Warehouse multi-RGV task planning method and device

Abstract

The invention discloses a warehouse multi-RGV task planning method and device, and belongs to the technical field of intelligent warehousing. According to the method, a multi-RGV task planning model with logistics efficiency and energy consumption cost as targets is established, and RGV and elevator task planning codes and algorithms are provided; the optimal task planning scheme with the minimum time cost and the minimum energy consumption cost required by transportation of all distributed cargos is completed through iterative optimization, and each RGV is controlled to execute the cargo transportation task through the optimal task planning scheme, so that the logistics transportation efficiency is improved, and the problems that the task planning of the current intelligent warehouse on the RGV is unreasonable, and the task planning efficiency is poor are solved. The energy consumption is high; and the transfer efficiency is low.

Description

Technical Field

[0001] This invention relates to a method and apparatus for planning multiple RGV tasks in a warehouse, belonging to the field of intelligent warehousing technology. Background Technology

[0002] Smart warehouses have been widely adopted, relying on RGVs to achieve unmanned logistics transfer, which greatly improves transfer efficiency compared with traditional manual handling. The rational scheduling of RGVs is crucial for smart warehouses. However, existing smart warehouses generally control RGVs with optimization algorithms, which have a single search mode and do not take into account whether the RGV is matched with the current task due to energy consumption and task duration. The mismatch between tasks will lead to additional energy consumption and extended task time, ultimately resulting in low logistics and transportation efficiency. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art and provide a warehouse multi-RGV task planning method and device. The method plans tasks based on the minimum energy consumption and time cost of RGVs, thereby improving logistics and transportation efficiency and solving the problem that current smart warehouses have unreasonable task planning for RGVs, which increases energy consumption and task duration and reduces transfer efficiency.

[0004] To solve the above-mentioned technical problems, the present invention is implemented using the following technical solution: This invention provides a warehouse multi-RGV task planning method, comprising: Obtain the storage data of the target warehouse and the location of the goods transportation tasks to be performed on the shelves of the target warehouse; Input the warehousing data and the location of the goods transportation tasks to be performed on the target warehouse shelves into the constructed task planning model to determine the optimal task planning scheme that minimizes the time and energy costs required to complete the transportation of all assigned goods. Based on the optimal task planning scheme, control each RGV to perform cargo transportation tasks.

[0005] Furthermore, the warehousing data includes the placement of goods on each shelf in the target warehouse, the initial position of each RGV, and the initial position of each elevator.

[0006] Furthermore, the construction of the well-constructed task planning model includes: Establish a warehouse scenario model; Based on the warehousing model, with the goal of minimizing the time and energy costs required for RGV to complete all assigned cargo transportation tasks, an objective function and constraints are established. Based on the objective function and constraints, an improved particle swarm optimization algorithm is used to find the optimal particle position, resulting in a task planning model that can output the optimal task planning scheme.

[0007] Furthermore, the establishment of the warehousing scenario model includes: The warehouse model includes a simplified racking model, an RGV model, and an elevator model; The simplified shelf model is expressed by the following formula: ; in: L Indicates that the shelves have a total of L layer, M This indicates that each shelf has M There are [number] goods placement locations, totaling [number] locations. The location for placing goods. Indicates shelf N No. j The location of goods placement; the location of shelves. The coordinates are , Indicates position The number of floors, Indicates position The number of columns; The RGV model is expressed by the following formula: ; in: The shelves are equipped with a total of m RGV, Indicates the first i One RGV, The coordinates are ; The elevator model is represented by the following formula: ; Among them: the elevator for each shelf The quantity is 2. This indicates the elevator on the left side of the shelf. This indicates the elevator on the right side of the shelf.

[0008] Furthermore, based on the warehousing model, with the objective of minimizing the time and energy costs required for RGV to complete all assigned cargo transportation tasks, an objective function and constraints are established, including: The objective function is expressed by the following formula: ; in: This represents a function to find the minimum value; This represents the weighted sum of the total time cost and the energy cost of the RGV in completing all cargo transportation tasks. This represents the total time cost for RGV to complete all cargo transportation tasks. This represents the energy cost of the RGV completing all tasks; Indicates the time cost weight. Indicates the weight of energy consumption cost. ; The constraint is expressed by the following formula: ; ; in: Indicates the first F The number of RGVs performing cargo transportation tasks on each shelf; Indicates the number on the shelf. F The RGV of the layer performs the first d The distance traveled on the track during a cargo transport mission. Indicates the number on the shelf. F The RGV of the layer performs the first d+ The distance traveled on the track during one cargo transport mission represents the total number of cargo transport missions.

[0009] Furthermore, the total time cost for RGV to complete all cargo transportation tasks can be expressed by the following formula: ; in: Indicates elevator Execute the Total runtime before each task; ; lift The time for traveling up and down, Indicates elevator Waiting at the designated shelf level Time, express Entering and exiting the elevator and the time for picking up the goods, Indicates elevator The time spent waiting to pick up goods on the first shelf; Indicates elevator Complete the first The time allotted for each task.

[0010] Furthermore, the energy cost of RGV completing all tasks can be expressed by the following formula: ; in: express Complete the task d Energy consumption; ; Indicates the first d The number of shelves corresponding to each task This indicates the first column of each shelf. This indicates the last column on each shelf. This indicates the energy consumption coefficient of the RGV during operation. express Energy consumption for entering and exiting the elevator.

[0011] Furthermore, the task planning model, which uses an improved particle swarm optimization algorithm to find the optimal particle positions based on the objective function and constraints, and obtains an optimal task planning scheme, includes: The number of shelf positions to be placed is used as the particle dimension, and the number of RGVs is used as the upper limit of particle position and velocity for encoding. The position vector of the encoded particle represents a set of task planning solutions, and each position element in the position vector represents a task scheme. In the preset update rounds, the particle position is continuously updated based on the particle velocity, and the optimal particle position is finally output as the optimal task planning scheme.

[0012] Furthermore, the step of continuously updating the particle position based on the particle velocity in a preset update cycle includes: Update particle positions using the following formula: ; ; in: Indicates the first The next update task is executed. d particle c Location, Indicates the first The next update task is executed. d particle c Location; Indicates the first The next update task is executed. d particle c speed, Indicates the first k The next update task is executed. d particle c speed; Indicates inertial weight, Indicates self-learning factor, Represents social learning factors; , This represents a random number that is uniformly distributed in the interval (0,1). This indicates that the individual extreme value corresponds to the current algebraic particle. c Individual optimal solution This represents the global extremum corresponding to the global optimal solution of the particle swarm in the current generation.

[0013] Another aspect of the present invention provides a warehouse multi-RGV task planning device, comprising: The acquisition module is used to acquire the storage data of the target warehouse and the location of the goods transportation task to be performed on the shelves of the target warehouse; The processing module is used to input the warehousing data and the location of the goods transportation tasks to be performed on the target warehouse shelves into the constructed task planning model, and determine the optimal task planning scheme that minimizes the time cost and energy cost required to complete the transportation of all assigned goods. The execution module is used to control each RGV to perform cargo transportation tasks according to the optimal task planning scheme.

[0014] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: 1. This invention inputs the location of warehousing data and the goods transportation tasks to be executed on the target warehouse shelves into a pre-constructed task planning model, determines the optimal task planning scheme that minimizes the time and energy costs required to complete the transportation of all assigned goods, and controls each RGV to execute goods transportation tasks through the optimal task planning scheme, thereby improving logistics transportation efficiency and solving the problem of unreasonable task planning for RGVs in current smart warehouses, which increases energy consumption and task duration and reduces transfer efficiency.

[0015] 2. This invention uses an improved particle swarm optimization algorithm to encode particles. The number of shelf positions to be placed is used as the particle dimension, and the number of RGVs is used as the upper limit of particle position and velocity. After encoding, the position vector of the particle represents a set of task planning solutions. Each position element in the position vector represents a task scheme. By introducing nonlinear weight coefficients and learning factors, the algorithm is prevented from getting trapped in local optima too early, and finally the global optimum is reached. Attached Figure Description

[0016] Figure 1 This is a flowchart of a warehouse multi-RGV task planning method provided by an embodiment of the present invention; Figure 2 This is a flowchart illustrating the construction process of the task planning model provided in this embodiment of the invention; Figure 3 This is a schematic diagram of a warehousing scenario model provided in an embodiment of the present invention; Figure 4This is a task planning diagram of a warehouse multi-RGV task planning method provided in an embodiment of the present invention; Figure 5 This is a schematic diagram comparing a warehouse multi-RGV task planning method provided in an embodiment of the present invention with existing algorithms. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and should not be used to limit the scope of protection of the present invention. Example

[0018] like Figure 1 As shown, a warehouse multi-RGV task planning method includes: Obtain the storage data of the target warehouse and the location of the goods transportation tasks to be performed on the shelves of the target warehouse; In this embodiment, the warehouse data includes the goods placement position of each shelf in the target warehouse, the initial position of each RGV, and the initial position of each elevator; The warehousing data and the locations of the goods transportation tasks to be performed on the target warehouse shelves are input into the constructed task planning model to determine the optimal task planning scheme that minimizes the time and energy costs required to complete the transportation of all assigned goods. The construction of the constructed task planning model includes: like Figure 2 As shown, a warehouse scenario model is established, which includes a simplified rack model, an RGV model, and an elevator model. The simplified shelf model is expressed by the following formula: ; in: L Indicates that the shelves have a total of L layer, M This indicates that each shelf has M There are [number] goods placement locations, totaling [number] locations. The location for placing goods. Indicates shelf N No. j The location of goods placement; the location of shelves. The coordinates are , Indicates position The number of floors, Indicates position The number of columns; The RGV model is expressed by the following formula: ; in: The shelves are equipped with a total of m RGV, Indicates the firsti One RGV, The coordinates are ; The elevator model is represented by the following formula: ; Among them: the elevator for each shelf The quantity is 2. This indicates the elevator on the left side of the shelf. This indicates the elevator on the right side of the shelf.

[0019] Based on the warehousing model, with the goal of minimizing the time and energy costs required for RGV to complete all assigned cargo transportation tasks, an objective function and constraints are established. The objective function is expressed by the following formula: ; in: This represents a function to find the minimum value; This represents the weighted sum of the total time cost and the energy cost of the RGV in completing all cargo transportation tasks. This represents the total time cost for RGV to complete all cargo transportation tasks. This represents the energy cost of the RGV completing all tasks; Indicates the time cost weight. Indicates the weight of energy consumption cost. ; The total time cost for RGV to complete all cargo transportation tasks is expressed by the following formula: ; in: Indicates elevator Execute the Total runtime before each task ; lift The time for traveling up and down, Indicates elevator Waiting on the designated shelf level Time, express Entering and exiting the elevator and the time of pickup, Indicates elevator The time spent waiting to pick up goods on the first shelf; Indicates elevator Complete the first The time required for each task; The energy cost of RGV completing all tasks can be expressed by the following formula: ; express Complete the task d energy consumption Indicates the total number of cargo transportation tasks; ; Indicates the first d The number of shelves corresponding to each task This indicates the first column of each shelf. This indicates the last column on each shelf. This indicates the energy consumption coefficient of the RGV during operation. express Energy consumption for entering and exiting the elevator; The shelving tracks are bidirectional, and each shelf location requires a pallet. However, if a pallet is already in front of the RGV when it performs a task, and the RGV cannot overcome this obstacle, the RGV cannot perform the delivery task. For example, on the first floor, the RGV travels from left to right and needs to deliver to location 6, but a pallet is already placed at location 4. Therefore, the RGV cannot reach location 6 to deliver the goods. The locations after location 4 can only be delivered from the other end using another RGV. The constraint is expressed by the following formula: ; ; in: Indicates the first F The number of RGVs performing cargo transportation tasks on each shelf; Indicates the number on the shelf. F The RGV of the layer performs the first d The distance traveled on the track during a cargo transport mission. Indicates the number on the shelf. F The RGV of the layer performs the first d+ The distance traveled on the track during a single cargo transport mission.

[0020] Based on the objective function and constraints, an improved particle swarm optimization algorithm is used to find the optimal particle positions, resulting in a task planning model that can output the optimal task planning scheme. Specifically: The number of shelf positions to be placed is used as the particle dimension, and the number of RGVs is used as the upper limit of particle position and velocity for encoding. The encoded particle position vector represents a set of task planning solutions, where each position element in the position vector represents a task plan, including: The particle's position range is [1, m The speed range is []. -m , m ]; No. k During the next update: The position vector of the particle is represented as Positional elements This represents a task plan; For example , It is a decimal. , indicating the first k The next update will include the task. d Assigned to and Go and execute; The velocity vector is represented as ; In a preset update cycle, the particle position is continuously updated based on the particle velocity, and the optimal particle position is finally output. This optimal particle position is used as the optimal task planning scheme, including: Update particle positions using the following formula: ; ; in: Indicates the first The next update task is executed. d particle c Location, Indicates the first The next update task is executed. d particle c Location; Indicates the first The next update task is executed. d particle c speed, Indicates the first k The next update task is executed. d particle c speed; Indicates inertial weight, Indicates self-learning factor, Represents social learning factors; , This represents a random number that is uniformly distributed in the interval (0,1). This indicates that the individual extreme value corresponds to the current algebraic particle. c Individual optimal solution This represents the global optimum solution for the current generation of the particle swarm corresponding to the global extremum; Based on the optimal task planning scheme, control each RGV to perform cargo transportation tasks.

[0021] In this embodiment, the warehouse scenario model has 3 shelves, with 7 positions on each shelf, 4 RGVs and 2 elevators; Initially, four RGVs were randomly placed on the shelf, each with a full charge. like Figure 3 As shown, the simplified shelf model has 3 layers, 7 positions per layer, for a total of 21 positions, represented as follows: The first layer is numbered 1-7, the second layer is numbered 8-14, and the third layer is numbered 15-21; The shelf is equipped with 4 RGVs, indicating that... ,set up m =4; There are elevators on both the left and right sides. This indicates two elevators, one on the left and one on the right. Initially, the RGV stops at its initial positions on both sides of each floor. Each floor has two initial positions, one near the left and one near the right elevator. Figure 3 Positions 1, 7, 8, 14, 15 and 21 are shown.

[0022] The initial random position coordinates of the four RGVs are shown in Table 1, and the task position coordinates are shown in Table 2; the initial population size is 100, and the number of algorithm update iterations is 300.

[0023] like Figure 4 As shown, simulation calculations were performed on the four RGVs in Table 1 and all task positions in Table 2. The optimal fitness value of the proposed method (IPSO) in this embodiment was 63.43. The optimal task planning scheme was obtained after 82 updates and iterations. The optimal task planning scheme is shown in Tables 3 and 4. The total time taken was 0.239422 seconds.

[0024] Table 1 Initial Random Positions of RGV

[0025] Table 2 Task Location Coordinates

[0026] Table 3 RGV Task Allocation Results and Number of Tasks Executed

[0027] Table 4. Task Allocation Results and Number of Tasks Performed for the Elevator

[0028] To verify the superiority of the method proposed in this embodiment, the method is compared with the existing PSO algorithm and PO-PSO algorithm in the task planning problem: like Figure 5 As shown, the fitness value of the method proposed in this embodiment is smaller than that of the PSO and PO-PSO algorithms, and the number of iterations required to find the optimal solution by the method proposed in this embodiment is also less than that of the PSO and PO-PSO algorithms.

[0029] Comparison of multiple tasks and multiple RGVs A comparison was made between different task sizes: Scale 1: 4 RGVs, 21 missions; Scale 2: 5 RGVs, 35 tasks; Scale 3: 7 RGVs, 63 tasks; The initial population size was set to 100, and the number of iterations was set to 300. The proposed method, the PO-PSO algorithm, and the PSO algorithm were used to solve the task planning problems of different scales. The algorithm was run 25 times for each scale of problem. The optimal fitness value, worst fitness value, average fitness value, maximum number of iterations to converge to the optimal solution, minimum number of iterations, average number of iterations to converge to the optimal solution, and average running time of the algorithm were recorded. The experimental results are shown in Tables 5 and 6.

[0030] Tables 5 and 6 present simulation analyses of three different scales of task planning problems from seven aspects, respectively. As can be seen from Table 5, when using the method proposed in this example to plan tasks for the system under the same task size, the total cost of all RGVs to complete all tasks, i.e., the fitness value, is lower than that of the PO-PSO and traditional PSO algorithms. From the optimal fitness value, worst fitness value, and average fitness value for different task sizes, it can be seen that the optimal and worst fitness values ​​of the method proposed in this example are not much different, and the average fitness value is also low, indicating that the IPSO algorithm is more stable in finding the optimal solution for task planning.

[0031] As can be seen from Table 6, under the same number of RGVs and task load, the maximum and minimum number of iterations at convergence of the PSO and PO-PSO algorithms differ significantly, and the average number of iterations is also relatively large; while the method proposed in this example performs more stably and excellently in these aspects. As the scale of the task increases, the computational load of the algorithm also increases. However, the method proposed in this example can achieve convergence faster than the PO-PSO and PSO algorithms, and the number of iterations at convergence is greatly reduced, thus reducing the computational load. As can be seen from Table 5, while the number of convergences is reduced, the fitness value is also lower, that is, the total cost is reduced, thus balancing efficiency and quality.

[0032] Table 5 Comparison of fitness values ​​of different algorithms for all tasks

[0033] Table 6 Comparison of average number of iterations and average iteration time at convergence of different algorithms

[0034] Example 2 A warehouse multi-RGV task planning device, comprising: The acquisition module is used to acquire the storage data of the target warehouse and the location of the goods transportation task to be performed on the shelves of the target warehouse; The processing module is used to input the warehousing data and the location of the goods transportation tasks to be performed on the target warehouse shelves into the constructed task planning model, and determine the optimal task planning scheme that minimizes the time cost and energy cost required to complete the transportation of all assigned goods. The execution module is used to control each RGV to perform cargo transportation tasks according to the optimal task planning scheme.

[0035] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage CD-ROMs, optical storage, etc.) containing computer-usable program code.

[0036] This application is described with reference to flowchart illustrations of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each step in the flowchart can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the processes. Figure 1 One or more processes or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0037] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 The function specified in one or more processes.

[0038] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 Steps of a specified function in one or more processes.

[0039] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims. All of these forms are within the protection scope of the present invention.

Claims

1. A warehouse multi-RGV task planning method, characterized in that, include: Obtain the storage data of the target warehouse and the location of the goods transportation tasks to be performed on the shelves of the target warehouse; Input the warehousing data and the location of the goods transportation tasks to be performed on the target warehouse shelves into the constructed task planning model to determine the optimal task planning scheme that minimizes the time and energy costs required to complete the transportation of all assigned goods. Based on the optimal task planning scheme, control each RGV to perform cargo transportation tasks.

2. The warehouse multi-RGV task planning method according to claim 1, characterized in that, The warehousing data includes the placement of goods on each shelf in the target warehouse, the initial position of each RGV, and the initial position of each elevator.

3. The warehouse multi-RGV task planning method according to claim 1, characterized in that, The construction of the pre-built task planning model includes: Establish a warehouse scenario model; Based on the warehousing model, with the goal of minimizing the time and energy costs required for RGV to complete all assigned cargo transportation tasks, an objective function and constraints are established. Based on the objective function and constraints, an improved particle swarm optimization algorithm is used to find the optimal particle position, resulting in a task planning model that can output the optimal task planning scheme.

4. The warehouse multi-RGV task planning method according to claim 3, characterized in that, The establishment of the warehousing scenario model includes: The warehouse model includes a simplified racking model, an RGV model, and an elevator model; The simplified shelf model is expressed by the following formula: ； in: L Indicates that the shelves have a total of L layer, M This indicates that each shelf has M There are [number] goods placement locations, totaling [number] locations. The placement of goods Indicates shelf N No. j The location of goods placement; the location of shelves. The coordinates are , Indicates position The number of floors, Indicates position number of columns The RGV model is expressed by the following formula: ； in: The shelves are equipped with a total of m RGV, Indicates the first i One RGV, The coordinates are ; The elevator model is represented by the following formula: ； Among them: the elevator for each shelf The quantity is 2. This indicates the elevator on the left side of the shelf. This indicates the elevator on the right side of the shelf.

5. The warehouse multi-RGV task planning method according to claim 4, characterized in that, Based on the warehousing model, with the objective of minimizing the time and energy costs required for RGV to complete all assigned cargo transportation tasks, an objective function and constraints are established, including: The objective function is expressed by the following formula: ； in: This represents a function to find the minimum value; This represents the weighted sum of the total time cost and the energy cost of the RGV in completing all cargo transportation tasks. This represents the total time cost for RGV to complete all cargo transportation tasks. This represents the energy cost of the RGV completing all tasks; Indicates the time cost weight. Indicates the weight of energy consumption cost. ; The constraint is expressed by the following formula: ； ； in: Indicates the first F The number of RGVs performing cargo transportation tasks on each shelf; Indicates the number on the shelf. F The RGV of the layer performs the first d The distance traveled on the track during a cargo transport mission. Indicates the number on the shelf. F The RGV of the layer performs the first d+ The distance traveled on the track during one cargo transport mission. This indicates the total number of cargo transportation tasks.

6. The warehouse multi-RGV task planning method according to claim 5, characterized in that, The total time cost for RGV to complete all cargo transportation tasks is expressed by the following formula: ； in: Indicates elevator Execute the Total runtime before each task; ； lift The time for traveling up and down, Indicates elevator Waiting at the designated shelf level Time, express Entering and exiting the elevator and the time for picking up the goods, Indicates elevator The time spent waiting to pick up goods on the first shelf; Indicates elevator Complete the first The time allotted for each task.

7. The warehouse multi-RGV task planning method according to claim 5, characterized in that, The energy cost of RGV completing all tasks can be expressed by the following formula: ； in: express Complete the task d Energy consumption; ； Indicates the first d The number of shelves corresponding to each task This indicates the first column of each shelf. This indicates the last column on each shelf. This indicates the energy consumption coefficient of the RGV during operation. express Energy consumption for entering and exiting the elevator.

8. The warehouse multi-RGV task planning method according to claim 3, characterized in that, The task planning model, based on the objective function and constraints, uses an improved particle swarm optimization algorithm to find the optimal particle positions, resulting in an optimal task planning scheme. This model includes: The number of shelf positions to be placed is used as the particle dimension, and the number of RGVs is used as the upper limit of particle position and velocity for encoding. The position vector of the encoded particle represents a set of task planning solutions, and each position element in the position vector represents a task scheme. In the preset update rounds, the particle position is continuously updated based on the particle velocity, and the optimal particle position is finally output as the optimal task planning scheme.

9. The warehouse multi-RGV task planning method according to claim 8, characterized in that, The step of continuously updating particle positions based on particle velocity in a preset update cycle includes: Update particle positions using the following formula: ； ； in: Indicates the first The next update execution task d particle c Location, Indicates the first The next update execution task d particle c Location; Indicates the first The next update execution task d particle c speed, Indicates the first k The next update execution task d particle c speed; Indicates inertial weight, Indicates self-learning factor, Represents social learning factors; , This represents a random number that is uniformly distributed in the interval (0,1). This indicates that the individual extreme value corresponds to the current algebraic particle. c Individual optimal solution This represents the global extremum corresponding to the global optimal solution of the particle swarm in the current generation.

10. A warehouse multi-RGV task planning device, characterized in that, include: The acquisition module is used to acquire the storage data of the target warehouse and the location of the goods transportation task to be performed on the shelves of the target warehouse; The processing module is used to input the warehousing data and the location of the goods transportation tasks to be performed on the target warehouse shelves into the constructed task planning model, and determine the optimal task planning scheme that minimizes the time cost and energy cost required to complete the transportation of all assigned goods. The execution module is used to control each RGV to perform cargo transportation tasks according to the optimal task planning scheme.

Images