Logistics scheduling method and system based on warehouse space planning

Abstract

The present invention provides a logistics scheduling method and system based on warehouse space planning, which relates to the field of logistics warehousing technology, including constructing a comprehensive objective function according to the warehouse space and in combination with the logistics orders to be scheduled; setting an initial set of layout schemes, and calculating the fitness value of each layout scheme in the set of layout schemes based on a layout optimization algorithm by solving the comprehensive objective function, performing iterative position update based on a position guidance coefficient and a comparison result with a preset guidance coefficient threshold to obtain an optimal space layout, and setting a warehouse space structure according to the optimal space layout; constructing a warehouse space model based on the warehouse space structure and in combination with the basic parameters of an AGV robot, and generating an optimal navigation path by parsing the demand instructions of the logistics order and through continuous decision-making of a dynamic potential energy navigation algorithm to complete the scheduling task of the logistics order.

Description

Technical Field

[0001] The present invention relates to the field of logistics warehousing technology, and in particular to a logistics scheduling method and system based on warehouse space planning. Background Art

[0002] The efficiency and convenience of logistics is a difficult problem that must be solved. However, the current logistics industry still has shortcomings such as low mechanization, high labor demand, and low intelligence. The resulting high logistics costs, low operating efficiency and other problems seriously restrict the development of the logistics industry and even the progress of the economy and society. In the field of logistics, with the development and application of intelligent delivery robots, modern logistics warehousing will use intelligent delivery robots instead of manual labor to transport goods.

[0003] CN202110694510.3 discloses a robot path planning method, device and robot, the method comprising collecting environmental information in a preset fixed partition and establishing a partition map; when a threshold is identified in the process of establishing the partition map, the threshold is marked in the partition map; according to the established partition map and the marked threshold, a cleaning path is planned to sequentially clean each partition to be cleaned divided by both sides of the threshold in the fixed partition; the fixed partition is cleaned accordingly according to the planned cleaning path. The present invention solves the problem that the existing cleaning path planning is unreasonable, resulting in frequent crossing of thresholds during cleaning and low cleaning efficiency;

[0004] In summary, the use of AGV robots to handle logistics orders for warehouse goods transportation will become the mainstream. On the basis of existing technologies, in order to make AGV robots operate more efficiently, various costs are coordinated, and by optimizing warehouse layout and precise route planning of AGV robots, accurate and fast logistics automation solutions are provided. The application of this invention can at least solve some of the problems of the existing technologies. Summary of the invention

[0005] The embodiments of the present invention provide a logistics scheduling method and system based on warehouse space planning, which can at least solve some problems in the prior art.

[0006] According to a first aspect of the embodiments of the present invention,

[0007] A logistics scheduling method based on warehouse space planning is provided, comprising:

[0008] According to the warehouse space and the logistics orders to be dispatched, a first objective function is constructed with the goal of minimizing the total logistics handling cost, a second objective function is constructed with the goal of maximizing the comprehensive correlation, a third objective function is constructed with the goal of maximizing the land area utilization rate, and a comprehensive objective function is constructed based on the first objective function, the second objective function and the third objective function;

[0009] An initial set of layout solutions is set, and based on the layout optimization algorithm, the fitness value of each layout solution in the set of layout solutions is calculated by solving the comprehensive objective function, and based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold, the position is updated iteratively to obtain the optimal spatial layout, and the warehouse space structure is set according to the optimal spatial layout;

[0010] Based on the warehouse space structure and combined with the basic parameters of the AGV robot, a warehouse space model is constructed. By parsing the demand instructions of the logistics order and through continuous decision-making of the dynamic potential energy navigation algorithm, the optimal navigation path is generated to complete the scheduling task of the logistics order.

[0011] In an optional embodiment,

[0012] Taking the minimization of total logistics handling cost as the goal, the first objective function is constructed including:

[0013]

[0014] Among them, f1 represents the result of the first objective function, g represents the total number of cargo batches, x represents the cargo batch index, Ac x Denotes the labor cost of the xth batch of goods, Dc x represents the maintenance and repair cost of the machinery and equipment for the xth batch of goods, d x represents the transportation distance of the xth batch of goods, w x Indicates the weight of the xth batch of goods;

[0015] With the goal of maximizing the comprehensive correlation, the second objective function is constructed including:

[0016]

[0017] Among them, i, j, l and k represent the functional area index, m represents the total number of functional areas, R ij represents the correlation coefficient between functional area i and functional area j, F ij represents the logistics frequency between functional areas i and j, F lk represents the logistics frequency between functional area l and functional area k, T represents the total number of time periods, t represents the time period index, O ijt represents the amount of operation between functional area i and functional area j in time period t, D ij represents the distance between functional area i and functional area j, max(D) represents the maximum distance between all functional areas, A ij represents the adjacency association degree between functional area i and functional area j, ε represents a constant to prevent division by 0, P ij represents the path convenience score between functional area i and functional area j, Lij represents the physical connection score between functional area i and functional area j, and f2 represents the result of the second objective function;

[0018] Taking the maximization of land area utilization as the goal, the third objective function is constructed including:

[0019]

[0020] Among them, f3 represents the result of the third objective function, A F,i represents the area of ​​functional area i, U F,i represents the land use rate of functional area i, n represents the total number of transportation paths, y represents the transportation path index, and A P,y represents the area of ​​the transport path y, U P,y represents the land use rate of transportation path y, A T represents the total land area, λ represents the adjustment coefficient, r represents the total number of land use constraints, s represents the land use constraint index, V s represents the penalty term of land use constraint s.

[0021] In an optional embodiment,

[0022] An initial set of layout schemes is set, and based on the layout optimization algorithm, the fitness value of each layout scheme in the set of layout schemes is calculated by solving the comprehensive objective function, and the iterative position update is performed based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold value to obtain the optimal spatial layout, including:

[0023] Based on the total number of preset layout schemes, an initial layout scheme set is set; based on the layout scheme set, the fitness value of each layout scheme in the layout scheme set is calculated through the comprehensive objective function, and the layout scheme corresponding to the current highest fitness is determined as the benchmark layout scheme;

[0024] Based on the reference layout scheme, determining the position information of other layout schemes in the layout scheme set, and randomly assigning position guidance coefficients to the other layout schemes, and determining the position update mode according to the comparison relationship between the product of the position guidance coefficient and the position information and a preset guidance coefficient threshold;

[0025] According to the position update mode, the position update movement is performed, and the corresponding fitness value is calculated, and according to the preset optimal threshold, a corresponding number of layout schemes are retained to form an updated layout scheme set;

[0026] Iterate the position update until the preset maximum number of iterations is reached, and select the layout scheme corresponding to the maximum fitness value as the optimal spatial layout.

[0027] In an optional embodiment,

[0028] Based on the reference layout, determining the position information of other layouts in the initialization layout scheme, and randomly assigning position guidance coefficients to the other layouts, and determining the position update method according to a comparison relationship between the product of the position guidance coefficient and the position information and a preset guidance coefficient threshold, including:

[0029] If the product is greater than the guidance coefficient threshold, the position information of other layouts is updated in a gradient descent manner, and the formula is as follows:

[0030] pos(tp+1)=I·cos(2πl)+pos(tp);

[0031] Among them, pos(tp+1) represents the position information at time tp+1, pos(tp) represents the position information at time tp, I represents the initial position of the initial optimal individual, and l represents the gradient descent coefficient;

[0032] If the product is less than or equal to the guiding coefficient threshold, the position information of other layouts is updated toward the direction of the reference layout, and the formula is as follows:

[0033] pos(tp+1)=[pos best (tp)-pos(tp)]·B·a;

[0034] Among them, pos best (tp) represents the global optimal position at time tp, B represents the update scale step, and a represents the swing factor.

[0035] In an optional embodiment,

[0036] By analyzing the demand instructions of logistics orders and making continuous decisions through the dynamic potential navigation algorithm, the optimal navigation path is generated, including:

[0037] Parse the logistics order, and according to the distribution of the logistics order, obtain the demand instruction corresponding to the logistics order, and determine the order destination of the AGV robot;

[0038] Based on the order destination and the obstacle position, combined with the current position of the AGV robot, the target gravity and obstacle repulsion of the AGV robot at the current position are calculated by constructing an attractive potential field and a repulsive potential field. The motion state of the AGV robot at the current position is obtained by calculating the mechanical resultant force vector. The entire motion process is iterated, and the optimal navigation path is determined based on the motion state of each position.

[0039] In an optional embodiment,

[0040] By constructing an attractive potential field and a repulsive potential field, the target gravitational force and the obstacle repulsive force of the AGV robot at the current position are calculated, and the mechanical resultant force vector calculation includes:

[0041] By setting the safety distance of the attractive potential field, reducing the gravity on the far target side and compensating the gravity on the near target side, a gravity calculation formula is constructed, which is as follows:

[0042]

[0043] Where Fg represents gravity, u g represents the attractive potential field constant of the distant target side, d(X,X G ) represents the Euclidean distance between the AGV robot position and the order destination, ds represents the safety distance of the attractive potential field, and v g represents the attractive potential field constant of the near-target side gravity, d(X,X B ) represents the Euclidean distance between the AGV robot position and the obstacle;

[0044] Based on the distance between the AGV robot and the order destination, an adjustment factor is constructed. By setting the maximum influence distance of the repulsive potential field, combining the adjustment factor and the repulsive potential field constant, a repulsive force calculation formula is constructed, and the formula is as follows:

[0045]

[0046] Among them, Fr represents repulsion, u r represents the repulsive potential field constant, de represents the maximum influence distance of the repulsive potential field, Represents the gradient of the distance between the AGV robot and the obstacle.

[0047] According to a second aspect of the embodiments of the present invention,

[0048] A logistics scheduling system based on warehouse space planning is provided, comprising:

[0049] The first unit is used to construct a first objective function based on the warehouse space and the logistics orders to be scheduled with the goal of minimizing the total logistics handling cost, to construct a second objective function with the goal of maximizing the comprehensive correlation, to construct a third objective function with the goal of maximizing the land area utilization rate, and to construct a comprehensive objective function based on the first objective function, the second objective function and the third objective function;

[0050] The second unit is used to set an initial layout scheme set, calculate the fitness value of each layout scheme in the layout scheme set by solving the comprehensive objective function based on the layout optimization algorithm, perform iterative position update based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold, obtain the optimal spatial layout, and set the warehouse space structure according to the optimal spatial layout;

[0051] The third unit is used to construct a warehouse space model based on the warehouse space structure and the basic parameters of the AGV robot, and to generate an optimal navigation path by parsing the demand instructions of the logistics order and making continuous decisions through the dynamic potential energy navigation algorithm to complete the scheduling task of the logistics order.

[0052] According to a third aspect of the embodiments of the present invention,

[0053] An electronic device is provided, comprising:

[0054] processor;

[0055] a memory for storing processor-executable instructions;

[0056] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0057] According to a fourth aspect of the embodiments of the present invention,

[0058] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0059] In an embodiment of the present invention, warehouse management faces multiple challenges, including reducing costs, improving efficiency and maximizing space utilization. A single objective function cannot fully reflect all requirements, and a comprehensive objective function can balance these diverse requirements. By adjusting the weight coefficients provided when constructing the comprehensive objective function, the operation of the warehouse can be flexibly optimized according to the actual operation strategy and priority, ensuring the best performance in different stages or situations, promoting the collaborative work of various systems and processes within the warehouse, and improving the overall operation efficiency. By optimizing the spatial layout, simplifying the logistics path, and reducing the transportation distance and time, the efficiency and response speed of warehouse operations can be improved. The optimized layout plan can effectively reduce the logistics transportation and warehousing costs, including labor costs, equipment operation and maintenance costs. The dynamic potential energy navigation algorithm is applied to the warehouse AGV robot path planning to improve the flexibility and adaptability of path planning, optimize the operating efficiency of AGV, reduce the possibility of collision and detours, and improve the overall efficiency of logistics scheduling. By dynamically adjusting the path to cope with changes in the warehouse environment, the logistics cost can be effectively reduced, the automation and intelligence level of warehouse operations can be improved, and the operational efficiency and response speed of the warehouse system can be enhanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flow chart of a logistics scheduling method based on warehouse space planning according to an embodiment of the present invention;

[0061] Figure 2 It is a structural schematic diagram of a logistics scheduling system based on warehouse space planning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0063] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0064] Figure 1 FIG. 1 is a flow chart of a logistics scheduling method based on warehouse space planning according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0065] S101. Based on the warehouse space and the logistics orders to be dispatched, a first objective function is constructed with the goal of minimizing the total logistics handling cost, a second objective function is constructed with the goal of maximizing the comprehensive correlation, a third objective function is constructed with the goal of maximizing the land area utilization rate, and a comprehensive objective function is constructed based on the first objective function, the second objective function and the third objective function;

[0066] The handling cost is the core part of logistics activities. The total logistics handling cost refers to the sum of all costs incurred in the process of moving goods in logistics production activities. It is related to the distance and weight of goods. In the process of moving goods, labor, machine equipment maintenance and repair costs will be incurred, which will affect the unit cost and thus the total logistics handling cost. Therefore, minimizing the total logistics handling cost is taken as the first objective function.

[0067] In the process of warehouse layout, in order to coordinate the activities of various functional areas and make the operation flow between the functional areas smooth and convenient, the comprehensive correlation between different functional areas is the key factor affecting the overall efficiency of the warehouse center. Therefore, the second objective function is established with the comprehensive correlation coefficient and the adjacency correlation as variables;

[0068] Theoretically, under the premise of minimizing the total logistics handling cost and maximizing the comprehensive correlation, the functional area can be placed randomly. However, the unrestricted and random placement of functional areas will cause a large amount of land waste in the storage center, increase the road width, and thus increase the handling cost. It is also not conducive to the construction and long-term development of smart warehousing. Therefore, maximizing the utilization rate of land area is taken as the third objective function.

[0069] According to the first objective function, the second objective function and the third objective function, the comprehensive objective function is constructed as a weighted sum, which optimizes the total logistics handling cost, comprehensive correlation and land area utilization at the same time. The comprehensive objective function reflects the multi-dimensional requirements of warehouse logistics system optimization, including cost efficiency, operation coordination and space optimization;

[0070] In this embodiment, warehouse management faces multiple challenges, including reducing costs, improving efficiency, and maximizing space utilization. A single objective function cannot fully reflect all needs, and a comprehensive objective function can balance these diverse needs. By adjusting the weight coefficients provided when constructing the comprehensive objective function, the operation of the warehouse can be flexibly optimized according to actual operating strategies and priorities to ensure optimal performance in different stages or situations, thereby promoting the collaborative work of various systems and processes within the warehouse and improving overall operational efficiency.

[0071] In an optional embodiment, with the goal of minimizing the total logistics handling cost, constructing the first objective function includes:

[0072]

[0073] Among them, g represents the total number of cargo batches, x represents the cargo batch index, Ac x Denotes the labor cost of the xth batch of goods, Dc x represents the maintenance and repair cost of the machinery and equipment for the xth batch of goods, d x represents the transportation distance of the xth batch of goods, w x Indicates the weight of the xth batch of goods;

[0074] With the goal of maximizing the comprehensive correlation, the second objective function is constructed including:

[0075]

[0076] Among them, i, j, l and k represent the functional area index, m represents the total number of functional areas, R ij represents the correlation coefficient between functional area i and functional area j, F ij represents the logistics frequency between functional areas i and j, F lk represents the logistics frequency between functional area l and functional area k, T represents the total number of time periods, t represents the time period index, Oijt represents the amount of operation between functional area i and functional area j in time period t, D ij represents the distance between functional area i and functional area j, max(D) represents the maximum distance between all functional areas, A ij represents the adjacency association degree between functional area i and functional area j, ε represents a constant to prevent division by 0, P ij represents the path convenience score between functional area i and functional area j, L ij represents the physical connection score between functional area i and functional area j;

[0077] Taking the maximization of land area utilization as the goal, the third objective function is constructed including:

[0078]

[0079] Among them, A F,i represents the area of ​​functional area i, U F,i represents the land use rate of functional area i, n represents the total number of transportation paths, y represents the transportation path index, and A P,y represents the area of ​​the transport path y, U P,y represents the land use rate of transportation path y, A T represents the total land area, λ represents the adjustment coefficient, r represents the total number of land use constraints, s represents the land use constraint index, V s represents the penalty term of land use constraint s.

[0080] The first objective function aims to minimize the total logistics handling cost by calculating the handling cost of each batch of goods, including labor costs and equipment maintenance costs, and multiplying it by the handling distance and weight of the goods. It combines the total number of batches of goods, the specific cost factors of each batch of goods, and the handling requirements to ensure that the overall handling cost is as low as possible;

[0081] The second objective function focuses on maximizing the comprehensive correlation between different functional areas in the warehouse, which is achieved by evaluating the logistics frequency, operation volume and physical distance between functional areas, taking into account the operation volume within the time period and the adjacency correlation between functional areas, and optimizing the collaborative work efficiency between different parts of the warehouse;

[0082] The third objective function aims to maximize the utilization rate of land area. By comparing the ratio of effectively used land area to total land area and subtracting the penalty caused by not meeting the land use constraints, it integrates the land use efficiency of functional areas and transportation paths, as well as the penalty for unreasonable land use, and encourages more efficient spatial planning and use.

[0083] According to the formula, by minimizing the total logistics handling cost, it is possible to reduce operating costs while maintaining service levels and improve economic benefits; maximizing the comprehensive correlation helps to improve the collaborative work efficiency of different functional areas within the warehouse, shorten processing time, and improve response speed; by maximizing the utilization rate of land area, it promotes more reasonable space planning and use, reduces land waste, and supports sustainable development;

[0084] S102. Set an initial set of layout solutions, calculate the fitness value of each layout solution in the set of layout solutions by solving the comprehensive objective function based on the layout optimization algorithm, perform iterative position update based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold, obtain the optimal spatial layout, and set the warehouse space structure according to the optimal spatial layout;

[0085] The position guidance coefficient specifically refers to a quantitative indicator used to evaluate the guidance or optimization potential of a layout solution relative to the current search environment, reflecting the adaptability or closeness of the current position of a layout solution relative to the optimal solution position. Through this coefficient, the algorithm can determine whether the layout solution should continue to move in the current direction or need to adjust its search direction, providing a quantitative guidance indicator for the layout solutions to help them identify and move towards more promising areas, so as to more effectively approach the global optimal solution;

[0086] The guidance coefficient threshold specifically refers to a preset judgment standard, which is used to determine the position update strategy of the layout scheme, and helps the layout optimization algorithm to determine whether each layout scheme needs to be updated in a gradient descent manner, or whether it should move closer to the global optimal solution;

[0087] Determine the initial spatial layout solution set of the warehouse, which is used as the starting point for optimization;

[0088] Apply the layout optimization algorithm to each layout plan, calculate the fitness value through the comprehensive objective function, where the plan with a high fitness value is closer to the optimization goal; use the position guidance coefficient to evaluate the proximity of each layout plan to the optimization goal, compare the position guidance coefficient of each plan with the preset guidance coefficient threshold, find a better solution, and update it repeatedly until the preset goal is reached to obtain the warehouse space layout plan. According to the obtained optimal space layout plan, actually set and adjust the warehouse space structure to achieve the optimization goal;

[0089] In this embodiment, based on the iterative update mechanism of the position guidance coefficient and the guidance coefficient threshold, the layout plan can dynamically adapt to different optimization objectives and constraints, thereby enhancing the system's adaptability to a changing environment; by optimizing the spatial layout, simplifying the logistics path, and reducing the transportation distance and time, the efficiency and response speed of warehouse operations are improved; the optimized layout plan can effectively reduce logistics transportation and warehousing costs, including labor costs, equipment operation and maintenance costs; and provides a systematic decision support tool to help managers make more scientific and reasonable decisions in warehouse space planning and layout adjustments.

[0090] In an optional embodiment, an initial set of layout solutions is set, and based on a layout optimization algorithm, a fitness value of each layout solution in the set of layout solutions is calculated by solving a comprehensive objective function, and an iterative position update is performed based on a position guidance coefficient and a comparison result with a preset guidance coefficient threshold value, so as to obtain an optimal spatial layout, including:

[0091] Based on the total number of preset layout schemes, an initial layout scheme set is set; based on the layout scheme set, the fitness value of each layout scheme in the layout scheme set is calculated through the comprehensive objective function, and the layout scheme corresponding to the current highest fitness is determined as the benchmark layout scheme;

[0092] Based on the reference layout scheme, determining the position information of other layout schemes in the layout scheme set, and randomly assigning position guidance coefficients to the other layout schemes, and determining the position update mode according to the comparison relationship between the product of the position guidance coefficient and the position information and a preset guidance coefficient threshold;

[0093] According to the position update mode, the position update movement is performed, and the corresponding fitness value is calculated, and according to the preset optimal threshold, a corresponding number of layout schemes are retained to form an updated layout scheme set;

[0094] Iterate the position update until the preset maximum number of iterations is reached, and select the layout scheme corresponding to the maximum fitness value as the optimal spatial layout.

[0095] According to the total number of preset layout schemes, a series of initial warehouse layout schemes are formulated, each of which represents an arrangement of goods, shelves, channels, etc. in the warehouse; the fitness value of each layout scheme is evaluated through a comprehensive objective function, that is, the comprehensive performance of factors such as total logistics handling cost, spatial comprehensive correlation and land area utilization is considered, and the scheme with the highest fitness value is selected as the benchmark layout scheme. For other layout schemes in the set, based on the relative position information with the benchmark layout scheme, the position guidance coefficient is randomly assigned, and the product of the position guidance coefficient and the position information is compared with the preset guidance coefficient threshold to determine the position update method of each scheme. ; According to the determined update method, the layout plan is adjusted to generate a new layout plan, and then the fitness value of the adjusted layout plan is recalculated; a certain number of plans with the highest fitness value are selected from the updated layout plan set, and the corresponding number is limited by the preset preferential threshold, forming a new layout plan set, and the next round of iterative update is carried out. Optionally, in the process of warehouse layout, if a new layout method is found, it can be added to the layout plan set and the next round of iteration is carried out together; until the preset maximum number of iterations is reached, or the improvement of the fitness value reaches a satisfactory level, finally, the layout plan with the largest fitness value is selected as the optimal spatial layout;

[0096] Through simulation and iteration methods, the warehouse layout is dynamically adjusted and optimized to ultimately achieve the goals of minimizing costs, maximizing efficiency and optimizing space. The most reasonable warehouse space layout is designed based on specific warehouse operation requirements and space constraints.

[0097] In this embodiment, by optimizing the warehouse layout, unnecessary logistics transportation distance and time are reduced, thereby directly reducing labor costs and equipment operating costs. This cost optimization has a significant effect on improving the overall profit margin of the enterprise; the optimized layout can ensure the efficient operation of the warehouse, for example, by reducing the storage and retrieval time of goods and optimizing the flow path of goods through reasonable layout, the speed of processing orders and the ability to respond to customer needs are improved; through scientific planning and utilization of available space, the optimal layout plan can maximize the efficiency of land and space use, improve the storage capacity of the warehouse, and avoid space waste; by adopting dynamic iteration and fitness-based optimization mechanism, the layout plan can flexibly adapt to different operational needs and changing market conditions, thereby enhancing the adaptability and flexibility of warehouse management.

[0098] In an optional embodiment, based on the reference layout, determining the position information of other layouts in the initialization layout scheme, and randomly assigning position guidance coefficients to the other layouts, and determining the position update mode according to a comparison relationship between the product of the position guidance coefficient and the position information and a preset guidance coefficient threshold value includes:

[0099] If the product is greater than the guidance coefficient threshold, the position information of other layouts is updated in a gradient descent manner, and the formula is as follows:

[0100] pos(tp+1)=I·cos(2πl)+pos(tp);

[0101] Among them, pos(tp+1) represents the position information at time tp+1, pos(tp) represents the position information at time tp, I represents the initial position of the initial optimal individual, and l represents the gradient descent coefficient;

[0102] If the product is less than or equal to the guiding coefficient threshold, the position information of other layouts is updated toward the direction of the reference layout, and the formula is as follows:

[0103] pos(tp+1)=[pos best (tp)-pos(tp)]·B·a;

[0104] Among them, pos best (tp) represents the global optimal position at time tp, B represents the update scale step, and a represents the swing factor.

[0105] If the product is greater than the guidance coefficient threshold, the position is updated by gradient descent, and the layout scheme will improve the position information according to a periodic adjustment method. During the update process, the position of the initial optimal individual is combined, and a periodic change amount calculated based on the gradient descent coefficient is added to adjust the position of the current layout. Through subtle and directional adjustments, the layout scheme is gradually optimized and approaches a better state;

[0106] When the product is less than or equal to the guidance coefficient threshold, the position of the current layout is adjusted directly to the direction of the benchmark layout solution. This is achieved by calculating the difference between the current position and the global optimal position and multiplying it by a fixed update ratio step size and swing factor. The swing factor provides update randomness and enhances the global search capability. If the current layout solution is close to the optimal solution, a larger step size is used to directly approach the optimal solution, accelerating the convergence of the optimization process.

[0107] In this embodiment, when the product is greater than the threshold of the guiding coefficient, the position update by the gradient descent method can realize dynamic adaptation and fine adjustment of the layout scheme. By using the principle of gradient descent, the layout position is adjusted by a periodic function, which provides a flexible and effective approximation strategy for layout optimization, which is particularly suitable for optimization from a relatively far initial point. When the product is less than or equal to the threshold of the guiding coefficient, the update strategy directly toward the optimal layout direction can accelerate the convergence of the layout scheme to the global optimal solution. By directly adjusting to the known best fitness position, the optimal layout can be quickly approached in the later stage of the optimization process, thereby improving the convergence speed of the algorithm. By introducing the swing factor and the update ratio step in the update strategy, the algorithm can introduce necessary randomness and flexibility while keeping approaching the optimal solution, which helps the algorithm avoid falling into the local optimal solution and improves the probability of finding the global optimal solution. The combined use of these two update strategies achieves a balance between exploration and utilization. The gradient descent method provides a method that helps the algorithm to jump out of the local optimal solution and improves the efficiency of finding the optimal solution.

[0108] S103. Based on the warehouse space structure and the basic parameters of the AGV robot, a warehouse space model is constructed, and the optimal navigation path is generated by analyzing the demand instructions of the logistics order and continuous decision-making of the dynamic potential navigation algorithm to complete the scheduling task of the logistics order;

[0109] The basic parameters of the AGV robot specifically include robot size, load capacity, driving speed, navigation method, battery life, etc. The basic parameters determine the applicability and efficiency of the AGV in a specific warehouse environment;

[0110] The dynamic potential energy navigation algorithm is a path planning method that calculates the target gravity and obstacle repulsion, calculates the rationality of the two forces, dynamically determines the motion state and direction of the AGV robot at each time point, generates a collision-free optimal path, and is suitable for complex and changeable warehouse environments;

[0111] Based on the actual layout of the warehouse and the operating characteristics of the AGV, a digital warehouse space model is constructed; according to the demand instructions of the logistics order, the starting position and target position of the AGV are determined; using the dynamic potential navigation algorithm, continuous decisions are made based on the current environment and AGV parameters to plan the optimal path from the starting point to the target point; the AGV completes the handling of the logistics order according to the planned path, and if there are environmental changes or obstacles, the algorithm will dynamically adjust the path;

[0112] In this embodiment, the dynamic potential energy navigation algorithm is applied to the path planning of the warehouse AGV robot to improve the flexibility and adaptability of the path planning, optimize the operating efficiency of the AGV, reduce the possibility of collisions and detours, and improve the overall efficiency of logistics scheduling; by dynamically adjusting the path to cope with changes in the warehouse environment, it can effectively reduce logistics costs and improve the automation and intelligence level of warehouse operations, thereby enhancing the operational efficiency and response speed of the warehouse system.

[0113] In an optional embodiment, by analyzing the demand instructions of the logistics order, the optimal navigation path is generated through continuous decision-making of the dynamic potential navigation algorithm, including:

[0114] Parse the logistics order, and according to the distribution of the logistics order, obtain the demand instruction corresponding to the logistics order, and determine the order destination of the AGV robot;

[0115] Based on the order destination and the obstacle position, combined with the current position of the AGV robot, the target gravity and obstacle repulsion of the AGV robot at the current position are calculated by constructing an attractive potential field and a repulsive potential field. The motion state of the AGV robot at the current position is obtained by calculating the mechanical resultant force vector. The entire motion process is iterated, and the optimal navigation path is determined based on the motion state of each position.

[0116] The dynamic potential energy navigation algorithm is constructed by analyzing the force of the AGV robot's position and avoiding unreachable problems and accidental collision problems through computational mechanics control;

[0117] In order to ensure that the AGV robot reaches the destination efficiently and avoids obstacles, the dynamic potential energy navigation algorithm optimizes the path by adjusting the attractive potential field and the repulsive potential field. The attractive potential field increases as the distance between the AGV and the target point increases, and vice versa, it decreases to ensure that the AGV reaches the target smoothly. The repulsive potential field increases as the distance between the AGV and the obstacle decreases to avoid collision. The dynamic potential energy navigation algorithm is used to calculate the force conditions of the AGV robot at various locations in the warehouse space, determine the kinematic state of the AGV robot at any position, and then plan the trajectory of the AGV robot. The trajectory of the AGV robot is the optimal path obtained by path planning using the dynamic potential energy navigation algorithm.

[0118] In this embodiment, the process provides accurate path planning for the AGV robot through a dynamic potential energy navigation algorithm, optimizes the navigation path of the AGV robot, reduces detours or stagnation caused by obstacles, improves logistics efficiency, ensures that the AGV robot completes logistics tasks safely and efficiently, enhances the adaptability and flexibility of the system, and improves the level of automation in warehouse operations.

[0119] In an optional embodiment, by constructing an attractive potential field and a repulsive potential field, the target gravitational force and the obstacle repulsive force of the AGV robot at the current position are calculated, and the calculation of the mechanical resultant force vector includes:

[0120] By setting the safety distance of the attractive potential field, reducing the gravity on the far target side and compensating the gravity on the near target side, a gravity calculation formula is constructed, which is as follows:

[0121]

[0122] Where Fg represents gravity, u g represents the attractive potential field constant of the distant target side, d(X,X G ) represents the Euclidean distance between the AGV robot position and the order destination, ds represents the safety distance of the attractive potential field, and v g represents the attractive potential field constant of the near-target side gravity, d(X,X B ) represents the Euclidean distance between the AGV robot position and the obstacle;

[0123] Based on the distance between the AGV robot and the order destination, an adjustment factor is constructed. By setting the maximum influence distance of the repulsive potential field, combining the adjustment factor and the repulsive potential field constant, a repulsive force calculation formula is constructed, and the formula is as follows:

[0124]

[0125] Among them, Fr represents repulsion, u r represents the repulsive potential field constant, de represents the maximum influence distance of the repulsive potential field, Represents the gradient of the distance between the AGV robot and the obstacle.

[0126] In order for the AGV robot to reach the target position smoothly, the gravity and the attractive potential field should increase as the distance between the car and the target point increases. When the car gradually approaches the target point, the gravity should gradually decrease. When the car reaches the target point, the gravity is 0. That is to say, the size of the attractive potential field should be proportional to the positive power of the distance of the AGV robot. In the attractive potential field, the direction in which the potential energy decreases fastest is the negative gradient direction in the attractive potential field. By setting the attractive potential field safety distance, the gravitational function is segmented. The attractive potential field constant can be appropriately increased in the part closer to the target point, which helps to solve the problem of unreachable targets caused by insufficient gravity when approaching the target point. In the part farther from the target point, the attractive potential field constant should be appropriately reduced to avoid collision with obstacles.

[0127] In order to enable the AGV robot to avoid obstacles smoothly, the size of the repulsive potential field needs to increase as the distance between the AGV robot and the obstacle decreases. When the distance between the AGV robot and the obstacle is 0, that is, the repulsive potential energy is infinite when they collide. When the AGV robot gradually moves away from the obstacle, the repulsive potential field gradually decreases. When the distance between the AGV robot and the obstacle increases to the point where there is no possibility of collision, the repulsive potential field is directly set to 0, and this distance is called the maximum influence distance of the repulsive potential field. Similar to the situation in the attractive potential field, the size of the repulsive force of the AGV robot is the negative gradient of the repulsive potential field. In the process of applying the repulsive force, as the AGV robot and the target As the punctuation point approaches, the magnitude of gravity is constantly decreasing. If the magnitude of repulsion is not reduced at this time, it is very easy to cause the repulsion to be greater than the gravity when approaching the target point, and the target point will be unreachable. If the repulsion potential field constant is directly reduced, the gravity will be extremely large at the beginning of path planning, and the AGV robot is likely to collide with obstacles at the far end of the target. In order to solve these two problems, it is necessary to ensure that the gravity of the AGV robot is always slightly greater than the repulsion in the case of non-collision during path planning. Therefore, the Euclidean distance between the current position of the AGV robot and the target point is used as a factor to make the change trends of gravity and repulsion basically consistent, and the repulsion potential field constant is adjusted to make the repulsion smaller than gravity.

[0128] In this embodiment, the navigation capability of the AGV robot in a complex warehouse environment is optimized to ensure that it reaches the destination efficiently and safely; the gravity and repulsion are dynamically adjusted to balance the behavior of the AGV between approaching the target point and avoiding obstacles, avoiding deadlock or too many detours in path planning; more accurate and flexible path planning is achieved, the operating efficiency of the AGV robot is improved, and logistics costs are reduced.

[0129] Figure 2 FIG. 1 is a schematic diagram of the structure of a logistics scheduling system based on warehouse space planning according to an embodiment of the present invention. Figure 2 As shown, the system comprises:

[0130] The first unit is used to construct a first objective function based on the warehouse space and the logistics orders to be scheduled with the goal of minimizing the total logistics handling cost, to construct a second objective function with the goal of maximizing the comprehensive correlation, to construct a third objective function with the goal of maximizing the land area utilization rate, and to construct a comprehensive objective function based on the first objective function, the second objective function and the third objective function;

[0131] The second unit is used to set an initial layout scheme set, calculate the fitness value of each layout scheme in the layout scheme set by solving the comprehensive objective function based on the layout optimization algorithm, perform iterative position update based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold, obtain the optimal spatial layout, and set the warehouse space structure according to the optimal spatial layout;

[0132] The third unit is used to construct a warehouse space model based on the warehouse space structure and the basic parameters of the AGV robot, and to generate an optimal navigation path by parsing the demand instructions of the logistics order and making continuous decisions through the dynamic potential energy navigation algorithm to complete the scheduling task of the logistics order.

[0133] According to a third aspect of the embodiments of the present invention,

[0134] An electronic device is provided, comprising:

[0135] processor;

[0136] a memory for storing processor-executable instructions;

[0137] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0138] According to a fourth aspect of the embodiments of the present invention,

[0139] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0140] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A logistics scheduling method based on warehouse space planning, characterized in that: include: According to the warehouse space and the logistics orders to be dispatched, a first objective function is constructed with the goal of minimizing the total logistics handling cost, a second objective function is constructed with the goal of maximizing the comprehensive correlation, a third objective function is constructed with the goal of maximizing the land area utilization rate, and a comprehensive objective function is constructed based on the first objective function, the second objective function and the third objective function; An initial set of layout solutions is set, and based on the layout optimization algorithm, the fitness value of each layout solution in the set of layout solutions is calculated by solving the comprehensive objective function, and based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold, the position is updated iteratively to obtain the optimal spatial layout, and the warehouse space structure is set according to the optimal spatial layout; Based on the warehouse space structure and the basic parameters of the AGV robot, a warehouse space model is constructed. By analyzing the demand instructions of the logistics order and continuously making decisions through the dynamic potential energy navigation algorithm, the optimal navigation path is generated to complete the scheduling task of the logistics order. Taking the minimization of total logistics handling cost as the goal, the first objective function is constructed including: Among them, g represents the total number of cargo batches, x represents the cargo batch index, Ac x Denotes the labor cost of the xth batch of goods, Dc x represents the maintenance and repair cost of the machinery and equipment for the xth batch of goods, d x represents the transportation distance of the xth batch of goods, w x Indicates the weight of the xth batch of goods; With the goal of maximizing the comprehensive correlation, the second objective function is constructed including: Among them, i, j, l and k represent the functional area index, m represents the total number of functional areas, R ij represents the correlation coefficient between functional area i and functional area j, F ij represents the logistics frequency between functional areas i and j, F lk represents the logistics frequency between functional area l and functional area k, T represents the total number of time periods, t represents the time period index, O ijt represents the amount of operation between functional area i and functional area j in time period t, D ij represents the distance between functional area i and functional area j, max(D) represents the maximum distance between all functional areas, A ij represents the adjacency association degree between functional area i and functional area j, ε represents a constant to prevent division by 0, P ij represents the path convenience score between functional area i and functional area j, L ij represents the physical connection score between functional area i and functional area j; Taking the maximization of land area utilization as the goal, the third objective function is constructed including: Among them, A F,i represents the area of ​​functional area i, U F,i represents the land use rate of functional area i, n represents the total number of transportation paths, y represents the transportation path index, and A P,y represents the area of ​​the transport path y, U P,y represents the land use rate of transportation path y, A T represents the total land area, λ represents the adjustment coefficient, r represents the total number of land use constraints, s represents the land use constraint index, V s represents the penalty term of land use constraint s; An initial set of layout schemes is set, and based on the layout optimization algorithm, the fitness value of each layout scheme in the set of layout schemes is calculated by solving the comprehensive objective function, and the iterative position update is performed based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold value to obtain the optimal spatial layout, including: Based on the total number of preset layout schemes, an initial layout scheme set is set; based on the layout scheme set, the fitness value of each layout scheme in the layout scheme set is calculated through the comprehensive objective function, and the layout scheme corresponding to the current highest fitness is determined as the benchmark layout scheme; Based on the reference layout scheme, determining the position information of other layout schemes in the layout scheme set, and randomly assigning position guidance coefficients to the other layout schemes, and determining the position update mode according to the comparison relationship between the product of the position guidance coefficient and the position information and a preset guidance coefficient threshold; According to the position update mode, the position update movement is performed, and the corresponding fitness value is calculated, and according to the preset optimal threshold, a corresponding number of layout schemes are retained to form an updated layout scheme set; Iterate the position update until the preset maximum number of iterations is reached, and select the layout scheme corresponding to the maximum fitness value as the optimal spatial layout; By analyzing the demand instructions of logistics orders and making continuous decisions through the dynamic potential navigation algorithm, the optimal navigation path is generated, including: Parse the logistics order, and according to the distribution of the logistics order, obtain the demand instruction corresponding to the logistics order, and determine the order destination of the AGV robot; Based on the order destination and the obstacle position, combined with the current position of the AGV robot, the target gravity and obstacle repulsion of the AGV robot at the current position are calculated by constructing an attractive potential field and a repulsive potential field. The motion state of the AGV robot at the current position is obtained by calculating the mechanical resultant force vector. The entire motion process is iterated, and the optimal navigation path is determined based on the motion state of each position.

2. The method according to claim 1, characterized in that Based on the reference layout, determining the position information of other layouts in the initialization layout scheme, and randomly assigning position guidance coefficients to the other layouts, and determining the position update method according to a comparison relationship between the product of the position guidance coefficient and the position information and a preset guidance coefficient threshold, including: If the product is greater than the guidance coefficient threshold, the position information of other layouts is updated in a gradient descent manner, and the formula is as follows: pos(tp+1)=I·cos(2πl)+pos(tp); Among them, pos(tp+1) represents the position information at time tp+1, pos(tp) represents the position information at time tp, I represents the initial position of the initial optimal individual, and l represents the gradient descent coefficient; If the product is less than or equal to the guiding coefficient threshold, the position information of other layouts is updated toward the direction of the reference layout, and the formula is as follows: pos(tp+1)=[pos best (tp)-pos(tp)]·B·a; Among them, pos best (tp) represents the global optimal position at time tp, B represents the update scale step, and a represents the swing factor.

3. The method according to claim 1, characterized in that By constructing an attractive potential field and a repulsive potential field, the target gravitational force and the obstacle repulsive force of the AGV robot at the current position are calculated, and the mechanical resultant force vector calculation includes: By setting the safety distance of the attractive potential field, reducing the gravity on the far target side and compensating the gravity on the near target side, a gravity calculation formula is constructed, which is as follows: Where Fg represents gravity, u g represents the attractive potential field constant of the distant target side, d(X,X G ) represents the Euclidean distance between the AGV robot position and the order destination, ds represents the safety distance of the attractive potential field, and v g represents the attractive potential field constant of the near-target side gravity, d(X,X B ) represents the Euclidean distance between the AGV robot position and the obstacle; Based on the distance between the AGV robot and the order destination, an adjustment factor is constructed. By setting the maximum influence distance of the repulsive potential field, combining the adjustment factor and the repulsive potential field constant, a repulsive force calculation formula is constructed, and the formula is as follows: Among them, Fr represents repulsion, u r represents the repulsive potential field constant, de represents the maximum influence distance of the repulsive potential field, Represents the gradient of the distance between the AGV robot and the obstacle.

4. A logistics scheduling system based on warehouse space planning, used to implement the logistics scheduling method based on warehouse space planning as described in any one of claims 1 to 3, characterized in that: include: The first unit is used to construct a first objective function based on the warehouse space and the logistics orders to be scheduled with the goal of minimizing the total logistics handling cost, to construct a second objective function with the goal of maximizing the comprehensive correlation, to construct a third objective function with the goal of maximizing the land area utilization rate, and to construct a comprehensive objective function based on the first objective function, the second objective function and the third objective function; The second unit is used to set an initial layout scheme set, calculate the fitness value of each layout scheme in the layout scheme set by solving the comprehensive objective function based on the layout optimization algorithm, perform iterative position update based on the position guidance coefficient and the comparison result with the preset guidance coefficient threshold, obtain the optimal spatial layout, and set the warehouse space structure according to the optimal spatial layout; The third unit is used to construct a warehouse space model based on the warehouse space structure and the basic parameters of the AGV robot, and to generate an optimal navigation path by parsing the demand instructions of the logistics order and making continuous decisions through the dynamic potential energy navigation algorithm to complete the scheduling task of the logistics order.

5. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method described in any one of claims 1 to 3.

6. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 3 is implemented.

Images