Intelligent logistics stacking robot intelligent control system

Abstract

The present application relates to the technical field of palletizing robot control, and relates to an intelligent logistics palletizing robot intelligent control system. The present application determines the spatial coordinates of a pallet corner point and a center point based on original point cloud data of a pallet area, determines the current spatial pose of the pallet according to the relative position vector of the corner point and the pallet center point, analyzes the level stability and area matching degree based on the feasible placement area of each candidate palletizing level in combination with package size data, screens the optimal palletizing level, analyzes the pressure stability of the candidate area based on the candidate area of the package to be palletized in the optimal palletizing level, corrects the placement point coordinates of the package to be palletized, and judges whether the palletizing operation trajectory is in conflict based on the real-time palletizing operation trajectory of the end joint of the robot during execution in combination with the planned palletizing operation trajectory, triggers local trajectory re-planning when in conflict, guarantees the continuity of palletizing operation execution, and improves the safety of robot automatic operation.

Description

Technical Field

[0001] This invention relates to the field of palletizing robot control technology, and specifically to an intelligent control system for an intelligent logistics palletizing robot. Background Technology

[0002] Intelligent logistics palletizing is an automated operation in the warehousing process. The accuracy of pallet posture, the stability of palletizing layers, and the safety of robot movement trajectories directly determine palletizing efficiency, package integrity rate, and operational stability. Pallets are prone to spatial displacement due to placement errors or uneven ground. Improper selection of palletizing layers can easily lead to center of gravity shift and package collapse. Conflicts in movement trajectories can cause equipment collisions and operational interruptions. Building a full-process intelligent control system can improve the automation, accuracy, and safety of logistics palletizing.

[0003] In the prior art, Chinese Patent Publication No. CN120903160A discloses an automatic palletizing method, device, terminal and storage medium. The method collects the package information of the package to be palletized, matches the optimal pallet type with the historical successful pallet type, segments the pallet point cloud through a preset segmentation model to obtain the pallet gap and generate the palletizing path, and corrects the gripping posture of the flexible gripper in real time through a torque sensor. Finally, the palletizing 3D model is bound to the AGV to realize automatic storage, improving the adaptability and space utilization of the palletizing system.

[0004] The existing technology has the following problems: 1. The existing technology completes palletizing by matching the pallet type and planning the path, but does not accurately locate and correct the actual spatial posture of the pallet. It lacks analysis of the spatial coordinates of the pallet corner point and center point as well as the pallet spatial deflection angle, which leads to a deviation between the placement point coordinates and the actual spatial state of the pallet, causing the pallet placement position to shift, which is prone to problems such as palletizing tilt and package falling, affecting the palletizing accuracy and space utilization.

[0005] 2. Existing technologies generate palletizing paths by calculating pallet gaps through point cloud segmentation, without performing layered screening analysis on the stability and area matching of the palletizing layers. This makes it easy for palletizing to collapse due to layer center of gravity shift and insufficient contact area, resulting in insufficient stability of the palletizing layers and increasing palletizing safety risks.

[0006] 3. Existing technology controls robotic arms to perform palletizing according to the palletizing path, but does not perform real-time conflict judgment and local trajectory replanning for the palletizing trajectory. When the real-time position of the end joint deviates from the planned trajectory during the robot's execution, it cannot be dynamically adjusted, which poses a risk of motion trajectory conflict and collision, leading to palletizing execution interruption and equipment safety hazards, reducing the continuity and safety of palletizing operations. Summary of the Invention

[0007] This invention aims to overcome the shortcomings of existing technologies and provide an intelligent control system for intelligent logistics palletizing robots. By accurately sensing pallet posture, stably selecting palletizing layers, correcting placement coordinates for pressure, and dynamically avoiding motion conflicts, the system improves the stability and safety of logistics palletizing.

[0008] The technical solution adopted by the present invention to solve its technical problem is as follows: The present invention provides an intelligent control system for an intelligent logistics palletizing robot, including a pallet pose determination module, a palletizing layer screening module, a placement coordinate generation module, and a motion conflict determination module.

[0009] The connections between the modules are as follows: the pallet pose determination module is connected to the palletizing level filtering module, and the placement coordinate generation module is connected to both the palletizing level filtering module and the motion conflict determination module.

[0010] The pallet pose determination module determines the spatial coordinates of the pallet corners and center point based on the original point cloud data of the pallet area. It calculates the pallet spatial deflection angle based on the relative position vectors of the corners and the center point of the pallet to determine the current spatial pose of the pallet.

[0011] The palletizing layer selection module determines the feasible placement area for each candidate palletizing layer based on the outline occupied by already palletized packages on the pallet. Combined with the size data of the packages to be palletized, it analyzes the stability of the layer and the area matching degree to select the optimal palletizing layer.

[0012] The placement coordinate generation module analyzes the pressure stability of candidate areas in the optimal palletizing layer based on the candidate areas to be palletized, and combines the current spatial pose of the pallet to correct and generate the placement point coordinates of the package to be palletized.

[0013] The motion conflict determination module performs trajectory planning by combining the coordinates of the placement point with the current pose of the robot's end effector, and outputs the palletizing trajectory. Based on the real-time palletizing trajectory of the robot's end joints during execution, it determines whether the palletizing trajectory conflicts. If the palletizing trajectory conflicts, it triggers local trajectory replanning.

[0014] Compared with the prior art, the present invention has the following beneficial effects: (1) Based on the original point cloud data of the pallet area, the present invention determines the spatial coordinates of the pallet corner point and the center point, and calculates the pallet spatial deflection angle according to the relative position vector of the corner point and the center point of the pallet, determines the current spatial pose of the pallet, realizes the accurate perception and positioning of the pallet spatial pose, eliminates the influence of pallet tilt and deflection on the placement accuracy, improves the accuracy of parcel palletizing and placement and space utilization, and reduces the risk of parcel falling.

[0015] (2) Based on the outline occupied by the already stacked packages on the pallet, the present invention determines the feasible placement area of ​​each candidate stacking level, analyzes the stability and area matching of the level by combining the size data of the packages to be stacked, and selects the optimal stacking level, realizes the hierarchical evaluation and stable selection of stacking levels, improves the spatial adaptability and structural stability of the stacking level selection, avoids the problem of package tipping and pallet center of gravity imbalance, and improves the stability and fault tolerance of the stacking operation.

[0016] (3) Based on the candidate areas of the package to be palletized in the optimal palletizing level, the present invention analyzes the pressure stability of the candidate areas, and generates the placement point coordinates of the package to be palletized by combining the current spatial pose correction of the pallet, thereby realizing the pressure stability screening and spatial pose correction of the placement point coordinates, reducing the local crushing and slippage problems of the package, ensuring the pressure balance of the palletizing structure, and improving the package integrity rate.

[0017] (4) This invention performs trajectory planning by placing the coordinates of the placement point and the current end pose of the robot, outputs the palletizing trajectory, and determines whether the palletizing trajectory conflicts based on the real-time palletizing trajectory of the end joint during the execution of the robot. When there is a conflict, the local trajectory is replanned to avoid the risk of motion trajectory conflict and collision, ensure the continuity of palletizing operation, and improve the safety of robot automated operation. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments 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.

[0019] Figure 1 This is a schematic diagram of the system module connections of the present invention; Figure 2 This is a schematic diagram of the steps for evaluating the hierarchical stability at different placement positions in this invention; Figure 3 This is a schematic diagram of the process for generating the coordinate correction of the placement point of the package to be palletized in this invention. Detailed Implementation

[0020] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention. Furthermore, it should be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale.

[0021] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use. Techniques, methods, and apparatus known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and apparatus should be considered part of the specification.

[0022] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0023] Please see Figure 1 As shown, the present invention provides an intelligent control system for an intelligent logistics palletizing robot, including a pallet pose determination module, a palletizing level screening module, a placement coordinate generation module, and a motion conflict determination module.

[0024] The connections between the modules are as follows: the pallet pose determination module is connected to the palletizing level filtering module, and the placement coordinate generation module is connected to both the palletizing level filtering module and the motion conflict determination module.

[0025] The pallet pose determination module determines the spatial coordinates of the pallet corners and center point based on the original point cloud data of the pallet area. It calculates the pallet spatial deflection angle based on the relative position vectors of the corners and the center point of the pallet to determine the current spatial pose of the pallet.

[0026] The palletizing layer selection module determines the feasible placement area for each candidate palletizing layer based on the outline occupied by already palletized packages on the pallet. Combined with the size data of the packages to be palletized, it analyzes the stability of the layer and the area matching degree to select the optimal palletizing layer.

[0027] The placement coordinate generation module analyzes the pressure stability of candidate areas in the optimal palletizing layer based on the candidate areas to be palletized, and combines the current spatial pose of the pallet to correct and generate the placement point coordinates of the package to be palletized.

[0028] The motion conflict determination module performs trajectory planning by combining the coordinates of the placement point with the current pose of the robot's end effector, and outputs the palletizing trajectory. Based on the real-time palletizing trajectory of the robot's end joints during execution, it determines whether the palletizing trajectory conflicts. If the palletizing trajectory conflicts, it triggers local trajectory replanning.

[0029] Considering that pallets may deviate from their standard spatial orientation at logistics operations due to uneven ground or accumulated errors from previous palletizing, current technologies do not accurately perceive the actual spatial orientation of the pallet and directly plan palletizing based on ideal pallet coordinates. This leads to discrepancies between the placement point coordinates and the actual spatial state of the pallet, causing pallet placement position offsets. Therefore, it is necessary to process the original point cloud data of the pallet area and extract geometric features to accurately determine the spatial coordinates of the pallet's corners and center point, and then calculate the pallet's spatial deflection angle.

[0030] Based on this, the specific implementation of determining the spatial coordinates of the pallet corner and center point in this invention includes: S111, performing distance statistics on the original point cloud data of the pallet area, removing outlier point clouds whose distance from neighboring point clouds is greater than a preset neighborhood distance threshold, and obtaining a preprocessed pallet point cloud dataset.

[0031] It should be noted that the preset neighborhood distance threshold is set based on the typical size of the pallet and the point cloud acquisition density. For example, for a standard 1200mm × 1000mm wooden pallet, the preset neighborhood distance threshold is set to 50mm; for plastic or metal pallets, the preset neighborhood distance threshold can be adjusted to 30mm based on the sparseness of the point cloud caused by surface reflection characteristics. When the distance between a point cloud and its nearest neighbor point cloud exceeds this threshold, the point cloud is determined to be an outlier and is removed.

[0032] S112. Perform plane fitting on the preprocessed pallet point cloud dataset to obtain an approximate pallet plane. Project the point cloud in the approximate pallet plane onto a two-dimensional plane to obtain a projection point set.

[0033] The plane fitting process employs a random sample consensus algorithm to fit the largest plane from the pallet point cloud dataset as the approximate plane of the pallet, and its plane normal vector represents the orientation of the upper surface of the pallet. The random sample consensus algorithm is a well-known existing technique and will not be elaborated upon here.

[0034] S113. Perform boundary tracking on the projection point set, extract the pallet edge contour, and filter the main edge direction and adjacent edge direction of the pallet based on the straight line segment features of the edge contour.

[0035] Preferably, in a specific embodiment of the present invention, the filtering method for the main side direction and the adjacent side direction is as follows: all straight line segments are extracted by fitting the straight line segments of the tray edge contour, the direction of the longest straight line segment is taken as the main side direction, and the direction of the straight line segment that is perpendicular to the main side direction is taken as the adjacent side direction.

[0036] S114. Take the intersection of the line segments corresponding to the main side direction and the adjacent side direction as the pallet corner point, and determine the spatial coordinates of the pallet center point based on the geometric distance between the pallet corner points and the intersection of the diagonals.

[0037] Considering that the spatial coordinates of the pallet corners and center point only reflect the pallet's geometric position and cannot characterize its tilt in three-dimensional space, existing technologies lack analysis of the pallet's spatial deflection angle, resulting in the inability to correct the subsequent placement point coordinates based on the pallet's actual tilt state. Therefore, it is necessary to construct a local coordinate system for the pallet and analyze the projected components of the relative position vector to calculate the pallet's rotation angle around each coordinate axis.

[0038] Based on this, the specific implementation of determining the current spatial pose of the tray in this invention includes: S121, constructing a local coordinate system for the tray with the spatial coordinates of the center point of the tray as the origin, and transforming the spatial coordinates of each corner point of the tray into the local coordinate system. The local coordinate system is constructed as follows: with the spatial coordinates of the center point of the tray as the origin, the main side direction as the positive X-axis, the adjacent side direction as the positive Y-axis, and the upward direction of the normal vector of the approximate plane of the tray as the positive Z-axis, a spatial coordinate system is formed.

[0039] S122. Extract the relative position vectors of each pallet corner point and the pallet center point, obtain the projection components of each relative position vector in the horizontal plane of the pallet local coordinate system, calculate the angle between the projection components corresponding to different pallet corner points and the pallet local coordinate system, and use it as the heading angle of the pallet around the vertical coordinate axis.

[0040] S123. Extract the projection components of each relative position vector in the vertical plane of the local coordinate system of the pallet, calculate the ratio of the height difference between different pallet corner points and the center point of the pallet to the horizontal distance, and obtain the pitch angle and roll angle of the pallet around the horizontal coordinate axis through arctangent calculation.

[0041] S124. Combine the heading angle, pitch angle and roll angle to form the current spatial attitude of the tray.

[0042] Preferably, in a specific embodiment of the present invention, the heading angle of the pallet around the vertical coordinate axis is the average value of the angle between the projection components corresponding to different pallet corner points and the X-axis direction of the local coordinate system of the pallet, which characterizes the rotational deflection of the pallet in the horizontal plane.

[0043] The pitch angle and roll angle are calculated as follows: the relative position vector is projected onto the XOZ vertical plane and the YOZ vertical plane respectively, the ratio of the height difference in the Z-axis direction to the horizontal distance in the X-axis direction is calculated and the pitch angle is obtained by arctangent calculation, the ratio of the height difference in the Z-axis direction to the horizontal distance in the Y-axis direction is calculated and the roll angle is obtained by arctangent calculation, which characterizes the degree of tilt of the pallet in the vertical plane.

[0044] This invention determines the spatial coordinates of the pallet corners and center point based on the original point cloud data of the pallet area, and calculates the pallet spatial deflection angle based on the relative position vector of the corners and the center point of the pallet to determine the current spatial pose of the pallet. This enables accurate perception and positioning of the pallet spatial pose, eliminates the impact of pallet tilting and deflection on placement accuracy, improves the accuracy of parcel palletizing and placement and space utilization, and reduces the risk of parcels falling.

[0045] Considering that the spatial distribution of the palletized packages on a pallet is hierarchical, with varying available space and support conditions at different heights, current technologies do not perform layered selection of palletizing layers. Instead, they only calculate gaps through overall point cloud segmentation, resulting in package placement positions that do not consider layer stability and area adaptability. Therefore, it is necessary to achieve optimal palletizing layer selection through a comprehensive evaluation of the palletized package outline extraction, feasible placement area determination, area matching, and layer stability.

[0046] Based on this, the specific implementation of the analysis of hierarchical stability and area matching degree in this invention includes: S211, based on the outline occupied by the already stacked packages on the pallet, extracting the projected area of ​​the occupied area of ​​the already stacked packages corresponding to each stacking height, and dividing adjacent stacking heights with the same projected area into the same candidate stacking level.

[0047] Specifically, the outline of the stacked packages is scanned at preset heights (such as 100mm or 150mm) along the vertical direction of the palletizing process. The projected area of ​​the stacked package occupancy area on the horizontal plane at each stacking height is counted. All stacking heights with the same projected area above the upper surface of the pallet are grouped into the same candidate stacking level to represent the occupancy distribution of the stacked packages at that level.

[0048] S212. Determine the effective placement area for each candidate palletizing level, and take the area within the effective placement area that is not covered by the already palletized and wrapped area as the feasible placement area for each candidate palletizing level.

[0049] It should be noted that the effective placement area refers to the allowed placement area of ​​the pallet in each candidate palletizing level corresponding to the adjacent candidate palletizing level. For example, if the candidate palletizing level adjacent to the upper surface of the pallet is designated as candidate palletizing level A, and the candidate palletizing level adjacent to the upper surface of candidate palletizing level A is designated as candidate palletizing level B, then the effective placement area of ​​candidate palletizing level A is the allowed placement area on the upper surface of the pallet, and the effective placement area of ​​candidate palletizing level B is the allowed placement area of ​​candidate palletizing level A.

[0050] The feasible placement area is the remaining continuous area after subtracting the area occupied by already palletized packages from the effective placement area, and is used to assess the placement feasibility of the packages to be palletized.

[0051] S213. Simulate placing the parcel to be palletized in different positions within the feasible placement areas of each candidate palletizing level, and obtain the bottom area of ​​the parcel to be palletized at different placement positions based on the size data of the parcel to be palletized.

[0052] S214. Compare the bottom area of ​​the package to be stacked with the area of ​​the geometric bounding rectangle of the corresponding feasible placement area, and calculate the area occupancy ratio of the package to be stacked in different placement positions as the area matching degree. The area occupancy ratio is the ratio of the bottom area of ​​the package to be stacked to the area of ​​the geometric bounding rectangle of the corresponding feasible placement area. The closer the ratio is to 1, the higher the area matching degree.

[0053] S215. Obtain the overall center of gravity of the pallet after the parcels to be palletized are placed in different positions, calculate the horizontal projection offset distance between the overall center of gravity of the pallet and the center point of the pallet, and evaluate the layer stability of different placement positions by combining the interlayer contact area ratio of the parcels to be palletized.

[0054] Preferably, such as Figure 2 As shown, in a specific embodiment of the present invention, the step of evaluating the hierarchical stability of different placement positions is as follows: S2151, based on the overall outline of the pallet after the pallet to be palletized is placed in different placement positions, determine the overall center of gravity of the pallet after different placement positions.

[0055] S2152. Compare the projections of the pallet's overall center of gravity and the pallet's center point onto a horizontal plane, and calculate the horizontal projection offset distance between the pallet's overall center of gravity and the pallet's center point. This horizontal offset of the center of gravity characterizes the degree of eccentricity of the pallet's load-bearing capacity; a larger offset indicates that the pallet is more prone to tipping over.

[0056] S2153. Calculate the interlayer contact area of ​​packages placed in different locations, and then calculate the ratio of the interlayer contact area to the bottom area of ​​the package at the corresponding location. This ratio is used as the percentage of the interlayer contact area. The interlayer contact area is the actual contact area between the bottom surface of the package to be stacked and the top surface of the already stacked package below it. A larger percentage of the interlayer contact area indicates a lower risk of slippage after the package is placed.

[0057] S2154. Based on the horizontal offset of the center of gravity and the proportion of interlayer contact area, the layer stability of different placement positions is obtained by weighted fusion according to the preset weight.

[0058] It should be noted that the formula for calculating the hierarchical stability is as follows: .

[0059] In the formula, For hierarchical stability, This is the horizontal offset of the center of gravity. The maximum allowable offset is set to half the length of the tray diagonal. This represents the percentage of interlayer contact area. The weighting coefficient for the horizontal offset of the center of gravity. This is the weighting coefficient for the proportion of interlayer contact area, and In this invention, and All values ​​are set to 0.5. Implementers can also adjust the weighting based on the characteristics of the package material, for example, increasing it for fragile items. The value of is important for improving the flexibility of wrapping. The value of .

[0060] Based on this, the specific implementation of selecting the optimal palletizing level in this invention includes: S221, based on the area matching degree and level stability of the package to be palletized placed in different positions in each candidate palletizing level, determining the area matching degree sequence number and level stability sequence number of different placement positions by sorting. Specifically, the area matching degree of all placement positions within the same candidate palletizing level is arranged in descending order and assigned a sequence number, and the level stability is arranged in descending order and assigned a sequence number.

[0061] S222. Assign values ​​to the area matching degree sequence number and the hierarchical stability degree sequence number for different placement locations, determine the comprehensive assignment for different placement locations, filter the area matching degree and hierarchical stability degree corresponding to the placement location with the maximum comprehensive assignment, and mark them as representative area matching degree and representative hierarchical stability degree.

[0062] The assignment method can be the reciprocal assignment method, that is, the smaller the serial number, the higher the assignment value. The comprehensive assignment value is the sum of the area matching degree assignment value and the hierarchical stability degree assignment value.

[0063] S223. Calculate the representative area matching degree and representative layer stability of the parcels to be palletized in each candidate palletizing layer. Similarly, determine the comprehensive assignment of each candidate palletizing layer by using the comprehensive assignment method for different placement positions. Select the candidate palletizing layer corresponding to the maximum comprehensive assignment as the optimal palletizing layer.

[0064] This invention determines the feasible placement area for each candidate palletizing level based on the outline occupied by already palletized packages on the pallet. It analyzes the stability and area matching of the level by combining the size data of the packages to be palletized, and selects the optimal palletizing level. This achieves layered evaluation and stable selection of palletizing levels, improves the spatial adaptability and structural stability of palletizing level selection, avoids problems such as package tipping and pallet center of gravity imbalance, and enhances the stability and fault tolerance of palletizing operations.

[0065] Considering the potential for uneven support within the candidate areas of the optimal palletizing layer, placing packages in weakly supported areas could easily lead to localized compressive stress concentration, causing package deformation or collapse. Current technology does not analyze the compressive stability of the candidate areas. Therefore, it is necessary to analyze the contact area of ​​support points within the candidate areas to select the placement points with the highest compressive stability, and then perform coordinate correction based on the pallet's spatial orientation.

[0066] Based on this, such as Figure 3 As shown, the specific implementation of correcting and generating the placement point coordinates of packages to be palletized in this invention includes: S31. Select all candidate areas for placing the package to be palletized from the feasible placement areas of the optimal palletizing level, and obtain the contact area of ​​different support points within the bottom coverage area of ​​the package to be palletized.

[0067] It should be noted that the candidate area is all areas within the feasible placement area that can completely accommodate the bottom outline of the package to be stacked; the support point is the actual contact position between the top surface of the already stacked package or the surface of the pallet within the candidate area and the bottom surface of the package to be stacked, and the contact area of ​​the support point is calculated by the projected area of ​​the local area where the support point is located.

[0068] S32. Based on the weight of the package to be stacked and the contact area of ​​different support points, calculate the local compressive stress of each support point, and mark the support points with local compressive stress lower than the package bearing pressure threshold as stable support points.

[0069] Preferably, in a specific embodiment of the present invention, the local compressive stress is the pressure value of the weight of the package to be stacked evenly distributed to each support point according to the contact area, and the local compressive stress is calculated as follows: .

[0070] In the formula, It is a local compressive stress. The load borne by the support point W represents the weight of the packages to be stacked, and n represents the total number of support points. The contact area of ​​the support point is the largest local compressive stress, indicating that the support point bears higher pressure and the risk of local crushing of the package is greater.

[0071] It should be noted that the package pressure threshold is set according to the packaging material of the package to be stacked. For example, for corrugated cardboard packaging, the pressure threshold is set to 0.15 MPa; for wooden box packaging, the pressure threshold is set to 0.5 MPa. When the local compressive stress at the support point exceeds the package pressure threshold, the support point is determined to have a risk of crushing and is marked as an unstable support point; otherwise, it is marked as a stable support point.

[0072] S33. The ratio of the contact area of ​​all stable support points to the total contact area is taken as the pressure stability. The center point of the candidate area with the highest pressure stability is selected as the placement point of the package to be stacked.

[0073] S34. Based on the current spatial pose of the pallet, perform coordinate correction on the placement point of the package to be palletized, and generate the placement point coordinates of the package to be palletized.

[0074] Specifically, a rotation matrix is ​​constructed based on the heading angle, pitch angle, and roll angle in the current spatial pose of the pallet. The placement point is then transformed in the local coordinate system of the pallet to obtain the corrected coordinates of the placement point, ensuring that the placement point accurately matches the actual spatial state of the pallet.

[0075] This invention analyzes the pressure-bearing stability of candidate regions within the optimal palletizing layer of a package to be palletized. It then generates the placement point coordinates of the package by correcting the current spatial pose of the pallet. This enables the screening of pressure-bearing stability and spatial pose correction of the placement point coordinates, reducing local crushing and slippage issues of packages, ensuring the pressure-bearing balance of the palletizing structure, and improving the package integrity rate.

[0076] Considering that palletizing robots are susceptible to dynamic obstacles and joint servo errors during execution, causing deviations between the actual and planned trajectories of the end-effectors, current technologies lack real-time conflict monitoring and dynamic adjustment mechanisms for the running trajectory. This easily leads to trajectory conflicts and collisions, resulting in palletizing interruptions and equipment safety hazards. Therefore, it is necessary to achieve real-time determination of trajectory conflicts and local replanning through real-time tracking and deviation analysis of the robot's end-effector positions.

[0077] Based on this, the specific implementation of the motion conflict determination module in this invention includes: S41, based on the spatial transformation relationship between the robot reference coordinate system and the pallet local coordinate system, transforming the coordinates of the placement point of the package to be palletized to the robot reference coordinate system.

[0078] The spatial transformation relationship is determined by a rotation and translation transformation matrix constructed from the spatial coordinates of the pallet's center point and the pallet's heading angle, pitch angle, and roll angle, thereby unifying the pallet's local coordinate system with the robot's reference coordinate system.

[0079] S42. Extract the current end pose of the robot's end joint, compare it with the coordinates of the placement point in the robot's reference coordinate system, and output the palletizing trajectory of the robot's end joint using path planning.

[0080] Preferably, in one specific embodiment of the present invention, the path planning algorithm employs cubic spline curve planning or obstacle avoidance path planning to generate a smooth and continuous palletizing trajectory, ensuring that the robot's end effector movement is free from abrupt changes and avoiding joint impacts. The cubic spline curve planning or obstacle avoidance path planning are both existing techniques well-known to those skilled in the art, and will not be described further in this invention.

[0081] S43. Obtain the real-time palletizing trajectory of the robot's end joint during execution. If the real-time palletizing trajectory coincides with the palletizing trajectory of the robot's end joint, it is determined that there is no conflict in the palletizing trajectory. Wherein, the palletizing trajectory coincides when the deviation distance between the real-time palletizing trajectory and the palletizing trajectory of the robot's end joint is less than or equal to the trajectory deviation distance.

[0082] Conversely, the deviation distance time sequence between the real-time palletizing trajectory and the palletizing trajectory of the robot's end joint is obtained. If there is a deviation distance in the deviation distance time sequence that is greater than the trajectory deviation distance, and the duration of the deviation exceeds the preset deviation duration, then it is determined that there is a conflict in the palletizing trajectory.

[0083] It should be noted that the trajectory deviation distance is set according to the positioning accuracy of the robot end effector, for example, 2cm; the preset deviation duration is set according to the palletizing operation cycle, for example, 500ms. When the deviation distance is greater than the trajectory deviation distance and the deviation duration exceeds the preset deviation duration, it indicates that the trajectory deviation has exceeded the normal control error range, and there is a risk of collision or trajectory conflict.

[0084] S44. Obtain the starting time of the motion conflict, extract the actual position of the robot's end joint corresponding to the starting time, and use it as the initial position of the local running trajectory.

[0085] The coordinates of the placement point of the package to be palletized are extracted from the palletizing operation trajectory and used as the end point of the local operation trajectory. The trajectory is replanned between the initial position and the end point of the local operation trajectory. The local operation trajectory is generated by path planning and the corresponding trajectory segment in the palletizing operation trajectory is replaced.

[0086] The planning method for the local running trajectory is the same as that for the palletizing running trajectory, ensuring that the replanned trajectory is smoothly connected to the remaining trajectory.

[0087] This invention plans the palletizing trajectory by combining the coordinates of the placement point with the current pose of the robot's end effector. Based on the real-time palletizing trajectory of the robot's end joints during execution, it determines whether there is a conflict in the palletizing trajectory. When a conflict occurs, it triggers local trajectory replanning to avoid motion trajectory conflicts and collision risks, ensure the continuity of palletizing operations, and improve the safety of automated robot operations.

[0088] 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.

[0089] 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.

[0090] 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.

[0091] 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.

[0092] Finally, 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. An intelligent control system for an intelligent logistics palletizing robot, characterized in that, include: The pallet pose determination module determines the spatial coordinates of the pallet corners and center point based on the original point cloud data of the pallet area, calculates the pallet spatial deflection angle based on the relative position vectors of the corners and the center point of the pallet, and determines the current spatial pose of the pallet. The palletizing level selection module determines the feasible placement area for each candidate palletizing level based on the outline occupied by already palletized packages on the pallet. Combined with the size data of the packages to be palletized, it analyzes the stability of the level and the area matching degree to select the optimal palletizing level. The placement coordinate generation module analyzes the pressure-bearing stability of candidate areas in the optimal palletizing layer based on the candidate areas to be palletized, and, combined with the current spatial pose of the pallet, corrects and generates the placement point coordinates of the package to be palletized. The motion conflict determination module performs trajectory planning by combining the coordinates of the placement point with the current pose of the robot's end effector, and outputs the palletizing trajectory. Based on the real-time palletizing trajectory of the robot's end joints during execution, it determines whether the palletizing trajectory conflicts. If the palletizing trajectory conflicts, it triggers local trajectory replanning.

2. The intelligent control system for an intelligent logistics palletizing robot according to claim 1, characterized in that, The method for determining the spatial coordinates of the pallet corner point and center point is as follows: Distance statistics are performed on the original point cloud data of the tray area, and outlier point clouds that are more than the preset neighborhood distance threshold are removed to obtain the preprocessed tray point cloud dataset. Plane fitting is performed on the preprocessed pallet point cloud dataset to obtain an approximate pallet plane. The point cloud in the approximate pallet plane is then projected onto a two-dimensional plane to obtain the projected point set. Boundary tracking is performed on the projection point set to extract the pallet edge contour. The main edge direction and adjacent edge direction of the pallet are selected based on the straight line segment features of the edge contour. The intersection of the line segments corresponding to the main side direction and the adjacent side direction is taken as the corner point of the pallet. The spatial coordinates of the center point of the pallet are determined based on the geometric distance between the corner points of the pallet and the intersection of the diagonals.

3. The intelligent control system for an intelligent logistics palletizing robot according to claim 1, characterized in that, The method for determining the current spatial pose of the tray is as follows: Construct a local coordinate system for the pallet with the spatial coordinates of the center point of the pallet as the origin, and transform the spatial coordinates of each corner point of the pallet into the local coordinate system of the pallet; Extract the relative position vectors of each pallet corner point and the pallet center point, obtain the projection components of each relative position vector in the horizontal plane of the pallet local coordinate system, calculate the angle between the projection components corresponding to different pallet corner points and the pallet local coordinate system, and use it as the heading angle of the pallet around the vertical coordinate axis; Extract the projection components of each relative position vector in the vertical plane of the local coordinate system of the pallet, calculate the ratio of the height difference between different pallet corner points and the pallet center point to the horizontal distance, and obtain the pitch angle and roll angle of the pallet around the horizontal coordinate axis through arctangent calculation; The heading angle, pitch angle, and roll angle are combined to form the current spatial attitude of the tray.

4. The intelligent control system for an intelligent logistics palletizing robot according to claim 1, characterized in that, The method for analyzing hierarchical stability and area matching is as follows: Based on the outline occupied by the already stacked packages on the pallet, the projected area of ​​the area occupied by the already stacked packages corresponding to each stacking height is extracted, and adjacent stacking heights with the same projected area are divided into the same candidate stacking level. Determine the effective placement area for each candidate palletizing level, and take the area within the effective placement area that is not covered by the already palletized and wrapped area as the feasible placement area for each candidate palletizing level. Simulate placing packages to be palletized in different positions within the feasible placement areas of each candidate palletizing level, and obtain the bottom area of ​​the packages to be palletized at different positions based on the size data of the packages to be palletized; The bottom area of ​​the package to be palletized is compared with the area of ​​the geometric bounding rectangle of the feasible placement area at the corresponding level. The area occupancy ratio of the package to be palletized in different placement positions is calculated as the area matching degree. Obtain the overall center of gravity of the pallet after the parcels are placed in different positions, calculate the horizontal projection offset distance between the overall center of gravity of the pallet and the center point of the pallet, and evaluate the layer stability of different placement positions by combining the interlayer contact area ratio of the parcels.

5. The intelligent control system for an intelligent logistics palletizing robot according to claim 4, characterized in that, The steps for evaluating the hierarchical stability at different placement locations are as follows: Based on the overall outline of the pallet after the parcels to be palletized are placed in different placement positions, the overall center of gravity of the pallet after different placement positions is determined; Compare the projections of the pallet's overall center of gravity and the pallet's center point onto a horizontal plane, and calculate the horizontal projection offset distance between the pallet's overall center of gravity and the pallet's center point, which is taken as the horizontal offset of the center of gravity. The interlayer contact area of ​​the parcels to be palletized is calculated after they are placed in different positions. The ratio of the interlayer contact area to the bottom area of ​​the parcels to be palletized at the corresponding positions is used as the percentage of interlayer contact area. Based on the horizontal offset of the center of gravity and the proportion of interlayer contact area, the layer stability of different placement positions is obtained by weighted fusion according to preset weights.

6. The intelligent control system for an intelligent logistics palletizing robot according to claim 4, characterized in that, The method for selecting the optimal palletizing level is as follows: Based on the area matching degree and level stability of the parcels to be palletized in different placement positions in each candidate palletizing level, the area matching degree sequence number and level stability sequence number of different placement positions are determined by sorting. Assign values ​​to the area matching degree index and the hierarchical stability index of different placement locations, determine the comprehensive assignment value of different placement locations, filter the area matching degree and hierarchical stability of the placement location corresponding to the maximum comprehensive assignment value, and mark them as representative area matching degree and representative hierarchical stability. The representative area matching degree and representative layer stability of the parcels to be palletized are statistically analyzed for each candidate palletizing layer. Similarly, the comprehensive assignment value of each candidate palletizing layer is determined, and the candidate palletizing layer corresponding to the largest comprehensive assignment value is selected as the optimal palletizing layer.

7. The intelligent control system for an intelligent logistics palletizing robot according to claim 1, characterized in that, The method for correcting and generating the placement point coordinates of the packages to be palletized is as follows: From the feasible placement area of ​​the optimal palletizing level, select all candidate areas for placing the package to be palletized, and obtain the contact area of ​​different support points within the bottom coverage area of ​​the package to be palletized; Based on the weight of the package to be palletized and the contact area of ​​different support points, the local compressive stress of each support point is calculated, and the support points with local compressive stress lower than the package bearing capacity threshold are marked as stable support points. The ratio of the contact area of ​​all stable support points to the total contact area is used as the compressive stability. The center point of the candidate area with the highest compressive stability is selected as the placement point for the package to be stacked. Based on the current spatial pose of the pallet, the coordinates of the placement point of the package to be palletized are corrected to generate the coordinates of the placement point of the package to be palletized.

8. The intelligent control system for an intelligent logistics palletizing robot according to claim 1, characterized in that, The steps for outputting the palletizing operation trajectory are as follows: Based on the spatial transformation relationship between the robot's reference coordinate system and the pallet's local coordinate system, the coordinates of the placement point of the package to be palletized are transformed to the robot's reference coordinate system; Extract the current end pose of the robot's end joint, compare it with the coordinates of the placement point in the robot's reference coordinate system, and output the palletizing trajectory of the robot's end joint using path planning.

9. The intelligent control system for an intelligent logistics palletizing robot according to claim 8, characterized in that, The method for determining whether the palletizing trajectory conflicts is as follows: The real-time palletizing trajectory of the robot's end joint during execution is obtained. If the real-time palletizing trajectory coincides with the palletizing trajectory of the robot's end joint, it is determined that there is no conflict in the palletizing trajectory. Conversely, the deviation distance time sequence between the real-time palletizing trajectory and the palletizing trajectory of the robot's end joint is obtained. If there is a deviation distance in the deviation distance time sequence that is greater than the trajectory deviation distance, and the duration of the deviation exceeds the preset deviation duration, then it is determined that there is a conflict in the palletizing trajectory.

10. The intelligent control system for an intelligent logistics palletizing robot according to claim 9, characterized in that, The steps for triggering the replanning of the local running trajectory are as follows: Obtain the starting time of the motion conflict, extract the actual position of the robot's end joint corresponding to the starting time, and use it as the initial position of the local running trajectory; The coordinates of the placement point of the package to be palletized are extracted from the palletizing operation trajectory and used as the end point of the local operation trajectory. The trajectory is replanned between the initial position and the end point of the local operation trajectory. The local operation trajectory is generated by path planning and the corresponding trajectory segment in the palletizing operation trajectory is replaced.

Images