Four-way shuttle vehicle scheduling method and system based on distributed space-time resource reservation

Abstract

The invention relates to a four-direction shuttle vehicle scheduling method and system based on distributed space-time resource reservation, and the method comprises the steps: generating an initial task allocation scheme based on a dynamic priority model, generating a path plan by a four-direction shuttle vehicle according to an allocated task, and reserving a space-time block sequence to lock path resources, and performing conflict resolution on the reserved space-time block, generating and executing a final execution path, and performing task allocation and path updating by using closed-loop feedback and dynamic re-optimization in the execution process. According to the method and the system, through dynamic task priority allocation, improved path planning algorithm and space-time resource reservation multi-mechanism cooperation, decentralized scheduling, conflict intelligent resolution and efficient path optimization in a three-dimensional library dynamic environment are realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of shuttle vehicle path planning, and particularly relates to a four-way shuttle vehicle scheduling method and system based on distributed space-time resource reservation. BACKGROUND

[0002] Intelligentization and automation have become the core driving force for the transformation and upgrading of manufacturing and logistics industries. Under this background, automated warehouse systems, especially dense warehouse systems with four-way shuttles as the core, can efficiently complete the inbound and outbound, handling and picking operations of goods due to their flexible moving ability in three-dimensional space. They have become the mainstream choice in the industry because of their significant advantages in improving space utilization, operation efficiency and reducing energy consumption.

[0003] However, the existing four-way shuttle scheduling technology, especially in large-scale, high-density and high-dynamic warehouse scenarios, generally has the following defects that need to be solved:

[0004] Traditional scheduling methods mostly use centralized architecture and rely on global path planning algorithms, that is, a central control unit is responsible for task allocation, path planning and conflict management of all shuttle vehicles. The central control unit needs to process multiple vehicle path conflicts in real time, which has high computational complexity and significant system response delay. The task allocation strategy lacks flexibility in dynamic environments and is difficult to adapt to scenarios with frequent changes in warehouse state or emergency task insertion. As the number of vehicles and task complexity increases sharply, the computing power demand of the central control unit grows exponentially, which easily forms a performance bottleneck, leading to system response delay and low decision-making efficiency. Moreover, once the central control unit is down, the entire warehouse system will be paralyzed.

[0005] Therefore, how to design a four-way shuttle scheduling method that can realize decentralized autonomous decision-making, intelligent task allocation, efficient dynamic path planning and reliable conflict resolution has become the key to improving the overall performance of intelligent warehouse systems and breaking through the development bottleneck of the industry. SUMMARY

[0006] One of the purposes of the present application is to at least solve one or more of the above problems existing in the prior art, in other words, one of the purposes of the present application is to provide a four-way shuttle scheduling method and system based on distributed space-time resource reservation that meets one or more of the aforementioned needs.

[0007] In order to achieve the above-mentioned purposes of the application, the following technical solutions are adopted:

[0008] In a first aspect, the present application provides a four-way shuttle scheduling method based on distributed space-time resource reservation, comprising:

[0009] Based on a preset dynamic priority model, task parameters received and real-time vehicle states of the four-way shuttles, task priorities of multiple four-way shuttles are calculated, and an initial task allocation scheme is generated according to the priorities;

[0010] The four-way shuttles generate path planning according to the initial task allocation scheme and a dynamic map of the warehouse environment, and discretize the planned paths into a sequence of space-time blocks containing time and space information, wherein each path planning prebooks space-time blocks in a distributed manner to lock path resources;

[0011] Whether there is a conflict in the space-time blocks prebooked by each four-way shuttle is detected, and if there is a conflict, a preset rule base is applied to resolve the conflict to generate a final execution path without conflict;

[0012] The four-way shuttles execute tasks according to the final execution path;

[0013] During the execution of the tasks, a closed-loop feedback and dynamic re-optimization are used to update the task allocation scheme and the path planning.

[0014] As a preferred embodiment, the conditions for setting priorities in the dynamic priority model include:

[0015] The first condition is task urgency, the second condition is the current power of each four-way shuttle, and the third condition is the distance between each four-way shuttle and the task target point.

[0016] As a further preferred embodiment, the dynamic re-optimization includes:

[0017] During the execution of the tasks, the task conflict rate is monitored, and the weight parameters in the dynamic priority model are dynamically adjusted, if the task conflict rate exceeds a preset conflict rate threshold, the weight of the first condition is increased and the weight of the third condition is decreased.

[0018] As a preferred embodiment, the closed-loop feedback and dynamic re-optimization include:

[0019] An simulated annealing algorithm is used to iteratively optimize the weighted sum of total task time, emergency response time and conflict rate.

[0020] As a preferred embodiment, the method for generating path planning includes:

[0021] A turning penalty coefficient is set for the turning path.

[0022] As a preferred embodiment, the method for generating path planning includes:

[0023] The empty four-way shuttles can pass under the obstacles in the storage locations.

[0024] As a preferred embodiment, the method of generating path planning comprises:

[0025] The storage location of the goods is regarded as a dynamic obstacle, and when the goods are taken out, the corresponding dynamic obstacle is removed;

[0026] When the obstacle of the dynamic map changes, incremental path re-planning is performed.

[0027] As a preferred embodiment, the reservation of the space-time block is verified and stored through a hash chain mechanism;

[0028] Each space-time block generates a unique hash value based on its own information and the hash value of its previous space-time block, forming a continuous hash chain for consistency verification of the resource state.

[0029] As a preferred embodiment, the rule base includes specific rules for resolving hoist resource occupation conflicts;

[0030] The specific rules include:

[0031] Allow the four-way shuttle car with a task urgency higher than a preset urgency threshold to have priority in using the hoist.

[0032] On the other hand, the present application also provides a four-way shuttle car scheduling system based on distributed space-time resource reservation, characterized in that it comprises:

[0033] A scheme generation module, based on a preset dynamic priority model, combines the received task parameters and the real-time vehicle state of the four-way shuttle car to calculate the task priority for multiple four-way shuttle cars, and generates an initial task allocation scheme according to the priority;

[0034] A four-way shuttle car, which generates path planning according to the initial task allocation scheme and the dynamic map of the warehouse environment, and discretizes the planned path into a space-time block sequence containing time and space information, wherein each path planning reserves the space-time block in a distributed manner to lock the path resources;

[0035] A conflict resolution module, which detects whether there is a conflict in the space-time block reserved by each four-way shuttle car, and if there is a conflict, applies a preset rule base to resolve the conflict and generates a final execution path without conflict;

[0036] A control module, which controls the four-way shuttle car to execute the task according to the final execution path;

[0037] A closed-loop feedback and dynamic re-optimization module, which uses closed-loop feedback and dynamic re-optimization to update the task allocation scheme and path planning during task execution.

[0038] Compared with the prior art, the four-way shuttle vehicle scheduling method and system based on distributed space-time resource reservation have the following beneficial effects:

[0039] The method and system generate paths in a distributed architecture, each four-way shuttle vehicle autonomously generates path planning according to the task allocation issued by the central master, and uses space-time block reservation to handle path conflicts, thereby reducing the huge computing pressure brought by the central master to generate all paths and handle conflicts. Further, the data consistency problem is solved by a hash chain, avoiding single point failure, so that the success rate and reliability of conflict detection and resolution are guaranteed.

[0040] Meanwhile, the task allocation uses a multi-dimensional dynamic priority model that integrates task urgency, vehicle power and target point distance, and monitors the system conflict rate through a closed-loop feedback mechanism, dynamically adjusts the weights of each dimension in the priority model, so that the system can focus on efficiency when traffic is smooth, and focus on completing urgent tasks first when traffic is congested, significantly improving the response speed of high-priority tasks and the load balancing degree of vehicles.

[0041] The path planning algorithm of the four-way shuttle vehicle also considers many situations such as obstacle changes, turning, re-planning, and fully utilizes the crossing ability of the four-way shuttle vehicle in the empty state, improving the flexibility and response speed of path planning. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 is a flowchart of the four-way shuttle vehicle scheduling method based on distributed space-time resource reservation of the embodiment of the present application.

[0043] Figure 2 is a rasterization schematic diagram of the multi-layer dynamic map of the embodiment of the present application;

[0044] Figure 3 is a schematic diagram of the path generated by the conventional algorithm;

[0045] Figure 4 is a schematic diagram of the path generated by the path planning algorithm of the embodiment of the present application when empty;

[0046] Figure 5 is a schematic diagram of the path generated by the path planning algorithm of the embodiment of the present application when full. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application.

[0048] The following description provides examples, and is not intended to limit the scope, applicability or examples set forth in the claims. Alterations and further modifications of the described elements are possible without departing from the scope of the inventive concept. Various examples can omit, substitute, or add various procedures or components as appropriate. For instance, the methods described can be performed in an order different than described, and various steps can be added, omitted, or combined. Also, features described with respect to some examples can be combined in other examples.

[0049] Embodiments of the present application provide a four-way shuttle scheduling method based on distributed space-time resource reservation, referring to Figure 1 , the flow includes steps S100-S400.

[0050] S100, based on the preset dynamic priority model, combining the received task parameters and the real-time vehicle state of the four-way shuttle, the task priority of multiple four-way shuttles is calculated, and the initial task allocation scheme is generated according to the priority.

[0051] In step S100, the method of the present application first receives the task list issued from the upper system, such as the warehouse management system, in order to realize the intelligent allocation of the task, the method of the present application does not use the simple FIFO mechanism, but calculates a dynamic priority value for each vehicle and the possible pairing of the task.

[0052] Among them, the priority value in the preset dynamic priority model is calculated according to the following formula:

[0053]

[0054] The priority model includes three conditions for judging priority, that is, E, task urgency in the formula; B, the current power of each four-way shuttle; D, the distance between each four-way shuttle and the task target point. 、 、 Respectively, the weight corresponding to the three conditions.

[0055] Among them, The value range is 1 (lowest) to 5 (highest), which is dynamically set according to the task attribute, and is used to represent the urgency of the task itself.

[0056] ; is the standardized power percentage, a value close to 1 indicates that the vehicle power is sufficient, and a value close to 0 indicates that the vehicle power is low and is not suitable for accepting new tasks, especially long-distance tasks.

[0057] ; is a normalized index based on Manhattan distance calculation, the closer the distance, the higher, used to measure the convenience of task execution, encourage nearby allocation.

[0058] In this embodiment, as a non-limiting initial setting, the weight can be assigned the following initial value This means that in normal circumstances, the most important task is the urgency, followed by the vehicle's endurance, and finally the distance.

[0059] S200, the four-way shuttle car generates a path planning according to the initial task allocation scheme and the dynamic map of the warehouse environment, and discretizes the planning path into a space-time block sequence containing time and space information.

[0060] In order to reduce the computing pressure of the centralized central control, the method of the application adopts the four-way shuttle car as the calculation subject of the path planning, and each four-way shuttle car autonomously generates a travel path for itself according to the task allocated to it in step S100.

[0061] Subsequently, each four-way shuttle car discretizes the path planning into a space-time block sequence containing time and space information, and then each path planning reserves the space-time block in a distributed manner to lock the path resources.

[0062] To achieve this goal, the method of the application first needs to build and maintain a warehouse environment dynamic map accessible to all vehicles, in which the obstacles are dynamically variable. Specifically, a storage location containing goods is defined as a temporary obstacle. When a shuttle car completes an inbound task, a SKU-containing pallet is added to the target location, and an obstacle marker is added to the corresponding location on the grid map in real time. Conversely, when an outbound task is completed, the pallet in a certain location is removed, and the obstacle marker is removed in real time. This mechanism ensures that the map accurately reflects the physical real-time state of the warehouse.

[0063] In some embodiments, as shown in Figure 2 The dynamic map is configured to have a multi-layer structure. Since the four-way car only supports four-way movement, and its path planning is calculated independently in each two-dimensional plane based on Manhattan distance, in order to reduce the complexity of calculation, the layer changing process is not included in the path planning, and the layer changing is realized by taking the elevator, which only needs to add the elevator entrance and exit of the corresponding floor in the path planning.

[0064] When generating the path planning, the four-way shuttle car uses a path planning algorithm that combines its own task target, current position, and obstacle and passable area information obtained from the dynamic map to generate an optimal path.

[0065] In some embodiments of the present application, the path planning algorithm comprises the following characteristics and constraints:

[0066] A turning penalty coefficient is set Since the path planning algorithm adopts Manhattan distance as the search strategy, the cost calculation model thereof causes the movement cost in four directions to be completely the same. This easily generates a ladder-type path containing a large number of 90-degree turns. If the direction is changed too much during the travel of the ladder-type path, the actual running time of the path will become slower. After the turning penalty coefficient is set, the path planning algorithm can consider the direction changing time and the acceleration and deceleration time, try to select a straight and longer route, avoid too many turns, and achieve a higher actual running speed.

[0067] The four-way shuttle vehicle in an empty state is enabled to pass under the storage location obstacle. Specifically, when the four-way shuttle vehicle is empty, the block of the storage location obstacle is equivalent to no obstacle, and thus, in the path planning algorithm of the present application, the passing rule of the obstacle can be passed when the four-way shuttle vehicle is in an empty travel mode.

[0068] When the obstacle of the dynamic map changes, incremental path re-planning is performed. Specifically, the path planning algorithm of the present application does not use the traditional A* algorithm based on static environment search, but uses an improved D* algorithm adapted to dynamic environment update. The algorithm only retains the reverse search part of the A* algorithm. When the obstacle environment changes and the path needs to be re-planned, the shortest path information previously searched can be used to trigger incremental path re-planning, rather than re-planning from the starting point again, thereby greatly reducing the search amount and achieving real-time response to SKU storage state changes.

[0069] According to the different lane environments of the four-way shuttle vehicle, the orientation strategy of a specific position is changed. Specifically, when the path planning algorithm of the present application is executed, due to the blocking of the shuttle vehicle guide rail, the four-way vehicle in the cluster block only performs bidirectional path search (vertical direction passing, transverse direction not passing), and the rest of the lane performs four-way search.

[0070] Figure 3 and Figure 4 is a comparison of the paths generated by the path planning algorithm of the present application and a conventional algorithm, wherein, Figure 3 is a schematic diagram of the path generated by the conventional algorithm, Figure 4 is a schematic diagram of the path generated by the path planning algorithm of the present application.

[0071] Figure 4 and Figure 5 is a comparison of the paths generated by the path planning algorithm of the present application in empty and full states, wherein, Figure 4 is a schematic diagram of the path generated by the path planning algorithm of the present application in an empty state, Figure 5Fig. 1 shows a schematic diagram of a path generated by the path planning algorithm of the present application under full load condition.

[0072] After each four-way shuttle generates its own path planning, the time required for a single trip task can be calculated by kinematic formula. The running time is given by the following formula:

[0073] a. Horizontal movement:

[0074] x-axis movement:

[0075]

[0076] y-axis movement:

[0077]

[0078] b. Vertical movement:

[0079]

[0080] wherein H represents the height of a single bin, L represents the length of a single bin, W represents the width of a single bin, N represents the number of axial paths reaching the maximum speed in a certain single trip task, M represents the total number of axial paths in a certain single trip task, represents the number of inflection points of horizontal movement in a single trip task, represents the time required for reversing during horizontal movement of the four-way shuttle, represents the time required for SKU storage or retrieval.

[0081] In summary, the first four-way shuttle completes the first single trip task in a total time The cumulative running time including horizontal movement and vertical movement is: ,

[0082] wherein represents only storage or retrieval task, represents storage task completed without returning to the storage or retrieval point on the first floor, and subsequent retrieval task is performed immediately in the vertical warehouse.

[0083] Then, the time calculation algorithm for a single trip task is used to calculate the space-time block occupied by the four-way shuttle during the execution of the path, and the occupied space-time block resources are reserved.

[0084] Specifically, the space-time block definition six-tuple can be: block = (coordinate x, coordinate y, layer height z, start time t Start , end time t End , vehicle ID, hash)

[0085] In some embodiments of the present application, the spatiotemporal block has a hash chain verification mechanism: each spatiotemporal block generates a unique hash value based on the SHA-256 algorithm, and the calculation input of the hash value includes not only all the information of the current spatiotemporal block (coordinates, time, vehicle ID, etc.), but also the hash value of the previous spatiotemporal block. In this way, all spatiotemporal blocks are linked into a continuous and indivisible hash chain in a cryptographic sense.

[0086] When the four-way shuttle vehicle reserves the spatiotemporal block, it shares the memory resource table with each other. Specifically, the table can be stored in partition by floor to improve query efficiency. When a vehicle needs to reserve resources, it will publish the hash chain generated by itself to the table, and other vehicles will read the existing reservation information from the table before planning the path, and verify the integrity of these hash chains by recalculating and comparing the hash values. Only the resource occupation information that passes the consistency verification is considered to be trusted. This mechanism ensures that all vehicles can make decisions based on a unified and accurate resource view even without a central coordinator.

[0087] S300, detecting whether the spatiotemporal blocks reserved by each four-way shuttle vehicle conflict, if there is a conflict, applying a preset rule base to resolve the conflict, and generating a final execution path without conflict.

[0088] In some embodiments of the present application, the rule base includes:

[0089] Same path conflict: the rear vehicle reduces the speed to 80% of the speed of the front vehicle, and maintains a minimum safety time interval (Δt ≥ 2 seconds);

[0090] Intersection node conflict: compare task priority, high-priority vehicles pass first, and low-priority vehicles wait or detour;

[0091] Opposite path conflict: high-priority vehicles pass first, and low-priority vehicles detour;

[0092] Elevator resource reservation: usually follow the FIFO queuing order; in further embodiments, an emergency task with an emergency degree greater than or equal to 4 can trigger preemption.

[0093] Specifically, the trigger conditions and solutions of these rules are shown in Table 1.

[0094]

[0095] Table 1

[0096] S400, controlling the four-way shuttle vehicle to perform the task according to the final execution path.

[0097] During the execution of the tasks, the closed-loop feedback and dynamic re-optimization are used to update the task allocation scheme and the path planning.

[0098] After the aforementioned task allocation, path planning and conflict resolution steps, each four-way shuttle vehicle obtains a final path that is conflict-free and can be immediately executed. In this step, each shuttle vehicle moves, picks up or unloads goods according to the final path.

[0099] During the execution of the tasks, the method of the present application uses a closed-loop feedback and dynamic re-optimization mechanism to continuously iterate and optimize the scheduling strategy to adapt to the changing warehouse state and task requirements.

[0100] In some embodiments of the present application, the process of closed-loop feedback and dynamic re-optimization includes:

[0101] An annealing algorithm is used to find a task allocation and execution order scheme with better overall performance, with the weighted sum of total task time, emergency response time and system conflict rate as the optimization target.

[0102] The annealing algorithm can include the following steps:

[0103] Define the optimization target function, which is the weighted sum of total task time, emergency response time and system conflict rate.

[0104] Set the initial solution, preferably the initial task allocation scheme generated in step S100.

[0105] Set the initial temperature and iterate the optimization, the iteration process includes but is not limited to: swapping the execution order of two tasks of the same four-way shuttle vehicle, swapping one task of two four-way shuttle vehicles, and transferring one task to another four-way shuttle vehicle.

[0106] The iteration will continue until the temperature reaches a preset value, or a preset maximum number of iterations is reached.

[0107] Through the execution of the above-mentioned secondary optimization process of the annealing algorithm, the method of the present application can review and adjust the global rationality of the task allocation on a more macro level, thereby generating a final task allocation scheme that performs more balanced and better in multiple core performance indicators, and updating the path planning based on it.

[0108] In some embodiments of the present application, in order to enable the system as a whole to intelligently adapt to different traffic conditions and prevent the overall efficiency from declining due to frequent path conflicts, the dynamic re-optimization also includes a dynamic adjustment mechanism for the weights of the dynamic priority model in step S100.

[0109] Specifically, the method of the present application monitors the task conflict rate in real time during the subsequent execution of the task, and a conflict rate threshold is preset. If the monitored real-time task conflict rate exceeds the threshold, it means that the current warehouse network may be congested. At this time, in order to prioritize tasks that can alleviate bottlenecks or must be completed urgently, step S400 will automatically adjust the weight parameters in the priority model, increase the task urgency weight, and reduce the distance weight.

[0110] Specifically, the task urgency weight may be increased to 0.6, and the distance weight may be reduced to 0.1.

[0111] At the same time, in order to prevent ordinary tasks from being in a starvation state due to a long time without high priority, the weight adjustment mechanism also includes a recovery function. When the task conflict rate decreases and falls below the threshold after adjustment, the increased weight will gradually return to the initial weight value.

[0112] Further, some embodiments also set a phased adjustment for the urgency weight adjustment. When the conflict rate increases by 10%, the urgency weight is increased by 0.05, and when the conflict rate decreases by 10%, the urgency weight is gradually adjusted back to the initial weight.

[0113] After the above adjustment, the influence of the urgency of the task on the priority calculation is significantly enhanced. When the method of the present application assigns a new task to an idle shuttle vehicle, it will prefer to select a task with high urgency, even if the target point of the task is relatively far away.

[0114] The present application also provides a four-way shuttle vehicle scheduling system based on distributed space-time resource reservation, comprising:

[0115] A scheme generation module calculates the task priority for multiple four-way shuttle vehicles based on a preset dynamic priority model, in combination with the received task parameters and the real-time vehicle state of the four-way shuttle vehicles, and generates an initial task allocation scheme according to the priority.

[0116] A four-way shuttle vehicle generates a path plan according to the initial task allocation scheme and a dynamic map of the warehouse environment, and discretizes the planned path into a space-time block sequence containing time and space information, wherein each path plan reserves space-time blocks in a distributed manner to lock path resources.

[0117] A conflict resolution module detects whether there is a conflict in the space-time blocks reserved by each four-way shuttle vehicle. If there is a conflict, the conflict resolution module applies a preset rule base to resolve the conflict and generates a final execution path without conflict.

[0118] A control module controls the four-way shuttle vehicle to execute the task according to the final execution path.

[0119] A closed-loop feedback and dynamic re-optimization module is used to update the task allocation scheme and path planning during the execution of the task by the four-way shuttle vehicle.

[0120] In the system, each four-way shuttle vehicle autonomously generates a path planning according to the task allocation issued by the central master, and uses a time-space block reservation method to handle path conflicts, thereby reducing the huge computing pressure brought by the central master to centrally generate all paths and handle conflicts. Further, the data consistency problem is solved by a hash chain, avoiding single point failure, so that the success rate and reliability of conflict detection and resolution are guaranteed.

[0121] At the same time, a multi-dimensional dynamic priority model of task allocation is used, which integrates task urgency, vehicle power, and distance to target point, and through a closed-loop feedback mechanism to monitor the system conflict rate, dynamically adjust the weight of each dimension in the priority model, so that the system can focus on efficiency when traffic is smooth, and focus on completing urgent tasks first when traffic is congested, significantly improving the response speed of high-priority tasks and the load balancing degree of vehicles.

[0122] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized in electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0123] The above is only an exemplary embodiment of the present disclosure, and cannot limit the scope of the present disclosure. That is, any equivalent changes and modifications made in accordance with the teachings of the present disclosure are still within the scope of the present disclosure. Those skilled in the art will easily think of embodiments of the present disclosure after considering the specification and practicing the disclosure herein. The present application is intended to cover any variations, uses or adaptive changes of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or conventional techniques in the art not disclosed in the present disclosure. The specification and examples are only considered as exemplary, and the scope and spirit of the present disclosure are defined by the claims.

Claims

1. A four-way shuttle vehicle dispatching method based on distributed space-time resource reservation, characterized in that, The method comprises the following steps: Based on a preset dynamic priority model, the task priorities of multiple four-way shuttles are calculated by combining the received task parameters and the real-time vehicle states of the four-way shuttles, and an initial task allocation scheme is generated according to the priorities; The four-way shuttles generate path plans according to the initial task allocation scheme and a dynamic map of the warehouse environment, and discretize the planned paths into a sequence of space-time blocks containing time and space information, wherein each path plan reserves the space-time blocks in a distributed manner to lock path resources; Whether there is a conflict in the space-time blocks reserved by each four-way shuttle is detected, and if there is a conflict, a preset rule base is applied to resolve the conflict to generate a final execution path without conflict; The four-way shuttles execute tasks according to the final execution path; During task execution, a closed-loop feedback and dynamic re-optimization are used to update the task allocation scheme and the path plan.

2. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 1, wherein, The conditions for setting priorities in the dynamic priority model include: A first condition, task urgency; a second condition, current power of each four-way shuttle; and a third condition, distance between each four-way shuttle and a task target point.

3. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 2, wherein, The dynamic re-optimization includes: During task execution, the task conflict rate is monitored, and the weight parameters in the dynamic priority model are dynamically adjusted, if the task conflict rate exceeds a preset conflict rate threshold, the weight of the first condition is increased, and the weight of the third condition is decreased.

4. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 1, wherein, The closed-loop feedback and dynamic re-optimization include: An annealing algorithm is used to iteratively optimize the weighted sum of total task time, emergency response time and conflict rate.

5. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 1, wherein, The method for generating path plans includes: A turning penalty coefficient is set for a turning path.

6. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 1, wherein, The method for generating path plans includes: The four-way shuttles without load can pass under a storage location obstacle.

7. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 1, wherein, The method for generating path plans includes: A storage location with goods is regarded as a dynamic obstacle, and when the goods are taken out, the corresponding dynamic obstacle is removed; When the obstacles of the dynamic map change, incremental path re-planning is performed.

8. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 1, wherein, The reservation of the space-time blocks is verified and stored through a hash chain mechanism; Each space-time block generates a unique hash value based on its own information and the hash value of its previous space-time block to form a continuous hash chain for consistency verification of resource state.

9. The four-way shuttle vehicle dispatching method based on distributed space-time resource reservation of claim 1, wherein, The rule base includes specific rules for resolving elevator resource occupation conflicts; The specific rules include: The four-way shuttles with task urgency higher than a preset urgency threshold are allowed to use the elevator preferentially.

10. A four-way shuttle vehicle dispatching system based on distributed space-time resource reservation, characterized in that, The method comprises the following steps: A scheme generation module calculates the task priorities of multiple four-way shuttles based on a preset dynamic priority model, combines the received task parameters and the real-time vehicle states of the four-way shuttles, and generates an initial task allocation scheme according to the priorities; The four-way shuttles generate path plans according to the initial task allocation scheme and a dynamic map of the warehouse environment, and discretize the planned paths into a sequence of space-time blocks containing time and space information, wherein each path plan reserves the space-time blocks in a distributed manner to lock path resources; Whether there is a conflict in the space-time blocks reserved by each four-way shuttle is detected, and if there is a conflict, a preset rule base is applied to resolve the conflict to generate a final execution path without conflict; The four-way shuttles execute tasks according to the final execution path; During task execution, a closed-loop feedback and dynamic re-optimization are used to update the task allocation scheme and the path plan. The four-way shuttle vehicle generates a path plan according to the initial task allocation scheme and a dynamic map of a warehouse environment, and discretizes the planned path into a time-space block sequence containing time and space information, wherein each path plan prebooks the time-space block in a distributed manner to lock path resources; A conflict resolution module detects whether the time-space blocks prebooked by each four-way shuttle vehicle have conflicts, and if there are conflicts, applies a preset rule base to resolve the conflicts and generates a final execution path without conflicts; A control module controls the four-way shuttle vehicle to execute tasks according to the final execution path; A closed-loop feedback and dynamic re-optimization module updates the task allocation scheme and the path plan using closed-loop feedback and dynamic re-optimization during the execution of the tasks by the four-way shuttle vehicle.

Images