Shelf access control method and system for heavy load stacking robot

Abstract

The application discloses a kind of heavy load stacking robot shelf access control method and system, it is related to intelligent warehousing equipment control technical field, the method includes: the contact force distribution information between target tray and shelf and pallet fork is acquired;According to the contact force distribution information, the support area of target tray and the barycentric projection position are determined;According to support area and barycentric projection position control pallet fork movement, make support area continuously migrate between shelf support area and pallet fork support area, and make barycentric projection position in the migration process be located in support area;Contact force distribution information is continuously acquired and support area is updated in the process of pallet fork movement, and according to the updated support area, adjust the movement of pallet fork, until the target tray is taken out or stored in is completed.The application continuously changes by real-time sensing bearing conversion process support area and adaptively adjusts the movement of pallet fork, avoids load mutation and attitude instability, improves the stability and safety of heavy load goods access process.

Description

Technical Field

[0001] This invention relates to the field of intelligent warehousing equipment control technology, and more specifically, to a rack storage and retrieval control method and system for a heavy-duty stacking robot. Background Technology

[0002] As intelligent warehousing systems evolve towards higher load capacities and higher storage levels, heavy-duty stacking robots are widely used for automated storage and retrieval of large palletized goods. Most existing rack storage and retrieval control methods for heavy-duty stacking robots rely on preset storage location coordinates and motion trajectories to control fork extension, lifting, and walking movements. Some technical solutions use position feedback, force feedback, or posture feedback to correct motion parameters, thereby improving storage and retrieval accuracy and reducing operational impact.

[0003] However, in actual storage and retrieval, the target pallet is not always in a fixed supported state. As the forks gradually extend into the bottom of the target pallet and perform lifting and lowering actions, the contact state between the target pallet, the rack support beam, and the forks continuously changes. The load-bearing relationship gradually transitions from rack support to joint support by the rack and forks, and then to fork support. During this process, the load-bearing area of ​​the rack and forks on the target pallet and the load distribution relationship also continuously change with the movement of the forks.

[0004] Existing control methods typically divide the above process into several discrete stages and control it based on preset positions or motion trajectories. They do not identify and adjust the continuous changes in the support area during the load conversion process in real time. This can lead to problems such as sudden changes in local load, changes in posture, and cargo slippage during the transition phase when the target pallet is removed from or placed on the shelf. This can affect the stability of the storage and retrieval process, especially under heavy load conditions.

[0005] Therefore, there is an urgent need to provide a rack storage and retrieval control method and system for heavy-duty stacking robots, so as to adjust the fork movement according to the continuous changes in the support state of the target pallet and improve the stability of the heavy-duty goods storage and retrieval process. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide a rack storage and retrieval control method and system for a heavy-duty stacking robot. By acquiring the contact force distribution information between the target pallet and the rack and forks, the support area and center of gravity projection position of the target pallet are determined, and the movement of the forks is controlled according to the changes in the support area and center of gravity projection position. This solves the problem that the existing rack storage and retrieval process does not dynamically adjust according to the continuous changes in the pallet support state and load distribution relationship, resulting in insufficient stability of the heavy-duty pallet during the load conversion process.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A rack storage and retrieval control method for a heavy-duty stacking robot includes the following steps: acquiring contact force distribution information between a target pallet and the rack and forks; determining the support area of ​​the target pallet and the center of gravity projection position of the target pallet based on the contact force distribution information, wherein the support area is composed of the effective support area of ​​the rack and / or forks for the target pallet; controlling the movement of the forks based on the support area and the center of gravity projection position, so that the support area continuously migrates between the rack support area and the fork support area, and the center of gravity projection position is located within the support area during the migration; continuously acquiring contact force distribution information and updating the support area during the fork movement, and adjusting the fork movement based on the updated support area until the target pallet is retrieved or stored.

[0008] In a preferred embodiment, acquiring the contact force distribution information between the target pallet and the rack and forks includes: acquiring the initial multidimensional force signal output by the multidimensional force sensor installed on the forks, and simultaneously acquiring the current motion state parameters of the forks; determining the dynamic structural resonance frequency range of the mast system based on the motion state parameters; performing adaptive filtering on the initial multidimensional force signal according to the dynamic structural resonance frequency range to acquire the steady-state multidimensional force signal; and calculating and extracting the contact force distribution information based on the steady-state multidimensional force signal.

[0009] In a preferred embodiment, the step of calculating and extracting contact force distribution information based on the steady-state multidimensional force signal includes: decomposing the steady-state multidimensional force signal into a vertical normal force component and a horizontal shear force component; calculating the frictional coupling interference torque generated by the horizontal shear force component at the sensor measurement center based on the structural dimension parameters of the fork and the horizontal shear force component; decoupling the total measured torque contained in the steady-state multidimensional force signal from the frictional coupling interference torque by difference, and extracting the equivalent vertical load torque; establishing a static equilibrium equation based on the vertical normal force component and the equivalent vertical load torque, solving for the equivalent contact point coordinates and force values ​​of the target pallet within the support area, and generating contact force distribution information.

[0010] In a preferred embodiment, determining the support area of ​​the target pallet based on the contact force distribution information includes: calculating the load-bearing ratio of each equivalent contact point based on the force value corresponding to each equivalent contact point; selecting effective contact points from each equivalent contact point based on the load-bearing ratio and the trend of load-bearing ratio changes at adjacent sampling times; constructing corresponding support boundaries based on the spatial positions of the effective contact points on the shelf side and the fork side respectively; and determining the support area of ​​the target pallet based on each support boundary.

[0011] In a preferred embodiment, the step of selecting effective contact points from the equivalent contact points includes: obtaining the load-bearing ratio sequence of each equivalent contact point within a continuous sampling period; determining the load-bearing change direction of each equivalent contact point based on the load-bearing ratio sequence; and identifying equivalent contact points whose load-bearing change direction is consistent with the migration direction of the support area as effective contact points.

[0012] In a preferred embodiment, determining the center of gravity projection position of the target pallet includes: mapping the spatial coordinates of the effective contact points on the shelf side and the fork side to the same reference coordinate system; determining the load center of the target pallet based on the transformed spatial coordinates of each effective contact point and the corresponding force value; and projecting the load center onto the bearing plane of the target pallet to determine the center of gravity projection position.

[0013] In a preferred embodiment, controlling the fork movement based on the support area and the center of gravity projection position includes: acquiring support area boundary change information and center of gravity projection position change information at continuous sampling times; determining the distance change trend between the center of gravity projection position and the support area boundary based on the support area boundary change information and the center of gravity projection position change information; and adjusting the fork extension speed and / or lifting speed when the distance change trend indicates that the center of gravity projection position is moving closer to the support area boundary, so that the center of gravity projection position remains within the support area.

[0014] In a preferred embodiment, adjusting the fork movement according to the updated support area includes: acquiring the change in the support area at adjacent sampling times; determining whether the support area has undergone discontinuous changes based on the change in the support area; and when the support area undergoes discontinuous changes, reducing the fork movement speed or controlling the fork to move in the opposite direction so that the center of gravity projection position is maintained within the support area again.

[0015] A rack access control system for a heavy-duty stacking robot includes: one or more processors; and a memory connected to the one or more processors, the memory storing a computer program that, when executed by the one or more processors, causes the one or more processors to perform any of the methods described above.

[0016] A computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the method described in any of the preceding claims.

[0017] This invention acquires contact force distribution information between the target pallet and the rack and forks to determine the pallet's support area and center of gravity projection position. Based on this information, it controls fork movement, enabling continuous adjustment of the pallet's support state as it transitions between rack and fork support. This avoids the abrupt changes in load transition caused by discrete control methods in existing technologies, reducing localized load impacts and cargo slippage due to support area changes, and improving the stability of heavy-duty cargo handling. Simultaneously, it continuously acquires contact force distribution information and updates the support area during fork movement, adjusting fork movement based on the updated area to maintain the center of gravity projection position within the support area during load transitions. This provides continuous constraint on the stable support state of the target pallet, reducing the risk of instability due to mismatch between the center of gravity and the support area, and improving the safety and reliability of heavy-duty stacking robot rack handling. Attached Figure Description

[0018] Figure 1 A flowchart illustrating a rack access control method for a heavy-duty stacking robot provided in an embodiment of the present invention; Figure 2 A schematic diagram of the structure of a rack access control system for a heavy-duty stacking robot provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the migration of the target pallet support area during a load-bearing conversion process, provided by an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1, Figure 1 This invention provides a rack access control method for a heavy-duty stacking robot, comprising the following steps: S1, obtain the contact force distribution information between the target pallet and the rack and forks; In this embodiment, when the heavy-duty stacking robot performs target pallet storage and retrieval operations, it first obtains the contact force distribution information between the target pallet and the rack and forks to characterize the change in load-bearing state of the target pallet during the transition between rack support state, rack and fork joint support state, and fork load-bearing state.

[0021] Specifically, the contact force distribution information includes the spatial coordinates of each contact position and the force value at that position. The controller collects the force information during the fork's load-bearing process through multi-dimensional force sensors installed on the forks, and processes the collected signals in conjunction with the fork motion state parameters to obtain the coordinates and force values ​​of the equivalent contact points between the target pallet and the forks; at the same time, it determines the coordinates and force values ​​of the equivalent contact points on the rack side based on the support relationship between the target pallet and the rack, thereby forming the overall contact force distribution information of the target pallet.

[0022] The specific steps are as follows: S11. After receiving the rack access control command, the controller activates the multi-dimensional force sensor installed on the forks to continuously sample the force state of the forks during the load-bearing process and obtain the initial multi-dimensional force signal.

[0023] In this embodiment, the multi-dimensional force sensor is a six-dimensional force sensor, which is installed on the two forks of the fork to detect the three-dimensional forces acting on the fork and the torque generated around the three coordinate axes.

[0024] The three-dimensional forces include: horizontal forces along the fork extension / retraction direction. Horizontal force along the width of the forks Normal force perpendicular to the fork bearing surface The torques in the three directions include: the torque about the X-axis. Torque about the Y-axis Torque about the Z-axis .

[0025] The controller reads the output signal of the multi-dimensional force sensor according to the preset sampling period and stores the force and torque values ​​corresponding to each sampling moment in the buffer area.

[0026] Simultaneously, the controller acquires the current motion state parameters of the forks. Specifically, these motion state parameters include fork extension / retraction displacement, fork lifting / lowering height, and corresponding motion speed. The fork extension / retraction displacement is acquired by an encoder installed on the drive end of the extension mechanism, the fork lifting / lowering height is acquired by a displacement sensor installed on the lifting mechanism, and the controller calculates the corresponding motion speed based on the position change at adjacent sampling times.

[0027] By synchronously acquiring multi-dimensional force signals and fork motion state parameters, the subsequent signal processing can be dynamically adjusted based on the current fork motion conditions.

[0028] S12. During the fork extension, lifting, and load conversion processes of the heavy-duty stacking robot, the mast structure will generate mechanical vibrations, causing structural vibration components to be mixed into the signals collected by the multi-dimensional force sensors. Therefore, in this embodiment, the dynamic structural resonance frequency range of the mast system under the current working condition is determined based on the fork motion state parameters, and the initial multi-dimensional force signal is filtered based on the dynamic structural resonance frequency range.

[0029] Specifically, the controller establishes the structural vibration characteristics of the gantry system under its current operating conditions based on the current fork lifting height and the load. The load is determined based on the vertical normal force component collected by a multi-dimensional force sensor.

[0030] The controller pre-stores a mapping relationship of structural parameters for the gantry system. This mapping relationship characterizes the natural frequencies of the structure under different lifting heights and load conditions. Based on the current fork lifting height and load, the controller queries this mapping relationship to obtain the natural frequencies of the structure under the current operating condition.

[0031] The controller queries the working condition mapping relationship based on the current lifting height and load weight to obtain the first-order natural frequency under the current working condition. and second-order natural frequency .

[0032] Furthermore, a frequency range is set with the inherent frequency as the center: (1) as well as: (2) in, The preset frequency extension value ranges from 5Hz to 15Hz, and is preferably 10Hz in this embodiment.

[0033] By combining the above frequency ranges, we obtain the dynamic structural resonance frequency range corresponding to the current operating condition.

[0034] Subsequently, the controller adjusts the digital filtering parameters according to the resonant frequency range of the dynamic structure to adaptively filter the initial multidimensional force signal. In this embodiment, a digital notch filter is used to suppress the vibration component corresponding to the resonant frequency of the dynamic structure. The transfer function of the notch filter is: (3) in, The normalized center frequency is determined based on the resonant frequency range of the dynamic structure. Notch filter bandwidth control parameters ( In this embodiment ), The unit delay operator represents delaying the signal by one sampling period. In the transfer function of a digital filter, Corresponding to the delay unit in discrete-time signal processing, The two-unit delay operator means delaying the signal by two sampling periods.

[0035] The controller obtains a steady-state multidimensional force signal after removing structural vibration interference through the above filtering process, including the horizontal force in the extension and contraction direction after removing vibration interference. Horizontal force in the width direction after filtering out vibration interference Vertical normal pressure after filtering out vibration interference and the steady-state torque in the corresponding direction , , .

[0036] S13. Due to the friction between the forks and the target pallet, the horizontal shear force generates an additional torque at the sensor's measurement center, affecting the accuracy of the contact position retrieval based on the torque. Therefore, this embodiment calculates the frictional coupling interference torque generated by the horizontal shear force based on the fork's structural dimensions.

[0037] Specifically, let the vertical distance between the sensor measurement center and the fork bearing surface be... The frictional coupling interference torques generated around the X-axis and around the Y-axis are respectively: (4) (5) in, The frictional coupling disturbance torque generated by the horizontal shear force around the X-axis. The frictional coupling interference torque generated by the horizontal shear force around the Y-axis.

[0038] The controller is based on the steady-state torque signal , After deducting the corresponding coupling torques, the equivalent load torque generated by the vertical normal force is obtained: (6) (7) in, The equivalent vertical load moment about the X-axis is... This is the equivalent vertical load moment about the Y-axis.

[0039] Furthermore, the controller bases its calculations on the vertical positive pressure component. and equivalent vertical load moment , Establish the static equilibrium equations. Based on these equations, solve for the coordinates of the equivalent contact point on the fork side. The position along the fork extension / retraction direction is: (8) The position along the width direction of the forks is: (9) This yields the coordinates of the equivalent contact point on the fork side. and the corresponding force values .

[0040] S14, the controller determines the contact position between the target pallet and the rack support beam based on the current storage location of the target pallet and the rack structure parameters.

[0041] Specifically, based on the positional parameters of the rack support beams and the positional relationship between the target pallet and the rack, the effective contact area between the target pallet and the rack support beams is determined. The geometric center of this effective contact area is taken as the equivalent contact point on the rack side. .

[0042] Target pallet total load The vertical normal force component is obtained by reading the multi-dimensional force sensor after the forks have fully loaded the target pallet. The rack-side force value is determined based on the difference between the total load on the target pallet and the force value on the fork side. (10) in, The target pallet total load, This represents the force values ​​on the fork side. This represents the force values ​​on the side of the shelf.

[0043] S15, the controller performs a unified coordinate transformation on the equivalent contact points on the fork side and the equivalent contact points on the rack side, so that all contact points are located in the same spatial coordinate system.

[0044] The contact force distribution information includes: spatial coordinates of the contact point; the corresponding force value at the contact point; and the contact point source type. The contact point source type includes fork-side contact points and rack-side contact points.

[0045] The controller encapsulates the above information to form contact force distribution information between the target pallet and the rack and forks, and sends it to the subsequent support area determination step to determine the real-time support area of ​​the target pallet.

[0046] S2, determine the support area of ​​the target pallet and the center of gravity projection position of the target pallet based on the contact force distribution information. The support area is composed of the effective support area of ​​the shelf and / or forks on the target pallet. In this embodiment, the controller receives the contact force distribution information output in step S1, analyzes the load-bearing ratio of each contact point and its dynamic change trend within a continuous sampling period, eliminates interference points caused by tray deformation or local loose connections, and extracts the effective contact points that truly participate in stable load bearing; then, based on the effective contact points, it reconstructs the real-time support area of ​​the target tray and calculates the center of gravity projection position used for stability control. Specifically, this includes the following steps: S21, assuming there are currently a total of The equivalent contact point, the first The force values ​​corresponding to each equivalent contact point are: The total load-bearing capacity of the target pallet in its current state is: (11) The controller calculates the load-bearing ratio corresponding to each equivalent contact point based on the proportional relationship between the force values ​​at each equivalent contact point and the total load. (12) in, Indicates the first The proportion of the target pallet load borne by each equivalent contact point.

[0047] The controller calculates the load ratio of each equivalent contact point to obtain the load ratio sequence of each equivalent contact point.

[0048] S22, the controller obtains the load ratio of each equivalent contact point at the current sampling time and reads the historical data of the load ratio of each equivalent contact point stored at the previous sampling time.

[0049] For the Each equivalent contact point has a change in its load-bearing ratio. Calculate using the following formula: (13) in, The carrying capacity ratio at the current sampling time. This represents the carrying capacity ratio at the previous sampling time.

[0050] Based on the load-bearing ratio of each equivalent contact point and the change in the load-bearing ratio Effective contact points are selected from all equivalent contact points.

[0051] Optionally, the filtering can be performed in the following manner: for the first An equivalent contact point, when its load ratio Greater than the preset lower limit threshold of the load ratio (Exemplary example taken in this embodiment) ), and the change in the bearing ratio When the symbol is consistent with the migration direction of the current support area, the equivalent contact point is determined to be a valid contact point.

[0052] During the picking process, the support area shifts from the shelf side to the fork side, and the load-bearing ratio of the fork side contact points shows a continuous upward trend. The load-bearing ratio of the contact points on the shelf side shows a continuous downward trend. The opposite is true during the delivery process.

[0053] Optionally, to further improve screening reliability, the controller may also require the load ratio of the contact point to continuously meet the above conditions over multiple consecutive sampling periods (e.g., 3 sampling periods) to avoid misjudgment caused by instantaneous signal fluctuations.

[0054] S23, based on the source type of each effective contact point (fork side or rack side), divide the effective contact points into a set of effective contact points on the fork side and a set of effective contact points on the rack side.

[0055] For the set of effective contact points on the fork side, the controller extracts the spatial coordinates of each effective contact point in a unified reference coordinate system. The horizontal projected coordinates of each effective contact point Using the vertex as the point, construct the fork side support boundary polygon. The support boundary polygon is a convex or non-convex polygon formed by sequentially connecting the horizontal projected coordinates of all effective contact points, and the area enclosed by its boundary is the effective support area on the fork side.

[0056] For the set of effective contact points on the shelf side, the controller uses the same method to extract the horizontal projection coordinates of each effective contact point, constructs the shelf side support boundary polygon, and the area enclosed by its boundary is the shelf side effective support area.

[0057] S24, the controller merges the fork-side support boundary polygon with the rack-side support boundary polygon to determine the support area of ​​the target pallet.

[0058] Optionally, the convex hull envelope of the controller's fork-side support boundary polygon and the shelf-side support boundary polygon is used as the support area boundary of the target pallet. The convex hull envelope refers to the smallest convex polygon that includes all vertices of the two support boundary polygons.

[0059] The support area is determined as follows: Extract the coordinates of all vertices of the fork-side support boundary polygon and the shelf-side support boundary polygon, and merge them into the same vertex set; calculate the convex hull of the vertex set to obtain the minimum convex polygon; determine the area enclosed by the minimum convex polygon as the support area of ​​the target pallet.

[0060] Alternatively, for certain implementation scenarios, the union, intersection, or circumscribed minimum rectangle of two supporting boundary polygons can be used as the boundary of the supporting region. Those skilled in the art can choose an appropriate boundary determination method according to actual control requirements.

[0061] During the picking process, as the forks gradually bear weight, the number and distribution range of effective contact points on the fork side gradually increase, while the number and distribution range of effective contact points on the rack side gradually decrease. The convex hull envelope of the support area gradually transitions from being dominated by the rack side support boundary to being dominated by the fork side support boundary, achieving continuous migration of the support area. During the placing process, the migration direction of the support area is reversed.

[0062] The controller stores the determined support area as a set of polygon vertex coordinates, which serves as input data for subsequent steps.

[0063] S25. Based on the selected effective contact points and their corresponding force values, determine the load center of the target pallet, and project the load center onto the bearing plane of the target pallet to obtain the center of gravity projection position.

[0064] Specifically, the controller maps the spatial coordinates of the effective contact points on the shelf side and the fork side to the same reference coordinate system. Assume there are a total of... The effective contact point, the first The coordinates of the effective contact points within the horizontal bearing surface are: The corresponding force value is Then the horizontal coordinate of the center of the load is Preferably, it is calculated using the following formula: (14) (15) in, For the first The coordinates of each effective contact point in the reference coordinate system.

[0065] Furthermore, the center of the load is projected vertically onto the target pallet bearing plane to obtain the projected position of the target pallet's center of gravity.

[0066] S3 controls the movement of the forks based on the support area and the center of gravity projection position, so that the support area continuously moves between the shelf support area and the fork support area, and the center of gravity projection position is located within the support area during the movement. like Figure 3 As shown, Figure 3 This is a schematic diagram illustrating the migration of the target pallet's support area during the load-bearing transition process. In this embodiment, during the storage and retrieval transition, the target pallet's load-bearing state does not directly switch from rack support to fork support, but rather involves a transitional phase where the rack and forks share the load.

[0067] exist Figure 3 In the main load-bearing stage of the rack shown in (a), the target pallet is mainly supported by the rack beams. The effective contact points on the rack side form the rack-side support area, while the fork side has not yet formed an effective load-bearing area. At this time, the real-time support area... It mainly consists of the shelf-side support area, and the center of gravity projection point C of the target pallet is located inside the real-time support area.

[0068] As the forks gradually extend into the bottom of the target pallet, Figure 3 As shown in (b), the forks begin to make effective contact with the target pallet, and part of the load is transferred from the rack side to the fork side. At this time, the effective contact points on the rack side and the effective contact points on the fork side together form the load-bearing constraint boundary, providing real-time support for the area. It is jointly determined by the rack-side support area and the fork-side support area, and continuously migrates as the fork's position changes.

[0069] Furthermore, such as Figure 3 As shown in (c), when the forks are fully inserted into the bottom of the target pallet and bear the main load, the effective contact point on the fork side forms the main support area, and the load-bearing capacity on the rack side gradually disappears, resulting in a real-time support area. It gradually transitions into an area primarily supported by the fork side.

[0070] During the aforementioned load conversion process, the controller continuously acquires the force information corresponding to the effective contact points on the rack side and the effective contact points on the fork side, and updates the support area in real time according to the changes in the load state of each contact point. The fork movement is adjusted according to the positional relationship between the target pallet center of gravity projection point C and the boundary of the support area, so that the center of gravity projection point C is always located in the real-time support area. internal.

[0071] Through the aforementioned support area migration process, when the target pallet switches between rack support and fork support, it does not rely on fixed position parameters for control, but forms dynamic support constraints based on the actual load-bearing state, thereby achieving stable control during the continuous change of load-bearing state.

[0072] In this embodiment, the controller controls the extension and / or lifting movements of the forks based on the target pallet support area and the center of gravity projection position determined in step S2, so that the center of gravity projection position of the target pallet remains within the real-time support area during the transition from the rack support state to the fork support state, or from the fork support state to the rack support state.

[0073] The support area is a dynamic load-bearing area formed by the combined action of the rack and the forks. As the position of the forks relative to the target pallet changes, the rack-side support area and the fork-side support area continuously migrate.

[0074] Specifically, the following steps are included: S31, the controller continuously acquires the support area information and the center of gravity projection position information determined in step S2 according to the preset sampling period.

[0075] The supporting region is represented by a polygonal region composed of multiple boundary points: (16) in, Indicates the first [unit] on the boundary of the supporting region at the current moment. The coordinates of the points.

[0076] The centroid projection position is represented as follows: (17) in, , These represent the coordinates of the target pallet's center of gravity projection point in the fork reference coordinate system.

[0077] The controller continuously acquires support area information and center of gravity projection position information corresponding to multiple sampling cycles, and forms a dynamic change sequence.

[0078] S32, based on the support region boundary and the centroid projection position at the current sampling time, calculate the shortest distance from the centroid projection position to the support region boundary.

[0079] Specifically, the support area boundary is a polygon, and the controller calculates the centroid projection position. The minimum of the shortest Euclidean distances to each side of the boundary polygon of the supporting region is taken as the safe distance at the current moment. : (18) in, , , To support the region boundary polygon number The equation coefficients of the line containing the edge are determined by the coordinates of the two endpoints of that edge. and The calculation yielded: , , The above coefficients satisfy... The equation of the straight line.

[0080] Furthermore, the controller calculates the safe distance sequence based on the continuous sampling time. To determine the changing trend of the safe distance.

[0081] Optionally, the trend of change of the safety distance can be determined in the following ways: Calculate the difference in safe distance between adjacent sampling times. ,like And within multiple consecutive sampling periods (preferably 3 to 5 periods) If all values ​​are negative, the center of gravity projection is determined to be closer to the boundary of the support area; if And within multiple consecutive sampling periods If all values ​​are positive, then the projected position of the center of gravity is determined to be far from the boundary of the support area; if If the value fluctuates around zero, it is determined that the projected position of the center of gravity and the boundary of the support area remain relatively stable.

[0082] S33, the controller determines whether the center of gravity projection position is approaching the boundary of the support area based on the distance change trend determined in S32.

[0083] When the controller determines that the center of gravity projection position is approaching the boundary of the support area, it adjusts the fork movement state to reduce the migration speed of the support area.

[0084] Specifically, the controller determines the current safe distance. The relationship between the distance and the preset safety distance is used to generate fork movement control commands: when At this time, the controller maintains the current fork movement speed; when When this occurs, the controller reduces the fork extension / retraction speed and / or lifting / lowering speed; when When this happens, the controller stops the movement in the current direction, or controls the forks to generate compensating movement in the opposite direction.

[0085] in, For warning distance, To limit the distance.

[0086] During the picking process, as the forks gradually extend into the bottom of the target pallet, the support area on the fork side gradually expands, while the support area on the shelf side gradually decreases. The controller adjusts the fork movement speed according to the real-time changes in the support area, so that the center of gravity projection position synchronously follows the migration of the support area.

[0087] During the loading process, the controller adjusts the fork movement in the reverse process, so that the target pallet can be smoothly transferred from the fork-supported state to the rack-supported state.

[0088] Through the above control, the target pallet's center of gravity projection position is always kept within the support area, avoiding pallet instability caused by the center of gravity going out of bounds.

[0089] S34, during the movement of the forks, the controller continuously compares the changes in the support area within adjacent sampling periods.

[0090] Specifically, based on the current support area Support area at the previous moment Calculate the change in the support area.

[0091] The changes in the support area include changes in the position of the boundary points of the support area, changes in the number of effective contact points, and changes in the shape of the support area boundary.

[0092] The positional change of the boundary point of the support region Calculate using the following formula: (19) in, For the first The coordinates of each boundary point at the current moment. For the first The coordinates of a boundary point at a given moment.

[0093] S35, the controller compares the current change in the support area with the preset change range to determine whether the support area has undergone discontinuous change.

[0094] The support area is considered to have undergone a discontinuous change when any of the following conditions are met: The number of valid contact points suddenly decreases within adjacent sampling periods; The boundary position of the support area changes beyond the preset range; The position of the center of gravity projection relative to the support area undergoes a sudden change.

[0095] The discontinuous change indicates that the actual load-bearing state of the target pallet is inconsistent with the expected support migration process.

[0096] For example, if the contact point on the shelf side suddenly fails during the picking process, and the fork side has not yet formed a stable support, the support area will change abruptly; or the center of gravity projection will move rapidly due to the uneven loading of the pallet, causing the center of gravity projection position to quickly approach or even exceed the boundary of the support area.

[0097] S36: When the controller determines that the support area has undergone discontinuous changes, it generates a correction control command.

[0098] Specifically: When the change in the support area is small, the controller reduces the fork extension and / or lifting speed to stabilize the support area again. When the degree of change in the support area increases further, the controller controls the forks to move in the opposite direction of the current movement direction, so that the target pallet can regain stable support. When the center of gravity projection position approaches or exceeds the boundary of the support area, the controller controls the forks to stop moving.

[0099] During the correction process, the controller continuously reacquires the support area and the center of gravity projection position, and adjusts the fork movement state according to the positional relationship between the two.

[0100] When detected Furthermore, when the support area resumes its continuous change state, the controller resumes normal access movement.

[0101] S37. During the fork movement, the controller continuously executes the steps S31 to S36 above to form a closed-loop control. At the same time, based on the change in the load ratio between the fork side and the rack side, it determines whether the target pallet has completed the load state transition.

[0102] The controller calculates the load ratio on the fork side and the rack side based on the set of effective contact points determined by S2.

[0103] Let the set of effective contact points on the fork side be... The set of effective contact points on the shelf side is ,but: Fork side load ratio: (20) Shelf side load ratio: (twenty one) in, This indicates the proportion of load borne by the forks. This indicates the proportion of load borne by the shelf.

[0104] During the picking process, as the forks gradually bear the load, the load-bearing ratio on the fork side changes. Gradually increase the load-bearing ratio of the rack side. Gradually decrease.

[0105] when Greater than the first carrying threshold and Less than the second bearing threshold, and At that time, the controller determines that the target pallet has completed the load transfer to the fork side.

[0106] During the loading process, as the forks are gradually released, the load-bearing ratio on the rack side... Gradually increase the load-bearing ratio on the fork side. Gradually decrease.

[0107] when Greater than the third bearing threshold and Less than the fourth load capacity threshold, and At that time, the controller determines that the target pallet has completed the transfer of load to the shelf side.

[0108] Once the controller determines that the load conversion is complete, it generates a load conversion completion signal, controls the forks to stop their extension and / or lifting movements, and ends the fork motion control phase of the current storage and retrieval operation.

[0109] In this way, the controller does not control the fork movement based on fixed position parameters, but dynamically constrains the fork movement process according to the real-time support area formed by the change of the target pallet's load-bearing state, so that the center of gravity projection position of the target pallet is always within the effective support area, thus realizing a stable transition between rack support and fork support.

[0110] Example 2: In this example, a rack access control system for a heavy-duty stacking robot is provided. The system is used to execute the rack access control method for the heavy-duty stacking robot described in Example 1.

[0111] like Figure 2 As shown, the rack access control system of the heavy-duty stacking robot includes: Contact information acquisition device 100, motion state acquisition device 200, control device 300, execution device 400, and storage device 500.

[0112] The contact information acquisition device 100, motion state acquisition device 200, control device 300, execution device 400, and storage device 500 are connected via wired or wireless communication.

[0113] (a) Contact information acquisition device The contact information acquisition device 100 is used to acquire contact force information between the target pallet and the shelf and the forks, and send the contact force information to the control device 300.

[0114] Specifically, the contact information acquisition device 100 includes a multi-dimensional force sensor disposed on the fork.

[0115] In this embodiment, the multi-dimensional force sensor is preferably a six-dimensional force sensor, which is installed on the two forks of the fork. The six-dimensional force sensor is used to detect the spatial force state of the target pallet acting on the fork, including force components along three directions of the sensor coordinate system and torque components around the three directions. The control device 300 determines the equivalent contact point and corresponding force value between the target pallet and the fork based on the force signals collected by the multi-dimensional force sensor, and generates contact force distribution information by combining the rack side support relationship.

[0116] (ii) Motion state acquisition device The motion state acquisition device 200 is used to acquire the current motion state parameters of the forks and send the motion state parameters to the control device 300. Specifically, the motion state acquisition device 200 includes a telescopic state detection unit and a lifting state detection unit. The telescopic state detection unit is used to detect the horizontal telescopic displacement and telescopic speed of the forks; the lifting state detection unit is used to detect the vertical lifting height and lifting speed of the forks.

[0117] In this embodiment, the telescopic state detection unit is preferably an encoding detection device installed on the telescopic drive mechanism; the lifting state detection unit is preferably a displacement detection device installed on the lifting mechanism.

[0118] The hardware structure, connection relationships, and signal flow of each component in this embodiment are described below. The specific data processing and control logic executed by each component has been described in detail in Embodiment 1, and will not be repeated here.

[0119] The system includes a controller, a multi-dimensional force sensor, a motion state detection device, a fork drive mechanism, and a storage device.

[0120] The multi-dimensional force sensor and the motion state detection device are respectively communicatively connected to the controller, the controller is controllably connected to the fork drive mechanism, and the storage device is connected to the controller. The control device 300 determines the current motion state of the fork based on the above motion state parameters and uses it to assist in judging the change process of the support area.

[0121] (iii) Control device The control device 300 is connected to the contact information acquisition device 100, the motion state acquisition device 200, the execution device 400, and the storage device 500, respectively. The control device 300 includes a processor and a memory connected to the processor. The memory stores a computer program, which, when executed by the processor, causes the processor to perform the following steps: Obtain information on the contact force distribution between the target pallet and the rack and forks; The support area and center of gravity projection position of the target pallet are determined based on the contact force distribution information. The fork movement is controlled based on the support area and the center of gravity projection position.

[0122] Specifically, the control device 300 determines the real-time contact state based on the data collected by the contact information acquisition device 100, determines the fork movement state based on the data collected by the movement state acquisition device 200, and generates fork movement control commands based on the real-time support state.

[0123] (iv) Actuating device The actuator 400 is connected to the control device 300 and is used to drive the fork movement according to the control commands output by the control device 300.

[0124] Specifically, the actuator 400 includes a telescopic drive unit and a lifting drive unit.

[0125] The telescopic drive unit is used to drive the forks to extend or retract horizontally; the lifting drive unit is used to drive the forks to rise or fall vertically.

[0126] The control device 300 adjusts the operating status of the telescopic drive unit and / or the lifting drive unit according to the positional relationship between the real-time support area and the center of gravity projection position.

[0127] (v) Storage device The storage device 500 is connected to the control device 300 and is used to store the control program and the data generated during the control process.

[0128] The data stored in the storage device 500 includes contact force distribution information, support area information, center of gravity projection position information, and fork motion control parameters.

[0129] The storage device 500 may employ a non-volatile storage medium.

[0130] (vi) System Working Process When executing the target tray access task: First, the contact information acquisition device 100 collects the contact force information between the target pallet and the forks; the control device 300 determines the current load-bearing state of the target pallet based on the contact force information and generates corresponding contact force distribution information.

[0131] Subsequently, the control device 300 determines the real-time support area and the projected position of the target pallet's center of gravity based on the contact force distribution information.

[0132] During the movement of the forks, the control device 300 continuously updates the support area and the center of gravity projection position, and adjusts the movement state of the actuator 400 according to the positional relationship between the two.

[0133] When the center of gravity projection position is detected to be approaching the boundary of the support area or when the support area undergoes discontinuous changes, the control device 300 adjusts the fork movement speed or direction to bring the target pallet back to a stable support state.

[0134] The above method enables dynamic and stable control of the target pallet during the transition between rack support and fork support states.

[0135] Example 3: This example provides a computer-readable storage medium.

[0136] The computer-readable storage medium stores a computer program, which, when executed by a processor, causes the processor to perform the rack access control method for the heavy-duty stacking robot described in Embodiment 1.

[0137] The computer-readable storage medium is a physical medium. The computer program is stored in the computer-readable storage medium in the form of source code, object code, or executable file.

[0138] In this embodiment, the computer-readable storage medium is a solid-state drive (SSD). In other embodiments, the computer-readable storage medium may also be a read-only memory (ROM), random access memory (RAM), hard disk, optical disc (CD-ROM), USB flash drive, SD card, or other non-volatile storage media.

[0139] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0140] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0141] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0142] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0143] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0144] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A rack access control method for a heavy-duty stacking robot, characterized in that, Includes the following steps: Obtain information on the contact force distribution between the target pallet and the rack and forks; The support area of ​​the target pallet and the center of gravity projection position of the target pallet are determined based on the contact force distribution information. The support area is composed of the effective support area of ​​the rack and / or forks for the target pallet. The movement of the forks is controlled based on the support area and the center of gravity projection position, so that the support area continuously moves between the rack support area and the fork support area, and the center of gravity projection position is located within the support area during the movement. During the fork movement, contact force distribution information is continuously acquired and the support area is updated. The fork movement is adjusted according to the updated support area until the target pallet is removed or stored.

2. The method according to claim 1, characterized in that, The acquisition of contact force distribution information between the target pallet and the rack and forks includes: Acquire the initial multidimensional force signal output by the multidimensional force sensor installed on the fork, and simultaneously acquire the current motion state parameters of the fork; Determine the dynamic structural resonance frequency range of the gantry system based on motion state parameters; The initial multidimensional force signal is adaptively filtered based on the resonant frequency range of the dynamic structure to obtain the steady-state multidimensional force signal. Contact force distribution information is calculated and extracted based on steady-state multidimensional force signals.

3. The method according to claim 2, characterized in that, The step of calculating and extracting contact force distribution information based on steady-state multidimensional force signals includes: The steady-state multidimensional force signal is decomposed into a vertical normal force component and a horizontal shear force component; Based on the structural dimensions of the forks and the horizontal shear force component, calculate the frictional coupling interference torque generated by the horizontal shear force component at the sensor measurement center; The total measured torque contained in the steady-state multidimensional force signal is decoupled from the frictional coupling interference torque by difference, and the equivalent vertical load torque is extracted. A static equilibrium equation is established based on the vertical normal force component and the equivalent vertical load moment. The coordinates of the equivalent contact point and the force values ​​of the target pallet within the support area are solved to generate contact force distribution information.

4. The method according to claim 3, characterized in that, Determining the support area of ​​the target pallet based on the contact force distribution information includes: Calculate the bearing ratio of each equivalent contact point based on the force value corresponding to each equivalent contact point; Based on the load-bearing ratio and the trend of load-bearing ratio changes at adjacent sampling times, effective contact points are selected from each equivalent contact point. Construct corresponding support boundaries based on the spatial location of the effective contact points on the shelf side and the fork side, respectively. The support area of ​​the target pallet is determined based on each support boundary.

5. The method according to claim 4, characterized in that, The step of selecting effective contact points from all equivalent contact points includes: Obtain the load-bearing ratio sequence of each equivalent contact point within a continuous sampling period; The direction of load change at each equivalent contact point is determined based on the load ratio sequence. The equivalent contact point where the direction of load change is consistent with the direction of support area migration is determined as the effective contact point.

6. The method according to claim 5, characterized in that, Determining the projected position of the center of gravity of the target pallet includes: Map the spatial coordinates of the effective contact points on the shelf side and the fork side to the same reference coordinate system; Based on the converted spatial coordinates of each effective contact point and the corresponding force values, the load center of the target pallet is determined. The center of load is projected onto the bearing plane of the target pallet to determine the projected position of the center of gravity.

7. The method according to claim 6, characterized in that, The control of fork movement based on the support area and the center of gravity projection position includes: Acquire information on the changes in the boundary of the support region and the changes in the projected position of the centroid at continuous sampling times; Based on the information on changes in the support area boundary and the information on changes in the center of gravity projection position, determine the trend of the distance change between the center of gravity projection position and the support area boundary; When the trend of distance change indicates that the center of gravity projection position is moving closer to the boundary of the support area, adjust the extension and / or lifting speed of the forks to keep the center of gravity projection position within the support area.

8. The method according to claim 7, characterized in that, The adjustment of fork movement based on the updated support area includes: Obtain the change in the support region at adjacent sampling times; Determine whether the support area has undergone discontinuous changes based on the amount of change in the support area; When the support area changes discontinuously, reduce the movement speed of the forks or control the forks to move in the opposite direction so that the center of gravity projection position is restored to the support area.

9. A rack access control system for a heavy-duty stacking robot, characterized in that, include: One or more processors; as well as A memory connected to the one or more processors, the memory storing a computer program that, when executed by the one or more processors, causes the one or more processors to perform the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, causes the processor to perform the method as described in any one of claims 1 to 8.

Images