Intelligent control method and system for storage equipment

Abstract

The invention relates to the technical field of storage equipment management, in particular to an intelligent control method and system for storage equipment, and aims to solve the problems of task allocation, path optimization and equipment collaboration in a complex storage scene. According to the method, a storage equipment function model comprising an upper-layer frame model and a lower-layer equipment model is constructed, task allocation and path planning are optimized by using a neural network model in combination with task requirements, and a storage equipment control instruction is generated. In the task execution process, task priorities and path planning are adjusted in real time to adapt to dynamic environment changes, and it is ensured that tasks are completed efficiently. According to the method, a task distribution diagram and a path planning scheme are further optimized through a dynamic task diagram and a dynamic energy field, the problems of equipment congestion, too high energy consumption and task delay are effectively solved, the collaboration efficiency and the resource utilization rate of storage equipment are improved, and the method is suitable for various complex storage scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of storage equipment management, and in particular to an intelligent control method and system for storage equipment. Background Art

[0002] With the development of intelligent warehousing systems, the automation and intelligent control of warehousing equipment has become the key to improving warehousing efficiency. However, existing technologies have obvious deficiencies in task allocation and path planning, especially in complex scenarios with multi-device collaboration and dynamic environmental changes. Problems such as low task allocation efficiency, path congestion, high equipment energy consumption, and task delays are prone to occur. In addition, traditional methods usually rely on static rules or simple optimization algorithms, which are difficult to adapt to dynamic task requirements and real-time environmental changes, resulting in low utilization of warehousing equipment resources, and the flexibility and intelligence level of the system need to be improved. Therefore, there is an urgent need for a more efficient and intelligent control method to achieve multi-device task collaboration and dynamic path optimization, and to improve the overall operating efficiency of the warehousing system.

[0003] For example, the Chinese patent application with publication number CN118941212A discloses an intelligent warehousing equipment and control system and method. It includes a WMS warehouse management module, an equipment control module connected to the WMS warehouse management module, an intelligent identification module, a security monitoring module, and a personnel management module; through automated management, the automated storage, retrieval and transportation of goods are realized, which greatly reduces manual intervention and improves the efficiency of warehousing operations. The automated and intelligent management methods reduce the dependence on manual labor, thereby reducing labor costs, and the automated equipment control module is used to realize the handling of goods, and the goods, equipment and environment are monitored in real time, which improves the safety of warehousing management; and the inventory management is optimized, the data of inventory goods is obtained in real time, and the efficient management and optimization of inventory are realized, the inventory level and cost are reduced, and the capital turnover efficiency is improved.

[0004] The above patent applications all have the problem raised by this background technology: it is difficult to achieve efficient task allocation and path optimization of storage equipment in a complex dynamic environment, resulting in low resource utilization, task delays and high equipment energy consumption. To solve the above problems, this application designs an intelligent control method and system for storage equipment. Summary of the invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent control method and system for storage equipment in view of the deficiencies of the prior art, aiming to solve the problems of task allocation, path optimization and equipment coordination in complex storage scenarios. The method constructs a storage equipment function model including an upper-level framework model and a lower-level equipment model, combines task requirements, and uses a neural network model to optimize task allocation and path planning, and generates storage equipment control instructions. During the task execution process, the task priority and path planning are adjusted in real time to adapt to dynamic environmental changes to ensure efficient completion of tasks. The method further optimizes the task distribution map and path planning scheme through a dynamic task graph and a dynamic energy field, effectively solving the problems of equipment congestion, excessive energy consumption and task delays, improving the coordination efficiency and resource utilization of storage equipment, and is suitable for a variety of complex storage scenarios.

[0006] To achieve the above object, the present invention provides the following technical solutions: An intelligent control method for storage equipment, the method comprising: Constructing a storage equipment function model according to the storage equipment information, wherein the storage equipment function model includes an upper framework model and a lower equipment model; Extract task requirements from the task queue; Constructing a neural network model, training the task requirements and the storage equipment function model through the neural network model, and outputting storage equipment control instructions; According to the storage equipment control instructions, the storage equipment is controlled to collaboratively execute tasks. During the task execution process, the task and path planning are adjusted according to real-time environmental data.

[0007] The construction of the storage equipment function model includes: Collect multi-source information related to storage equipment in the storage area; Building an upper framework model according to the types of storage equipment, the upper framework model includes categories of dynamic equipment, static equipment and interactive equipment and functional collaboration relationships between categories; Constructing a lower-layer equipment model based on multi-source information of each storage equipment, wherein the lower-layer equipment model includes a sub-model of a single storage equipment, and the sub-model includes equipment function parameters, equipment real-time status, and task capability characteristics; The upper-level framework model also includes a task scheduling center, which is configured with the neural network model and is used to interact with the lower-level equipment model, obtain status information and task requirements of the lower-level equipment, and generate storage equipment control instructions based on the status information and task requirements.

[0008] The neural network model comprises: The task optimization layer is used to optimize the task allocation scheme of the storage equipment according to the upper framework model, under the constraint condition of satisfying the priority of the storage tasks, with the goal of maximizing the efficiency of task allocation, and output a task distribution map; The path optimization layer is used to optimize the path plan of the dynamic equipment according to the lower-layer equipment model, under the constraints of the equipment operation status and the storage environment, with the goal of minimizing the path consumption, and output the path planning plan; The task execution layer is used to update the adaptability of the task distribution map and the path planning solution through iterative calculation, and output storage equipment control instructions.

[0009] The task requirements include task type, task priority and completion time limit, and the optimization of the task allocation scheme for storage equipment includes: Constructing a dynamic task graph and an allocation objective function, and generating an initial task allocation scheme according to the edge weights of the dynamic task graph and the allocation objective function, wherein the dynamic task graph is composed of task nodes and device nodes, and the edge weights are calculated according to task requirements, state information, and environmental state; Verify the adaptability of the initial task allocation plan, identify failed tasks that do not meet the adaptation conditions, adjust the edge weights of the dynamic task graph according to the task allocation relationship of the failed tasks, and regenerate the task allocation plan; The regenerated task allocation scheme is iteratively verified until all task allocation relationships meet the adaptation conditions and the task distribution graph is output.

[0010] The optimization of the path scheme of the dynamic device includes: According to the real-time status of dynamic equipment, the target task location and the storage environment information, a dynamic energy field including equipment nodes and path nodes is constructed, wherein the dynamic energy field is characterized by the energy value of each path node, and the energy value is calculated as follows: , in, Represents the energy value from dynamic device n to path node m, represents the path length factor, Indicates the remaining completion time of dynamic device n, and represents the adjustment parameter, represents the real-time congestion factor of path node m, represents the logarithmic function with base 2, Indicates the number of devices currently existing on path node m. represents the environmental complexity factor, represents the energy consumption of dynamic device n at path node m, Indicates the current remaining power of dynamic device n; Traversing low energy value paths according to the dynamic energy field, generating an initial path planning scheme with the goal of minimizing path consumption; According to the initial path planning scheme, identifying whether there is a conflicting path for the dynamic device, wherein the conflicting path means that multiple devices select the same path node in the same time period, or the congestion risk in the path overlap area exceeds a preset threshold; According to the task priority of the tasks corresponding to the dynamic devices, the initial path planning scheme is adjusted based on the principle that high-priority devices first occupy the path nodes with low energy values; According to the adjusted initial path planning scheme, the path consumption of each dynamic device is recalculated, the path planning scheme is updated according to the path consumption, the adaptability of the adjusted path is verified, and if all dynamic devices meet the task priority and energy consumption optimization goals, the path planning scheme is output.

[0011] The adjusting the initial path planning scheme includes: Delay the time planning of the path of low-priority devices, set waiting points in the path conflict area, and allow high-priority devices to pass first; When the waiting time exceeds the set threshold or the path node is unavailable due to congestion, the path of the low-priority device is reallocated and the suboptimal low-energy value path node is selected; When the task completion deadline of a low-priority device is approaching or the reallocated path affects the task completion deadline, the path priority of the low-priority device is increased and path node resources are released for the low-priority device to pass.

[0012] The task execution layer includes: Optimize the allocation relationship between task nodes and device nodes to meet the needs of high-priority tasks to the greatest extent possible while meeting the optimization goal of consistent task completion time and task priority; Recalculate the energy values ​​of path nodes. When path conflicts, node congestion, and insufficient device energy are detected, replan the path and select low-energy path nodes. The control instructions include the path node sequence of the dynamic equipment, the cargo position operation instructions of the static equipment and the task processing sequence of the interactive equipment.

[0013] The adjusting of the task and path planning according to the real-time environment data includes: When the environmental data of the storage area affects the execution conditions of the task node, the priority of the task is adjusted, and the target equipment allocation plan of the task node is updated according to the environmental changes of the path node; According to the dynamic device density in the environment and the real-time status of the path nodes, the access order of the path nodes is adjusted, the total energy value of the path is re-evaluated through the environmental impact factor, and the path node sequence is re-planned; During the operation of the device, when the real-time environmental data prevents the dynamic device from completing the real-time task, the running task of the current dynamic device is suspended, the execution condition of the running task is changed, and the task node is switched to other dynamic devices, wherein the task switching includes transferring the task from the current device to the idle device and reallocating the device when the task priority changes; According to the energy consumption status of dynamic equipment, when the energy of the equipment cannot meet the task execution requirements, guide the equipment to hand over the task.

[0014] An intelligent control system for storage equipment, the system comprising a task scheduling module, a path planning module, a real-time monitoring module and a feedback optimization module; The task scheduling module is used to obtain the task requirements in the warehouse task queue, including the task type, task priority and completion time limit, generate a dynamic task graph according to the task requirements, and allocate the tasks to the adapted dynamic devices, static devices and interactive devices according to the optimization results of the task allocation scheme; The path planning module is used to construct a dynamic energy field to generate a path planning solution based on the real-time status of the dynamic equipment, the target task location and the storage environment information, and detect conflicting paths in the path planning, optimize the conflicting paths, and generate an updated path node sequence; The real-time monitoring module is used to collect the operating status of dynamic devices, static devices and interactive devices in real time, detect abnormal situations in task and path execution, and handle the abnormal situations.

[0015] The task scheduling module includes: A task extraction unit is used to extract task requirement information from the task queue, including task type, task priority and completion time limit; The task distribution generation unit is used to generate a dynamic task graph according to the task requirements and the storage equipment function model, and allocate task nodes to the adapted dynamic equipment, static equipment and interactive equipment; The task priority adjustment unit is used to dynamically adjust the task priority according to the task deadline and real-time environmental data to ensure that high-priority tasks are allocated and executed first.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention constructs a functional model of storage equipment, combines task requirements and real-time environmental data, and uses a neural network model to dynamically optimize task allocation and path planning, thereby achieving efficient collaborative control of storage equipment; 2. The present invention can quickly respond to changes in the storage environment by adjusting task priorities and path planning in real time, ensuring the flexibility and stability of task execution. In the case of insufficient equipment energy consumption or task obstruction, it provides a task switching and path replanning mechanism to maximize the resource utilization rate and task completion efficiency of storage equipment, and is suitable for complex storage scenarios with multi-device collaboration and high dynamics. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings: Figure 1 This is a flow chart of the intelligent control method for storage equipment according to Embodiment 1 of the present invention; Figure 2 This is a functional model structure diagram of storage equipment in Example 1 of the present invention; Figure 3 This is a neural network structure diagram of Example 1 of the present invention; Figure 4 This is a schematic diagram of the task optimization process of Example 1 of the present invention; Figure 5 This is a schematic diagram of the path planning process of Embodiment 1 of the present invention; Figure 6 This is a module diagram of the intelligent control system for storage equipment according to embodiment 2 of the present invention. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0019] Example 1

[0020] See also Figure 1 The present invention provides an embodiment: an intelligent control method for storage equipment, wherein the specific steps of the method are as follows: S1: Construct a storage equipment function model based on storage equipment information; In this step, the upper framework model and the lower device model are constructed according to the device type. The upper framework model describes the functional collaboration logic between dynamic devices, static devices and interactive devices by classification, and the lower device model establishes detailed functional parameters, real-time status and historical record data models for a single device. It can clearly define the device capabilities and operating status.

[0021] S2: extract task requirements from the task queue; In this step, the task queue is parsed to extract the core requirement parameters of each task, including task type, target location, priority, and completion deadline.

[0022] S3: Build a neural network model, train the task requirements and storage equipment function model through the neural network model, and output storage equipment control instructions; In this step, based on the task requirements and the functional model of the storage equipment, a neural network model is designed and trained, which includes the task optimization layer, the path optimization layer and the task execution layer. The task optimization layer trains and optimizes the matching relationship between the task requirements and the equipment status through the dynamic task graph and the allocation objective function, and generates a task distribution graph; the path optimization layer generates a path planning solution based on the real-time status of the dynamic equipment, the energy value of the path node and the task priority; the task execution layer optimizes the task distribution and path planning results through real-time feedback iteration, and finally generates the storage equipment control instructions. This step realizes the intelligent optimization of complex task allocation and path planning problems through the neural network model, and improves the efficiency of task allocation and the ability to minimize path consumption.

[0023] S4: Adjust the task and path planning based on real-time environmental data; In this step, dynamic data in the warehousing environment is collected in real time, including the congestion of path nodes, obstacle locations, equipment energy consumption status, and task progress data. When real-time data indicates that the task distribution map and path planning scheme are no longer compatible, the adaptability of task execution is ensured by dynamically adjusting the task distribution relationship and the order of path planning nodes. For example, when a path node is congested or the equipment cannot pass through, the energy value of the path node is dynamically adjusted to guide the equipment to detour to a suboptimal path; when task priorities or environmental changes lead to equipment task conflicts, the task node is dynamically switched to other devices. In this way, it is possible to effectively respond to sudden environmental changes and equipment status updates during task execution, and ensure the efficiency and stability of the warehousing system.

[0024] Although there have been some studies on intelligent control methods for storage equipment in the prior art, in actual applications, traditional task allocation and path planning methods are difficult to achieve efficient task and equipment coordination in complex dynamic environments. Specifically, existing methods usually use static task allocation and path planning technology, relying on fixed rules or limited real-time data. In multi-task scenarios, traditional technologies are difficult to dynamically optimize task allocation, especially when factors such as task priority, equipment status, and path resource conflicts change. The allocation results are prone to failure, resulting in delays in high-priority tasks and waste of resources; Furthermore, in this embodiment, the above-mentioned technical defects are effectively solved through the collaborative design of the task optimization layer, the path optimization layer and the task execution layer. For example, in a multi-task scenario, it is assumed that the warehousing system needs to complete 10 tasks at the same time, including 5 high-priority tasks, and the existing technology may cause some low-priority task devices to seize high-quality resources due to the static nature of task allocation, thereby causing high-priority tasks to be delayed. The present invention constructs a dynamic task graph, dynamically adjusts task allocation according to task priority and device status, and ensures that high-priority devices obtain resources first. For example, based on the edge weights of task nodes and device nodes in the dynamic task graph, this embodiment can reallocate high-priority tasks to the most suitable devices in real time, and adjust the execution order of low-priority tasks at the same time, significantly improving the overall efficiency of task allocation.

[0025] See also Figure 2 , the functional model structure diagram of the storage equipment in the embodiment of the present invention, the specific steps of S1 are as follows: S1.1: Collect multi-source information related to storage equipment in the storage area; The multi-source information includes information of dynamic devices, information of static devices, information of interactive devices and environmental information; Specifically, the current position, power status, task load, moving speed and task response time of dynamic equipment (such as AGV, robots) are collected; the storage capacity, cargo distribution, storage location status (empty or occupied) and cargo type of static equipment (such as shelves) are obtained; the task processing capability, load status, operation path and current task queue of interactive equipment (such as sorting machines, conveyor belts) are collected; the temperature and humidity of the storage area, path congestion, obstacle location and other dynamic environmental information that affects task and path planning are collected.

[0026] S1.2: construct an upper framework model according to the type of storage equipment, wherein the upper framework model includes categories of dynamic equipment, static equipment and interactive equipment and functional collaboration relationships between the categories; Dynamic equipment is mainly responsible for task execution and path planning. Its functional collaboration includes the equipment's mobility (such as speed and turning radius), task execution capability (such as load capacity and response time) and path planning constraints (such as obstacle avoidance capability).

[0027] Static equipment is mainly used for cargo storage management, and its functional collaboration includes cargo storage capacity, cargo type distribution and access frequency.

[0028] Interactive devices are mainly used for task processing, and their functional collaboration relationships include task processing speed, task load capacity, and interaction rules with dynamic devices (such as the order of cargo transportation).

[0029] Through the construction of the upper-level framework model, not only the classification and functional definition of the equipment are clarified, but also the collaborative relationship between the equipment is established. This functional abstraction and collaborative definition provides global guidance for task allocation and path planning, improving the execution efficiency of system tasks and the utilization of equipment resources.

[0030] S1.3: construct a lower-layer equipment model based on the multi-source information of each storage equipment, wherein the lower-layer equipment model includes a sub-model of a single storage equipment, and the sub-model includes equipment function parameters, equipment real-time status, and task capability characteristics; Equipment function parameters are used to describe the equipment's operating capabilities, including the moving speed and load capacity of dynamic equipment; the cargo storage capacity and cargo type of static equipment; and the task processing capability and operating speed of interactive equipment. The real-time status of the equipment includes the current location, power status, and current load of dynamic equipment; the cargo location status (empty or full status, cargo type) of static equipment; and the current task queue and task completion status of interactive equipment. The task capability characteristic is defined as the ability boundary of the equipment to complete tasks, such as the upper limit of the task response time of dynamic equipment, the cargo access constraints of static equipment, and the upper limit of the task load of interactive equipment.

[0031] By constructing the lower-level device model, the operating capability and real-time status of each device can be accurately described, providing detailed data support for subsequent task allocation and path planning.

[0032] S1.4: The upper-level framework model also includes a task scheduling center, which is configured with the neural network model for interacting with the lower-level equipment model, obtaining status information and task requirements of the lower-level equipment, and generating storage equipment control instructions based on the status information and task requirements.

[0033] See also Figure 3 , a structural diagram of a neural network model according to an embodiment of the present invention, wherein the neural network model comprises: The task optimization layer is used to optimize the task allocation scheme of the storage equipment according to the upper framework model, under the constraint condition of satisfying the priority of the storage tasks, with the goal of maximizing the efficiency of task allocation, and output a task distribution map; Specifically, the task optimization layer receives task requirement parameters (including task type, priority, and completion time limit) and current device status information (such as device idle status, current load, power, etc.), and scores the compatibility between task nodes and device nodes by constructing a dynamic task graph to ensure that task nodes are preferentially allocated to high-priority, high-fit device nodes. At the same time, the task optimization layer can dynamically allocate low-priority tasks based on the actual operating capabilities of the device to ensure full utilization of device resources. In this way, the task optimization layer significantly improves the overall efficiency of task allocation, ensuring timely response to high-priority tasks and reasonable execution of low-priority tasks.

[0034] The path optimization layer is used to optimize the path plan of the dynamic equipment according to the lower-layer equipment model, under the constraints of the equipment operation status and the storage environment, with the goal of minimizing the path consumption, and output the path planning plan; Specifically, when a path node is unavailable due to environmental changes (such as new obstacles or equipment congestion), the path optimization layer dynamically adjusts the priority of the path node to guide the device to select a low-energy and unobstructed suboptimal path. In addition, this layer also combines the real-time power status of the device to ensure that the path consumption of the device does not exceed the remaining power before the task is completed, thereby effectively avoiding the risk of task interruption due to insufficient power. Through the dynamic adjustment of the path optimization layer, not only can the path consumption be reduced and the equipment operation efficiency be improved, but also the collaborative optimization of multi-device path planning can be achieved in complex environments.

[0035] The task execution layer is used to update the adaptability of the task distribution map and the path planning solution through iterative calculation, and output storage equipment control instructions.

[0036] When the task distribution map or path planning scheme is not suitable due to changes in equipment status or environmental data, the task execution layer will iteratively optimize the distribution relationship between tasks and equipment and adjust the path planning scheme in real time. For example, when the equipment of a task node cannot be executed due to a fault, the task execution layer will immediately reallocate the task to other idle equipment and adjust its path node sequence to avoid path conflicts. During the path planning process, when the path node is congested or the task execution delay of the dynamic equipment exceeds the threshold, the task execution layer ensures the smoothness of the overall task execution by adjusting the task priority or path access order. In addition, the task execution layer combines real-time monitoring data to generate specific operation instructions for storage equipment, including the path node sequence of dynamic equipment, the cargo position operation instructions of static equipment, and the task processing order of interactive equipment, to ensure efficient coordination of tasks and path planning. Through the feedback optimization module, the task execution layer can collect and analyze the actual task execution results, and input the result data into the neural network model for parameter update, thereby improving the system's adaptability to complex environments and task execution efficiency.

[0037] Furthermore, in this embodiment, the collaborative design of the task optimization layer, the path optimization layer, and the task execution layer solves the problems of low efficiency in task allocation, insufficient adaptability of path planning, and poor real-time performance of task execution in a multi-task environment. In the task allocation stage, the dynamic task graph and priority rules are used to achieve efficient allocation of task resources; in the path planning stage, the dynamic energy field model is used to minimize and dynamically adjust path consumption in combination with task priority and path node status; in the task execution stage, iterative optimization and real-time instruction generation are used to improve the system's ability to adapt to changes in the environment and equipment status, thereby significantly improving the collaborative efficiency and task completion quality of storage equipment in complex dynamic scenarios.

[0038] See also Figure 4 , the task optimization flow chart of the embodiment of the present invention, the task optimization layer includes: Constructing a dynamic task graph and an allocation objective function, and generating an initial task allocation scheme according to the edge weights of the dynamic task graph and the allocation objective function, wherein the dynamic task graph is composed of task nodes and device nodes, and the edge weights are calculated according to task requirements, state information, and environmental state; Specifically, each task node represents a task requirement (including task type, priority, completion time limit, etc.), and the device node represents the function and status of dynamic devices, static devices or interactive devices (including current device location, load status, power level and availability, etc.). By connecting task nodes and device nodes with edges, a complete task graph is formed. The edge weight is calculated according to the adaptability between the task and the device, including the priority of the task, the operating status of the device and the real-time environmental conditions. For example, the higher the task priority, the more suitable the device operating status and the shorter the task completion path, the higher the edge weight, indicating that the adaptability between the task and the device is higher. The construction of the dynamic task graph can intuitively reflect the adaptability relationship between tasks and devices, solving the problem of lack of flexibility in traditional task allocation methods. The allocation objective function is designed according to the principle of maximizing the efficiency of warehouse task allocation. Its optimization goal is to minimize the delay time of high-priority tasks and the imbalance of device task allocation, while considering the impact of environmental factors on task execution. In this way, it can ensure that high-priority tasks obtain the optimal device resource allocation and improve the overall task execution efficiency.

[0039] The edge weight calculation formula is: , in, represents the edge weight between task i and device j, Indicates the edge weight corresponding to the case where device j is a dynamic device, represents the edge weight corresponding to the case where device j is a static device, Indicates the corresponding edge weight when device j is an interactive device.

[0040] The calculation formula for the edge weight corresponding to the dynamic device is: , in, , and Indicates the weight coefficient corresponding to the dynamic equipment, which is set according to the actual storage system requirements. represents the task priority of task i, It represents the estimated completion time of device j to complete task i, which is calculated based on the historical completion time of device j. represents the distance from device j to the initial location of task i, Represents the path sensitivity coefficient, which is used to adjust the impact of path distance on weight. represents the power of device j, represents the maximum power of device j, Indicates the current load of device j, represents the maximum load of equipment j; The calculation formula for the edge weight corresponding to the static device is: , in, and Indicates the weight coefficient corresponding to the static equipment, which is set according to the actual storage system requirements. Indicates the availability of the storage space of static equipment j, which is identified by a binary parameter. represents the access frequency of static device j, represents the matching degree between the environmental conditions of static device j and the environmental requirements of task i, Indicates the size matching between static device j and task i, with a value range of [0,1]; The calculation formula for the edge weight corresponding to the interactive device is: , in, , and Indicates the weight coefficient corresponding to the interactive device, which is set according to the actual storage system requirements. represents the task queue length of device j, Represents the sensitivity coefficient of the task queue, represents the historical task completion rate of device j.

[0041] Furthermore, the device nodes with the highest edge weights are preferentially assigned to the task nodes. By traversing the task nodes, the allocation order of the device nodes is dynamically adjusted according to the task priority and device status to ensure that high-priority tasks are met first. For example, when the target device of a high-priority task is fully loaded, it is switched to another device node with the second highest adaptability to avoid task delays. The generated initial task allocation plan can quickly cover all task nodes, and according to the optimized design of edge weights, it can achieve rapid adaptation of tasks and devices, ensuring the efficiency and rationality of the initial task allocation.

[0042] Verify the adaptability of the initial task allocation plan, identify failed tasks that do not meet the adaptation conditions, adjust the edge weights of the dynamic task graph according to the task allocation relationship of the failed tasks, and regenerate the task allocation plan; Specifically, the main contents of adaptability verification include: whether the task completion time meets the task time limit, whether the equipment load exceeds the allowable range, and whether the environmental status restricts the execution of the task. For example, when a dynamic device assigned to a task cannot reach the task location on time due to congestion of the path node, the task is identified as a failed task. By analyzing the task requirements of the task node and the real-time status of the device node, adaptability verification can quickly identify failed tasks and adjust their task allocation relationships, thereby avoiding resource waste and system efficiency reduction caused by task failure.

[0043] In this embodiment, by reducing the edge weight value between the failed task and the current device node, the allocation priority of the device node is reduced, and the suboptimal device node is re-searched in the dynamic task graph. During the reallocation process, the allocation order is dynamically adjusted according to the task priority to ensure that the allocation of high-priority tasks is not affected by low-priority tasks. For example, when the failed task is a low-priority task, the allocation relationship of the low-priority task will be adjusted first, and high-quality device resources will be reserved for high-priority tasks.

[0044] The regenerated task allocation scheme is iteratively verified until all task allocation relationships meet the adaptation conditions and the task distribution graph is output.

[0045] The regenerated task allocation plan needs to be verified again by the adaptability verification module to ensure that the adjusted task allocation relationship meets the adaptation conditions. If there are still failed tasks, repeat the process of dynamic task graph adjustment and regeneration of task allocation plan until all task allocation relationships pass the adaptability verification. Through multiple rounds of iterations, the task allocation plan can be gradually optimized to solve the allocation conflicts and resource competition problems in multi-task scenarios. The final task distribution graph is based on the allocation relationship between task nodes and device nodes as the core output.

[0046] See also Figure 5, a flow chart of path planning in an embodiment of the present invention, wherein the path planning layer comprises: According to the real-time status of the dynamic equipment, the target task location and the storage environment information, a dynamic energy field including equipment nodes and path nodes is constructed, wherein the dynamic energy field is characterized by the energy value of each path node; Specifically, the construction of the dynamic energy field is carried out by collecting the current location of the equipment, the target location of the task, and the status information of the path nodes in the storage environment. The equipment node contains the current load, power status and task execution progress of the equipment, and the path node contains information such as path length, congestion status, obstacle distribution and task urgency. The energy value of the path node comprehensively considers the path length factor, task priority factor, path congestion factor and equipment energy consumption factor, and reflects the comprehensive cost of the path by weighing these factors. For example, when a path node is severely congested or there are high-energy consumption devices, its energy value will increase significantly, reducing the priority of the device passing; while for the target path node of a high-priority task, its energy value will be significantly reduced due to the effect of the priority factor, thereby giving priority to guiding the dynamic device to select the path. Through the construction of this dynamic energy field, real-time self-adaptation of path selection can be achieved to ensure a comprehensive balance between task priority and path consumption.

[0047] The energy value is calculated as follows: , in, Represents the energy value from dynamic device n to path node m, Represents the path length factor, which is calculated based on the straight-line distance from dynamic device n to path node m. Indicates the remaining completion time of dynamic device n, and Represents the adjustment parameter, which is used to adjust the weight of congestion factor and environmental complexity. represents the real-time congestion factor of path node m, represents the logarithmic function with base 2, Indicates the number of devices currently existing on path node m. Represents the environmental complexity factor, which is used to measure the impact of the number of path obstacles, slope, and temperature and humidity on path selection. represents the energy consumption of dynamic device n at path node m, Indicates the current remaining power of dynamic device n.

[0048] The low-energy value path is traversed according to the dynamic energy field, and an initial path planning scheme is generated with the goal of minimizing path consumption. During the traversal process, in order to ensure the integrity of path planning, path continuity constraints are added according to the operating characteristics of the equipment to avoid large-scale detours caused by the equipment selecting secondary paths.

[0049] According to the initial path planning scheme, identifying whether there is a conflicting path for the dynamic device, wherein the conflicting path means that multiple devices select the same path node in the same time period, or the congestion risk in the path overlap area exceeds a preset threshold; The identification of conflicting paths is based on the conflict detection of the time dimension and space dimension of the path node. The time dimension determines whether there are multiple devices planning to pass through the same node in the same time period by calculating the estimated arrival time of the device; the space dimension determines whether the path node is overloaded by the device density and congestion threshold of the path node. When a conflicting path is detected, the conflict area is marked and the path adjustment mechanism is started.

[0050] According to the task priority of the tasks corresponding to the dynamic devices, the initial path planning scheme is adjusted based on the principle that high-priority devices first occupy the path nodes with low energy values; The adjusting the initial path planning scheme includes: Delay the time planning of the path of low-priority devices, set waiting points in the path conflict area, and allow high-priority devices to pass first; When the waiting time exceeds the set threshold or the path node is unavailable due to congestion, the path of the low-priority device is reallocated and the suboptimal low-energy value path node is selected; When the task completion deadline of a low-priority device is approaching or the reallocated path affects the task completion deadline, the path priority of the low-priority device is increased and path node resources are released for the low-priority device to pass.

[0051] Specifically, the system first analyzes the dynamic energy field changes of the conflicting paths, guides the devices to select suboptimal paths by increasing the energy values ​​of the path nodes of low-priority tasks, and recalculates the path consumption of low-priority devices based on the matching degree between the task nodes and the path nodes. During the selection of suboptimal paths, the system gives priority to path nodes with lower energy values ​​in the path energy field that can reach the target location within the task completion time. In addition, to ensure the real-time performance of high-priority device tasks, the path adjustment phase also recalculates the energy values ​​of the path nodes in combination with real-time environmental data, and dynamically adjusts the priority weights of the path nodes, thereby releasing conflicting node resources for high-priority devices.

[0052] According to the adjusted initial path planning scheme, the path consumption of each dynamic device is recalculated, the path planning scheme is updated according to the path consumption, the adaptability of the adjusted path is verified, and if all dynamic devices meet the task priority and energy consumption optimization goals, the path planning scheme is output.

[0053] The adaptability verification includes the following key indicators: First, whether the task priority of the device is met to ensure that the completion time of high-priority tasks is better than that of low-priority tasks; second, whether the energy consumption of the path planning is within the power range of the device to prevent the failure of the device task due to excessive energy consumption; third, whether the congestion of the path node is reduced to the system preset safety threshold, thereby ensuring the traffic capacity of the path node. If the verification is passed, the path planning plan will be sent to the device control module to guide the dynamic device to execute the planned task.

[0054] The task execution layer includes: Optimize the allocation relationship between task nodes and device nodes to meet the needs of high-priority tasks to the greatest extent possible while meeting the optimization goal of consistent task completion time and task priority; Recalculate the energy values ​​of path nodes. When path conflicts, node congestion, and insufficient device energy are detected, replan the path and select low-energy path nodes. The control instructions include the path node sequence of the dynamic equipment, the cargo position operation instructions of the static equipment and the task processing sequence of the interactive equipment.

[0055] The specific steps of S4 are as follows: S4.1: When the environmental data of the storage area affects the execution conditions of the task node, adjust the priority of the task and update the target device allocation plan of the task node according to the environmental changes of the path node; For example, when it is detected that the temperature in a certain task target area exceeds the safe range for cargo storage, or when the increase in humidity may affect the quality of the cargo, the priority of the task will be increased, and the task will be transferred from the inefficient device to the device with the best response capability to ensure timely completion of the task. The path planning module will further adjust the target device path involved in the task allocation in real time according to the environmental changes of the path nodes, such as the appearance of obstacles or traffic restrictions at a certain path node, to ensure the effectiveness of the path planning. At this time, the device nodes and path nodes are re-optimized in combination with the dynamic task graph to ensure the adaptability of the task allocation. Through this adjustment strategy, not only the dynamic adaptability of the task is improved, but also the system's fault tolerance to complex environmental changes is enhanced to avoid task delays or failures.

[0056] S4.2: According to the dynamic device density in the environment and the real-time status of the path nodes, the access order of the path nodes is adjusted, the total energy value of the path is re-evaluated through the environmental impact factor, and the path node sequence is re-planned; Specifically, an environmental impact factor is set for the path node. For example, the environmental impact factor of the congested path is dynamically increased, and low-congestion paths are preferentially allocated to dynamic devices. At the same time, the energy of the path nodes is re-evaluated, and nodes with lower path energy are preferentially added to the path planning sequence. At this time, the dynamic energy field model comes into play and the path sequence is replanned by adjusting the energy value of the path nodes. For example, when the device density is too high, the energy value of the congested path node is increased to guide other devices to choose a suboptimal path. In this way, the access load of the path nodes can be effectively dispersed to avoid path conflicts, and the total energy value of the path planning can be reduced, thereby improving the efficiency of the overall path planning of the system.

[0057] S4.3: During the operation of the device, when the real-time environmental data prevents the dynamic device from completing the real-time task, suspend the running task of the current dynamic device, change the execution condition of the running task, and switch the task node to other dynamic devices, wherein the task switching includes transferring the task from the current device to the idle device and reallocating the device when the task priority changes; Specifically, when real-time environmental data indicates that a dynamic device cannot complete a task due to path obstruction, abnormal device status (such as insufficient power) or other environmental restrictions (such as path obstacles), the task execution module will trigger the task switching mechanism. First, the running task of the current device is suspended, and the remaining execution status of the task is recorded, such as the completion progress of the task, the current location of the device, etc. Then, by evaluating the idle status, task adaptability and responsiveness of other dynamic devices in real time, the task node is transferred to the most suitable idle device. For example, when it is detected that a device is delayed in execution due to path node obstruction, the task switching module reallocates the task to another device with idle resources and plans a new path for the device. The task switching mechanism can also dynamically adjust the matching relationship between devices and tasks according to the task priority. When the priority of a task is increased, the task node can be switched from an inefficient device to an efficient device, thereby ensuring the timely completion of the task. Through this task switching strategy, the system can effectively respond to emergencies in a dynamic environment and reduce the task failure rate and device downtime.

[0058] S4.4: When it is monitored in real time that the remaining energy of a dynamic device is lower than the set threshold and cannot support the completion of the current task, the task execution layer will automatically trigger the task handover mechanism. Specifically, first, the current task of the low-energy device is suspended, and the task node is assigned to a backup device with sufficient energy; secondly, a charging path is planned for the low-energy device to guide it to enter the charging station first to restore its operating capacity. During the task handover process, the system will dynamically evaluate the compatibility of the device and the task to ensure that the device after the task handover can complete the task within the remaining time. For example, in a task scenario, a dynamic device cannot complete a long-distance task due to insufficient power in path planning. The task handover mechanism reallocates the task to another device with high load capacity and replans an efficient path for it. The introduction of the task handover strategy can not only avoid task failures due to insufficient power of the device, but also extend the working time of the device, and improve the overall operating efficiency and task completion rate of the system.

[0059] Example 2

[0060] See also Figure 6 , the present invention provides an embodiment: an intelligent control system for storage equipment, the system comprising a task scheduling module, a path planning module, a real-time monitoring module and a feedback optimization module; The task scheduling module is used to obtain the task requirements in the warehouse task queue, including the task type, task priority and completion time limit, generate a dynamic task graph according to the task requirements, and allocate the tasks to the adapted dynamic devices, static devices and interactive devices according to the optimization results of the task allocation scheme; The path planning module is used to construct a dynamic energy field to generate a path planning solution based on the real-time status of the dynamic equipment, the target task location and the storage environment information, and detect conflicting paths in the path planning, optimize the conflicting paths, and generate an updated path node sequence; The real-time monitoring module is used to collect the operating status of dynamic devices, static devices and interactive devices in real time, detect abnormal situations in task and path execution, and handle abnormal situations; The feedback optimization module is used to collect actual feedback data according to the task execution results, input the feedback data into the neural network model, and update the parameter configuration of the model.

[0061] The task scheduling module includes: A task extraction unit is used to extract task requirement information from the task queue, including task type, task priority and completion time limit; The task distribution generation unit is used to generate a dynamic task graph according to the task requirements and the storage equipment function model, and allocate task nodes to the adapted dynamic equipment, static equipment and interactive equipment; The task priority adjustment unit is used to dynamically adjust the task priority according to the task deadline and real-time environmental data to ensure that high-priority tasks are allocated and executed first.

[0062] The path planning module comprises: An energy field construction unit, used to construct a dynamic energy field including device nodes and path nodes according to the real-time status of the dynamic device, the target task location and the storage environment information, wherein the dynamic energy field is characterized by the energy value of each path node; An initial path planning unit, configured to traverse a low energy path according to the dynamic energy field and generate an initial path planning solution with the goal of minimizing path consumption; A path detection unit, configured to identify whether a dynamic device has a conflicting path according to the initial path planning scheme, wherein the conflicting path indicates that multiple devices select the same path node in the same time period, or the congestion risk in the path overlap area exceeds a preset threshold; The path adjustment unit is used to adjust the initial path planning scheme according to the task priority of the task corresponding to the dynamic device, based on the principle that high-priority devices preferentially occupy low-energy value path nodes, recalculate the path consumption of each dynamic device according to the adjusted initial path planning scheme, update the path planning scheme according to the path consumption, verify the adaptability of the adjusted path, and output the path planning scheme if all dynamic devices meet the task priority and energy consumption optimization goals.

[0063] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. An intelligent control method for storage equipment, characterized in that: The method comprises: Constructing a storage equipment function model according to the storage equipment information, wherein the storage equipment function model includes an upper framework model and a lower equipment model; Extract task requirements from the task queue; Constructing a neural network model, training the task requirements and the storage equipment function model through the neural network model, and outputting storage equipment control instructions; According to the storage equipment control instructions, the storage equipment is controlled to collaboratively execute tasks. During the task execution process, the task and path planning are adjusted according to real-time environmental data.

2. The intelligent control method for storage equipment according to claim 1 is characterized in that: The construction of the storage equipment function model includes: Collect multi-source information related to storage equipment in the storage area; Building an upper framework model according to the types of storage equipment, the upper framework model includes categories of dynamic equipment, static equipment and interactive equipment and functional collaboration relationships between categories; Constructing a lower-layer equipment model based on multi-source information of each storage equipment, wherein the lower-layer equipment model includes a sub-model of a single storage equipment, and the sub-model includes equipment function parameters, equipment real-time status, and task capability characteristics; The upper-level framework model also includes a task scheduling center, which is configured with the neural network model and is used to interact with the lower-level equipment model, obtain status information and task requirements of the lower-level equipment, and generate storage equipment control instructions based on the status information and task requirements.

3. According to claim 1, the intelligent control method for storage equipment is characterized in that: The neural network model comprises: The task optimization layer is used to optimize the task allocation scheme of the storage equipment according to the upper framework model, under the constraint condition of satisfying the priority of the storage tasks, with the goal of maximizing the efficiency of task allocation, and output a task distribution map; The path optimization layer is used to optimize the path plan of the dynamic equipment according to the lower-layer equipment model, under the constraints of the equipment operation status and the storage environment, with the goal of minimizing the path consumption, and output the path planning plan; The task execution layer is used to update the adaptability of the task distribution map and the path planning solution through iterative calculation, and output storage equipment control instructions.

4. According to the intelligent control method for storage equipment in claim 3, the task requirements include task type, task priority and completion time limit, characterized in that: The optimization of the task allocation scheme for storage equipment includes: Constructing a dynamic task graph and an allocation objective function, and generating an initial task allocation scheme according to the edge weights of the dynamic task graph and the allocation objective function, wherein the dynamic task graph is composed of task nodes and device nodes, and the edge weights are calculated according to task requirements, state information, and environmental state; Verify the adaptability of the initial task allocation plan, identify failed tasks that do not meet the adaptation conditions, adjust the edge weights of the dynamic task graph according to the task allocation relationship of the failed tasks, and regenerate the task allocation plan; The regenerated task allocation scheme is iteratively verified until all task allocation relationships meet the adaptation conditions and the task distribution graph is output.

5. The intelligent control method for storage equipment according to claim 3 is characterized in that: The optimization of the path scheme of the dynamic device includes: According to the real-time status of the dynamic equipment, the target task location and the storage environment information, a dynamic energy field including equipment nodes and path nodes is constructed, wherein the dynamic energy field is characterized by the energy value of each path node; Traversing the low energy path according to the dynamic energy field, generating an initial path planning scheme with the goal of minimizing path consumption; According to the initial path planning scheme, identifying whether there is a conflicting path for the dynamic device, wherein the conflicting path means that multiple devices select the same path node in the same time period, or the congestion risk in the path overlap area exceeds a preset threshold; According to the task priority of the tasks corresponding to the dynamic devices, the initial path planning scheme is adjusted based on the principle that high-priority devices first occupy the path nodes with low energy values; According to the adjusted initial path planning scheme, the path consumption of each dynamic device is recalculated, the path planning scheme is updated according to the path consumption, the adaptability of the adjusted path is verified, and if all dynamic devices meet the task priority and energy consumption optimization goals, the path planning scheme is output.

6. The intelligent control method for storage equipment according to claim 5 is characterized in that: The adjusting the initial path planning scheme includes: Delay the time planning of the path of low-priority devices, set waiting points in the path conflict area, and allow high-priority devices to pass first; When the waiting time exceeds the set threshold or the path node is unavailable due to congestion, the path of the low-priority device is reallocated and the suboptimal low-energy value path node is selected; When the task completion deadline of a low-priority device is approaching or the reallocated path affects the task completion deadline, the path priority of the low-priority device is increased and path node resources are released for the low-priority device to pass.

7. The intelligent control method for storage equipment according to claim 3 is characterized in that: The task execution layer includes: Optimize the allocation relationship between task nodes and device nodes to meet the needs of high-priority tasks to the greatest extent possible while meeting the optimization goal of consistent task completion time and task priority; Recalculate the energy values ​​of path nodes. When path conflicts, node congestion, and insufficient device energy are detected, replan the path and select low-energy path nodes. The control instructions include the path node sequence of the dynamic equipment, the cargo position operation instructions of the static equipment and the task processing sequence of the interactive equipment.

8. The intelligent control method for storage equipment according to claim 1 is characterized in that: The adjusting of the task and path planning according to the real-time environment data includes: When the environmental data of the storage area affects the execution conditions of the task node, the priority of the task is adjusted, and the target equipment allocation plan of the task node is updated according to the environmental changes of the path node; According to the dynamic device density in the environment and the real-time status of the path nodes, the access order of the path nodes is adjusted, the total energy value of the path is re-evaluated through the environmental impact factor, and the path node sequence is re-planned; During the operation of the device, when the real-time environmental data prevents the dynamic device from completing the real-time task, the running task of the current dynamic device is suspended, the execution condition of the running task is changed, and the task node is switched to other dynamic devices, wherein the task switching includes transferring the task from the current device to the idle device and reallocating the device when the task priority changes; According to the energy consumption status of dynamic equipment, when the energy of the equipment cannot meet the task execution requirements, guide the equipment to hand over the task.

9. An intelligent control system for storage equipment, used to implement the intelligent control method for storage equipment according to any one of claims 1 to 8, characterized in that: The system includes a task scheduling module, a path planning module, a real-time monitoring module and a feedback optimization module; The task scheduling module is used to obtain the task requirements in the warehouse task queue, including the task type, task priority and completion time limit, generate a dynamic task graph according to the task requirements, and allocate the tasks to the adapted dynamic devices, static devices and interactive devices according to the optimization results of the task allocation scheme; The path planning module is used to construct a dynamic energy field to generate a path planning solution based on the real-time status of the dynamic equipment, the target task location and the storage environment information, and detect conflicting paths in the path planning, optimize the conflicting paths, and generate an updated path node sequence; The real-time monitoring module is used to collect the operating status of dynamic devices, static devices and interactive devices in real time, detect abnormal situations in task and path execution, and handle the abnormal situations.

10. The intelligent control system for storage equipment according to claim 9, characterized in that: The task scheduling module includes: A task extraction unit is used to extract task requirement information from the task queue, including task type, task priority and completion time limit; The task distribution generation unit is used to generate a dynamic task graph according to the task requirements and the storage equipment function model, and allocate task nodes to the adapted dynamic equipment, static equipment and interactive equipment; The task priority adjustment unit is used to dynamically adjust the task priority according to the task deadline and real-time environmental data to ensure that high-priority tasks are allocated and executed first.

Images