Full-automatic stacking method and system

Abstract

The invention provides a full-automatic stacking method and system.The method is completed on the basis of a stacking robot, physical parameter information of goods to be stacked is obtained, a single-layer stacking reference scheme is generated on the basis of the physical parameter information and stacking environment parameters, and a final stacking scheme is generated on the basis of the single-layer stacking reference scheme; based on the area where the goods are located and the target stacking area, an optimal operation path is planned for the stacking robot, and automatic stacking operation is executed according to the final stacking scheme; the optimal stacking scheme can be automatically generated according to the physical characteristics of different cargos, and the method is suitable for stacking operation of cargos of various specifications and has wide application prospects.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent manufacturing, and in particular relates to a fully automatic palletizing method and system. Background Art

[0002] Palletizing is an important part of logistics warehousing. Traditional manual palletizing methods have problems such as high labor intensity, low efficiency, and high safety risks. Although various types of palletizing robots have appeared on the market, the existing automatic palletizing solutions still have the following problems: the generation of palletizing solutions lacks systematicity and scientificity, and often only considers simple space utilization and ignores loading stability; most existing technologies adopt fixed palletizing modes and cannot flexibly adjust the palletizing solutions according to the characteristics of different goods; in the multi-layer palletizing process, there is a lack of effective assessment of the overall stability, which can easily lead to dumping or damage of goods; the path planning of the palletizing robot is not optimized enough, and there are problems such as long paths and uneven movements, which affect work efficiency.

[0003] Therefore, there is an urgent need for an automatic palletizing solution that can both ensure high space utilization efficiency and ensure the stability of goods. Summary of the invention

[0004] The present invention provides a fully automatic palletizing method and system, which comprehensively considers space utilization and loading stability, so that the generated palletizing scheme can achieve the purpose of ensuring both high space utilization efficiency and stability of goods.

[0005] The specific plan is as follows:

[0006] In a first aspect, the present invention provides a fully automatic palletizing method, which is performed based on a palletizing robot and comprises the following steps:

[0007] Step S1, obtaining physical parameter information of goods to be palletized, wherein the physical parameter information includes three-dimensional size, weight, and pallet specifications of the goods.

[0008] Step S2, generating a single-layer palletizing benchmark solution based on the physical parameter information and the palletizing environment parameters, and generating a final palletizing solution based on the single-layer palletizing benchmark solution.

[0009] Step S3, based on the area where the goods are located and the target palletizing area, plan the optimal operation path for the palletizing robot, and perform automatic palletizing operations according to the final palletizing plan.

[0010] Furthermore, the generation process of the single-layer palletizing benchmark scheme includes:

[0011] Step S21, based on the three-dimensional dimensions of the goods and the specifications of the pallet, calculate feasible arrangements and combinations of the goods, and the length, width and height of each item cannot exceed the pallet;

[0012] Step S22, calculating the space utilization and loading stability of each arrangement and combination;

[0013] The formula for the space utilization rate η of a single pallet is: Among them, L i , W i is the length and width of the ith cargo, L 0 , W 0 is the length and width of the pallet, n is the number of single-layer goods on the pallet;

[0014] The loading stability takes into account the uniformity of pressure distribution of single-layer palletizing. Where P is the uniformity of pressure distribution. The closer it is to 1, the better the uniformity of pressure distribution. p is the standard deviation of the pressure distribution of a single layer of cargo, μ p is the average pressure of a single layer of cargo;

[0015] Step S23, comprehensively considering space utilization and loading stability, selecting the optimal arrangement and combination as a single-layer palletizing benchmark solution;

[0016] Calculate the value of each permutation and combination η(P-1), and the permutation and combination scheme corresponding to the minimum value is the single-layer palletizing benchmark scheme.

[0017] Furthermore, generating a final palletizing solution based on the single-layer palletizing benchmark solution includes:

[0018] Step S24, based on the single-layer palletizing benchmark solution, calculate the center of gravity distribution when multiple layers are stacked;

[0019] Step S25, generating an overall stability index, and selecting a solution that meets the overall stability threshold and has the highest number of layers as the final palletizing solution;

[0020] Among them, the expression of the overall stability index S evaluation model is: A i is the supported area of ​​the i-th layer of cargo, i.e., the overall surface area of ​​the next layer of cargo, A 0 is the pallet area, m is the total number of layers, ΔG is the gravity center offset distance, which refers to the offset of the gravity center of the cargo relative to the center of the pallet. The center point of the pallet is the origin (0, 0, 0). Among them, (xc, yc, zc) is the center of gravity of the cargo, and the calculation formula is:

[0021] Among them, mi is the mass of the i-th cargo, and (xi, yi, zi) is the coordinate of the center of gravity of the i-th cargo.

[0022] The overall stability threshold S thresholdThe calculation method is: S threshold =1.5×Kh×Kw×Km, where Kh is the altitude correction factor, Where H is the total stacking height, Wmin is the minimum side length of the pallet; Kw is the weight correction factor Where W is the total weight, Wmax is the maximum allowable weight; Km is the material property coefficient, which is determined according to the properties of the goods.

[0023] Furthermore, performing automatic palletizing operations according to the selected palletizing scheme includes: transferring the goods to the pallet according to the final palletizing scheme; planning an optimal operation path for the palletizing robot, and the palletizing robot performs cargo handling and palletizing according to the optimal operation path.

[0024] Furthermore, the optimal operation path is determined according to the fitness function of the path, and the palletizing operation area is evenly divided into fixed grids according to the area where the goods are located and the target palletizing area, and the side length of the grid is not less than the minimum value that the palletizing robot can pass through;

[0025] The grids without obstacles are marked as plannable areas. In the plannable areas, the grids that the palletizing robot must pass through or stay in from the starting point of cargo transfer to the target palletizing area are marked as path nodes. The number of path nodes is recorded as b. Then, for the i-th path planning l of the palletizing robot i , the mathematical expression of its fitness function is:

[0026] Among them, s(a, a+1) is the distance from the adjacent path node a to its a+1 in the path planning sequence, μ is the weight coefficient of the smoothness penalty term, 0.05<μ<0.25, p(l i ) is the path smoothness, and the mathematical expression of path smoothness is:

[0027] Among them, θ(a, a+1, a+2) represents the path l i In the figure, the angle formed by three consecutive nodes is calculated using the dot product formula of the vector:

[0028] For all the path planning of the palletizing robot, the path corresponding to the minimum fitness function value is selected, which is the optimal operation path; the mathematical expression of the path corresponding to the minimum fitness function value is: Min(f(l 1 )…f(l i )…f(l g )); where g is the total number of paths.

[0029] In a second aspect, the present invention provides a fully automatic palletizing system for executing the method described in the first aspect, wherein the system comprises: a cargo information acquisition module, a palletizing plan generation module, and a path planning execution module connected in sequence.

[0030] The cargo information acquisition module is used to acquire physical parameter information of the cargo to be palletized, wherein the physical parameter information includes the three-dimensional size, weight and pallet specifications of the cargo.

[0031] The palletizing scheme generating module is used to generate a single-layer palletizing reference scheme based on the physical parameter information and the palletizing environment parameters, and to generate a final palletizing scheme based on the single-layer palletizing reference scheme.

[0032] The palletizing scheme generation module is also used to: calculate feasible cargo arrangement and combination methods based on the three-dimensional dimensions of the cargo and the specifications of the pallet; calculate the space utilization and loading stability of each arrangement and combination method; and comprehensively consider the space utilization and loading stability to select the optimal arrangement and combination as the single-layer palletizing benchmark scheme.

[0033] Based on the single-layer palletizing benchmark solution, calculate the center of gravity distribution when multiple layers are stacked; generate an overall stability index, and select the solution with the highest overall stability as the final palletizing solution.

[0034] The path planning execution module is used to plan the optimal operation path for the palletizing robot based on the area where the goods are located and the target palletizing area, and control the palletizing robot to perform automatic palletizing operations according to the final palletizing plan.

[0035] Furthermore, the path planning execution module is specifically used to:

[0036] The palletizing operation area is evenly divided into fixed grids, and the side length of the grid is not less than the minimum value that the palletizing robot can pass through.

[0037] The grids without obstacles are marked as plannable areas, and the grids in the plannable areas that the palletizing robot must pass through or stay in from the cargo transfer starting point to the target palletizing area are marked as path nodes.

[0038] The fitness values ​​of all possible paths are calculated based on the path fitness function, the path with the minimum fitness function value is selected as the optimal operation path, and the palletizing robot is controlled to complete the palletizing operation along the optimal operation path.

[0039] The path fitness function includes two evaluation indicators: path length and path smoothness. The path smoothness is evaluated by calculating the angle formed by three consecutive nodes in the path.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] The method of the present invention not only considers the uniformity of pressure distribution of single-layer palletizing, but also pays attention to the distribution of center of gravity when multiple layers are stacked, thereby effectively improving the reliability of palletizing operations, and designs a path planning method based on grid division. By comprehensively considering the fitness function of path length and smoothness, it is possible to generate an optimal motion trajectory for a palletizing robot, which significantly improves the efficiency and safety of palletizing operations. The method of the present invention has strong versatility and adaptability, can automatically generate an optimal palletizing scheme according to the physical characteristics of different goods, is suitable for palletizing operations of goods of various specifications, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flow chart of the fully automatic palletizing method of the present invention;

[0043] Figure 2 It is a schematic diagram of the composition of the full-automatic palletizing system of the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0045] It should be noted that the fully automatic palletizing method of the present invention is not only applicable to the palletizing of goods of regular shapes, but also to the palletizing of goods of irregular shapes. For example, in practical applications, for goods of irregular shapes, they can be simplified by their circumscribed rectangles, and automatic stacking can be achieved while ensuring the stability of the palletizing.

[0046] Example 1

[0047] like Figure 1 As shown, the fully automatic palletizing method of the present invention is shown, and the steps of the method are specifically as follows:

[0048] Step S1, obtaining physical parameter information of goods to be palletized, wherein the physical parameter information includes three-dimensional size, weight, and pallet specifications of the goods.

[0049] This embodiment uses a machine vision system to obtain physical parameter information of the goods to be palletized. The hardware configuration of the vision system used in this embodiment is as follows:

[0050] Four high-resolution industrial cameras (resolution ≥ 2 million pixels) are installed above the palletizing area, using a matrix layout to achieve full-area coverage; eight structured light projectors are configured to generate structured light stripe patterns, and four area array laser rangefinders are used to assist in measuring the height of goods; the industrial light source system, including a ring LED light source and a strip light source, ensures uniform lighting.

[0051] The machine vision processing flow includes: (1) image preprocessing, using adaptive histogram equalization for image enhancement, using median filtering and Gaussian filtering to eliminate noise, performing perspective transformation correction, and eliminating perspective distortion; (2) cargo identification and positioning, using deep learning-based target detection algorithms (such as improved YOLOv5) to identify cargo, using contour extraction and rectangle fitting to obtain cargo contours, and combining Hough transform to detect cargo edge features; (3) three-dimensional information reconstruction, using structured light stripe deformation patterns to reconstruct cargo three-dimensional contours, using binocular stereo vision to measure cargo height, and improving three-dimensional reconstruction accuracy based on multi-view information fusion; (4) parameter information extraction, calculating the length and width of cargo through the mapping relationship between pixel coordinates and actual dimensions, combining structured light and laser ranging data to obtain accurate height values, establishing a cargo posture estimation model, and obtaining spatial position and orientation information.

[0052] The Zhang Zhengyou calibration method is used to calibrate the internal and external parameters of the camera, and the standard calibration board is used to verify the system accuracy. An illumination compensation mechanism is established to cope with illumination changes in different environments and realize automatic exposure and white balance adjustment. The position and posture changes of the goods during transportation are monitored in real time to detect whether the goods have been dumped or damaged. The palletizing quality is evaluated in real time, and any abnormalities are promptly fed back to the control system. Through the above machine vision solutions, the system can achieve a measurement accuracy of ±2mm for the size of the goods, a position detection error within ±3mm, a posture recognition accuracy of >99%, and a processing speed of 30 frames per second.

[0053] It realizes the fast, accurate and contactless acquisition of the physical parameters of the goods, providing a reliable data basis for the subsequent generation of palletizing solutions and path planning.

[0054] Step S2, generating a single-layer palletizing benchmark solution based on the physical parameter information and the palletizing environment parameters, and generating a final palletizing solution based on the single-layer palletizing benchmark solution.

[0055] Palletizing environment parameters include but are not limited to: the space limit height of the palletizing area; the ground bearing capacity; the robot's operating range; environmental temperature and humidity and other factors; these parameters will directly affect the generation of the palletizing plan. For example, in a cold storage environment, it is necessary to consider reserving appropriate gaps between goods to ensure the circulation of cold air.

[0056] The generation process of the single-layer palletizing benchmark scheme includes:

[0057] Step S21, based on the three-dimensional dimensions of the goods and the specifications of the pallet, calculate feasible arrangements and combinations of the goods, and the length, width and height of each item cannot exceed the pallet;

[0058] In addition to considering the basic length, width and height constraints, the orientation requirements of the goods are also considered. For example, goods marked "upward" must maintain a specific direction; considering the fragility of the goods, fragile goods are placed on the upper layer; considering the partition placement requirements of different batches of goods; in practice, intelligent optimization methods such as genetic algorithms or ant colony algorithms can be used to search for the optimal permutation and combination.

[0059] Step S22, calculating the space utilization and loading stability of each arrangement and combination;

[0060] The formula for the space utilization rate η of a single pallet is: Among them, L i , W i is the length and width of the ith cargo, L 0 , W 0 is the length and width of the pallet, and n is the number of single-layer goods on the pallet. The space utilization calculation formula mainly considers the plane utilization. In practical applications, in order to be more accurate, the volume utilization can also be considered.

[0061] The loading stability takes into account the uniformity of pressure distribution of single-layer palletizing. Where P is the uniformity of pressure distribution. The closer it is to 1, the better the uniformity of pressure distribution. p is the standard deviation of the pressure distribution of a single layer of cargo, μ p is the average pressure of a single layer of cargo.

[0062] For example, for a pallet loaded with four boxes of the same weight, if the four boxes are completely symmetrically distributed, the standard deviation of the pressure distribution σ p Close to 0, at this time P value is close to 1, indicating that the pressure distribution is most uniform; if the four boxes are concentrated on one side of the tray, then σ p If the pressure is larger, the P value will be significantly reduced, indicating that the pressure distribution is uneven. The pressure distribution can be monitored in real time through a pressure sensor.

[0063] Step S23, comprehensively considering space utilization and loading stability, selecting the optimal arrangement and combination as a single-layer palletizing benchmark solution;

[0064] Calculate the value of each permutation and combination η(P-1), and the permutation and combination scheme corresponding to the minimum value is the single-layer palletizing benchmark scheme. In addition to using η(P-1) as an evaluation index, a weighted summation method can also be used: Score = w1×η+w2×P, where w1 and w2 are weight coefficients, and the initial setting is 0.5. They can be dynamically adjusted according to actual needs. For example, for heavy-loaded goods, the value of w2 can be appropriately increased to emphasize stability; for light-loaded goods, the value of w1 can be increased to emphasize space utilization.

[0065] The generating of the final palletizing scheme based on the single-layer palletizing benchmark scheme comprises:

[0066] Step S24, based on the single-layer palletizing benchmark solution, calculate the center of gravity distribution when multiple layers are stacked;

[0067] Step S25, generating an overall stability index, and selecting a solution that meets the overall stability threshold and has the highest number of layers as the final palletizing solution;

[0068] The friction coefficient between goods; the deformation characteristics of goods of different materials; the impact of environmental vibration; the impact force that the goods may be subjected to during transportation, etc., all these factors will affect the stability of the stacking; however, considering all the above factors will inevitably lead to the disadvantages of excessive calculation and overly complex model. Therefore, in this embodiment, the above factors are ignored when calculating the center of gravity distribution of multi-layer stacking.

[0069] The expression of the overall stability index S evaluation model is: A i is the supported area of ​​the i-th layer of cargo, i.e., the overall surface area of ​​the next layer of cargo, A 0 is the pallet area, m is the total number of layers, ΔG is the gravity center offset distance, which refers to the offset of the gravity center of the cargo relative to the center of the pallet. The center point of the pallet is the origin (0, 0, 0). Among them, (xc, yc, zc) is the center of gravity of the cargo, and the calculation formula is:

[0070] Among them, mi is the mass of the i-th cargo, and (xi, yi, zi) is the coordinate of the center of gravity of the i-th cargo.

[0071] Take a three-layer palletizing as an example: Assuming the pallet size is 1200×1000mm, the first layer of 4 boxes completely covers the pallet, and the second layer of 3 boxes has a total area of ​​0.9m 2 The total area of ​​the two boxes on the third floor is 0.6m 2, and the total center of gravity is offset by 10 cm, then S = (1.2 × 1 + 0.9 + 0.6) / (1.2) × 1 / 0.1 = 20.8. By comparing the S values ​​of different solutions, the best solution is selected.

[0072] The coordinates of the center of gravity of the cargo can be obtained in the following ways: for cargo of regular shape, it can be approximately calculated through the geometric center; for cargo of irregular shape, it can be obtained through three-dimensional scanning combined with center of gravity testing equipment; for cargo with uneven internal density, the center of gravity position needs to be determined by professional equipment.

[0073] The overall stability threshold S threshold The calculation method is: S threshold =1.5×Kh×Kw×Km, where Kh is the altitude correction factor, Where H is the total stacking height, Wmin is the minimum side length of the pallet; Kw is the weight correction factor Where W is the total weight, Wmax is the maximum allowable weight; Km is the material property coefficient, which is determined according to the properties of the goods, hard packaging: 1.0, soft packaging: 1.2, fragile goods: 1.3; select the overall stability threshold S≥S threshold The solution with the highest number of layers is selected as the final palletizing solution.

[0074] In a further strict management scheme, the palletizing scheme is considered safe if and only if all of the following conditions are met: 1) S ≥ S threshold ; 2) (each layer); 3) ΔG ≤ 0.1 × min (L0, W0); However, in actual cases, conditions 2) and 3) can basically be met, unless the types and sizes of goods vary greatly. Therefore, in this embodiment, only the overall stability threshold is considered.

[0075] Step S3, based on the area where the goods are located and the target palletizing area, plan the optimal operation path for the palletizing robot, and perform automatic palletizing operations according to the final palletizing plan.

[0076] The automatic palletizing operation according to the selected palletizing scheme includes: transferring the goods to the pallet according to the final palletizing scheme; planning an optimal operation path for the palletizing robot, and the palletizing robot performs cargo handling and palletizing according to the optimal operation path.

[0077] In actual operations, the goods can be stored in partitions in advance to reduce the round-trip distance of the robot; suitable grasping tools, such as vacuum suction cups, mechanical grippers, etc., can be selected according to the weight and volume characteristics of the goods; a cache area can be set up to enable parallel operation of goods preprocessing and palletizing operations; the palletizing quality can be monitored in real time through the visual system, and timely adjustments can be made if abnormalities are found.

[0078] The optimal operation path is determined according to the fitness function of the path. The palletizing operation area is evenly divided into fixed grids according to the area where the goods are located and the target palletizing area. The side length of the grid is not less than the minimum value that the palletizing robot can pass through. For example, for a 10m×8m operation area, considering that the passing width of the palletizing robot is 1.2m, the area can be divided into 0.6m×0.6m grids to ensure that the robot can safely pass through two adjacent grids.

[0079] Mark the grids without obstacles as the plannable areas. When marking the plannable areas, in addition to considering static obstacles, the dynamic working range of other operating equipment, personnel activity areas, safety buffer areas, temporary storage areas, etc. should also be considered. These areas should be properly marked in the grid map to ensure the safety of path planning; in the plannable area, the grids that the palletizing robot must pass through or stay in from the starting point of cargo transfer to the target palletizing area are marked as path nodes. The number of path nodes is recorded as b. Then, for the i-th path planning l of the palletizing robot i , the mathematical expression of its fitness function is:

[0080] Among them, s(a, a+1) is the distance from the adjacent path node a to its a+1 in the path planning sequence, μ is the weight coefficient of the smoothness penalty term, 0.05<μ<0.25, p(l i ) is the path smoothness, and the mathematical expression of path smoothness is:

[0081] Among them, θ(a, a+1, a+2) represents the path l i In the figure, the angle formed by three consecutive nodes is calculated using the dot product formula of the vector:

[0082] There are three consecutive nodes A(0,0), B(1,1), and C(2,0), then we can calculate the vectors AB(1,1) and BC(1,-1), and use the dot product formula to calculate the angle θ=90°. The smoothness penalty value in this case is π 2 / 4, indicating that this is a larger turn. In practical applications, such large-angle turns should be avoided as much as possible to improve the smoothness of movement.

[0083] For all the path planning of the palletizing robot, the path corresponding to the minimum fitness function value is selected as the optimal operation path; the mathematical expression of the path corresponding to the minimum fitness function value is: Min(f(l 1 )…f(l i )…f(l g )); where g is the total number of paths.

[0084] Take the finished product warehouse of a large beverage manufacturing company as an example. The company handles the palletizing task of about 50,000 boxes of various beverages every day. Before adopting the fully automatic palletizing method of the present invention, the company mainly relied on manual labor and traditional forklifts for palletizing, which had the following problems: the manual labor intensity was high, and an average of 10-15 workers were required to work in three shifts every day; the manual palletizing efficiency was low, and only 150 boxes of palletizing could be completed per hour on average; the palletizing quality was unstable, and accidents such as stacking tilt and collapse occurred from time to time; the labor cost was high, and it was difficult to recruit workers, and the mobility was strong; there were many hidden dangers in the operation safety, and industrial accidents often occurred.

[0085] When adopting the fully automatic palletizing method of the present invention, the system configuration is as follows: 4 palletizing robots are used to be responsible for the palletizing tasks of products of different specifications; 6 parallel conveyor lines are set up to realize automatic diversion and conveying of products; 12 groups of 3D visual inspection systems are equipped to monitor the product posture and palletizing quality in real time; and a 1,000-square-meter intelligent palletizing operation area is established, which is divided into 200 standard grid units.

[0086] The palletizing efficiency of a single robot reaches 450 boxes per hour, which is three times the efficiency of manual labor; four robots work continuously for 24 hours, with a daily processing capacity of more than 40,000 boxes; product turnover time is shortened from an average of 4 hours to 1.5 hours; warehouse utilization rate is increased by 35%, and the effective storage space is increased by 2,000 square meters; stacking stability is significantly improved, and no collapse accidents have occurred within 6 months; pallet space utilization rate is increased by 20%, and each pallet can be stacked 2 more layers on average; labor costs are saved by more than 1 million yuan per year; the storage area saving effect is significant, saving 600,000 yuan in rental costs annually.

[0087] In addition, it should be noted that for the path planning of the palletizing robot, you can also use heuristic algorithms such as the A* algorithm to perform preliminary path search, apply genetic algorithms or particle swarm algorithms to optimize the initial path, establish a path database, store commonly used paths for quick call, and implement a real-time path dynamic adjustment mechanism to deal with emergencies; at the same time, you can also combine machine learning methods to continuously optimize the path planning strategy through historical data.

[0088] Example 2

[0089] like Figure 2 As shown in the figure, it is a schematic diagram of the composition of the fully automatic palletizing system of the present invention, and the system includes: a cargo information acquisition module, a palletizing scheme generation module and a path planning execution module connected in sequence. The system adopts a modular design, and the modules exchange data through standardized interfaces, which is convenient for system maintenance and upgrading. The system can also flexibly add other functional modules according to actual needs, such as a quality monitoring module, a data analysis module, etc. In addition, each module can be deployed in a distributed manner to improve the reliability and processing efficiency of the system.

[0090] The cargo information acquisition module is used to acquire the physical parameter information of the cargo to be palletized, including the three-dimensional size, weight and pallet specifications of the cargo. The module may include a 3D visual sensor system consisting of multiple industrial cameras and structured light sensors for quickly acquiring cargo dimensions; a high-precision electronic scale that supports dynamic weighing; a barcode scanner for identifying cargo information; and an industrial computer for data processing and storage. All devices achieve real-time data transmission via industrial Ethernet.

[0091] The palletizing scheme generation module is used to generate a single-layer palletizing benchmark scheme based on the physical parameter information and palletizing environmental parameters, and to generate a final palletizing scheme based on the single-layer palletizing benchmark scheme. The module adopts a three-layer architecture design: 1) Data layer: responsible for the storage and management of physical parameters and environmental parameters; 2) Algorithm layer: including various palletizing algorithms and optimization strategies; 3) Interface layer: providing a standard interface for interaction with other modules. The module also integrates a human-computer interaction interface to support visual display and manual intervention of the scheme.

[0092] The palletizing solution generation module is also used to: calculate feasible cargo arrangement and combination methods based on the three-dimensional size of the cargo and the specifications of the pallet; calculate the space utilization and loading stability of each arrangement and combination method; comprehensively consider the space utilization and loading stability, and select the optimal arrangement and combination as the single-layer palletizing benchmark solution; use multi-threaded parallel computing technology to improve the efficiency of solution generation. Specifically, it includes: using GPU to accelerate three-dimensional space calculations; using heuristic algorithms to quickly screen feasible solutions; establishing a solution evaluation database to store historical optimal solutions for reference; and realizing real-time simulation verification functions, which can preview the palletizing effect in a 3D environment.

[0093] Based on the single-layer palletizing benchmark solution, calculate the center of gravity distribution when multiple layers are stacked; generate an overall stability index, and select the solution with the highest overall stability as the final palletizing solution.

[0094] The system can also provide the following extended functions: support optimization of mixed palletizing of different types of goods; provide a dynamic adjustment mechanism for the weights of multiple evaluation indicators; integrate environmental factor impact assessment models; support the generation of customized palletizing solutions for special requirements, such as considering the temperature requirements and moisture-proof requirements of goods.

[0095] The path planning execution module is used to plan the optimal operation path for the palletizing robot based on the area where the goods are located and the target palletizing area, and control the palletizing robot to perform automatic palletizing operations according to the final palletizing plan.

[0096] The path planning execution module is specifically used for:

[0097] The palletizing operation area is evenly divided into fixed grids, and the side length of the grid is not less than the minimum value that the palletizing robot can pass through.

[0098] The grid without obstacles is marked as a plannable area, and the grids in the plannable area that the palletizing robot must pass through or stay in from the starting point of cargo transfer to the target palletizing area are marked as path nodes. Grid division can also adopt adaptive grid division, dynamically adjusting the grid size according to the characteristics of the area; multi-resolution grid map, using finer grids in important areas; probabilistic grid map, considering environmental uncertainty; supporting real-time map updates and responding to dynamic obstacles.

[0099] Based on the path fitness function, the fitness values ​​of all possible paths are calculated, the path with the minimum fitness function value is selected as the optimal operation path, and the palletizing robot is controlled to complete the palletizing operation along the optimal operation path. Set a safety buffer zone to ensure a safe distance from obstacles; detect dynamic obstacles in real time, including personnel and other equipment; define different levels of safety areas to achieve differentiated management; support remote emergency stop and manual operation mode.

[0100] The path fitness function includes two evaluation indicators: path length and path smoothness. The path smoothness is evaluated by calculating the angle formed by three consecutive nodes in the path.

[0101] In actual applications, the system adopts a modular and distributed architecture design, and data is exchanged between functional modules through standardized interfaces. The core data processing unit of the system uses a high-performance industrial server equipped with a GPU accelerator card, which can realize the rapid processing of complex algorithms. The system software adopts a B / S architecture, supports remote access and monitoring, and facilitates managers to grasp the system operation status in real time. In terms of data storage, the system uses a distributed database to support real-time data backup and historical tracing to ensure data security and reliability.

[0102] In terms of communication architecture, the system uses a multi-layer communication network: the field device layer uses industrial Ethernet and PROFINET fieldbus to ensure real-time performance; the control layer uses industrial Ethernet to support high-speed data transmission; the management layer uses standard Ethernet to facilitate system integration and expansion. At the same time, the system supports the OPC UA protocol, which can easily interact with other automation equipment and systems.

[0103] In terms of system reliability, multiple redundancy designs are adopted: key sensors use dual redundant configuration; the control system uses a hot backup solution; and the communication network uses a ring network topology. The system also has a complete fault diagnosis and processing mechanism, which can automatically detect and locate faults and handle them accordingly. For common faults, the system can automatically switch to the backup solution to continue working, ensuring production continuity.

[0104] In terms of security, the system implements multi-layer security protection: physical security is achieved through security gratings, protective doors and other equipment; data security is ensured through encrypted transmission and access control; functional safety is achieved through a security PLC that complies with the IEC61508 standard. The system also supports hierarchical authorization management, and users at different levels have different operating permissions.

[0105] In terms of system scalability, a rich set of expansion interfaces are reserved: the hardware interface supports the access of various sensors and actuators; the software interface supports secondary development and function expansion; the communication interface supports integration with upper-level systems such as MES and ERP. The control algorithm of the system adopts a plug-in design, which can be easily updated and optimized.

[0106] In terms of human-computer interaction, the system provides a rich interactive interface: on-site operation uses an industrial touch screen with a simple and intuitive interface; remote monitoring supports Web access and can be operated through PC and mobile terminals; the system status is displayed through multimedia, including 2D / 3D visualization, sound and light alarms, etc.

[0107] In terms of maintenance management, the system has complete maintenance functions: supporting equipment operation status monitoring and predictive maintenance; providing detailed operation logs and reporting functions; and having remote diagnosis and online upgrade capabilities. The system also integrates a knowledge base system to provide maintenance personnel with troubleshooting guidance.

[0108] In general, the fully automatic palletizing system of this embodiment realizes the intelligentization and automation of palletizing operations through advanced hardware configuration, scientific software architecture, and reliable control strategy, and has good practicality and promotion value. The system has significant advantages in improving production efficiency, reducing labor intensity, and ensuring operation quality, and can be widely used in automation transformation and upgrading in logistics warehousing, manufacturing and other fields. At the same time, the modular design and standardized interface of the system also provide a good foundation for future functional expansion and technical upgrades.

[0109] It should be noted that those skilled in the art should understand that various changes and equivalent substitutions may be made to the present invention without departing from the scope of the present invention. In addition, various modifications may be made to the present invention for specific situations or materials without departing from the scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed, but should include all embodiments falling within the scope of the claims of the present invention.

Claims

1. A fully automatic palletizing method, which is based on a palletizing robot and is characterized in that: The method comprises the following steps: Step S1, obtaining physical parameter information of the goods to be palletized, wherein the physical parameter information includes the three-dimensional size, weight, and pallet specifications of the goods; Step S2, generating a single-layer palletizing benchmark scheme based on the physical parameter information and the palletizing environment parameters, and generating a final palletizing scheme based on the single-layer palletizing benchmark scheme; The generation process of the single-layer palletizing benchmark scheme includes: Step S21, based on the three-dimensional dimensions of the goods and the specifications of the pallet, calculate feasible arrangements and combinations of the goods, and the length, width and height of each item cannot exceed the pallet; Step S22, calculating the space utilization and loading stability of each arrangement and combination; Step S23, comprehensively considering space utilization and loading stability, selecting the optimal arrangement and combination as a single-layer palletizing benchmark solution; Calculate the value of each permutation and combination η(P-1), and the permutation and combination scheme corresponding to the minimum value is the single-layer palletizing benchmark scheme; Generating a final palletizing scheme based on the single-layer palletizing benchmark scheme includes: Step S24, based on the single-layer palletizing benchmark solution, calculate the center of gravity distribution when multiple layers are stacked; Step S25, generating an overall stability index, and selecting a solution that meets the overall stability threshold and has the highest number of layers as the final palletizing solution. Step S3, based on the area where the goods are located and the target palletizing area, plan the optimal operation path for the palletizing robot, and perform automatic palletizing operations according to the final palletizing plan.

2. The fully automatic palletizing method according to claim 1, characterized in that: The formula for the space utilization rate η of a single pallet is: Among them, L i , W i is the length and width of the i-th cargo, L0, W0 are the length and width of the pallet, and n is the number of single-layer cargo on the pallet; The loading stability takes into account the uniformity of pressure distribution of single-layer palletizing. Where P is the uniformity of pressure distribution. The closer it is to 1, the better the uniformity of pressure distribution. p is the standard deviation of the pressure distribution of a single layer of cargo, μ p is the average pressure of a single layer of cargo.

3. The fully automatic palletizing method according to claim 2, characterized in that: The expression of the overall stability index S evaluation model is: A i is the supported area of ​​the i-th layer of cargo, that is, the overall surface area of ​​the next layer of cargo, A0 is the pallet area, m is the total number of layers, ΔG is the center of gravity offset distance, which refers to the offset of the center of gravity of the cargo relative to the center of the pallet. The center point of the pallet is the origin (0, 0, 0), then Among them, (xc, yc, zc) is the center of gravity of the cargo, and the calculation formula is: Among them, mi is the mass of the i-th cargo, and (xi, yi, zi) is the coordinate of the center of gravity of the i-th cargo.

4. The fully automatic palletizing method according to claim 3, characterized in that: The overall stability threshold S threshold The calculation method is: S threshold =1.5×Kh×Kw×Km, where Kh is the altitude correction factor, Where H is the total stacking height, Wmin is the minimum side length of the pallet; Kw is the weight correction factor Where W is the total weight, Wmax is the maximum allowable weight; Km is the material property coefficient, which is determined according to the properties of the goods; select the one that meets the overall stability threshold S≥S threshold The solution with the highest number of layers is selected as the final palletizing solution.

5. The fully automatic palletizing method according to claim 4, characterized in that: The automatic palletizing operation according to the selected palletizing scheme includes: transferring the goods to the pallet according to the final palletizing scheme; planning an optimal operation path for the palletizing robot, and the palletizing robot performs cargo handling and palletizing according to the optimal operation path.

6. The fully automatic palletizing method according to claim 5, characterized in that: The optimal operation path is determined according to the fitness function of the path, and the palletizing operation area is evenly divided into fixed grids according to the area where the goods are located and the target palletizing area, and the side length of the grid is not less than the minimum value that the palletizing robot can pass through; The grids without obstacles are marked as plannable areas. In the plannable areas, the grids that the palletizing robot must pass through or stay in from the starting point of cargo transfer to the target palletizing area are marked as path nodes. The number of path nodes is recorded as b. Then, for the i-th path planning of the palletizing robot, i , the mathematical expression of its fitness function is: Among them, s(a, a+1) is the distance from the adjacent path node a to its a+1 in the path planning sequence, μ is the weight coefficient of the smoothness penalty term, 0.05<μ<0.25, p(l i ) is the path smoothness, and the mathematical expression of path smoothness is: Among them, θ(a, a+1, a+2) represents the path l i The angle formed by three consecutive nodes.

7. The fully automatic palletizing method according to claim 6, characterized in that: The angle formed by three consecutive nodes is calculated using the dot product formula of the vector: For all the path planning of the palletizing robot, the path corresponding to the minimum fitness function value is selected as the optimal operation path; the mathematical expression of the path corresponding to the minimum fitness function value is: Min(f(l1_…f(l i )…f(l g )); where g is the total number of paths.

8. A fully automatic palletizing system for executing the method according to any one of claims 1 to 7, characterized in that: The system comprises: a cargo information acquisition module, a palletizing scheme generation module and a path planning execution module connected in sequence; The cargo information acquisition module is used to acquire physical parameter information of the cargo to be palletized, wherein the physical parameter information includes the three-dimensional size, weight and pallet specifications of the cargo; The palletizing scheme generating module is used to generate a single-layer palletizing benchmark scheme based on the physical parameter information and the palletizing environment parameters, and to generate a final palletizing scheme based on the single-layer palletizing benchmark scheme; The stacking scheme generation module is also used to: calculate feasible cargo arrangement and combination methods based on the three-dimensional size of the cargo and the specifications of the pallet; calculate the space utilization rate and loading stability of each arrangement and combination method; comprehensively consider the space utilization rate and loading stability, and select the optimal arrangement and combination as the single-layer stacking benchmark plan; Based on the single-layer palletizing benchmark solution, calculate the center of gravity distribution when multiple layers are stacked; generate an overall stability index and select the solution with the highest overall stability as the final palletizing solution; The path planning execution module is used to plan the optimal operation path for the palletizing robot based on the area where the goods are located and the target palletizing area, and control the palletizing robot to perform automatic palletizing operations according to the final palletizing plan.

9. The fully automatic palletizing system according to claim 8, characterized in that: The path planning execution module is specifically used for: The palletizing operation area is evenly divided into fixed grids, wherein the side length of the grid is not less than the minimum value through which the palletizing robot can pass; Marking the grid without obstacles as a plannable area, and marking the grids in the plannable area that the palletizing robot must pass through or stay in from the cargo transfer starting point to the target palletizing area as path nodes; The fitness values ​​of all possible paths are calculated based on the path fitness function, the path with the minimum fitness function value is selected as the optimal operation path, and the palletizing robot is controlled to complete the palletizing operation along the optimal operation path.

10. The fully automatic palletizing system according to claim 9, characterized in that: The path fitness function includes two evaluation indicators: path length and path smoothness. The path smoothness is evaluated by calculating the angle formed by three consecutive nodes in the path.

Images