefficient B1, and then corrects the standard air pressure P0 to B1 times, finally obtaining the actual lifting air pressure adapted to the current working condition (actual air pressure = B1 × P0), where the correction formula is B1 = c * A1 / A0; B1 is the air pressure correction coefficient, which can quantify the air pressure correction range. Its value directly determines the degree of deviation between the actual lifting air pressure and the standard air pressure P0. When B1>1, the actual air pressure is higher than the standard air pressure, and the lifting power is enhanced; when B1<1, the actual air pressure is lower than the standard air pressure, and the lifting power is weakened; when B1=1, the actual air pressure is consistent with the standard air pressure, and no correction is required.

[0034] c is a pre-input adjustment coefficient, with a value range of 1.0-1.2. Its function is to fine-tune the overall amplitude of air pressure correction, avoiding excessive deviation of B1 due to fluctuations in A1 / A0, and ensuring the stability and rationality of air pressure correction. When the material of plate 2 is brittle, easily deformed, or has a light degree of adhesion, the value of c can be adjusted to 1.0-1.1 to avoid excessive air pressure correction amplitude leading to excessive lifting force and damage to plate 2. When plate 2 has high hardness, severe adhesion, or a large deviation between L1 and L0, the value of c can be adjusted to 1.1-1.2 to ensure that the air pressure correction amplitude is sufficient to provide enough lifting force to break the adhesion.

[0035] The above A1 / A0 is the threshold correction ratio, which reflects the degree of deviation between the corrected threshold A1 and the original fixed threshold A0. When A1 / A0>1, it means that the current threshold is higher than the original threshold, and it is necessary to increase the lifting force so that the deformation size of the plate 2 reaches the corrected threshold A1 in a short period of time, thereby quickly breaking the interlayer adhesion, shortening the material picking cycle and improving the material picking efficiency. When A1 / A0<1, it means that the current threshold is lower than the original threshold, and the lifting force can be appropriately reduced to avoid the deformation size of the plate 2 exceeding the A1 value excessively after stepping, which would cause energy waste of the pneumatic module, bending and cracking of the end of the plate 2, or even the vacuum suction cup detaching from the surface of the plate 2.

[0036] Therefore, this invention achieves the matching of lifting air pressure and the corrected threshold, solving the problem of "power mismatch after threshold correction" mentioned above. It ensures that the lifting power can accurately adapt to the current adhesion and deformation characteristics of the board 2, and can provide sufficient power to quickly reach the corrected threshold for deformation size, effectively breaking the interlayer adhesion, while avoiding the adverse effects of excessively high or low air pressure; excessively high air pressure leads to damage to the board 2 and abnormal equipment load, while excessively low air pressure leads to low efficiency.

[0037] In one embodiment of the present invention, when both the initial deformation dimension d0 (i.e., the maximum deformation dimension at both ends of the board 2 detected by the visual recognition module after the first end lifting action) and the deformation dimension d1 after the first step (i.e., the maximum deformation dimension at both ends of the board 2 detected by the visual recognition module after the end suction cup group 5 completes the first stepping movement away from the middle suction cup group 4 and performs the suction and lifting action again) are less than the threshold A0, it indicates that the current fixed step distance does not match the deformation characteristics of the board 2, and the step distance correction program needs to be started. At this time, the control module, based on the deformation coefficient a, combined with the step distance reference threshold Z0, the adjustment coefficient e, and the deformation coefficient reference threshold a0, corrects the subsequent step distance to a personalized step distance Z1 adapted to the current working condition using a preset formula, Z1=e*Z0*a0 / a; Wherein, Z1 is the corrected step distance, a personalized step distance adapted to the deformation coefficient a and adhesion level of the current board 2, replacing the original fixed step distance Z, used for the subsequent stepping movement of the end suction cup assembly 5, ensuring that each stepping movement accurately increases the deformation size of the board 2, balancing efficiency and safety; Z0 is the reference threshold for the step distance Z, a fixed benchmark value obtained through experimental calibration based on wood boards 2 of conventional size, material, and adhesion level, reflecting the optimal step distance under normal working conditions; e is the pre-input adjustment coefficient, strictly set within the range of 1.0-1.2, its function is to fine-tune the correction range of the step distance, avoiding over-adjustment of Z1 due to small fluctuations in the deformation coefficient a and the deformation coefficient reference threshold a0. Large deviations ensure the stability of the step distance correction. When the material of plate 2 is brittle and easily deformed, or the adhesion is light, the value of e can be adjusted to 1.0-1.1 to avoid excessive deformation of plate 2 due to excessively large step distance after correction. When plate 2 has high hardness, severe adhesion, and is not easily deformed, the value of e can be adjusted to 1.1-1.2 to ensure that the step distance after correction is large enough to effectively increase the lifting torque, increase the deformation size of plate 2, and accelerate the breaking of adhesion. 'a' is the deformation coefficient, reflecting the degree of adhesion and deformation characteristics of plate 2. The larger 'a' is, the lighter the adhesion and the easier it is to deform; the smaller 'a' is, the heavier the adhesion and the less easily it is to deform. 'a0' is the deformation coefficient reference threshold, reflecting the average influence of step adjustment on the deformation of plate 2 under normal working conditions. Therefore, when a increases, it indicates that the current board 2 has good flexibility, is easy to deform, and has low interlayer adhesion strength, making the step adjustment effect more significant. At this time, Z1 will decrease synchronously, and in conjunction with the weakened lifting force mentioned above, it will further avoid the deformation size of board 2 after stepping exceeding the A1 value, causing energy waste of the pneumatic module, bending and cracking of the end of board 2, or even the vacuum suction cup detaching from the surface of board 2. When a decreases, the current board 2 has strong rigidity, is not easy to deform, and has high interlayer adhesion strength, resulting in a larger corrected Z1. In conjunction with the enhanced lifting force mentioned above, the deformation size of board 2 reaches the corrected threshold A1 in a short period of time, thereby quickly breaking the interlayer adhesion, thus shortening the material picking cycle and improving material picking efficiency.

[0038] In summary, the automatic storage and retrieval method for wooden boards 2 in the intelligent three-dimensional silo 1 proposed in this invention addresses the problems of interlayer negative pressure adhesion, material smearing during vacuum suction cup retrieval, and damage to the boards 2 that easily occur when smooth boards 2 with surface treatment are stacked. The method employs a central suction cup group 4 and a movable end suction cup group 5 working together. First, the boards 2 are adsorbed as a whole. Then, by fixing the central suction cup and lifting the end suction cups, the ends of the boards 2 are deformed to break the interlayer negative pressure. The deformation size is detected in real time using visual recognition. When the deformation amount does not reach the debonding threshold, the end suction cups are controlled to gradually move outward to increase the lifting torque. This lifting is repeated until the adhesion is released. Simultaneously, by calculating the deformation coefficient of the two lifting operations, the debonding judgment threshold, end lifting air pressure, and step distance are adaptively corrected. This allows the system parameters to dynamically match the boards 2 with different materials, thicknesses, surface conditions, and adhesion strengths. While avoiding deformation, chipping, and scratches of the boards 2, this method achieves stable and efficient single-sheet retrieval, ensuring the continuous and reliable operation of the automated production line.

[0039] In one embodiment, the intelligent three-dimensional silo 1 is typically equipped with multiple independent compartments to meet the automated production needs of classifying and storing wood-based panels 2 of different specifications, materials, and surface treatment processes, and scheduling them in different areas. Since the panels 2 stored in the same compartment are usually from the same batch, of the same material, of the same size, and with the same surface treatment process, their interlayer adhesion strength, deformation characteristics, and other working parameters are highly consistent. Therefore, when the robotic arm 3 performs the first material removal operation on the stacked panels 2 in a certain compartment, it will strictly follow steps S1.2 to S1.3 to complete the initial adsorption and end-lifting de-adhesion action. After the top layer of panels 2 in the compartment is successfully removed, the control module will automatically record and save the position information corresponding to the compartment, the actual lifting air pressure used by the pneumatic module, the final distance between the end suction cup group 5 and the middle suction cup group 4, and the de-adhesion judgment threshold A1 after adaptive correction. The control module will establish a corresponding mapping relationship between the above compartment position information and parameters such as actual air pressure, final distance, and threshold A1 to form a working parameter mapping table specific to the compartment. Before all stacked boards 2 in the bin are completely emptied, the parameters in the mapping table will be retained and directly applied to every subsequent material retrieval operation in the bin. There is no need to repeat the adaptive identification and correction process. The historical data of the bin is only cleared and updated when all boards 2 in the bin are removed, it is ready to be re-loaded, or the batch of boards 2 is replaced. This enables the reuse and inheritance of parameters under the same working conditions, greatly improving the stability and efficiency of continuous material retrieval and reducing repeated debugging and malfunctions.

[0040] Within the same compartment, as each sheet of material 2 is removed, the overall height of the remaining stacked sheets of material 2 continuously decreases. The stacking pressure on the upper sheet of material 2 gradually decreases, making it easier to fully expel interlayer air. The adhesion between adjacent sheets of material 2 becomes tighter, and the adhesive force formed by the interlayer negative pressure significantly increases, gradually increasing the difficulty of material removal. To match this objective working condition change pattern, a dynamic adjustment rule is set for the final actual lifting air pressure of the pneumatic module in the parameter mapping table corresponding to the compartment, so that the actual lifting air pressure increases linearly with the decrease in the height of the sheet of material 2. By gradually increasing the lifting driving force of the end suction cup group 5, it is ensured that the deformation generated by the end lifting can continuously meet the de-adhesion judgment threshold requirement, thereby effectively breaking the interlayer negative pressure that continuously increases with the decrease in the number of remaining sheets of material 2. This avoids problems such as the inability to remove adhesion, material removal failure, or damage to the sheet of material 2 during later material removal due to constant air pressure, ensuring that the entire material removal process from the first to the last sheet of material 2 within the same compartment is stable, reliable, and smooth.

[0041] In an optional embodiment, after the robotic arm 3 performs an end-lifting action, the deformation dimension d0 obtained in the initial detection is less than the initial debonding threshold A0. Furthermore, after the end suction cup group 5 completes its first stepping movement away from the middle suction cup group 4 and is lifted again, the deformation dimension d1 is still less than the corresponding debonding threshold. This indicates that the interlayer adhesion strength of the material to be removed, 2, is significantly higher than under normal working conditions. Relying solely on conventional step adjustments is insufficient to quickly achieve effective deformation and break the interlayer negative pressure. Therefore, in other embodiments, the control module automatically enters an enhanced debonding mode, limiting the total number of stepping adjustments for the end suction cup group 5 to three. This enhances the debonding capability while ensuring controllable material handling cycle time and without significantly extending the work cycle. In the third stepping action, the end suction cup group 5 is directly controlled to step to the preset optimal end position, maintaining the distance between the end suction cup group 5 and the middle suction cup group 4 within the range of 1 / 5 to 2 / 5 of the length of the material 2. This maximizes the lifting torque without causing excessive bending or cracking of the material 2, thereby significantly improving the end-deformation debonding effect. Meanwhile, the process uses the actual air pressure B1×P0 after adaptive correction as the lifting force, and the corrected threshold A1 as the debonding judgment standard. This allows the lifting air pressure, judgment threshold, and suction cup spacing to work together to match the deformation characteristics and adhesion strength of the current board 2. While ensuring that the board 2 is not damaged, the interlayer adhesion is quickly, stably, and in one go, ensuring the success rate of material removal.

[0042] In the suction cup positioning step S1.1, the standard distance L between the end suction cup group 5 and the middle suction cup group 4 is set to 1 / 10-1 / 5 of the length of the board 2. This range is the optimal range determined after a large number of tests and working condition verifications, taking into account the structural strength of the wooden board 2 itself, the end lifting deformation requirements, and the adsorption stability requirements. Limiting the standard distance L to 1 / 10-1 / 5 of the length of the board 2 serves two purposes. First, setting the standard distance L to 1 / 10-1 / 5 of the length of the board 2 ensures that the end suction cup assembly 5 is in a suitable initial working position, providing a reasonable starting basis for subsequent step adjustment. This ensures that the lifting torque of the end suction cup assembly 5 can increase steadily and gradually during the step-out movement, making the deformation of the end of the board 2 increase in a controllable manner, thereby accurately matching the requirements for breaking interlayer adhesion. Second, this initial distance avoids insufficient step adjustment range and slow torque increase due to an excessively small initial spacing, and also avoids insufficient step adjustment margin and premature excessive deformation of the board 2 due to an excessively large initial spacing, ensuring a smooth and effective subsequent step adjustment process. Meanwhile, this adaptability range can accommodate the initial positioning requirements of different lengths of sheet material 2, enabling subsequent adaptive adjustment processes such as stepping distance, lifting air pressure, and deformation threshold to be executed uniformly. This improves the universality and consistency of the entire stepping adjustment system for sheet material 2 of different sizes, providing a stable premise for reliable stepping of the end suction cup group 5, accurate deformation detection, and efficient release of adhesion.

[0043] S2. Material Storage Operation: The control module receives the material storage instruction and smoothly transfers the stacked boards 2 to the empty compartment of the intelligent three-dimensional material storage silo 1 through the transfer component, avoiding collisions and scratches to the boards 2 during the transfer process; after the material storage is completed, the control module records the specifications of the boards 2 and the stacking height of the compartment, which is convenient for subsequent material retrieval operations.

[0044] S3. Self-learning and mapping of bin parameters: In the intelligent three-dimensional material bin 1, when the robotic arm 3 performs its first material retrieval of stacked plates 2 in a certain bin, the robotic arm 3 normally executes steps S1.2-S1.3. After successfully retrieving the top plate 2 in the bin, the control module records the position information of the bin, the final actual air pressure of the pneumatic module, the final distance between the end suction cup group 5 and the middle suction cup group 4, and the corrected threshold A1. A mapping table is created between the position information, the actual air pressure value, the final distance, and the threshold A1. Subsequent material retrieval from this bin directly calls the parameters in the mapping table, eliminating the need for repeated suction cup position adjustments and air pressure corrections, thus improving material retrieval efficiency. In the mapping table, the final actual air pressure of the pneumatic module increases linearly as the height of the plate 2 being retrieved decreases. This is because the lower the stack height of the plates 2, the greater the pressure on the lower plates 2, and the higher the interlayer adhesion strength. By linearly increasing the air pressure, it ensures that plates 2 at different heights can effectively break the adhesion, achieving stable material retrieval.

[0045] This invention also discloses an intelligent three-dimensional silo automatic storage and retrieval device for realizing an intelligent three-dimensional silo automatic storage and retrieval method. The device consists of three main parts: an intelligent three-dimensional silo 1, at least one set of robotic arms 3, and a transfer component. The parts work together to realize the fully automated operation of the entire process from storage and retrieval of the sheet material 2 to transfer. The robotic arm 3, as the execution component, has a structural design that directly determines the material retrieval accuracy and adaptability. Specifically, it includes an execution end, a middle suction cup group 4, at least two sets of end suction cup groups 5, a pneumatic module, a vision recognition module, and a control module. The components have clear division of labor and interconnected signals to form a complete closed-loop control system.

[0046] Specifically, the intelligent three-dimensional material storage silo 1 serves as the storage carrier for the sheet material 2, and is equipped with several independent compartments. These compartments can be divided according to the specifications, color, and material of the sheet material 2, providing a foundation for accurate subsequent material retrieval. The transfer component uses a conveyor or mechanical pushing mechanism, primarily responsible for the smooth transfer of the sheet material 2 during storage operations, preventing collisions and scratches during transfer and ensuring the surface quality of the sheet material 2. The robotic arm 3 undertakes the tasks of material retrieval and transfer, and the collaborative workflow of its components corresponds to the methods described above. The visual recognition module is activated first, accurately identifying the length, stacking height, and surface condition of the material to be picked up (2), and transmitting the identified real-time data to the control module. The control module, acting as the "control center" of the entire device, quickly calculates the standard positions of the middle suction cup group 4 and the end suction cup group 5 based on the received data, and sends commands to the actuator to drive both suction cup groups to move synchronously down to the surface of the material 2, completing the initial adsorption. Subsequently, the control module commands the pneumatic module to activate, using a preset standard air pressure to drive the end suction cup group 5 upwards individually, while simultaneously controlling the middle suction cup group 4 to remain fixed, achieving the action of "middle fixed, end lifted." During this process, the visual recognition module monitors the material in real time... The deformation dimensions at both ends of the board 2 are detected, and the data is fed back to the control module. The control module determines whether the board 2 has detached based on the comparison between the deformation dimensions and the threshold. If it has not detached, the adaptive correction process is automatically started, adjusting parameters such as air pressure, step distance, and threshold, and repeating the adsorption and lifting actions until accurate detachment is achieved. After the material is picked up, the control module drives the robotic arm 3 to transfer the board 2 to the preset position, and records the relevant parameters to form a bin parameter mapping table, providing a basis for reuse in subsequent material picking. During the storage operation, the control module instructs the transfer component to smoothly transfer the board 2 to the empty bin. After the storage is completed, the bin information is recorded, realizing the automated control of the entire storage and retrieval process.

[0047] It should be noted that when a component is said to be "fixed to" another component, it can be directly attached to the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "upper," "lower," "left," "right," "front," "back," and similar expressions used in this document are for illustrative purposes only.

[0048] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for automatic storage and retrieval of intelligent three-dimensional silos, characterized in that, Includes the following steps: S1. Material Retrieval Operation: The intelligent three-dimensional material silo receives the material retrieval command, the robotic arm moves to the top of the stack of boards to be retrieved, the vision recognition module identifies the length of the top layer of boards and transmits it to the control module; S1.1 Suction Cup Positioning: The control module controls the end suction cup group to move to the standard suction position according to the length of the board. The distance between the standard position and the position of the middle suction cup group is the standard distance L. The standard distance L is positively correlated with the length of the board. S1.2 Initial Adsorption: The actuator drives the middle and end suction cup assemblies to move down synchronously to complete the initial adsorption; S1.3, End Lifting: The pneumatic module drives the end suction cup assembly to lift with standard air pressure P0, while the middle suction cup assembly is fixed, lifting only the two ends of the sheet metal; S1.4 Deformation Detection and Judgment: The visual recognition module identifies the largest deformation size at both ends of the board and compares this deformation size with the threshold A0. If the deformation size is greater than or equal to the threshold A0, the robotic arm lifts the board as a whole and transfers it to the preset position to complete the material handling. If the deformation size is less than the threshold A0, the end suction cup group is de-aired and reset. The end suction cup group moves away from the middle suction cup group by a step distance Z. Each time the end suction cup group moves, the actions of S1.2-S1.3 are repeated until the deformation size is greater than or equal to the threshold A0, and then the board is lifted as a whole to complete the material handling. S2, Material Storage Operation: Move stacked boards to empty storage locations by transferring components.

2. The intelligent three-dimensional silo automatic storage and retrieval method according to claim 1, characterized in that, In step S1.4, when the initial deformation size d0 is less than the threshold A0, and the deformation size d1 of the end suction cup assembly is still less than the threshold A after the first step movement, the control module calculates the deformation coefficient a = (d1-d0) / d0, and corrects the threshold A0 to A1 = b*A0*a0 / a through the deformation coefficient a, where a0 is the reference threshold of the deformation coefficient a, and b is the pre-input adjustment coefficient.

3. The intelligent three-dimensional silo automatic storage and retrieval method according to claim 2, characterized in that, During the stepping process in step S1.4, the control module corrects the standard air pressure P0 in the pneumatic module to B1 times, and B1=c*A1 / A0, where c is the adjustment coefficient.

4. The intelligent three-dimensional silo automatic storage and retrieval method according to claim 3, characterized in that, When the initial deformation dimension d0 is less than the threshold A0, and the deformation dimension d1 of the end suction cup assembly is still less than the threshold A after the first step movement, the control module corrects the subsequent step distance Z1=e*Z0*a0 / a according to the deformation coefficient a, where Z0 is the reference threshold for the step distance Z, and e is the adjustment coefficient.

5. The intelligent three-dimensional silo automatic storage and retrieval method according to claim 4, characterized in that, The intelligent automated material storage unit has several storage compartments. When the robotic arm first picks up the stacked boards in the compartment, it executes steps S1.2-S1.

3. After successfully removing the top board in the compartment, the control module records the compartment's position information, the final actual air pressure of the pneumatic module, the final distance between the end suction cup group and the middle suction cup group, and the threshold A1. The control module then creates a mapping table between the position information, the actual air pressure value, the final distance, and the threshold A1. The data for the compartment is only updated after all the stacked boards in the compartment have been removed.

6. The intelligent three-dimensional silo automatic storage and retrieval method according to claim 5, characterized in that, In the mapping table, the final actual air pressure of the pneumatic module increases linearly as the height of the plate being removed decreases.

7. The intelligent three-dimensional silo automatic storage and retrieval method according to claim 3, characterized in that, When the initial deformation dimension d0 is less than the threshold A0, and the deformation dimension d1 of the end suction cup group is still less than the threshold A after the first step movement, the control module will control the total number of steps to be three, and during the third step, control the end suction cup group to step to the end position. The distance between the end position and the middle suction cup group is 1 / 5 to 2 / 5 of the length of the board.

8. The intelligent three-dimensional silo automatic storage and retrieval method according to claim 1, characterized in that, In step S1.1, the standard distance L is 1 / 10 to 1 / 5 of the length of the board.

9. An intelligent automated storage and retrieval device for a three-dimensional silo, used to implement the intelligent automated storage and retrieval method for a three-dimensional silo as described in any one of claims 1-8, characterized in that, include: The intelligent automated storage and retrieval system is equipped with several storage compartments for storing sheet metal. At least one set of robotic arms, which are used to grab and transfer sheet metal from the corresponding compartment; The robotic arm includes an actuator, a central suction cup assembly, at least two sets of end suction cup assemblies, a pneumatic module, a vision recognition module, and a control module. Transfer components are used to move stacked boards to corresponding storage locations; The control modules are all electrically connected to the pneumatic module, vision recognition module, robotic arm, and transfer components.

Images