An unmanned trolley pile reversing method and device fusing height constraint

Abstract

The present application relates to the technical field of crane unstacking, and discloses an unmanned crane unstacking method and equipment fusing height limit constraint, which comprises the following steps: obtaining the initial state of the warehouse area, target demand and height limit threshold; generating an initial population with the target steel plate moving sequence code; in the decoding stage, if there is an obstacle steel plate above the target steel plate to be moved, then according to the current height of the candidate moving-in stack position, whether it is an idle stack position, whether it is over the limit after moving in and other state information, the moving difficulty evaluation value is calculated; the obstacle steel plate is moved into the stack position with the minimum evaluation value, and the unstacking step sequence is generated; the population is iteratively updated by using the artificial bee colony algorithm, and the unstacking scheme with the optimal fitness is output. By introducing the height limit constraint and the moving difficulty evaluation mechanism, the present application effectively avoids the risk of stack overheight, reduces the redundancy of unstacking steps, and significantly improves the operation efficiency of the unmanned crane under strict height limit constraint.

Description

Technical Field

[0001] This invention relates to the field of overhead crane stacking technology, and more specifically, to an unmanned overhead crane stacking method and equipment that integrates height restriction constraints. Background Technology

[0002] In the field of steel logistics and intelligent warehousing, steel plates are typically stored in warehouses using multi-layer mixed stacking. As downstream cutting and processing tasks are assigned, unmanned overhead cranes are usually deployed to sort specific target steel plates scattered across multiple initial stacks, according to the needs of different cutting batches. These plates are then moved to designated target stacks for further processing or warehouse organization. During this process, because the number of steel plates in the initial stacks varies and target plates are often pressed down by non-target plates, the unmanned overhead cranes must first move the obstructing plates to other empty or non-empty stacks before retrieving the target plates. This constitutes a typical constrained combinatorial optimization problem, the core objective of which is to complete the task with the fewest possible stacking operations while satisfying actual operational constraints.

[0003] Existing technologies typically employ traditional genetic algorithms or simple rule-based heuristics to solve this type of steel plate repositioning problem, searching for movement paths through simulated evolution or greedy strategies. However, due to the rugged and chaotic fitness landscape of the solution space for this type of problem, with no obvious patterns, traditional algorithms are prone to getting trapped in local optima during the search process. More importantly, actual warehousing environments impose strict height constraints on stacking, and existing path planning methods often lack a dynamic evaluation mechanism for this physical constraint. When handling the removal position of obstacle steel plates, they fail to fully consider their impact on the difficulty of subsequent repositioning steps. Therefore, existing technologies lack a dynamic evaluation mechanism for warehouse height constraints when planning repositioning paths, resulting in high redundancy in the generated repositioning schemes, low operational efficiency, and an inability to obtain the optimal solution under strict height constraints. Summary of the Invention

[0004] To overcome the shortcomings of existing stacking techniques, such as high redundancy, low operational efficiency, and inability to obtain the optimal solution under strict height restrictions, this invention proposes the following technical solution: Firstly, this invention proposes a method for unmanned overhead crane stacking that integrates height-limited constraints, comprising: Obtain the initial stack status information of the warehouse area, the sorting requirements of the target steel plates, and the height limit threshold information of each stack. Based on the sorting requirements of the target steel plate, the movement order of the target steel plate is used as a decision variable for encoding to generate an initial population containing several individuals. Decode the individuals in the population. When there is an obstacle steel plate above the target steel plate to be moved, obtain the current height of each candidate moving position, whether it is an empty position, and whether it exceeds the height limit threshold after moving. The movement difficulty evaluation value of each candidate stacking position is calculated based on the status information. Move the obstacle steel plate into the feasible stacking position with the lowest moving difficulty evaluation value until there are no obstacle steel plates above the target steel plate to be moved. Then move the target steel plate to be moved into the designated target stacking position and generate the corresponding stacking step sequence. The fitness value of an individual is calculated based on the generated stacking step sequence, and the initial population is iteratively updated using the artificial bee colony algorithm until a preset termination condition is met. The final stacking step corresponding to the individual with the best fitness value is then output.

[0005] As a preferred technical solution, the movement order of the target steel plate is encoded as a decision variable to generate an initial population containing several individuals, including: Calculate the layer distance between the target steel plates at the bottom and top of the stack, and convert the layer distance into a first numerical component; The total number of target steel plates contained within the layer distance interval between the bottom target steel plate and the top target steel plate is counted, and the total number of target steel plates is converted into a second numerical part. The first numerical part and the second numerical part are superimposed to form the clearance difficulty value of the stack position. Then, according to the clearance difficulty value in ascending order, individual codes containing the movement order of the target steel plate are generated by the removal operation.

[0006] As a preferred technical solution, the movement order of the target steel plate is encoded as a decision variable to generate an initial population containing several individuals, including: Obtain the hierarchical position information of each target steel plate in its respective stack; Based on the hierarchical location information, calculate the hierarchical difference between each target steel plate and the top layer of its stack position; Select the target steel plate with the smallest layer difference to add to the moving sequence. If there are multiple target steel plates with the same layer difference, randomly select one of them to add to the moving sequence, until all target steel plates are added to the moving sequence, forming an individual code.

[0007] As a preferred technical solution, the step of calculating the movement difficulty evaluation value of each candidate moving position based on the status information includes: Determine whether the current height of the current candidate pallet position exceeds the height limit threshold information. If it exceeds the limit, assign a first basic difficulty coefficient to the current candidate pallet position. If it does not exceed the limit, determine whether the current candidate pallet position is an empty pallet position that does not contain the target steel plate to be sorted. If it is an empty pallet position, assign a second basic difficulty coefficient to the current candidate pallet position. If it is not an empty pallet position, assign a third basic difficulty coefficient to the current candidate pallet position. Wherein, the value of the first basic difficulty coefficient is greater than the value of the third basic difficulty coefficient, and the value of the third basic difficulty coefficient is greater than the value of the second basic difficulty coefficient. Obtain the specific height value of the current candidate stacking position and convert it into a correction coefficient in decimal form. Add the correction coefficient to the assigned basic difficulty coefficient to obtain the final movement difficulty evaluation value of the current candidate stacking position.

[0008] As a preferred technical solution, the process of moving the obstacle steel plate into the feasible moving position with the lowest moving difficulty evaluation value satisfies the following constraints when performing the moving operation:

[0009]

[0010] In the formula, Indicates at time The first reservoir area The first stack position layer If it exists, the value is 1; otherwise, it is 0. Represents a time set, Represents the set of stack positions. Represents a hierarchical set. Indicates the maximum number of levels; Indicates at time The first reservoir area The first stack position The specific numerical designation of the steel plate on the upper layer; It is a positive integer constant.

[0011] As a preferred technical solution, the process of moving the target steel plate to be moved into the designated target stack position satisfies the following constraint formula:

[0012]

[0013]

[0014]

[0015] in, Indicates at time , will be located in the stack position No. The steel plates of the layer are moved to the stack. No. The value indicates whether the layer action has occurred; if it has, the value is 1, otherwise it is 0. Indicates at time The first reservoir area The first stack position layer ; For state variables, the number is _____. Whether the target steel plate is ultimately located at the target stack position The Layer (1 if yes, 0 if no); For the set of all target steel plates; The starting number of the target stack position This represents the maximum number of stack positions. This represents the maximum number of levels. The target is the total number of steel plates. The starting number of the target stack location. Indicates stack position This belongs to the target stacking area. For the final moment.

[0016] As a preferred technical solution, the calculation of the fitness value of an individual based on the generated stacking step sequence includes: Calculate the total number of stacking steps using the following formula:

[0017] In the formula, This represents the target value for the total number of stacking steps; Indicates the maximum time step; Indicates the maximum number of stack positions; Indicates the maximum number of levels; Let be the decision variable, representing the time interval . Located at the stack position No. The steel plates of the layer are moved to the stack. No. layer; The cumulative number of times the steel plate placement height exceeded the height limit threshold during the statistical decoding process; The cumulative number of times is weighted to obtain a penalty value. The penalty value is then added to the total number of stacking steps to obtain the individual's penalty fitness value.

[0018] As a preferred technical solution, the initial population is iteratively updated using the artificial bee colony algorithm, including performing evolutionary operator operations during the hired bee stage or the follower bee stage. The execution of the evolution operator operation includes: Obtain the length of the individual code that needs to be evolved, where the length of the individual code is equal to the number of target steel plates; The number of mutation points required for the evolutionary operation is determined based on the length of the individual code. In the individual coding, a gene position with an equal number of mutation sites is randomly selected; Insertion or exchange operations are performed on the codes at the selected gene locations to generate updated individuals.

[0019] As a preferred technical solution, the iterative update of the initial population using the artificial bee colony algorithm further includes performing a restart operation during the scout bee phase and performing a neighborhood search operation after the iteration is completed. The restart operation includes: when an individual has not been updated within a preset number of times, triggering a scout bee to check the individual's encoding structure; if the individual's encoding is a non-sequential structure, then performing a reverse structure to sequence operation to generate a new individual; if the individual's encoding is already a sequential structure, then randomly generating a new individual to replace it; adding both the individual before replacement and the new individual after replacement to an external archive set. The neighborhood search operation includes: stopping the main loop after the iterative update meets the preset termination condition, merging the current population with the individuals in the external archive set to form a temporary population; performing the reverse structure sequentialization operation on each individual in the temporary population to generate a corresponding neighborhood search individual; calculating the total number of stacking steps during the decoding process of the individual encoding, and recording the source stack position number and target stack position number corresponding to each move operation to form a detailed step list; selecting the optimal individual based on the total number of stacking steps, and outputting the detailed step list corresponding to the optimal individual as the final stacking scheme.

[0020] In a second aspect, the present invention also proposes an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the operations performed by the unmanned overhead crane stacking method with integrated height constraints as described in any of the embodiments of the first aspect.

[0021] The beneficial effects of the present invention include at least the following: This invention introduces a movement difficulty evaluation mechanism that integrates height-limit constraints during the decoding stage. Instead of using random or simple greedy strategies to select the movement position for obstacle steel plates above the target steel plate, it comprehensively considers the current height of each candidate movement position, whether it is an idle position, and whether it exceeds the height limit threshold after movement. It calculates and selects the position with the lowest movement difficulty evaluation value as the temporary storage point for the obstacle steel plate. This effectively avoids invalid paths where the height of the position exceeds the threshold due to simply pursuing movement, ensuring that the algorithm can still find feasible and high-quality solutions under strict height constraints. This solves the problem of existing technologies easily getting trapped in local optima or having no solution under height restrictions. Furthermore, by prioritizing the selection of the lowest difficulty value (e.g., idle or low-height positions), the number of subsequent secondary handling of the obstacle steel plate is effectively reduced, significantly lowering the redundancy of the overall repacking steps. Combined with the global iterative update of the artificial bee colony algorithm, the sorting task is ultimately completed with the fewest repacking operations, greatly improving the operational efficiency of the unmanned overhead crane. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the unmanned overhead crane stacking method that integrates height limit constraints provided in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the encoding logic provided in an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the clearance priority rule logic provided in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of the upper-level priority rule logic provided in an embodiment of the present invention.

[0026] Figure 5 This is a complete decoding flowchart provided for an embodiment of the present invention.

[0027] Figure 6 This is a schematic diagram of the stack movement difficulty ranking logic provided in an embodiment of the present invention, which integrates height limit constraints.

[0028] Figure 7 The flowchart of the artificial bee colony algorithm provided in the embodiment of the present invention is shown.

[0029] Figure 8 The graph shows the number of iterations for different core counts provided in the embodiments of the present invention.

[0030] Figure 9 This is a normalized line graph showing the number of iterations for different core counts provided in an embodiment of the present invention.

[0031] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0032] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred technical solutions. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred technical solutions are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0033] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0034] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0035] Example 1 This embodiment proposes an unmanned overhead crane stacking method that integrates height-limited constraints, such as... Figure 1 As shown, Figure 1 This is a flowchart illustrating an unmanned overhead crane stacking method that integrates height restriction constraints, as provided in this embodiment. The method includes the following steps: S1: Obtain the initial stacking status information of the warehouse area, the sorting requirements of the target steel plate, and the height limit threshold information of each stacking location; S2: Based on the sorting demand information of the target steel plate, the movement order of the target steel plate is used as a decision variable for encoding to generate an initial population containing several individuals. S3: Decode individuals within the population. When there is an obstacle steel plate above the target steel plate to be moved, obtain the current height of each candidate moving position, whether it is an empty position, and whether it exceeds the height limit threshold after moving. S4: Calculate the movement difficulty evaluation value of each candidate moving position based on the status information; S5: Move the obstacle steel plate into the feasible stacking position with the lowest moving difficulty evaluation value until there are no obstacle steel plates above the target steel plate to be moved. Move the target steel plate to be moved into the designated target stacking position and generate the corresponding stacking step sequence. S6: Calculate the fitness value of an individual based on the generated stacking step sequence, and use the artificial bee colony algorithm to iteratively update the initial population until the preset termination condition is met, and output the final stacking step corresponding to the individual with the best fitness value.

[0036] Understandably, by introducing a movement difficulty evaluation mechanism that incorporates height-limit constraints during the decoding stage, the algorithm no longer uses a random or simple greedy strategy to select the movement position for the obstacle steel plate above the target steel plate to be moved. Instead, it comprehensively considers the current height of each candidate movement position, whether it is an idle position, and whether it exceeds the height limit threshold after movement. It calculates and selects the position with the smallest movement difficulty evaluation value as the temporary storage point for the obstacle steel plate. This can avoid invalid paths caused by simply pursuing movement, resulting in the height of the position exceeding the threshold. This ensures that the algorithm can still find feasible and high-quality solutions under strict height constraints, solving the problem that existing technologies are prone to getting trapped in local optima or having no solution under height-limit conditions. On the other hand, by prioritizing the selection of the lowest difficulty value (such as an idle or low-height position), the number of secondary handling of the obstacle steel plate is effectively reduced, significantly reducing the redundancy of the overall repacking steps. Combined with the global iterative update of the artificial bee colony algorithm, the sorting task is finally completed with the fewest repacking steps, greatly improving the operating efficiency of the unmanned overhead crane.

[0037] Example 2 This embodiment is an improvement on the unmanned overhead crane stacking method that integrates height limit constraints proposed in Embodiment 1.

[0038] like Figure 2 As shown, Figure 2 This is a schematic diagram of the encoding logic provided in an embodiment of the present invention. In this embodiment, the actual physical stacking state is mapped to a numerical encoding sequence. Since there are no duplicate steel plates in the initial stack position, and the movement order of the target steel plates determines the final stacking steps, the ID sequence of the global target steel plates is used as a decision variable. The encoding example shows how a specific stacking state is transformed into a set of numerical sequences representing movement priorities, which is an individual in the algorithm.

[0039] In this embodiment, the movement order of the target steel plate is used as a decision variable for encoding to generate an initial population containing several individuals, including: Calculate the layer distance between the target steel plates at the bottom and top of the stack, and convert the layer distance into a first numerical component; The total number of target steel plates contained within the layer distance interval between the bottom target steel plate and the top target steel plate is counted, and the total number of target steel plates is converted into a second numerical part. The first numerical part and the second numerical part are superimposed to form the clearance difficulty value of the stack position. Then, according to the clearance difficulty value in ascending order, individual codes containing the movement order of the target steel plate are generated by the removal operation.

[0040] It should be noted that, in order to accurately quantify the difficulty of clearing and convert it into a sortable value, this embodiment sets the influence of layer distance to the thousandths place and the influence of the target quantity to the millionths place. Here, priority is given to stacks where the target steel plates are physically closer to the top layer (smaller layer distance, lower difficulty value). When the layer distance is the same, priority is given to stacks where the target steel plates are more densely packed (more quantity, larger reciprocal, but less overall impact than layer distance), thereby accelerating the formation of empty stacks. Furthermore, height restriction factors and empty stack factors are not considered in this initialization phase, as this process is only used to generate initial codes; global height restriction constraints will be strictly controlled in the subsequent decoding stage.

[0041] like Figure 3 As shown, Figure 3 This is a schematic diagram of the clearance priority rule logic provided in an embodiment of the present invention. This rule is used to generate high-quality initial population individuals. During the calculation process, the layer distance (the distance of the target steel plate from the top layer) is considered in the thousandths place, and the target quantity (the number of other target steel plates contained within the interval) is considered in the millionths place. Figure 3 As can be seen from the example, the algorithm will prioritize processing those stacks that are close to the top layer and have relatively concentrated target distribution (for example, the difficulty of clearing out in the figure is ranked as 3<4<1<2), thereby accelerating the formation of empty stacks.

[0042] In this embodiment, the movement order of the target steel plate is used as a decision variable for encoding to generate an initial population containing several individuals, including: Obtain the hierarchical position information of each target steel plate in its respective stack; Based on the hierarchical location information, calculate the hierarchical difference between each target steel plate and the top layer of its stack position; Select the target steel plate with the smallest layer difference to add to the moving sequence. If there are multiple target steel plates with the same layer difference, randomly select one of them to add to the moving sequence, until all target steel plates are added to the moving sequence, forming an individual code.

[0043] It should be noted that although theoretically the movement order of the target steel plates is unordered in the encoding, in actual physical stacking, the steel plates on the upper layer must be moved out before the steel plates on the lower layer. Therefore, this embodiment simulates this physical constraint by calculating the layer difference, prioritizing the addition of target steel plates with less obstruction (i.e., closest to the top layer) to the sequence, thus generating initial individuals that better conform to the actual operational logic. If multiple target steel plates have the same layer difference from the top layer, it means they are all in a relatively movable state in their respective stack positions. In this case, random selection not only does not violate the physical logic but also increases the diversity of the population, helping the algorithm escape local optima. Figure 4 As shown, Figure 4 This is a schematic diagram of the upper-level priority rule logic provided in an embodiment of the present invention. Figure 4 The logic of the upper-layer priority rule is illustrated. In the initial environment, although the logical order of the target steel plates is unrestricted, physically they must follow the rule of moving the upper layers first and then the lower layers. The illustrated process calculates the difference in the number of layers between each target steel plate and the top layer of its stack, and prioritizes adding the target steel plate with the smallest difference in the number of layers (i.e., the topmost one) to the coding sequence (e.g., removing the topmost steel plate 6 or 7 from the stack first), thereby generating an initial individual that conforms to the physical operation logic.

[0044] like Figure 5 As shown, Figure 5 This is a complete decoding flowchart provided in the embodiments of the present invention. Figure 5 This embodiment demonstrates the complete decoding logic and process. The process uses the target steel plate as the main loop: first, the target steel plate to be moved is determined based on the encoding; second, it is determined whether there are any obstruction steel plates above it; if so, a sub-loop for moving the obstruction steel plate is entered, and the obstruction steel plate is moved to the optimal position by calculating the moving difficulty (primarily based on the empty stack position, and secondarily based on the stack height); when there are no obstructions above the target steel plate, it is moved into the final target stack position, until all target steel plates are in place. In this embodiment, the calculation of the moving difficulty evaluation value of each candidate stack position based on the state information includes: Determine whether the current height of the current candidate pallet position exceeds the height limit threshold information. If it exceeds the limit, assign a first basic difficulty coefficient to the current candidate pallet position. If it does not exceed the limit, determine whether the current candidate pallet position is an empty pallet position that does not contain the target steel plate to be sorted. If it is an empty pallet position, assign a second basic difficulty coefficient to the current candidate pallet position. If it is not an empty pallet position, assign a third basic difficulty coefficient to the current candidate pallet position. Wherein, the value of the first basic difficulty coefficient is greater than the value of the third basic difficulty coefficient, and the value of the third basic difficulty coefficient is greater than the value of the second basic difficulty coefficient. Obtain the specific height value of the current candidate stacking position and convert it into a correction coefficient in decimal form. Add the correction coefficient to the assigned basic difficulty coefficient to obtain the final movement difficulty evaluation value of the current candidate stacking position.

[0045] It should be noted that, in order to transform the complex warehouse conditions into an executable decision-making basis for the algorithm, this embodiment sets a first basic difficulty coefficient of 2 for infeasible conditions exceeding the height limit threshold, representing the highest penalty; sets a second basic difficulty coefficient of 0 for the optimal condition of a completely idle stack without exceeding the height limit; and sets a third basic difficulty coefficient of 1 for the ordinary condition of a non-target steel plate already present without exceeding the height limit. Furthermore, to differentiate the merits of stacks of different heights under the same conditions, the actual height of the stack is processed to a decimal with ten-thousandths as a correction coefficient.

[0046] like Figure 6 As shown, Figure 6 This is a schematic diagram of the stack movement difficulty ranking logic provided in an embodiment of the present invention, which integrates height limit constraints. Figure 6 This paper demonstrates a method for calculating the difficulty of stack relocation by incorporating height restrictions. Different base difficulty coefficients (2, 0, and 1) are assigned to different stack relocation states (e.g., insufficient remaining height, empty stacks without target steel plates, and non-empty stacks with target steel plates). Combining the actual height of the stack (as a decimal correction), a precise relocation difficulty value is finally calculated. In the illustrated example, the algorithm intelligently selected the empty stack position with the lowest difficulty as the moving position for the obstacle steel plate by comparing the difficulty values, thereby effectively avoiding the risk of exceeding the height limit and reducing the subsequent stacking cost.

[0047] In this embodiment, the following constraints must be met when moving the obstacle steel plate into the feasible moving position with the lowest moving difficulty evaluation value:

[0048]

[0049] In the formula, Indicates at time The first reservoir area The first stack position layer If it exists, the value is 1; otherwise, it is 0. Represents a time set, Represents the set of stack positions. Represents a hierarchical set. Indicates the maximum number of levels; Indicates at time The first reservoir area The first stack position The specific numerical designation of the steel plate on the upper layer; It is a positive integer constant.

[0050] In this embodiment, moving the target steel plate to be moved into the designated target stack position satisfies the following constraint formula:

[0051]

[0052]

[0053]

[0054] in, Indicates at time , will be located in the stack position No. The steel plates of the layer are moved to the stack. No. The value indicates whether the layer action has occurred; if it has, the value is 1, otherwise it is 0. Indicates at time The first reservoir area The first stack position layer ; For state variables, the number is _____. Whether the target steel plate is ultimately located at the target stack position The Layer (1 if yes, 0 if no); For the set of all target steel plates; The starting number of the target stack position This represents the maximum number of stack positions. This represents the maximum number of levels. The target is the total number of steel plates. The starting number of the target stack location. Indicates stack position This belongs to the target stacking area. For the final moment.

[0055] It should be noted that the above constraint formula ensures that the target steel plate appears only once in the target stack, the target steel plate is assigned to a single position within a target stack, and steel plates cannot be transferred between target stacks. Restrictions are placed on the movable steel plates. First, a top-level constraint is set: when in two different stacks, the position to be moved out must contain a steel plate, and the position to be moved in must be empty; subsequently, the same constraints are applied to four different cases: ① When the layer being moved in is the bottom layer and the layer being moved out is not the top layer; ② When the layer being moved in is not the bottom layer and the layer being moved out is not the top layer; ③ When the layer being moved in is the bottom layer and the layer being moved out is the top layer; ④ When the layer being moved in is not the bottom layer, and the layer being moved out is the top layer; The above constraints ensure the correctness of each steel plate stacking. Based on this, further constraints are applied to ensure that the mathematical model can solve for the optimal result.

[0056] In this embodiment, calculating the fitness value of an individual based on the generated stacking step sequence includes: Calculate the total number of stacking steps using the following formula:

[0057] In the formula, This represents the target value for the total number of stacking steps; Indicates the maximum time step; Indicates the maximum number of stack positions; Indicates the maximum number of levels; Let be the decision variable, representing the time interval . Located at the stack position No. The steel plates of the layer are moved to the stack. No. layer; The cumulative number of times the steel plate placement height exceeded the height limit threshold during the statistical decoding process; The cumulative number of times is weighted to obtain a penalty value. The penalty value is then added to the total number of stacking steps to obtain the individual's penalty fitness value.

[0058] It should be noted that, in order to effectively guide the population to avoid solutions exceeding the height limit during the evolutionary process, this embodiment designs two fitness functions. One is a penalized fitness function used for internal algorithm evolution. Through weighted penalties, the fitness of individuals that violate the height limit significantly decreases, making them more likely to be eliminated by the algorithm. The other is a complete fitness function used for the final output. It records the source and target stack numbers for each move in detail while calculating the number of steps, without applying penalty weights. Its purpose is to output the real physical operation steps and status for direct use by the subsequent unmanned crane control system.

[0059] like Figure 7 As shown, Figure 7 This is a flowchart of the artificial bee colony algorithm provided in an embodiment of the present invention. Figure 7 This diagram illustrates the execution flow of the improved artificial bee colony algorithm used in this embodiment. The algorithm mainly includes population initialization, the hired bee phase (searching for new food sources), the follower bee phase (selecting and searching based on fitness), the scout bee phase (restarting solutions that have stalled), and the neighborhood search phase. The illustrated flow clearly shows how the algorithm, through iterative iteration, utilizes a dual fitness function for evaluation and an adaptive evolutionary operator to ultimately output the globally optimal stacking solution.

[0060] In this embodiment, the iterative update of the initial population using the artificial bee colony algorithm includes performing evolutionary operator operations during the hired bee stage or the follower bee stage. The execution of the evolution operator operation includes: Obtain the length of the individual code that needs to be evolved, where the length of the individual code is equal to the number of target steel plates; The number of mutation points required for the evolutionary operation is determined based on the length of the individual code. In the individual coding, a gene position with an equal number of mutation sites is randomly selected; Insertion or exchange operations are performed on the codes at the selected gene locations to generate updated individuals.

[0061] It should be noted that due to the vast differences in scale of the steel plate stacking problem, a fixed mutation strategy is insufficient to balance the breadth and depth of the search. Therefore, this embodiment employs an adaptive adjustment strategy, where the number of mutation points M is positively correlated with the current individual code length Len. Specifically, when the code length Len is long, the algorithm automatically increases the number of insertion or swapping points M to expand the scope of the neighborhood search; conversely, when the code is short, the number of mutation points is reduced to perform a refined local search.

[0062] In this embodiment, the iterative update of the initial population using the artificial bee colony algorithm further includes performing a restart operation during the scout bee phase and performing a neighborhood search operation after the iteration is completed. The restart operation includes: when an individual has not been updated within a preset number of times, triggering a scout bee to check the individual's encoding structure; if the individual's encoding is a non-sequential structure, then performing a reverse structure to sequence operation to generate a new individual; if the individual's encoding is already a sequential structure, then randomly generating a new individual to replace it; adding both the individual before replacement and the new individual after replacement to an external archive set. The neighborhood search operation includes: stopping the main loop after the iterative update meets the preset termination condition, merging the current population with the individuals in the external archive set to form a temporary population; performing the reverse structure sequentialization operation on each individual in the temporary population to generate a corresponding neighborhood search individual; calculating the total number of stacking steps during the decoding process of the individual encoding, and recording the source stack position number and target stack position number corresponding to each move operation to form a detailed step list; selecting the optimal individual based on the total number of stacking steps, and outputting the detailed step list corresponding to the optimal individual as the final stacking scheme.

[0063] It should be noted that the reverse structure reordering operation in this embodiment is a special repair mechanism for deadlocks or poor solutions. When the scout bee discovers that an individual has not been updated for a long time and its encoding presents a non-sequential, disordered structure, the algorithm will force a rearrangement to conform to the basic physical output logic, thereby repairing an infeasible or poor solution into a feasible one, which serves as a restart point. Simultaneously, the use of an external archive set ensures that the historical best solutions generated in each iteration are not lost. By combining the current population with the archive set after the iteration ends and performing another neighborhood search, potential better stacking paths can be further explored, ensuring the robustness of the final output solution.

[0064] Example 3 Based on Example 2, this embodiment further details the data processing logic, anomaly detection mechanism, and multi-process computing resource configuration scheme of the unmanned overhead crane stacking method that integrates height limit constraints in actual system applications.

[0065] In this embodiment, it is assumed that the input of the algorithm is datas=[[[5,6,1],[2,7,3],[9,4]],[[2,4,1],[5,6,7]],{"1":8,"2":30,"3":10,"4":8,"5":20,"6":50,"7":12,"8":16,"9":8},[78,52,16,46,82],10,6], and the input data structure contains six parts: Initial warehouse status: stored in list form, such as [[5,6,1],[2,7,3],[9,4]]. Sublists represent initial warehouse stack positions, where numbers represent steel plate numbers. The list order (e.g., from left to right) corresponds to the physical stacking order of steel plates from bottom to top. Target steel plates and their allocation: stored in list form, such as [[2,4,1],[5,6,7]]. Sublists represent target stack positions, numbers represent target steel plate numbers, and the list order corresponds to the stacking requirements from bottom to top. Steel plate thickness data dictionary: The dictionary is in the form of {"1":8,"2":30,"3":10,"4":8,"5":20,"6":50,"7":12,"8":16,"9":8}, where "1":8 represents that the thickness of steel plate No. 1 is 8mm (Note that quotation marks need to be added to the steel plate number in the WebService call, otherwise it cannot be recognized). Case height limit list: stored in list form, for example [78,52,16,46,82], where the first half corresponds to the height limit threshold of the initial stack position and the second half corresponds to the height limit threshold of the target stack position, in mm. The steel plates on the stack position cannot exceed their corresponding height limit. Computation time: Sets the upper limit of the algorithm's execution time in seconds, which can be defined by the user; Number of cores: Sets the number of CPU cores used for multi-process parallel computing, in units of individual cores, which can be defined by the user.

[0066] In this embodiment, the algorithm's output is result=[7,[[3,4],[1,4],[2,3],[2,5],[1,5],[2,4],[1,5]],1704], which is encapsulated into an information packet containing three dimensions: Optimal number of steps: The specific number of steps obtained after solving this case, such as 7 in the example; Detailed step sequence: The specific movement instructions are recorded in list form, such as [[3,4],[1,4],[2,3]...]. Each instruction contains a stack position number. For example, [3,4] means moving the top layer steel plate of stack position 3 to stack position 4 (the stack position number is "initial stack position + target stack position"). Since the stacking steps only need to consider the stack position numbers, and the steel plate that can be moved by each action is fixed and unique (i.e., the top layer), there is no need to output the specific steel plate number. Algorithm loop algebra: Records the actual number of loops (i.e., iteration algebras) completed by the algorithm within a specified time, such as 1704 in the example.

[0067] To ensure the stability of the algorithm, the system performs the following five levels of data anomaly checks before initializing the population. If there is a problem with the initial input data, the output will be a corresponding text message: Duplicate check: Check if there are duplicate numbers in the initial set of steel plates and the target set of steel plates. If so, output the prompts "Duplicate target exists in the initial set of steel plates, please check the data!" and "Duplicate target exists in the target set of steel plates, please check the data!" respectively. Existence check: Checks whether all target steel plates are included in the initial storage area. If the target steel plate is not included in the initial environment, the target cannot be found, and the output message is "A target steel plate does not exist in the initial environment. Please check the data!". Thickness data integrity check: Traverse all steel plate numbers and check if the data dictionary contains the corresponding thickness data. If some steel plates have no thickness data, the height cannot be calculated, and the output message is "There are steel plates with no thickness data, the height cannot be calculated, please check the data!". Height limit matching verification: First, determine whether the height limit data matches the number of stack positions. If they do not match, output the message "The number of stack positions in the given data does not match the height limit data. Please check the data!". Physical state compliance verification: Next, it is determined whether the initial state exceeds the height limit requirement. If it does not meet the requirement, the output prompt is "The steel plate in the initial stack exceeds the height limit. Please check the data!"; Finally, it is determined whether the target meets the height limit requirement. If it does not meet the requirement, the output prompt is "The steel plate in the target stack exceeds the height limit. Please check the data!".

[0068] Because many heuristic rules in this project yielded excellent solutions under sufficient height constraints, the computational stability for these cases was very high, resulting in excellent results. Even in the extreme case of insufficient height constraints, the multi-process computation still yielded relatively good results. Therefore, it was ultimately decided not to add any more machine learning or deep reinforcement learning modules. Due to the addition of multi-process computation, an algorithm input component related to the number of CPU cores was added. Detailed segmented solutions were performed for 105 special cases, with a solution time of 300 seconds and the algebra output every 60 seconds. Plotting the data as a line graph along both horizontal and vertical dimensions determined that the optimal number of CPU cores was 6 when the population size was 30. Figure 8 As shown, Figure 8 This is a line graph showing the number of iterations for different core counts provided in an embodiment of the present invention. Figure 8 The graph shows the computational algebra line graphs for five different numbers of cores. It can be seen that, except for a few points, most of them are dark green (6 cores) at the top, indicating that the algorithm with 6 cores iterates the most times in the same amount of time. Figure 9 This is a normalized line graph showing different core counts and iteration times provided in the embodiments of the present invention. Figure 9 It can be seen that most of the peak values ​​recede at the 6-core position, indicating that the 6-core algorithm iterates the most times in the same amount of time.

[0069] Example 3 Figure 10 This is a schematic diagram of the structure of the electronic device 100 provided in this embodiment. The electronic device 100 includes: a memory 101, a processor 102, and a computer program stored in the memory 101 and executable on the processor 102.

[0070] When the processor 102 executes the program, it implements the unmanned overhead crane stacking method with integrated height limit constraints provided in the above embodiments.

[0071] Furthermore, the electronic device 100 also includes a communication interface 103 for communication between the memory 101 and the processor 102.

[0072] The memory 101 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0073] If the memory 101, processor 102, and communication interface 103 are implemented independently, then the communication interface 103, memory 101, and processor 102 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 10 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0074] Optionally, in a specific implementation, if the memory 101, processor 102, and communication interface 103 are integrated on a single chip, then the memory 101, processor 102, and communication interface 103 can communicate with each other through an internal interface.

[0075] Processor 102 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention.

[0076] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described unmanned overhead crane stacking method with fused height constraints.

[0077] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0078] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "N" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0079] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more N executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of the invention includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as should be understood by those skilled in the art to which embodiments of the invention pertain.

[0080] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any of the following techniques known in the art, or a combination thereof: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (FPGAs), field-programmable gate arrays (FPGAs), etc.

[0081] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium. When executed, the program includes one or a combination of the steps of the method embodiments.

[0082] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A method for unmanned overhead crane stacking that integrates height restriction constraints, characterized in that, include: Obtain the initial stack status information of the warehouse area, the sorting requirements of the target steel plates, and the height limit threshold information of each stack. Based on the sorting requirements of the target steel plate, the movement order of the target steel plate is used as a decision variable for encoding to generate an initial population containing several individuals. Decode the individuals in the population. When there is an obstacle steel plate above the target steel plate to be moved, obtain the current height of each candidate moving position, whether it is an empty position, and whether it exceeds the height limit threshold after moving. The movement difficulty evaluation value of each candidate stacking position is calculated based on the status information. Move the obstacle steel plate into the feasible stacking position with the lowest moving difficulty evaluation value until there are no obstacle steel plates above the target steel plate to be moved. Then move the target steel plate to be moved into the designated target stacking position and generate the corresponding stacking step sequence. The fitness value of an individual is calculated based on the generated stacking step sequence, and the initial population is iteratively updated using the artificial bee colony algorithm until a preset termination condition is met. The final stacking step corresponding to the individual with the best fitness value is then output.

2. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 1, characterized in that, The movement order of the target steel plate is encoded as a decision variable to generate an initial population containing several individuals, including: Calculate the layer distance between the target steel plates at the bottom and top of the stack, and convert the layer distance into a first numerical component; The total number of target steel plates contained within the layer distance interval between the bottom target steel plate and the top target steel plate is counted, and the total number of target steel plates is converted into a second numerical part. The first numerical part and the second numerical part are superimposed to form the clearance difficulty value of the stack position. Then, according to the clearance difficulty value in ascending order, individual codes containing the movement order of the target steel plate are generated by the removal operation.

3. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 1, characterized in that, The movement order of the target steel plate is encoded as a decision variable to generate an initial population containing several individuals, including: Obtain the hierarchical position information of each target steel plate in its respective stack; Based on the hierarchical location information, calculate the hierarchical difference between each target steel plate and the top layer of its stack position; Select the target steel plate with the smallest layer difference to add to the moving sequence. If there are multiple target steel plates with the same layer difference, randomly select one of them to add to the moving sequence, until all target steel plates are added to the moving sequence, forming an individual code.

4. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 1, characterized in that, The step of calculating the movement difficulty evaluation value of each candidate stacking position based on the status information includes: Determine whether the current height of the current candidate pallet position exceeds the height limit threshold information. If it exceeds the limit, assign a first basic difficulty coefficient to the current candidate pallet position. If it does not exceed the limit, determine whether the current candidate pallet position is an empty pallet position that does not contain the target steel plate to be sorted. If it is an empty pallet position, assign a second basic difficulty coefficient to the current candidate pallet position. If it is not an empty pallet position, assign a third basic difficulty coefficient to the current candidate pallet position. Wherein, the value of the first basic difficulty coefficient is greater than the value of the third basic difficulty coefficient, and the value of the third basic difficulty coefficient is greater than the value of the second basic difficulty coefficient. Obtain the specific height value of the current candidate stacking position and convert it into a correction coefficient in decimal form. Add the correction coefficient to the assigned basic difficulty coefficient to obtain the final movement difficulty evaluation value of the current candidate stacking position.

5. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 4, characterized in that, The following constraints must be met when moving the obstacle steel plate into the feasible moving position with the lowest moving difficulty evaluation value: In the formula, Indicates at time The first reservoir area The first stack position layer If it exists, the value is 1; otherwise, it is 0. Represents a time set, Represents the set of stack positions. Represents a hierarchical set. Indicates the maximum number of levels; Indicates at time The first reservoir area The first stack position The specific numerical designation of the steel plate on the upper layer; It is a positive integer constant.

6. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 1, characterized in that, The process of moving the target steel plate to be moved into the designated target stack position satisfies the following constraint formula: in, Indicates at time , will be located in the stack position No. The steel plates of the layer are moved to the stack. No. The value indicates whether the layer action has occurred; if it has, the value is 1, otherwise it is 0. Indicates at time The first reservoir area The first stack position layer ; For state variables, the number is _____. Whether the target steel plate is ultimately located at the target stack position The layer; For the set of all target steel plates; The starting number of the target stack position This represents the maximum number of stack positions. This represents the maximum number of levels. The target is the total number of steel plates. The starting number of the target stack location. Indicates stack position This belongs to the target stacking area. For the final moment.

7. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 1, characterized in that, The calculation of the fitness value of an individual based on the generated stacking step sequence includes: Calculate the total number of stacking steps using the following formula: In the formula, This represents the target value for the total number of stacking steps; Indicates the maximum time step; Indicates the maximum number of stack positions; Indicates the maximum number of levels; Let be the decision variable, representing the time interval . Located at the stack position No. The steel plates of the layer are moved to the stack. No. layer; The cumulative number of times the steel plate placement height exceeded the height limit threshold during the statistical decoding process; The cumulative number of times is weighted to obtain a penalty value. The penalty value is then added to the total number of stacking steps to obtain the individual's penalty fitness value.

8. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 1, characterized in that, The initial population is iteratively updated using an artificial bee colony algorithm, including performing evolutionary operator operations during the hired bee stage or the follower bee stage. The execution of the evolution operator operation includes: Obtain the length of the individual code that needs to be evolved, where the length of the individual code is equal to the number of target steel plates; The number of mutation points required for the evolutionary operation is determined based on the length of the individual code. In the individual coding, a gene position with an equal number of mutation sites is randomly selected; Insertion or exchange operations are performed on the codes at the selected gene locations to generate updated individuals.

9. The unmanned overhead crane stacking method integrating height restriction constraints according to claim 8, characterized in that, The method of iteratively updating the initial population using the artificial bee colony algorithm also includes performing a restart operation during the scout bee phase and performing a neighborhood search operation after the iteration is completed. The restart operation includes: when an individual has not been updated within a preset number of times, triggering a scout bee to check the individual's encoding structure; if the individual's encoding is a non-sequential structure, then performing a reverse structure to sequence operation to generate a new individual; if the individual's encoding is already a sequential structure, then randomly generating a new individual to replace it; adding both the individual before replacement and the new individual after replacement to an external archive set. The neighborhood search operation includes: stopping the main loop after the iterative update meets the preset termination condition, merging the current population with the individuals in the external archive set to form a temporary population; performing the reverse structure sequentialization operation on each individual in the temporary population to generate a corresponding neighborhood search individual; calculating the total number of stacking steps during the decoding process of the individual encoding, and recording the source stack position number and target stack position number corresponding to each move operation to form a detailed step list; selecting the optimal individual based on the total number of stacking steps, and outputting the detailed step list corresponding to the optimal individual as the final stacking scheme.

10. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the operations performed by the unmanned overhead crane stacking method with fused height constraints as described in any one of claims 1 to 9.

Images