Stacking-based distributed scheduling system and path planning method thereof

Abstract

The invention relates to the technical field of stacking, in particular to a stacking-based distributed scheduling system and a path planning method thereof, comprising a distributed scheduling architecture, an autonomous task allocation module, a cooperative work module and a load balancing module, the distributed scheduling architecture comprises a task management layer capable of dynamically adjusting a task allocation strategy according to real-time order data and a scheduling layer composed of a plurality of scheduling nodes, and each scheduling node can monitor the state and task progress of the stacker in real time and dynamically adjust task allocation and path planning according to a monitoring result. The stackers execute according to task allocation and path planning instructions of a scheduling layer, the autonomous task allocation module configured to each stacker can independently decide task paths and execution sequences according to task priorities, stacker positions and current states, so that the stackers can quickly respond to task changes, and dependence on a central scheduling system is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of stacking, and more specifically, to a distributed scheduling system based on stacking and a path planning method thereof. Background Art

[0002] The receipt and shipment of goods in the vertical warehouse are achieved through stacking equipment. The stacker is an automated equipment used to stack goods vertically. It can efficiently transport and stack goods to save space and improve logistics efficiency.

[0003] During operation, the stacker outputs execution instructions according to the control machine. After receiving the task, the stacker plans the path to prepare for execution. The stacker moves to the target cargo location along the specified path, lifts the goods to the specified height, and accurately places them on the predetermined cargo location. After the stacker completes the operation, the warehouse management system will update the inventory information in real time, record the storage location, time and other information of the goods, and ensure the accuracy of the inventory data. During the entire receiving and shipping process, the system will monitor the operating status of the stacker in real time to ensure the accuracy and safety of the operation. Once a fault or abnormal situation occurs, the system will promptly issue an alarm and take corresponding measures. In addition, for safety, the stacker is equipped with anti-collision sensors and emergency stop devices to ensure that collisions or other dangerous situations are avoided during operation.

[0004] However, the existing path planning cannot adapt to the dynamically changing warehouse environment in real time, such as sudden order changes or temporarily occupied channels, which may cause the stacker to take the wrong path and reduce efficiency. In order to solve this problem, the current method of changing tracks is used for regulation. For example, the method and system for multi-lane operation of a stacker disclosed in announcement number CN 113501242 A obtains the operation status and task coordinates of the stacker, creates an equipment map according to the operation status of the stacker, creates a task map according to the task coordinates, and mobilizes the stacker to process tasks based on the equipment map, task map and task strategy; when the task source position, target position and stacker position are all in the same lane, the stacker in the lane is mobilized to process the task, otherwise the stacker is mobilized to change tracks, and the task is processed after the track change is completed. Considering the priority, distance and time comprehensively, the task weight value is calculated, and the task with the highest weight value is given priority.

[0005] Another example is announcement number CN115344050A, which discloses a stacker path planning method based on an improved clustering algorithm. The method first maps the task nodes into the time domain, proposes a time distance scale (Length of time, Lt) indicator, and applies the clustering idea based on Lt. Then, in response to the problem of possible allocation conflicts in the clustering results, an iterative elimination strategy is proposed, and finally the stacker path planning based on the clustering idea of ​​the time distance scale is realized.

[0006] The two existing technologies mentioned above can flexibly respond to environmental changes through track-changing strategies, while the latter can optimize path selection through time-domain clustering.

[0007] Although the track-changing strategy can flexibly respond to environmental changes, the track-changing operation itself increases the complexity of the system. Frequent track-changing may make system scheduling complicated, especially when multiple stacking cranes are operating at the same time, which may cause track contention, scheduling conflicts and other problems, thereby reducing overall efficiency.

[0008] Although the time-domain clustering algorithm can optimize the path in a static environment, the response speed and flexibility of the algorithm may be insufficient when faced with real-time dynamic changes (such as sudden increase in orders, temporary closure of channels, etc.). Especially when tasks change frequently, the recalculation of the clustering algorithm may cause the real-time performance of the system to deteriorate.

[0009] In summary, there is a need for a path planning algorithm that can respond to changes in the warehouse environment in real time to reduce path planning failures caused by order changes or channel occupancy. Introducing a dynamic path adjustment mechanism enables stackers to quickly adapt to environmental changes, avoid unnecessary trips, and improve operating efficiency. Developing a distributed scheduling system allows multiple stackers to make independent decisions and work together, reducing the burden of central scheduling and improving the flexibility and scalability of the system. Summary of the invention

[0010] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a distributed scheduling system and path planning method for stacking that can efficiently manage and coordinate the task execution of multiple stackers, realize path optimization and task allocation in a dynamic environment, and improve the overall efficiency and reliability of warehousing operations.

[0011] To achieve the above object, the present invention provides the following technical solutions: A distributed scheduling system based on stacking equipment includes the following modules: Distributed scheduling architecture: including task management layer, scheduling layer and execution layer, The task management layer can dynamically adjust the task allocation strategy based on real-time order data and warehouse environment, and distribute the adjusted tasks to the scheduling layer; The scheduling layer consists of several scheduling nodes. Each scheduling node communicates with the stacker of the execution layer through a real-time communication protocol to coordinate task execution and optimize path planning to avoid path conflicts and resource contention. It can also monitor the status of the stacker and task progress in real time, and dynamically adjust task allocation and path planning based on the monitoring results. The execution layer includes the actual stacker operation. The stacker performs operations such as material handling, stacking, and picking up goods according to the task assignment and path planning instructions of the scheduling layer. During the task execution, the operation status and task progress can be fed back to the scheduling layer in real time. The scheduling layer optimizes and adjusts the task based on the feedback information. Autonomous task allocation module: configured in each stacker, used to independently decide the task path and execution order according to the priority, location and current status of the task; Collaborative work module: adopts real-time communication protocol to achieve low-latency information sharing and task collaboration between stackers, avoiding path conflicts and resource contention; Load balancing module: monitors the workload of each stacker in real time, dynamically adjusts task allocation, and ensures load balance among stackers.

[0012] The present invention is further configured as follows: the autonomous task allocation module includes task path planning and task execution sequence management, and realizes task coordination and resource sharing among multiple stackers through a collaborative work module to ensure the coordination and consistency of global task execution. When the autonomous task allocation module is interrupted in communication or the collaborative work module is unavailable, it can make independent decisions and continue to execute the current task. When the communication is restored, the collaborative work module will resynchronize the task status and coordinate the adjustment of the task path and execution order.

[0013] The present invention is further configured as follows: the collaborative work module ensures that multiple stackers effectively cooperate in the same task space to avoid path conflicts and resource contention based on path conflict detection, priority management, resource locking mechanism and dynamic task reallocation.

[0014] The present invention is further configured as follows: the load balancing module includes a redundancy mechanism, and when a certain stacker fails, the standby stacker can automatically take over the task to ensure the continuous operation of the system.

[0015] The present invention is further configured as follows: the path conflict detection comprises: After planning the path, the stacker shares the path information with other stackers through a real-time communication protocol. Each stacker makes a path prediction based on the shared path information and automatically adjusts the path or speed to avoid conflicts. Priority management includes: when multiple stackers need to share the same channel, the system dynamically allocates access priorities based on the urgency of the task and the location of the stacker. The stackers with lower priorities automatically adjust their paths or wait to ensure priority execution of critical tasks. The resource locking mechanism includes: before entering the shared resource area, the stacker requests to lock the resource through real-time communication, and after receiving the lock notification, other stackers avoid entering the locked resource area until the lock is released; The dynamic task redistribution includes: when a stacker cannot continue to perform the current task due to task change or failure, the system reallocates the task to other stackers through real-time communication. After receiving the new task, the reallocated stacker immediately adjusts the path and executes the new task to ensure the timely completion of the task.

[0016] A path planning method based on stacking includes the following steps: S1. Initialize algorithm-related parameters and data, including the operating status of the stacker, warehouse environment data, order information, etc., to establish a basic data model for path planning; S2, collects the operation data of the stacker and the warehouse environment change data in real time, and transmits these data to the path planning module; S3, when a path obstruction or task change is detected, the dynamic path adjustment mechanism is triggered, and an incremental path planning algorithm is used to replan only the affected path segments; S4. Combine machine learning algorithms to predict future order changes or channel occupancy, and perform route optimization and task allocation in advance; S5. Output the optimized path planning result and update the execution instructions of the stacker in real time.

[0017] The present invention is further configured as follows: the dynamic path adjustment mechanism includes: Based on the real-time environmental data, the new optimal path is quickly calculated through the incremental A* algorithm; During the path adjustment process, priority is given to the urgency of the task, the current status of the stacker crane, and the minimum range of path changes to reduce job interruption time.

[0018] The present invention is further configured as follows: the prediction model adopts deep learning and is trained based on historical order data, warehouse operation records, etc. to improve the prediction accuracy of future order fluctuations and environmental changes.

[0019] The present invention is further configured as follows: the stacker is provided with a GPS positioning system, an IMU (inertial measurement unit), and a power monitoring sensor, which collects the operating status data of the stacker in real time, and transmits the data to the path planning module after being processed by the on-board computing unit.

[0020] The present invention is further configured as follows: the collected data must be preliminarily processed locally before being transmitted to the path planning module through a real-time communication protocol, including noise filtering, data fusion and format conversion, and some of the data will be stored in a database for subsequent analysis and training of machine learning algorithms.

[0021] Compared with the shortcomings of the prior art, the beneficial effects of the present invention are: The autonomous task allocation module configured in each stacker can independently decide the task path and execution order according to the task priority, stacker location and current status. This independence enhances the flexibility of the system, enables the stacker to respond quickly to task changes, reduces dependence on the central scheduling system, and ensures efficient completion of tasks. When the path is blocked or the task changes, the incremental path planning algorithm can quickly adjust the path, reduce the interruption time of the operation, and ensure the continuity and efficiency of the task in a dynamic environment.

[0022] Combined with machine learning algorithms, the system can predict future order changes or channel occupancy, and optimize routes and assign tasks in advance. By analyzing historical data and real-time environmental changes, the system can more accurately predict future situations and optimize scheduling in advance, thereby improving the accuracy of task execution. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram of the distributed scheduling system module of the present invention; Figure 2 This is a schematic diagram of the path planning method steps of the present invention; Figure 3 Detailed steps of the path planning method of the present invention are shown in FIG. DETAILED DESCRIPTION

[0024] Reference Figures 1 to 3 The embodiments of the present invention are further described.

[0025] A distributed scheduling system based on stacking equipment includes a distributed scheduling architecture, an autonomous task allocation module, a collaborative work module and a load balancing module.

[0026] The distributed scheduling architecture includes the task management layer, the scheduling layer and the execution layer.

[0027] The task management layer, as the central control module of the system, can dynamically adjust the task allocation strategy based on the real-time order data and warehouse environment information, so as to flexibly respond to situations such as changes in order volume, warehouse layout adjustments or temporary occupation of channels. The optimized task allocation strategy will be quickly distributed to the scheduling layer to ensure that each task can be assigned to the most suitable stacker at the most appropriate time, maximizing the efficiency and resource utilization of the system.

[0028] The scheduling layer is composed of multiple distributed scheduling nodes. (In the scheduling system, a scheduling node is an independent unit responsible for managing and allocating tasks. Each node can be regarded as a small control center, which is responsible for the scheduling and management of specific tasks within a certain range.) Through the distributed design of these nodes, the system can improve parallel processing capabilities, reduce the risk of single point failures, and provide a good foundation for system expansion.

[0029] Each dispatching node communicates with the stacker at the execution layer through a real-time communication protocol. Based on the real-time monitoring data, the dispatching node can independently plan the path and optimize the stacker's driving path in real time to avoid path conflicts and resource contention. If it is detected that a path is occupied or another stacker is in use, the dispatching node can adjust the path immediately and select an alternative.

[0030] The scheduling node continuously monitors the stacker's operating status and task progress, including parameters such as position, speed, load, and power. If the stacker's status changes (such as low power or encountering an obstacle), the scheduling node can respond immediately and adjust the task or path.

[0031] The execution layer includes the actual stacker operation. The stacker performs operations such as material handling, stacking, and picking according to the task assignments and path planning instructions of the scheduling layer. During the execution of the task, it can provide real-time feedback on the operation status and task progress to the scheduling layer. The scheduling layer optimizes and adjusts the task based on the feedback information.

[0032] Each stacker is equipped with a GPS positioning system, IMU (inertial measurement unit), and power monitoring sensor to collect the stacker's operating status data in real time, which is then processed by the on-board computing unit and transmitted to the path planning module.

[0033] Autonomous task allocation module: It is configured in each stacker and is used to independently decide the task path and execution order according to the priority, location and current status of the task.

[0034] The autonomous task allocation module includes a task path planning module and a task execution sequence management module, and realizes task collaboration and resource sharing among multiple stackers through a collaborative work module, ensuring the coordination and consistency of global task execution.

[0035] When receiving multiple tasks, the autonomous task allocation module will automatically arrange the execution order of the tasks according to the urgency, importance and location of the stacker. High-priority tasks will be processed first to ensure that critical tasks are not delayed.

[0036] Task execution sequence management not only involves the arrangement of tasks, but also includes how to handle new tasks or task changes during task execution. The module can flexibly adjust the task sequence to optimize the efficiency and accuracy of task completion.

[0037] The autonomous task allocation module selects the shortest or safest path based on the current status of the stacker crane, task priority, and warehouse layout to avoid obstacles, reduce travel time, and ensure that the task can be completed on time.

[0038] During the path planning process, the module will take into account the current load of the stacker, the health status of the equipment, and the traffic conditions in the warehouse. By dynamically adjusting the path, the stacker can adapt to the real-time changing environment and improve the reliability and efficiency of task execution.

[0039] When the system is disconnected from communication or the collaborative module is unavailable, the module can rely on local task path planning and execution sequence management functions to make independent decisions and continue to execute the current task. This capability ensures the continuity and reliability of the system in extreme situations.

[0040] When communication is restored or the collaborative work module is available again, the autonomous task allocation module will immediately resynchronize the task status with other stackers and system modules. Through the collaborative work module, the task progress, path information and resource usage of all stackers will be re-updated to ensure the coordination of each stacker in the system. After resynchronization, the collaborative work module will coordinate and adjust the task path and execution order, correct the path deviation or task conflict that may occur due to communication interruption, and restore the normal operation of the system.

[0041] Collaborative work module: adopts real-time communication protocol (Wi-Fi, Zigbee, or 5G network) to ensure that stackers can quickly exchange task status, path information and resource occupancy, and realize low-latency information sharing and task collaboration between stackers. The collaborative work module ensures that multiple stackers can effectively cooperate in the same task space based on path conflict detection, priority management, resource locking mechanism and dynamic task redistribution to avoid path conflicts and resource contention. For example, when multiple stackers need to use the same channel, the collaborative work module will coordinate the passage order of each stacker to ensure the optimal utilization of channel resources and avoid task delays caused by contention.

[0042] The collaborative work module also supports resource sharing between stackers, such as shared charging stations or specific storage areas. When a resource is occupied, the collaborative work module will notify other stackers to adjust their tasks or wait to avoid resource conflicts.

[0043] Path conflict detection includes: after planning the path, the stacker will share the path information with other stackers through the real-time communication protocol. Each stacker will make path predictions based on the shared path information and automatically adjust the path or speed to avoid conflicts. If a possible path conflict is detected, the collaborative work module will instruct the relevant stackers to automatically adjust the path or speed to avoid the conflict. This adjustment may include selecting an alternative path, slowing down and waiting, or speeding up to pass, so as to ensure the smooth execution of the task.

[0044] Priority management includes: when multiple stackers need to share the same channel, the system dynamically assigns access priorities based on the urgency of the task and the location of the stacker. For stackers with lower priorities, the collaborative work module will instruct them to automatically adjust the path or wait until the high-priority task is completed. This mechanism avoids resource contention and task delays, ensuring that all tasks in the system can be executed efficiently as planned.

[0045] The resource locking mechanism includes: when the stacker needs to use shared resources (such as charging stations or specific storage areas), the collaborative work module will request to lock the resource through a real-time communication protocol before the stacker enters the resource area. After receiving the resource locking notification, other stackers avoid entering the locked resource area until the lock is released. This resource locking mechanism effectively avoids the situation where multiple stackers occupy the same resource at the same time, reducing task conflicts and delays caused by resource contention. At the same time, through the priority management of resource locking, the collaborative work module can optimize the utilization efficiency of resources and ensure that tasks are completed as planned.

[0046] The dynamic task redistribution includes: when a stacker cannot continue to perform the current task due to task changes (such as task cancellation, priority adjustment) or equipment failure, the collaborative work module will immediately reallocate the task to other available stackers through the real-time communication protocol. After receiving the task, the newly assigned stacker can quickly adjust its own task path and execution order to ensure that the task is completed in time. The dynamic task redistribution mechanism ensures the task continuity of the system in the event of unexpected situations and avoids interruption of task execution. Through rapid task redistribution and path adjustment, the system can resume normal operation in the shortest time.

[0047] Load balancing module: collects the workload data of each stacker in real time. These data include the number of tasks currently performed by the stacker, the complexity of the tasks, the path length, the running speed, the load weight and the energy consumption status of the stacker. Through the real-time collection of these data, the load balancing module can fully understand the current working status of each stacker, thus providing an accurate basis for subsequent task allocation and adjustment.

[0048] When the load on some stackers is too high and other stackers are idle or lightly loaded, the load balancing module will transfer some tasks from the high-load stackers to the low-load stackers, thereby achieving a reasonable distribution of tasks.

[0049] When new task demands suddenly increase, the module will quickly allocate tasks based on the current status of each stacker to avoid overloading or idleness of the stacker.

[0050] The load balancing module includes a redundancy mechanism. When a stacker fails or cannot continue to perform tasks, the system will immediately start the redundancy mechanism. The backup stacker will automatically take over the unfinished tasks of the failed stacker to ensure that tasks can be executed continuously and the operation of the entire system will not be affected by the failure of a single device. The triggering of the redundancy mechanism is automatic and does not require human intervention. The system can respond as soon as the failure is detected and start the backup stacker to take over the task.

[0051] A path planning method based on stacking includes the following steps: The first step is to initialize the algorithm related parameters and data. Basic data model for path planning: Before path planning begins, relevant parameters and data need to be initialized to establish the basic data model for path planning. This includes the operating status of the stacker: such as the current position of the stacker, load conditions, battery power, and equipment health status; warehouse environmental data: including the layout of the warehouse, accessible paths, the status of the storage area (whether it is occupied), and environmental conditions such as the temperature and humidity of the warehouse; order information: obtaining current and future order data, including order priority, required execution time, destination location, etc.; these data form the basis of path planning and are used to guide subsequent path calculation and optimization processes.

[0052] Step 2: Collect the operation data of the stacker and the warehouse environment change data in real time Real-time data collection and transmission: When the stacker is performing tasks, the system will collect its operating data and warehouse environment change data in real time. These data include: the stacker's instant location, operating speed, path status; the location of obstacles in the warehouse, channel occupancy, temperature and humidity and other environmental parameters; After data collection is completed, the system transmits the data to the path planning module through a real-time communication protocol (such as Wi-Fi, Zigbee or 5G network) to ensure that the path planning can be adjusted based on the latest data.

[0053] Local data preprocessing: Before data is transmitted to the path planning module, it will first be processed locally, including: Noise filtering: remove errors or interference data that may exist in the sensor acquisition process to ensure data accuracy; Data fusion: fuse data from multiple sensors to provide more complete and accurate information on the status of the stacker and the environment; Format conversion: unify the data formats output by different sensors so that the path planning module can process them efficiently; Data storage: some processed data will be stored in the database for subsequent analysis and training of machine learning algorithms. By accumulating historical data, the system can continuously optimize and improve the accuracy of path planning.

[0054] The data are then transmitted to the path planning module. The collected data must be processed locally before being transmitted to the path planning module via a real-time communication protocol, including noise filtering, data fusion, and format conversion. Some of the data will be stored in a database for subsequent analysis and training of machine learning algorithms.

[0055] Step 3: Dynamic Path Adjustment Mechanism Incremental path planning algorithm: When the system detects that the path is blocked or the task requirements have changed, the dynamic path adjustment mechanism is immediately triggered. This mechanism uses an incremental A* algorithm to replan the affected path segments without recalculating the entire path. This approach can significantly reduce the computing resources and time required for path adjustment.

[0056] Priority factors for dynamic path adjustment: Task urgency: Prioritize high-priority tasks to be completed on time, and even adjust the paths of low-priority tasks when necessary; Current state of the stacker: Consider the stacker's load, battery charge, operating status, etc. to ensure that path planning adapts to the stacker's actual capabilities; Minimum range of path change: Minimize adjustments to existing paths to reduce job interruption time and the impact on stacker operations, thereby improving the continuity of task execution.

[0057] Step 4: Combine machine learning algorithms for prediction and optimization Machine learning prediction model: The system combines machine learning algorithms to predict future order changes or channel occupancy by analyzing historical order data, warehouse operation records, etc. The prediction model is trained using deep learning methods, and the prediction accuracy of future order fluctuations and environmental changes is improved by continuously updating and optimizing the model.

[0058] Optimize the path and task allocation in advance: Based on the prediction results, the system can optimize the path in advance and adjust the task allocation strategy. This allows the stacker to be prepared before the actual demand occurs, ensuring the optimal state of path and resource utilization when the task is executed.

[0059] Step 5: Output the optimized path planning results Output and execution of path planning results: After completing the path planning and optimization, the system will output the final path planning results and update the execution instructions of the stacker in real time. The stacker will operate according to the latest path instructions to ensure that the task is completed efficiently as planned.

[0060] During the task execution, the system will continuously monitor the operating status of the stacker crane and trigger path adjustment or optimization again when necessary to cope with dynamic changes in the environment.

[0061] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Common changes and substitutions made by those skilled in the art within the scope of the technical solution of the present invention should be included in the protection scope of the present invention.

Claims

1. A distributed scheduling system based on stacking equipment, characterized in that: Includes the following modules: Distributed scheduling architecture: including task management layer, scheduling layer and execution layer, The task management layer can dynamically adjust the task allocation strategy based on real-time order data and warehouse environment, and distribute the adjusted tasks to the scheduling layer; The scheduling layer consists of several scheduling nodes. Each scheduling node communicates with the stacker of the execution layer through a real-time communication protocol to coordinate task execution and optimize path planning to avoid path conflicts and resource contention. It can also monitor the status of the stacker and task progress in real time, and dynamically adjust task allocation and path planning based on the monitoring results. The execution layer includes the actual stacker operation. The stacker performs material handling, stacking, and picking operations according to the task assignment and path planning instructions of the scheduling layer. During the execution of the task, the operation status and task progress can be fed back to the scheduling layer in real time. The scheduling layer optimizes and adjusts the task based on the feedback information. Autonomous task allocation module: configured in each stacker, used to independently decide the task path and execution order according to the priority, location and current status of the task; Collaborative work module: adopts real-time communication protocol to achieve low-latency information sharing and task collaboration between stackers, avoiding path conflicts and resource contention; Load balancing module: monitors the workload of each stacker in real time, dynamically adjusts task allocation, and ensures load balance among stackers.

2. The distributed scheduling system based on stacking as claimed in claim 1, characterized in that: The autonomous task allocation module includes task path planning and task execution sequence management, and realizes task coordination and resource sharing among multiple stackers through the collaborative work module to ensure the coordination and consistency of global task execution. When the autonomous task allocation module is interrupted in communication or the collaborative work module is unavailable, it can make independent decisions and continue to execute the current task. When the communication is restored, the collaborative work module will resynchronize the task status and coordinate the adjustment of the task path and execution order.

3. The distributed scheduling system based on stacking as claimed in claim 2, characterized in that: The collaborative work module ensures that multiple stackers can cooperate effectively in the same task space to avoid path conflicts and resource contention based on path conflict detection, priority management, resource locking mechanism and dynamic task redistribution.

4. The distributed scheduling system based on stacking as claimed in claim 3, characterized in that: The load balancing module includes a redundancy mechanism. When a stacker fails, the standby stacker can automatically take over the task to ensure the continuous operation of the system.

5. The distributed scheduling system based on stacking as claimed in claim 4, characterized in that: The path conflict detection comprises: After planning the path, the stacker shares the path information with other stackers through a real-time communication protocol. Each stacker makes a path prediction based on the shared path information and automatically adjusts the path or speed to avoid conflicts. Priority management includes: when multiple stackers need to share the same channel, the system dynamically allocates access priorities based on the urgency of the task and the location of the stacker. The stackers with lower priorities automatically adjust their paths or wait to ensure priority execution of critical tasks. The resource locking mechanism includes: before entering the shared resource area, the stacker requests to lock the resource through real-time communication, and after receiving the lock notification, other stackers avoid entering the locked resource area until the lock is released; The dynamic task redistribution includes: when a stacker cannot continue to perform the current task due to task change or failure, the system reallocates the task to other stackers through real-time communication. After receiving the new task, the reallocated stacker immediately adjusts the path and executes the new task to ensure the timely completion of the task.

6. A path planning method based on stacking, characterized in that: The following steps are involved: S1. Initialize algorithm-related parameters and data, including the operating status of the stacker, warehouse environment data, order information, etc., to establish a basic data model for path planning; S2, collects the operation data of the stacker and the warehouse environment change data in real time, and transmits these data to the path planning module; S3, when a path obstruction or task change is detected, the dynamic path adjustment mechanism is triggered, and an incremental path planning algorithm is used to replan only the affected path segments; S4. Combine machine learning algorithms to predict future order changes or channel occupancy, and perform route optimization and task allocation in advance; S5. Output the optimized path planning result and update the execution instructions of the stacker in real time.

7. The stacking-based path planning method according to claim 4, characterized in that: The dynamic path adjustment mechanism includes: Based on the real-time environmental data, the new optimal path is quickly calculated through the incremental A* algorithm; During the path adjustment process, priority is given to the urgency of the task, the current status of the stacker crane, and the minimum range of path changes to reduce job interruption time.

8. The stacking-based path planning method according to claim 7, characterized in that: The prediction model uses deep learning and is trained based on historical order data, warehouse operation records, etc. to improve the prediction accuracy of future order fluctuations and environmental changes.

9. The stacking-based path planning method according to claim 6, characterized in that: The stacker is equipped with a GPS positioning system, an IMU (inertial measurement unit), and a power monitoring sensor to collect the stacker's operating status data in real time, and transmit the data to the path planning module after being processed by the on-board computing unit.

10. The stacking-based path planning method according to claim 9, characterized in that: The collected data must be processed locally before being transmitted to the path planning module via a real-time communication protocol, including noise filtering, data fusion, and format conversion. Some of the data will be stored in a database for subsequent analysis and training of machine learning algorithms.

Images