Four-way vehicle scheduling method and system based on dense storage shelf

Abstract

The invention discloses a four-way vehicle scheduling method and system based on dense storage shelves, and the method comprises the following steps: constructing a scene model of a dense storage shelf system, the scene model comprising a three-dimensional dot matrix model composed of a plurality of lattice point locations; based on the scene model, constructing a floor change algorithm model used for scheduling the four-way vehicle to migrate among different floors and a same-floor path planning algorithm model used for scheduling the four-way vehicle to execute tasks in the same floor; the central control platform monitors a task pool and an equipment pool of each floor in real time, and when a task needs to be executed, a four-way vehicle of a proper floor is allocated to the task based on a floor change algorithm model; and based on the same-layer path planning algorithm model, planning an optimal path from the current position to the task target position for the four-way vehicle executing the task, and controlling the four-way vehicle to move along the optimal path to complete the operation. By constructing the layer-changing algorithm model and the same-layer algorithm model and carrying out cooperative scheduling by the central control platform, the defects in the prior art are effectively overcome.

Description

Technical Field

[0001] This invention relates to the field of automated logistics technology, specifically to a four-way vehicle scheduling method and system based on high-density storage racks. It is used for vehicle allocation, route planning, and command control for four-way vehicles operating on high-density storage racks. Background Technology

[0002] With the rapid development of the manufacturing and e-commerce logistics industries, enterprises have placed higher demands on the storage density, inbound and outbound efficiency, and flexible production capabilities of automated storage and retrieval systems (AS / RS). Against this backdrop, the four-way shuttle automated storage and retrieval system (AS / RS) has emerged. The four-way shuttle (hereinafter referred to as "four-way shuttle") is an intelligent handling device capable of freely moving along longitudinal and transverse tracks within the racking plane. In conjunction with a hoist, it can achieve cross-floor operations. Compared to traditional aisle stacker crane AS / RS, it offers significant advantages such as flexible layout, high space utilization, strong system scalability, and relatively low construction costs.

[0003] However, the efficiency of a four-way vehicle system is highly dependent on the intelligence level of its dispatching and control system. In practical applications, existing four-way vehicle dispatching methods still have many shortcomings, which restrict further improvement in the overall efficiency of the system. These shortcomings are mainly reflected in the following aspects: Rigid vehicle allocation and task scheduling strategies: Many existing systems employ simple, static allocation rules, such as fixed vehicle service areas or first-come, first-served. This approach lacks a global perspective and cannot dynamically allocate vehicle resources based on the real-time workload of each floor. This easily leads to uneven workload distribution, where some floors experience a large backlog of tasks due to insufficient vehicles, while vehicles on other floors remain idle, severely impacting system throughput and equipment utilization.

[0004] Path planning algorithms fail to adequately consider equipment characteristics: for four-way vehicles, turning is a relatively time-consuming operation that can cause congestion. Existing path planning algorithms (such as the classic A* algorithm or Dijkstra's algorithm) typically use path length as the sole optimization objective when applied to four-way vehicle scheduling, ignoring the actual impact of the number of turns on execution efficiency. This results in short planned paths that may not be time-optimal due to multiple turns, failing to fully utilize the performance of the four-way vehicles.

[0005] Lack of coordinated dynamic optimization capabilities: Existing solutions often treat layer-change scheduling and intra-layer path planning as two independent problems. This decoupled scheduling approach cannot achieve global optimization at the system level. For example, when deciding whether to dispatch a vehicle to change layers, factors such as path congestion on the target layer, the vehicle's own task queue, and the turning costs in the execution path are not comprehensively considered, leading to scheduling decisions that are locally optimal but globally suboptimal.

[0006] Poor environmental adaptability: In complex warehousing environments, path congestion and the insertion of temporary tasks frequently occur. Many existing systems lack effective dynamic obstacle avoidance and real-time replanning mechanisms. Once anomalies occur, they can easily lead to vehicle waiting, deadlocks, or task execution interruptions, reducing the robustness and operational efficiency of the system.

[0007] Therefore, there is an urgent need in this field for an efficient and intelligent four-way vehicle scheduling method that can comprehensively consider vehicle load, floor task balance, path turning cost, and dynamic environmental changes, in order to solve the above-mentioned technical problems and fully tap the potential of four-way vehicle automated warehouses. Summary of the Invention

[0008] The purpose of this invention is to provide a four-way vehicle scheduling method and system based on dense storage racks to address the aforementioned problems and improve warehouse operation efficiency.

[0009] The technical solution of the present invention is as follows: A four-way vehicle scheduling method based on dense storage racks, executed by a central control platform, includes the following steps: Construct a scenario model for a dense storage racking system. The scenario model includes a three-dimensional dot matrix model consisting of multiple storage locations. Based on the scenario model, a floor-switching algorithm model for scheduling four-way vehicles to migrate between different floors is constructed, as well as a same-floor path planning algorithm model for scheduling four-way vehicles to perform tasks within the same floor. The central control platform monitors the task pool and equipment pool of each floor in real time. When there is a task to be executed, it assigns a four-way vehicle to the appropriate floor based on the floor-switching algorithm model. Based on the same-layer path planning algorithm model, the optimal path from the current position to the target position is planned for the four-way vehicle to perform the task, and the four-way vehicle is controlled to move along the optimal path to complete the task.

[0010] Furthermore, the construction of the layer-switching algorithm model includes: Calculate each floor Expected completion time of the task The calculation formula is: , in, For task pool The number of tasks that have not yet started. For equipment pool The number of devices in; The average execution time of a single task; responding to at least one floor. Meets the preset layer switching trigger conditions, from Select an available four-way vehicle from the lowest floor and dispatch it to the target floor via a floor-changing elevator.

[0011] Furthermore, the layer switching triggering conditions include: Number of devices per floor Expected completion time ; Or the estimated completion time of the floor When the value exceeds the waiting threshold.

[0012] By employing the methods described above, the overall system throughput and operational efficiency are significantly improved. The layer-switching algorithm dynamically calculates the estimated completion time of tasks on each floor and intelligently triggers layer-switching requests, achieving dynamic load balancing for the four-way vehicles across different floors. This effectively avoids the problem of uneven vehicle workload, ensuring that floors with backlogged tasks receive timely vehicle support, thereby drastically shortening the average response and completion times of tasks and enhancing the overall inbound and outbound throughput capacity of the warehousing system.

[0013] Furthermore, the same-level path planning algorithm model is an improved A* algorithm, and its cost function is: , in, From the starting point to the current node The actual cost, To start from the current node The estimated cost to reach the target point This is the time cost coefficient for a single turn. From the starting point to the current node The cumulative number of turns.

[0014] Furthermore, when using the improved A* algorithm for path search, the search range is limited to the four orthogonal neighborhood directions of the current node. When expanding the search nodes, the system determines whether the node is passable based on the current cargo status of the four-way vehicle and the shelf status of the target location, and only passable nodes are added to the search queue.

[0015] By employing the methods described above, the quality of path planning is optimized to achieve true time optimization. In same-level path planning, by introducing a turning cost factor into the A* algorithm, the objective of path planning is optimized from the traditional "shortest distance" to "time optimization." This allows the four-way vehicle to proactively choose paths with fewer turns and shorter actual time when performing tasks, reducing unnecessary start-stop and turning operations. This not only improves the execution speed of a single task but also reduces the mechanical wear of the equipment.

[0016] Furthermore, when the layer-switching algorithm model assigns a suitable floor for a four-way vehicle to a task, if there is an available four-way vehicle on the floor where the task is located, the algorithm model based on the same floor path planning is used to calculate the cost of each available four-way vehicle to reach the task starting point, and the four-way vehicle with the lowest cost is selected to execute the task.

[0017] Furthermore, when the central control platform monitors the task pools and equipment pools on each floor in real time, it also monitors the status and environmental changes of the four-way vehicle in real time, and dynamically replans the optimal path when path blockage or task change is detected.

[0018] This application also includes a four-way vehicle scheduling system based on dense storage racks, used to implement a four-way vehicle scheduling method based on dense storage racks, including: The model building module is used to build scene models, layer-switching algorithm models, and same-layer path planning algorithm models; The central control platform is communicatively connected to multiple four-way vehicles and level-changing elevators in the warehouse. The central control platform includes: The status monitoring unit is used to monitor the task pools and device pools on each floor in real time. The scheduling decision unit is used to make vehicle allocation and route planning decisions based on the layer-switching algorithm model and the same-layer path planning algorithm model. The instruction issuing unit is used to issue movement, picking up, unloading and layer changing instructions to the four-way vehicle and the layer changing elevator.

[0019] By implementing the aforementioned system, the overall utilization efficiency of space and equipment resources is improved. This application uses a global perspective to monitor and schedule the task pool and equipment pool in real time, enabling the limited four-way vehicle resources to be efficiently and flexibly shared and utilized throughout the entire automated storage and retrieval system (including all floors). On the one hand, it minimizes vehicle idle time and increases the utilization rate of individual vehicles; on the other hand, it enhances the coverage and service capabilities of the entire dense storage space, thereby achieving higher space utilization and return on equipment investment.

[0020] This application also includes an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of a four-way vehicle scheduling method based on dense storage racks.

[0021] This application also includes a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of a four-way vehicle scheduling method based on dense storage racks.

[0022] Compared with existing technologies, the advantages of this invention are: 1. Enhanced System Adaptability and Robustness: When performing path search, the same-layer algorithm model dynamically determines the path's feasibility based on the cargo status of the four-way vehicles (empty or loaded) and the shelf status of the points (whether there are goods or not), effectively avoiding path conflicts and deadlocks and ensuring smooth workflow. Simultaneously, the central control platform's real-time monitoring and dynamic path planning mechanism enable the system to cope with unexpected situations such as temporary task insertions and path congestion, demonstrating good environmental adaptability and operational reliability. 2. Achieving Intelligent and Global Optimization of Scheduling Decisions: This application organically integrates floor-changing scheduling with same-floor path planning, with a central control platform making unified decisions. This collaborative optimization mechanism breaks through the limitations of the traditional "each doing its own thing" scheduling approach, comprehensively considering multiple dimensions such as floor load, vehicle location, and path cost in each scheduling decision, thereby achieving system-level global efficiency optimization rather than local optima. Attached Figure Description

[0023] Figure 1 This is a model diagram of the dense shelving scenario in this application.

[0024] Figure 2 This is a dot matrix diagram of a single layer of high-density shelving in this application.

[0025] Figure 3 This is a schematic diagram of the layer replacement model in this application.

[0026] Figure 4 This is a schematic diagram of the flat-floor model of this application.

[0027] Figure 5 This is a schematic diagram illustrating the path search direction for this application.

[0028] Figure 6 This is the control flowchart of the central control platform of this application.

[0029] Figure 7 This is an example diagram of a layer-changing scenario in this application.

[0030] Figure 8 This is a flowchart of the layer replacement operation in this application.

[0031] Figure 9 This is an example diagram of a scene on the same floor as described in this application.

[0032] Figure 10 This is a flowchart of the same-level operations in this application.

[0033] Figure 11 This is a schematic diagram of the path taken by the four-way vehicle to the starting point of the mission.

[0034] Figure 12 This is a schematic diagram of the path taken by the four-way vehicle to the mission endpoint. Detailed Implementation

[0035] It should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0036] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0037] Please see Figure 1-12 A four-way vehicle scheduling method based on dense storage racks includes the following steps: Scenario Model Construction: Build a 3D scene model based on the design drawings. Construct the entire high-density shelving area as a cuboid, defining one corner of the first floor as the origin (this can be selected based on the actual warehouse conditions). Starting from the origin, each storage compartment is a point. For example... Figure 1 As shown, the bitmap is split into layers, with each layer forming a separate bitmap. For example... Figure 2 As shown.

[0038] Building the Algorithm Model: The algorithm model is divided into a layer-switching algorithm model and a same-layer algorithm model. The layer-switching algorithm model is used to calculate the timing, cost, and floor of the layer switch. The same-layer algorithm model is used to select the optimal path and the optimal vehicle, as well as for dynamic programming of obstacle avoidance.

[0039] Layer-changing algorithm model: Layer-changing for four-way vehicles is generally carried out using a layer-changing lifting machine, such as... Figure 3 As shown.

[0040] Each floor has its own task pool. and equipment pool Task pool The number of tasks not yet started is Equipment pool The number of devices in Average execution time of a single task (This can be dynamically calculated and updated based on historical task execution times); Based on the above parameters, the estimated completion time for all tasks in the task pool for this floor is calculated. The formula is as follows: , When the number of equipment on the floor At that time, the estimated completion time .

[0041] Floor task count At that time, the estimated completion time .

[0042] Regarding the estimated completion time Categorize and process: when When the time comes, the highest priority response is given, triggering a layer switching request; when If the response exceeds the waiting threshold, trigger a layer switching request. when At that time, the vehicle is in standby mode, based on other layers. The decision is whether to change the management level.

[0043] After triggering the layer switch, search The vehicle on the floor with the lowest value.

[0044] Same-level algorithm model: Same-level algorithms are derived from path search algorithms, such as... Figure 4 As shown.

[0045] A single-level scene can be transformed into a grid map. Adjacent grids are connected to form a path. Locations with storage attributes are shelves. Grids corresponding to the entrance and exit are entrance and exit points. Four-way vehicles occupy one grid.

[0046] Warehouse operations such as inbound, outbound, and repositioning involve relationships between four-way vehicle locations, shelf locations, and inbound / outbound locations. Ultimately, these relationships are transformed into relationships between three points: the current point, the starting point, and the target point. Finding the optimal path between these three points maximizes warehouse operational efficiency.

[0047] The in-plane path search algorithm is A*, and the formula is as follows: , Since warehouse operations are often complex, the original A* algorithm is insufficient to meet engineering requirements. Therefore, the original algorithm is improved as follows: Since a four-way vehicle can only move in four directions, the waypoint only needs to search for four adjacent points, ignoring the four points at an angle. This reduces the search range and improves efficiency. Figure 5 As shown.

[0048] Due to the four-way nature of vehicles, turning takes a relatively long time; therefore, the cost of turning needs to be considered. Let tturn represent the time for one turn, and fturn represent the number of turns. The improved formula is as follows: , Since the shelving area can be in either a loaded or unloaded state, the four-way vehicle can also be in a loaded or empty state. When the four-way vehicle is empty, it can freely move through the shelving area; when it is loaded, it cannot move through the shelving area. Therefore, when traversing points, it is necessary to determine whether to add the point to the search queue based on its state.

[0049] Central Control Platform: As the central control center, the central control platform scans the task pool and equipment pool in real time. Based on algorithm models, it performs optimal scheduling and control of the equipment to maximize warehouse efficiency. The control flow of the central control platform is as follows: Figure 6 As shown.

[0050] The central control platform first starts the model, loads the task pool and equipment pool, and scans these pools in real time. When a task needs to be performed, the task pool and equipment pool are input as parameters into the layer-change model for calculation. If a layer change is required, a suitable four-way vehicle is found based on the model output. The floor and four-way vehicle location information are input into the same-floor model to calculate the path. Finally, the four-way vehicle is instructed to move, and the calculation is repeated until the vehicle is finally in place. On the other hand, if there are tasks to be performed, it checks if there is a four-way vehicle that meets the operation conditions. If not, it waits for the next calculation. If there is, the task location and four-way vehicle location are input into the same-floor model to calculate the shortest path, and instructions for the four-way vehicle to move, pick up goods, and unload goods are issued. The calculation is repeated until the final operation is completed.

[0051] This application also includes a four-way vehicle scheduling system based on dense storage racks, used to implement a four-way vehicle scheduling method based on dense storage racks, including: The model building module is used to build scene models, layer-switching algorithm models, and same-layer path planning algorithm models; The central control platform is communicatively connected to multiple four-way vehicles and level-changing elevators in the warehouse. The central control platform includes: The status monitoring unit is used to monitor the task pools and device pools on each floor in real time. The scheduling decision unit is used to make vehicle allocation and route planning decisions based on the layer-switching algorithm model and the same-layer path planning algorithm model. The instruction issuing unit is used to issue movement, picking up, unloading and layer changing instructions to the four-way vehicle and the layer changing elevator.

[0052] This application also includes an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of a four-way vehicle scheduling method based on dense storage racks.

[0053] This application also includes a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of a four-way vehicle scheduling method based on dense storage racks.

[0054] The following examples further illustrate this application.

[0055] Layer changing operation: such as Figure 7 As shown, the 6-layer task pool TP6 has two tasks, the 6-layer device pool EP6 has no four-way vehicles, the 1-layer task pool TP1 has one task, and the 1-layer device pool EP1 has two vehicles. In this scenario, the layer switching involves the following steps: Figure 8 As shown.

[0056] Step S1: The task pool at level 1 is The equipment pool is Number of tasks Number of devices The 6-layer task pool is The equipment pool is Number of tasks Number of devices Assuming the average execution time of a single task Assuming the estimated task completion time trigger threshold is 600 seconds; based on the above parameters, calculate the estimated completion time for all tasks at levels 1 and 6: Estimated completion time for the task on level 1 ; Estimated completion time for the 6th floor task ; Step S2: Analyze the calculation results, layer 1 If the time is less than the threshold of 600 seconds, it is determined that no layer change is needed; 6 layers If a layer change is needed, a layer change request is triggered.

[0057] Step S3: After triggering the layer change, search For the vehicle on the floor with the lowest value, in this scenario, select the available four-way vehicle on floor 1.

[0058] Step S4: Select the floor-changing elevator closest to the four-way vehicle, issue a four-way vehicle movement command to drive the four-way vehicle to work, move into the floor-changing elevator, and then issue a floor-changing elevator lifting command to lift the four-way vehicle from the first floor to the sixth floor. After the elevator lifts to the sixth floor, issue a four-way vehicle movement command to exit the elevator, and the floor-changing operation ends.

[0059] Same-level operations: such as Figure 9 As shown, there is a task where goods need to be moved to the black shelf location for storage. In this scenario, the same-floor operation includes the following steps: Figure 10 As shown: Step S1: Retrieve the starting position of the task (19,1); the position of car #1 is (24,16), requiring 2 turns to reach the starting point; the position of car #2 is (14,2), requiring 1 turn to reach the starting point. Assuming the turning cost is 2, substituting it into the algorithm model, the result shows that car #2 has the lowest cost, so car #2 is selected to execute the task. Figure 11 As shown.

[0060] Step S2: Based on the path calculated by the model, issue a movement command for the vehicle. After the vehicle reaches the target, issue a pickup command for the vehicle. Step S3: Read the status of the trolley and determine whether the trolley has completed picking up the goods. If not, issue a trolley movement or picking up instruction based on the current working status of the trolley. If it has been completed, proceed to the next step.

[0061] Step S4: Extract the endpoint position (8,13) and the position of car #1 (19,1). Assuming the turning cost is 2, substitute these values ​​into the algorithm model to calculate the path planning result. Figure 12 As shown.

[0062] Step S5: Based on the path calculated by the model, issue a trolley movement command. After the trolley reaches the target, issue a trolley unloading command.

[0063] Step S6: Read the status of the trolley and determine whether the trolley has completed unloading. If not, issue a trolley movement or unloading instruction based on the current working status of the trolley. If the unloading is completed, end the operation.

[0064] The embodiments described above merely illustrate specific implementation methods of this application, and while the descriptions are detailed and specific, they should not be construed as limiting the scope of protection of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the technical solution of this application, and these modifications and improvements all fall within the scope of protection of this application.

Claims

1. A four-way vehicle scheduling method based on dense storage racks, characterized in that, The method is executed by a central control platform and includes the following steps: Construct a scenario model for a dense storage racking system. The scenario model includes a three-dimensional dot matrix model consisting of multiple storage locations. Based on the scenario model, a floor-switching algorithm model for scheduling four-way vehicles to migrate between different floors is constructed, as well as a same-floor path planning algorithm model for scheduling four-way vehicles to perform tasks within the same floor. The central control platform monitors the task pool and equipment pool of each floor in real time. When there is a task to be executed, it assigns a four-way vehicle to the appropriate floor based on the floor-switching algorithm model. Based on the same-layer path planning algorithm model, the optimal path from the current position to the target position is planned for the four-way vehicle to perform the task, and the four-way vehicle is controlled to move along the optimal path to complete the task.

2. The four-way vehicle scheduling method based on dense storage racks according to claim 1, characterized in that, The construction of the layer-switching algorithm model includes: Calculate each floor Expected completion time of the task The calculation formula is: ， in, For task pool The number of tasks that have not yet started. For equipment pool The number of devices in; The average execution time of a single task; responding to at least one floor. Meets the preset layer switching trigger conditions, from Select an available four-way vehicle from the lowest floor and dispatch it to the target floor via the floor-changing elevator.

3. The four-way vehicle scheduling method based on dense storage racks according to claim 2, characterized in that, The layer switching triggering conditions include: Number of devices per floor Expected completion time ; Or the estimated completion time of the floor When the value exceeds the waiting threshold.

4. The four-way vehicle scheduling method based on dense storage racks according to claim 1, characterized in that, The same-level path planning algorithm model is an improved A* algorithm, and its cost function is: ， in, From the starting point to the current node The actual cost, To start from the current node The estimated cost to reach the target point This is the time cost coefficient for a single turn. From the starting point to the current node The cumulative number of turns.

5. A four-way vehicle scheduling method based on dense storage racks according to claim 4, characterized in that, When using the improved A* algorithm for path search, the search range is limited to the four orthogonal neighborhood directions of the current node. Furthermore, when expanding the search nodes, the system determines whether the node is passable based on the current cargo status of the four-way vehicle and the shelf status of the target location, and only passable nodes are added to the search queue.

6. The four-way vehicle scheduling method based on dense storage racks according to claim 1, characterized in that, When the layer-switching algorithm model assigns a suitable floor for a four-way vehicle to a task, if there is an available four-way vehicle on the floor where the task is located, it prioritizes using the same-floor path planning algorithm model to calculate the cost of each available four-way vehicle reaching the task's starting point, and selects the four-way vehicle with the lowest cost to execute the task.

7. The four-way vehicle scheduling method based on dense storage racks according to claim 1, characterized in that, When the central control platform monitors the task pools and equipment pools on each floor in real time, it also monitors the status and environmental changes of the four-way vehicle in real time, and dynamically replans the optimal path when it detects path blockage or task changes.

8. A four-way vehicle dispatching system based on dense storage racks, characterized in that, A four-way vehicle scheduling method based on dense storage racks as described in any one of claims 1-7 includes: The model building module is used to build scene models, layer-switching algorithm models, and same-layer path planning algorithm models; The central control platform is communicatively connected to multiple four-way vehicles and level-changing elevators in the warehouse. The central control platform includes: The status monitoring unit is used to monitor the task pools and device pools on each floor in real time. The scheduling decision unit is used to make vehicle allocation and route planning decisions based on the layer-switching algorithm model and the same-layer path planning algorithm model. The instruction issuing unit is used to issue movement, picking up, unloading and layer changing instructions to the four-way vehicle and the layer changing elevator.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of a four-way vehicle scheduling method based on dense storage racks as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the steps of a four-way vehicle scheduling method based on dense storage racks as described in any one of claims 1-7.

Images