A warehouse point management method and system based on an industrial internet

Abstract

The application relates to the technical field of industrial internet, and discloses a warehouse point management method and system based on industrial internet, which comprises the following steps: acquiring the basic space topology structure of a warehouse area and temporary ground pile occupation information in the area where a target warehouse-out point is located through industrial internet; instantiating the temporary ground pile occupation information into a temporary topology constraint object independent of the basic space topology structure, and establishing a mapping relationship between the temporary topology constraint object and the basic space topology structure; based on the mapping relationship, superimposing a dynamic temporary traffic constraint layer on the basic space topology structure, so that the basic space topology structure and the dynamic temporary traffic constraint layer jointly constitute a hierarchical topology view; performing connectivity verification on the target warehouse-out point on the hierarchical topology view; and determining the execution path or the task state of the warehouse-out task according to the connectivity verification result, which can effectively solve the path planning mismatch problem caused by the lag of the topology version and improve the automatic execution efficiency of the warehouse operation.

Description

Technical Field

[0001] This application relates to the field of industrial internet technology, and more specifically, to a warehouse location management method and system based on industrial internet. Background Technology

[0002] In warehouse location management within the Industrial Internet architecture, it's typically necessary to determine the reachability of target locations based on the maintained spatial topology relationships of fixed storage locations, aisles, and lanes, and to generate outbound task paths for equipment such as automated guided vehicles (AGVs) or forklifts. This topology-based path planning and task output is a crucial foundation for ensuring the orderly flow of warehouse operations. However, in actual warehousing operations, especially during large-volume arrivals or wave operations, some goods are often temporarily stacked on the side of aisles or in front of warehouses. In this case, the temporary stacking locations physically occupy adjacent areas or part of the passageway space. In the Industrial Internet's process of determining the reachability of target fixed storage locations and outputting equipment paths based on spatial topology adjacency relationships, the change in the physical state of these temporary stacking locations is often not promptly included in the Industrial Internet's spatial topology version updates. The system will still rely on the old topology, which does not include the temporary occupancy status, for reachability assessments. This leads to a situation where, although the industrial internet outputs the reachability conclusions and task paths for fixed storage locations to devices in a logically valid manner, the devices may encounter obstacles or need to temporarily detour when traveling along the predetermined path to the temporary storage location. Existing conventional methods mainly rely on changes in fixed physical structures to trigger topology updates, lacking dynamic mapping of dynamic temporary occupancy boundaries, making it difficult to identify local mismatches between the on-site physical state and the logical topology.

[0003] Therefore, in outbound tasks, when a temporary storage location has actually occupied the adjacent area of ​​the channel but has not entered the industrial internet spatial topology, the industrial internet will still determine that the target fixed storage location is reachable and output the equipment path because the topology version does not reflect the temporary occupation. When the equipment executes according to the path, it is blocked by the temporary storage location, which in turn affects the executability of the outbound task and the acceptance of the location task. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this application provides a warehouse location management method and system based on the Industrial Internet, which can solve the technical problem in existing technologies where equipment path planning and actual on-site traffic conditions are mismatched due to the failure to update the spatial topology in a timely manner when temporary storage occupancy is not addressed.

[0005] Firstly, this application provides a warehouse location management method based on the Industrial Internet, including: The Industrial Internet is used to obtain the basic spatial topology of the warehouse area and the temporary storage occupancy information within the area where the target outbound point is located. The temporary heap occupancy information is instantiated into a temporary topological constraint object that is independent of the basic spatial topology, and a mapping relationship between the temporary topological constraint object and the basic spatial topology is established. Based on the mapping relationship, a dynamic temporary access constraint layer is superimposed on the basic spatial topology, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view. Based on the dynamic temporary access constraint layer, additional access costs are added to the channel segments in the basic spatial topology, and connectivity checks are performed on the target outbound points in the hierarchical topology view. The execution path or task status of the outbound task is determined based on the connectivity verification results. After the effective lifecycle of the temporary topology constraint object ends or according to the task status of the outbound task, the basic spatial topology is restored to the original connectivity state before the dynamic temporary access constraint layer is superimposed.

[0006] Optionally, the steps for obtaining the basic spatial topology of the storage area and the temporary storage occupancy information within the target storage location area through the Industrial Internet include: The industrial internet platform obtains the basic spatial topology structure, which is organized in the form of a graph and includes storage location nodes, channel intersection nodes, platform handover nodes, and channel and alleyway segments connecting each node. Acquire monitoring video streams, LiDAR scanning data, and temporary cargo release event logs through industrial internet edge computing nodes; Spatial fusion verification is performed on the monitoring video stream, LiDAR scanning data, and temporary cargo release event logs to identify temporary occupancy awareness events. The temporary occupancy awareness events and the coordinates of the physical contour polygons of the temporary occupancy awareness events are used as temporary land storage occupancy information.

[0007] Optionally, the step of instantiating temporary heap occupancy information into temporary topological constraint objects independent of the underlying spatial topology includes: Create a temporary topological constraint object to record the coordinates of the physical contour polygon and its lifecycle state for temporary occupancy-aware events, wherein the temporary topological constraint object is stored independently of the underlying spatial topology structure.

[0008] Optionally, the steps for establishing the mapping relationship between temporary topological constraint objects and the underlying spatial topology include: Using the geometric center of the physical contour polygon coordinates recorded by the temporary topological constraint object as the center, and a preset search radius, a spatial range search is performed in the basic spatial topology structure to filter out all topological nodes and topological edges that fall within the preset search radius as potentially affected topological objects. A mapping relationship is established between temporary topological constraint objects and basic spatial topology. The temporary topological constraint objects are associated with the channel segments and storage nodes in the screened potentially affected topological objects and recorded in the mapping relationship.

[0009] Optionally, the step of overlaying a dynamic temporary access constraint layer onto the basic spatial topology based on the mapping relationship, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view, includes: A dynamic temporary access constraint layer is constructed based on the mapping relationship, which includes temporary topological constraint objects and channel segment edges and storage location nodes in potentially affected topological objects. Using the basic spatial topology as the bottom layer and the dynamic temporary access constraint layer as the upper layer to carry all temporary topology constraint objects, a hierarchical topology view is formed in which the basic spatial topology and the dynamic temporary access constraint layer are logically superimposed but physically separated in storage.

[0010] Optionally, the step of adding passage costs to the channel segment edges in the basic spatial topology based on the dynamic temporary passage constraint layer includes: For each channel segment edge in the dynamic temporary access constraint layer, extract the standard physical width and safe access width threshold of the channel segment edge in the basic spatial topology; The ideal passage space area of ​​this passage segment edge is constructed based on the standard physical width and the safe passage width threshold; Calculate the overlap area between the temporary topological constraint object and the ideal passage space region based on the coordinates of the physical contour polygon recorded by the temporary topological constraint object; If the area of ​​the overlapping region is zero, mark the edge of the channel segment as having no effect and remove it from the mapping relationship; If the area of ​​the overlapping region is greater than zero, calculate the maximum intrusion depth of the overlapping region in the direction perpendicular to the centerline of the channel segment, and calculate the remaining effective passage width of the channel segment after being temporarily occupied and squeezed based on the difference between the standard physical width and the maximum intrusion depth. When the remaining effective passage width is less than the safe passage width threshold, a preset blocking penalty value is applied to the edge of the passage segment; When the remaining effective passage width is greater than or equal to the safe passage width threshold but less than the standard physical width, a preset passage resistance penalty value is added to the edge of the passage segment. When the remaining effective passage width is greater than or equal to the standard physical width, no additional passage cost is added to the edge of the passage segment.

[0011] Optionally, the step of performing connectivity verification on the target outbound point in the hierarchical topology view includes: Extract a local subgraph from the hierarchical topology view that includes all alternative connected paths from the starting point of the outbound task to the target outbound point. The preset blocking penalty value or preset passage resistance penalty value of each channel segment edge in the dynamic temporary passage constraint layer is superimposed on the corresponding channel segment edge in the local subgraph. The channel segments with preset blocking penalty values ​​are used as disconnected edges in connectivity verification, the channel segments with preset passage resistance penalty values ​​are used as high-cost passable edges in connectivity verification, and the channel segments without additional passage costs are used as default passable edges in connectivity verification. A connected path from the starting point of the outbound task to the target outbound point is searched on the local subgraph. If a connected path is found, the target outbound point is determined to have passed the connectivity check; otherwise, the target outbound point is determined to have failed the connectivity check.

[0012] Optionally, the steps for determining the execution path or task status of the outbound task based on the connectivity verification result include: When the target outbound point is determined to have passed the connectivity check, the searched connectivity path is smoothed and semantically transformed to generate the execution path instruction. When the target outbound location fails the connectivity check, the task status of the outbound task is marked as suspended, and it is checked whether there are other alternative storage location nodes that store the same material information and are topologically connected.

[0013] Optionally, the step of restoring the basic spatial topology to its original connectivity state before the dynamic temporary access constraint layer is superimposed after the effective lifetime of the temporary topology constraint object ends or according to the task status of the outbound task includes: When the effective lifecycle of a temporary topology constraint object ends or when the outbound task status is completed, all temporary topology constraint objects are destroyed, and the passage cost attached to the channel segment edge in the basic space topology is revoked, so that the basic space topology is restored to the original connected state before the dynamic temporary passage constraint layer was superimposed.

[0014] Secondly, this application also discloses a warehouse location management system based on the Industrial Internet, including: The information acquisition module is used to acquire the basic spatial topology of the storage area and the temporary storage occupancy information within the area where the target outbound point is located through the Industrial Internet. The object instantiation module is used to instantiate temporary heap occupancy information into temporary topological constraint objects that are independent of the basic spatial topology, and to establish a mapping relationship between the temporary topological constraint objects and the basic spatial topology. The view building module is used to overlay a dynamic temporary access constraint layer on the basic spatial topology based on the mapping relationship, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view. The connectivity verification module is used to perform connectivity verification on the target outbound point on the hierarchical topology view by adding passage costs to the channel segment edges in the basic spatial topology based on the dynamic temporary passage constraint layer. The task decision and topology restoration module is used to determine the execution path or task status of outbound tasks based on the connectivity verification results, and restore the basic spatial topology to the original connectivity state before the dynamic temporary access constraint layer was superimposed, after the effective life cycle of the temporary topology constraint object ends or according to the task status of the outbound task.

[0015] This application constructs a hierarchical topology view, overlaying temporary storage occupancy information as a dynamic constraint layer onto the basic spatial topology, achieving efficient perception and agile mapping of dynamic changes in the warehouse site. By calculating the degree of compression of aisle width by temporary occupancy and adding corresponding passage costs, the accessibility of target locations can be assessed in real time. This allows for the avoidance of obstructed areas during the path planning stage, solving the path planning mismatch problem caused by lagging topology version updates in existing technologies. This improves the automation level of warehouse operations, the success rate of task execution, and the safety of equipment operation. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a warehouse location management method based on the Industrial Internet, provided as an embodiment of this application.

[0017] Figure 2 This is a schematic diagram of a warehouse location management system based on the Industrial Internet, provided as an embodiment of this application.

[0018] Labeling Explanation: 210 Information Acquisition Module; 220 Object Instantiation Module; 230 View Construction Module; 240 Connectivity Verification Module; 250 Task Decision and Topology Recovery Module. Detailed Implementation

[0019] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0020] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0021] In intelligent warehousing and logistics systems under the modern Industrial Internet architecture, automated warehouse management relies on the high-precision spatial topology relationships of fixed storage locations, aisles, and lanes maintained within the system. This topology is not only the basis for the warehouse control center to determine the accessibility of rack locations, but also the core basis for planning the operation trajectories and issuing control commands for handling equipment such as Automated Guided Vehicles (AGVs), Autonomous Mobile Robots (AMRs), or automated forklifts. However, in actual warehousing operations, facing high-intensity flow conditions such as periodic concentrated arrivals and wave-like inbound and outbound picking operations, on-site workers or handling mechanisms often temporarily place some palletized goods in the form of ground stacks on the edge of aisles, the side of main roads, or temporary storage areas in front of warehouses. These temporary ground stacks physically occupy the adjacent space or effective passage width of the aisles. However, in the existing conventional Industrial Internet point management logic, the spatial topology model is usually only maintained and iterated infrequently based on permanent changes to fixed physical facilities (such as rack installation, column modification, and aisle planning), lacking an agile response mechanism for short-term, high-frequency dynamic temporary occupancy boundaries on-site. When temporary storage piles encroach on the passageway but are not synchronized to the system topology model, the industrial internet hub, based on the old version of the topology, will still determine that the target fixed storage location is physically unobstructed and issue operation instructions to the handling equipment to cross the area. This causes the equipment to encounter physical storage piles after proceeding to the site according to the instructions, resulting in emergency stops, forced on-site replanning, or even task deadlocks, severely disrupting the continuity of outbound operations and the reliability of point-to-point connection.

[0022] Firstly, see [the following] Figure 1 This application discloses a warehouse location management method based on the Industrial Internet, including: The Industrial Internet is used to obtain the basic spatial topology of the warehouse area and the temporary storage occupancy information within the area where the target outbound point is located. The temporary heap occupancy information is instantiated into a temporary topological constraint object that is independent of the basic spatial topology, and a mapping relationship between the temporary topological constraint object and the basic spatial topology is established. Based on the mapping relationship, a dynamic temporary access constraint layer is superimposed on the basic spatial topology, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view. Based on the dynamic temporary access constraint layer, additional access costs are added to the channel segments in the basic spatial topology, and connectivity checks are performed on the target outbound points in the hierarchical topology view. The execution path or task status of the outbound task is determined based on the connectivity verification results. After the effective lifecycle of the temporary topology constraint object ends or according to the task status of the outbound task, the basic spatial topology is restored to the original connectivity state before the dynamic temporary access constraint layer is superimposed.

[0023] The basic spatial topology refers to the graphical model used to characterize the fixed facilities and physical spatial connectivity of the storage site. It is usually composed of nodes representing physical locations such as fixed storage locations, passage intersections, and entry / exit platforms, as well as edges representing the physical paths that can be traversed between nodes and their static physical properties (such as passage width and rated passage direction).

[0024] Temporary storage occupancy information refers to the spatial range, location, and time and status description data of non-fixed static obstacles such as goods and material pallets temporarily stored at the physical operation site of warehousing.

[0025] Temporary topology constraint objects refer to entity data structures generated after object-oriented modeling of physical temporary heaps at the logical level. They are stored and managed independently of the underlying basic spatial topology structure and carry attributes such as the geometric boundaries and lifecycle of temporary occupancy. The role of this object is to transform short-term physical occupancy that originally only existed on-site into logical entities that can be directly read.

[0026] The dynamic temporary access constraint layer is a logical layer superimposed on the basic spatial topology. It is specifically designed to carry temporary topological constraint objects and the dynamic access restrictions imposed on adjacent topological edges and nodes, decoupling from and cooperating with the underlying basic topology. This layer does not rewrite the basic spatial topology itself, but participates in the calculation as an additional constraint during task judgment. Therefore, it can quickly respond to changes in the field without causing frequent changes to the basic topology version.

[0027] A layered topology view is a global spatial topology representation that is logically composed of a bottom-level baseline physical topology and an upper-level dynamic temporary access constraint layer. It retains the stability of the baseline physical topology while also having the flexibility to represent dynamic temporary disturbances.

[0028] The passage cost refers to the numerical value used in graph search or path optimization algorithms to quantify the spatial resistance, time consumption, or safety risks that a transport device needs to incur to pass through a specific channel segment edge or access a specific node.

[0029] First, an industrial internet is used to establish a convergence link between on-site physical perception and basic static information, continuously capturing the static basic map structure of the storage area and local dynamic occupancy data. Among them, the basic spatial topology structure is preferably periodically synchronized to the central or edge cache by the storage management system and storage control system, and the temporary stacking occupancy information is preferably provided by high-position video, vehicle-mounted LiDAR, manual loading logs and equipment execution feedback.

[0030] Subsequently, when a temporary occupancy is detected, the underlying basic spatial topology is not directly modified. Instead, the physical occupancy is abstracted into an independent temporary topology constraint object, and an association index is established between this object and the affected unit of the underlying topology. This association index preferably indicates at least the affected channel segment edge identifier, storage location node identifier, impact type, and impact status.

[0031] Subsequently, a hierarchical topology view composed of a bottom baseline layer and an upper constraint layer is constructed in memory. Before executing a specific outbound task, the spatial influence of the temporary constraint object is extracted and converted into a quantified passage cost, which is then superimposed on the topology edge. On this hierarchical topology view, the reachability and connectivity algorithm is used to verify the target outbound point. The verification preferably first determines whether the target storage node still has working space, and then determines whether there is a continuous passable path from the current device starting point to the target point.

[0032] Finally, depending on whether the check is smooth, restricted, or completely blocked, a feasible path is adaptively output or the task state is adjusted. Furthermore, after temporary occupancy is eliminated or the task loop is closed, constraints are seamlessly revoked, achieving zero-intrusion rollback of the topology. This approach is because temporary stacking is a high-frequency, short-term, localized disturbance. Directly rewriting the basic spatial topology structure could easily lead to frequent topology version changes, difficulties in rollback, and state pollution under multi-task concurrency. Using a layered overlay approach allows for timely incorporation of dynamic changes into the task decision-making chain without compromising the stability of the underlying layer.

[0033] Specifically, the basic spatial topology can be pre-configured and loaded into an in-memory database using a standard topology graph file format (e.g., adjacency list or adjacency matrix). Each node contains three-dimensional global spatial coordinates (X, Y, Z), and each edge contains connection endpoint identifiers and standard physical passage widths. Furthermore, to ensure consistency between this basic spatial topology and the actual on-site layout, a unified topology master data is formed through joint modeling using on-site warehouse survey data, rack installation diagrams, aisle planning diagrams, and storage location coding information from the warehouse management system. When permanent changes occur on-site, such as rack additions or removals, aisle modifications, or platform location adjustments, the master data is updated by the maintenance process. Temporary storage occupancy information can be accessed through multiple sources, including fixed scanning units deployed on-site, and manual reporting of loading coordinates and approximate occupancy radii via handheld terminals, all connected to the industrial internet data bus. More specifically, the fixed scanning unit can be a surveillance camera installed above the passageway or a laser scanning device deployed at key intersections; the handheld terminal can be a forklift-mounted tablet, a PDA terminal, or an industrial tablet, and its reported content preferably includes the cargo release area identification, cargo release time, approximate shape of the cargo, and operator confirmation information.

[0034] More specifically, when instantiating the temporary heap occupancy information and establishing the mapping relationship, an independent memory block is allocated in the memory dynamic model area to generate a temporary topology constraint object instance containing a unique identifier (ID), occupancy space envelope data (such as the minimum bounding rectangle or center point coordinates plus influence radius), and a timestamp.

[0035] Furthermore, the temporary topology constraint object preferably includes a lifecycle status field, a sensing source field, a confidence status field, and a release trigger condition field, wherein the lifecycle status can be at least distinguished as pending evaluation, in effect, in verification, and completed. To establish a mapping relationship, based on the spatial coordinates of the temporary topology constraint object, the spatial geometric segments of all channel segments in the basic spatial topology are traversed, and the Euclidean spatial distance between the geometric segments and the temporary occupancy envelope is calculated. When the distance is less than a preset spatial impact threshold, a binding record is inserted into the mapping relationship table, establishing a key-value pair mapping between the temporary topology constraint object ID and the ID of the channel segment edge. To ensure the feasibility of this preset spatial impact threshold, this threshold is preferably not arbitrarily given, but rather determined jointly based on the standard channel width, equipment outline width, safety lateral margin, and sensing boundary error.

[0036] In other words, it is preferable to first read the standard physical width of the corresponding channel segment edge from the basic spatial topology, then read the safe passage width requirement of the currently executing device or the main executing device from the equipment file, and combine this with the possible positioning deviation of the sensing boundary to obtain a search range for initially screening potentially affected edges. For storage location nodes, a similar mapping method can be used to establish a mapping, that is, to determine whether the temporary occupancy envelope enters the adjacent range of the working space or turning space directly in front of the storage location. If it does, an association record between the temporary topology constraint object and the storage location node is added to the mapping relationship table. The mapping relationship table preferably includes at least the constraint object ID, topology object ID, topology object type, impact status, creation time, and last update time, so that the constraint layer can quickly read and revoke these fields.

[0037] When constructing the hierarchical topology view and adding access costs, the underlying basic spatial topology is used as the resident graph structure, and the upper dynamic temporary access constraint layer maintains the set of instantiated temporary topology constraint objects and the mapping table. Before performing connectivity verification, the affected channel segments in the mapping table are read, and additional impedance weights are assigned to the affected channel segments based on the relative amount of space temporarily occupied by the intruding channel. For example, if the intrusion causes access difficulties, the original basic access distance weight is multiplied by a penalty coefficient or the resistance weight is directly accumulated; if the intrusion causes the channel to be unable to accommodate equipment access, a maximum value is assigned to represent logical blockage.

[0038] Specifically, it is preferable to first construct the ideal passage space region for each potentially affected passage segment edge, and then project the physical contour of the temporary topological constraint object onto the same plane coordinate system to determine whether there is actual overlap between the two. If there is no overlap, it only indicates that the temporary pile is located near the passage but has not invaded the effective passage space. In this case, the edge can be marked as having no impact and the initial screening mapping can be removed. If there is overlap, it is further determined whether the remaining effective passage width after invasion still meets the safe passage requirements of the current equipment. If the remaining effective passage width is insufficient, the edge is marked as impassable; if it is still passable but the space is significantly compressed, the edge is marked as restricted passage, and corresponding penalty weights and necessary speed limit marks are added. For the target storage location node, it is also preferable to determine whether the temporary pile invades the necessary working space in front of the storage location; if the invasion causes the equipment to be unable to complete the picking, turning or positioning actions, the node is marked as unoperable. The connectivity verification is performed by using a graph traversal algorithm (such as Dijkstra's algorithm) on the layered topological view superimposed with the passage cost to calculate the minimum cost path from the current equipment's starting point to the target outbound point. Furthermore, to improve computational efficiency, it is preferable to first extract a local subgraph around the starting point and target point of the current task to form a restricted topological slice oriented towards the current task, and then perform path search on the slice.

[0039] If the connectivity check successfully finds a continuous path with a total cost less than the blocking threshold, the path is converted into a motion control coordinate sequence and sent to the operating equipment. If the check result indicates that the path is completely blocked or the total cost exceeds the allowable threshold, the outbound task is put into a suspended or blocked state to prevent the equipment from blindly deploying. Furthermore, when the path contains restricted passage edges, it is preferable to include the execution semantic information of the restricted road segment in the issued path instructions, such as speed limit requirements, local obstacle avoidance high sensitivity prompts, or no-passing prompts, so that the equipment can adopt a more conservative control strategy when passing through the temporary stack adjacent area.

[0040] When the verification result is unreachable, in addition to putting the task into a suspended or blocked state, the inventory substitution search logic can be further triggered to query whether there are other alternative fixed storage locations that store the same goods and are currently topologically reachable. If they exist, the task target is redirected to the alternative storage location and the connectivity verification is re-executed. If they do not exist, the task is kept suspended and a blocking alarm message is pushed to the on-site management terminal. The alarm content preferably includes the coordinates of the blocked area, the edge identifier of the affected passage segment, and the impact range of the temporary pile. When the temporary pile is moved and a cleanup completion trigger signal is received, the corresponding temporary topology constraint object is deregistered in the dynamic temporary access constraint layer, and the corresponding key-value record in the mapping table is cleared. The passage segment edge in the underlying basic spatial topology structure is then automatically restored to its original basic weight, achieving a complete restoration of the original connectivity state. The cleanup completion trigger signal preferably comes from at least one of three sources: first, the associated cleanup task completion receipt; second, the multi-source sensing does not detect the occupying entity again within a continuous time window; and third, the verification scan confirmation result triggered after timeout.

[0041] By introducing connectivity verification based on hierarchical topology view during the outbound task planning stage, the above technical solution ensures that the issued path instructions truly reflect the capacity of the physical channels on site. This eliminates problems such as equipment obstruction, on-site congestion, and task stagnation caused by the mismatch between static topology and dynamic on-site occupancy, thereby improving the reliability and smoothness of industrial internet warehouse location management.

[0042] Furthermore, the steps for obtaining the basic spatial topology of the storage area and the temporary storage occupancy information within the target storage location area include: The industrial internet platform obtains the basic spatial topology structure, which is organized in the form of a graph and includes storage location nodes, channel intersection nodes, platform handover nodes, and channel and alleyway segments connecting each node. Acquire monitoring video streams, LiDAR scanning data, and temporary cargo release event logs through industrial internet edge computing nodes; Spatial fusion verification is performed on the monitoring video stream, LiDAR scanning data, and temporary cargo release event logs to identify temporary occupancy awareness events. The temporary occupancy awareness events and the coordinates of the physical contour polygons of the temporary occupancy awareness events are used as temporary land storage occupancy information.

[0043] Specifically, the industrial internet platform maintains a baseline map structure that includes warehouse location nodes, aisle intersection nodes, platform handover nodes, and aisle and lane segment edges. During data acquisition, surveillance cameras deployed above the aisles collect high-level video streams, while moving material handling equipment (such as AGVs) collects point cloud data via vehicle-mounted LiDAR during mobile scanning. Manual forklifts' onboard tablets then report real-time delivery event logs. Upon receiving these three types of data streams, edge computing nodes first perform a unified spatial coordinate system transformation, mapping the video pixel coordinates, local radar point cloud coordinates, and log coordinates to the warehouse's global world coordinate system (geocentric coordinate system or factory measurement coordinate system). To achieve this coordinate unification, it is preferable to pre-establish the correspondence between camera extrinsic calibration results, radar installation pose parameters, coordinates reported by manual terminals, and warehouse area codes.

[0044] The preset mapping relationship can be obtained through on-site calibration. For example, several marker points with known absolute coordinates are selected on the warehouse floor, and these marker points are photographed using a camera. The mapping from pixels to ground coordinates is then completed by combining parameters such as installation height and pitch angle. For vehicle-mounted LiDAR, the local scan coordinates are converted to the vehicle coordinate system through equipment installation pose calibration, and then converted to the warehouse global coordinate system by combining the vehicle's real-time positioning results. For manual logs, the coordinates are mapped to a unified world coordinate system by selecting the area code, the temporary storage area number, or manually inputting the coordinates at the terminal. For the video stream, background subtraction combined with target detection algorithms is used to extract the foreground boundary of the stacked goods image and project it onto the two-dimensional plane of the ground. Here, it is preferable to first use the background stabilization characteristics of the fixed camera to extract the newly added foreground area, and then combine the target detection results to filter out high dynamic targets such as personnel and forklifts, retaining only the stacked goods outline that is continuously present in non-standard shelf locations.

[0045] For point cloud data, after filtering out ground and known shelf point clouds, Euclidean clustering analysis is performed on the remaining obstacle point clouds to extract the convex hull polygons of the ground's two-dimensional projection. The filtering of known shelf point clouds can be performed by matching and eliminating them based on the spatial model of fixed facilities in the basic spatial topology, while the removal of mobile device point clouds can be performed based on the device's own shape template and real-time pose. For the delivery log, its specified coordinates and expected dimensions are parsed. Edge computing nodes calculate the spatial overlap of the above three types of spatial polygons within the same time window (e.g., calculating the spatial intersection-union ratio, IoU). When the overlap exceeds a set threshold or the multi-source data spatial verification is consistent, a temporary occupancy awareness event is generated, and the coordinate sequence of convex hull polygon vertices obtained from the final fusion calculation (e.g., a series of clockwise arranged two-dimensional absolute coordinate points [X1,Y1],[X2,Y2],...,[Xn,Yn]) is encapsulated as temporary stacking occupancy information. The threshold settings here are preferably configured based on the on-site perception accuracy, camera coverage angle, radar point cloud density, and manual log coordinate granularity. If a uniform value cannot be predetermined, different levels such as threshold A and threshold B can be configured according to the region type. In addition to the above consistency confirmation method, when a temporary release event is explicitly reported manually and the reporting area is adjacent to any abnormal area in the video or point cloud, a temporary occupancy perception event can also be directly confirmed to generate, so as to balance timeliness and reliability.

[0046] Furthermore, the step of instantiating temporary heap occupancy information into temporary topological constraint objects independent of the underlying spatial topology includes: Create a temporary topological constraint object to record the coordinates of the physical contour polygon and its lifecycle state for temporary occupancy-aware events, wherein the temporary topological constraint object is stored independently of the underlying spatial topology structure.

[0047] Specifically, a separate temporary constraint object manager is maintained in memory. When the temporary occupancy awareness event and its physical contour polygon coordinates are received from the previous stage, the constraint manager calls the object-oriented data structure constructor to instantiate a temporary topological constraint object. The data structure of this object encapsulates geometric contour polygon attributes containing an array of vertex coordinates, lifecycle state enumeration variables with an initial state of being active, and metadata such as generation timestamp and awareness source confidence.

[0048] Furthermore, temporary topology constraint objects preferably include the following fields: an object type field, used to identify whether the object belongs to a temporary storage area, faulty equipment, or other temporary occupancy type; an area of ​​effect field, used to record its storage partition, channel number, or warehouse front area number; an impact summary field, used to record the number of currently mapped affected channel segments and warehouse nodes; and a release condition field, used to record the triggering basis for the object to be released subsequently, such as the completion of a cleanup task, the disappearance of continuous sensing, or confirmation by a review scan. This instance is allocated in storage in a memory set or dedicated cache table independent of the basic spatial topology graph structure, without directly inserting any intrusive physical entities into the node list or edge list of the basic topology graph. To facilitate subsequent retrieval and lifecycle management, the dedicated cache table is preferably indexed according to the object identifier, area identifier, and status field; when a task verification needs to be performed on a certain area, all temporary topology constraint objects in effect or under verification within that area can be quickly read.

[0049] Furthermore, after an object is created, it is not immediately treated as a globally valid constraint. Instead, it first enters the mapping and impact assessment process. Only after confirming that it has indeed intruded into the valid passage space of the channel or the necessary working space of the storage location is its status updated to a valid constraint object participating in task judgment. The reason for this approach is that although temporary occupancy awareness events reflect the existence of abnormal stacking on site, not all abnormal stacking will affect the current task path. Only after spatial mapping and impact assessment can we avoid misjudging occupancy that is close to the channel but has not intruded into the channel as a topological constraint that must be blocked.

[0050] The above technical solutions ensure the high stability of the underlying static basic topology and provide object-oriented operation support for the upper layer to flexibly apply, dynamically modify, and losslessly cancel temporary access constraints.

[0051] Furthermore, the steps for establishing the mapping relationship between temporary topological constraint objects and the underlying spatial topology include: Using the geometric center of the physical contour polygon coordinates recorded by the temporary topological constraint object as the center, and a preset search radius, a spatial range search is performed in the basic spatial topology structure to filter out all topological nodes and topological edges that fall within the preset search radius as potentially affected topological objects. A mapping relationship is established between temporary topological constraint objects and basic spatial topology. The temporary topological constraint objects are associated with the channel segments and storage nodes in the screened potentially affected topological objects and recorded in the mapping relationship.

[0052] Specifically, the coordinates of the physical contour polygon recorded in the temporary topological constraint object represent the actual planar projection boundary of the on-site pile. The average coordinates of all vertices of the polygon or the center of the circumscribed rectangle are calculated using a polygon geometry algorithm and used as the geometric center. If the temporary occupancy boundary comes from video detection results, the edge nodes first convert the pixel boundaries in the image into unified world coordinates of the warehouse site through camera calibration parameters, and then output a two-dimensional vertex sequence. If the temporary occupancy boundary comes from the LiDAR point cloud, the fixed shelves and mobile equipment body point clouds are first filtered out, and then the remaining static obstacle points are clustered and their ground projection convex hulls are extracted as polygon boundaries.

[0053] If the temporary occupancy boundary is reported by a manual forklift terminal, a rectangular or polygonal outline can be generated based on the reported coordinates and a preset occupancy template. The preset search radius is not an arbitrary empirical value, but is calculated by comprehensively superimposing the maximum physical width of the standard aisle in the warehouse, the outer envelope radius of typical handling equipment, and the system error tolerance of the sensing and positioning system. This ensures that all aisle segments and storage location nodes that may interfere with the stack are completely covered. During the deployment phase, a search radius parameter table can be generated based on the warehouse area configuration file. For example, different search radii can be configured for the main aisle area, branch aisle area, and temporary storage area in front of the storage area. When the standard aisle of a certain area is wider and the turning envelope of the equipment is larger, the corresponding search radius increases accordingly. Spatial range retrieval is preferably based on a spatial index tree (such as a two-dimensional R-Tree or spatial grid index) pre-built from the basic spatial topology structure. By quickly intersecting the circular or circumscribed rectangular envelope frame formed by the geometric center and the preset search radius with the spatial index tree, topological nodes and topological edges falling within the range are instantly filtered as potentially affected topological objects.

[0054] The basic spatial topology here is preferably derived from fixed topology data maintained by the warehouse management system and warehouse control system. This includes at least fixed shelf location nodes, aisle intersection nodes, platform handover nodes, and the aisle and lane segments connecting these nodes, with each node and edge carrying uniform coordinate information. The mapping relationships are maintained in the form of a memory-mapped table. Each record in the table contains a temporary constraint object identifier, the associated aisle or storage location node identifier, an association timestamp, and an initial mapping status field. To facilitate subsequent detailed evaluation, the mapping relationship table can further record auxiliary fields such as object origin, region, most recent update time, and whether geometric verification has been completed. This process of first retrieving and then associating allows for the compression of the massive amount of primitives across the entire site into a local candidate set, and then the establishment of structured relationships within that candidate set. This reduces computational load and provides a clear entry point for subsequent edge-by-edge and point-by-point impact analysis.

[0055] This specific implementation plan provides a clear structured data carrier for subsequent accurate assessment of the degree of compression of each channel segment and the operational availability of the target storage location, ensuring the pertinence of spatial interference judgment and the efficiency of calculation.

[0056] Furthermore, the steps of superimposing a dynamic temporary access constraint layer on the basic spatial topology based on the mapping relationship, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view, include: A dynamic temporary access constraint layer is constructed based on the mapping relationship, which includes temporary topological constraint objects and channel segment edges and storage location nodes in potentially affected topological objects. Using the basic spatial topology as the bottom layer and the dynamic temporary access constraint layer as the upper layer to carry all temporary topology constraint objects, a hierarchical topology view is formed in which the basic spatial topology and the dynamic temporary access constraint layer are logically superimposed but physically separated in storage.

[0057] Specifically, the design philosophy of the layered graph model is that the baseline layer is resident and read-only, while the constraint layer is dynamic and detachable. The dynamic temporary access constraint layer is an independent logical layer opened in system memory. It does not contain complete static channel and node definitions for the entire field, but only maintains instances of temporary topological constraint objects that are currently in effect or awaiting evaluation, as well as a high-dimensional constraint attribute dictionary of the affected channel segments and storage nodes associated with these objects through the aforementioned mapping relationship. This high-dimensional constraint attribute dictionary preferably includes object identifier, constraint status, associated edge or node identifier, blocking flag, passage resistance penalty value, speed limit parameter, creation time, last update time, and release condition flag. The underlying basic spatial topology structure serves as the sole real source of facts for the entire fixed infrastructure and remains resident and read-only in physical memory. This basic spatial topology structure can be maintained by a central graph database, relational database, or graph computing engine, and the static attributes in its node and edge tables are not directly rewritten during daily task calculations due to temporary heaps.

[0058] When a topology calculation is initiated, the underlying graph data is not modified. Instead, the constraint parameters recorded in the dynamic temporary access constraint layer are logically superimposed onto the corresponding primitives of the basic spatial topology in the calculation execution stack, synthesizing a hierarchical topology view with dynamic accessibility assessment capabilities in real time. First, the basic nodes and edges within the task-related local area are read. Then, the corresponding dynamic constraint records are found according to the mapping relationship. Finally, the state fields in the constraint records are overwritten or appended to the local primitive copy, forming a view snapshot visible only to the current calculation process. This process ensures that even if multiple temporary constraints act on different regions simultaneously, or multiple temporary constraint objects are updated concurrently in the same region, the consistency of the underlying basic spatial topology will not be compromised. Furthermore, when a temporary topology constraint object enters a released state, simply deleting or invalidating the corresponding record in the dynamic temporary access constraint layer automatically restores subsequent calculations to the underlying default state, without requiring complex rollbacks to the underlying graph database.

[0059] Through this specific implementation plan, the underlying basic spatial topology can maintain extremely high stability and concurrent reading efficiency for a long time, while the upper dynamic constraint layer can dynamically load and unload materials in real time according to the placement and removal of materials on site, providing a guarantee for the safe and lossless restoration of the original state of the topology.

[0060] Furthermore, the steps of adding passage costs to the channel segment edges in the basic spatial topology based on the dynamic temporary passage constraint layer include: For each channel segment edge in the dynamic temporary access constraint layer, extract the standard physical width and safe access width threshold of the channel segment edge in the basic spatial topology; The ideal passage space area of ​​this passage segment edge is constructed based on the standard physical width and the safe passage width threshold; Calculate the overlap area between the temporary topological constraint object and the ideal passage space region based on the coordinates of the physical contour polygon recorded by the temporary topological constraint object; If the area of ​​the overlapping region is zero, mark the edge of the channel segment as having no effect and remove it from the mapping relationship; If the area of ​​the overlapping region is greater than zero, calculate the maximum intrusion depth of the overlapping region in the direction perpendicular to the centerline of the channel segment, and calculate the remaining effective passage width of the channel segment after being temporarily occupied and squeezed based on the difference between the standard physical width and the maximum intrusion depth. When the remaining effective passage width is less than the safe passage width threshold, a preset blocking penalty value is applied to the edge of the passage segment; When the remaining effective passage width is greater than or equal to the safe passage width threshold but less than the standard physical width, a preset passage resistance penalty value is added to the edge of the passage segment. When the remaining effective passage width is greater than or equal to the standard physical width, no additional passage cost is added to the edge of the passage segment.

[0061] Specifically, for each associated channel segment edge, the centerline segment coordinates and standard physical width W are extracted from the underlying topology, and the safe passage width threshold W_safe of the main equipment is read from the field execution equipment parameter file (this threshold includes the equipment body width plus the legally mandated safety clearance margin on both sides). This equipment parameter file can be pre-registered when the equipment is connected to the industrial internet platform, such as recording AGV type, vehicle width, minimum safe lateral clearance, recommended speed limit range, etc.; the corresponding safe passage width threshold can be selected according to the main equipment type of the area to avoid misjudgment caused by different equipment using the same threshold. The ideal passage space area is a two-dimensional rectangular polygon formed by extending W / 2 to both sides with the channel centerline as the reference axis. If the channel is a regular straight line segment, the rectangular area can be directly generated based on the centerline start point, end point, and width; if the channel has corners or broken lines in some areas, the corresponding ideal passage space areas can be constructed separately according to the multiple channel segment edges split from the edge table. Using the polygon Boolean intersection algorithm in two-dimensional computational geometry, the intersection of the pile outline polygon and the ideal passage space rectangle of the channel is calculated to obtain the overlapping area.

[0062] If the overlap area is zero, it means the pile has not encroached on the passageway, and the redundant mapping is immediately removed. If the overlap area is greater than zero, all boundary points of the overlapping polygon are extracted, and a vertical projection is made onto the passageway centerline. The maximum projection distance is calculated as the maximum intrusion depth D. First, the direction vector of the passageway centerline is read to determine the lateral measurement direction perpendicular to it; then, the boundary points of the overlapping area are traversed, and the lateral offset relative to the centerline is calculated point by point; finally, the largest offset is taken as the deepest intrusion degree of the temporary pileway into the passageway. The remaining effective passageway width W_r after the passageway is squeezed is then precisely quantified as W_r = WD.

[0063] Subsequently, a three-level cost judgment is performed: if W_r < W_safe, it indicates that the physical passage is severely occupied and cannot be passed safely, and a preset blocking penalty value (for example, set to an infinite impedance weight INF) is assigned to this passage edge, so that it behaves as a logical disconnection in the topology algorithm; if W_safe ≤ W_r < W, it indicates that although the passage is narrowed, it still has passage conditions, and a preset passage resistance penalty value is assigned to this edge according to the compression ratio of the width (for example, a positive penalty weight ΔCost is added on the basis of the basic length weight), so as to prompt the path algorithm to prefer a wider alternative route, and allow speed-limited passage when there is no alternative route; if W_r ≥ W, it is regarded that the passage is not compressed, and the original basic cost remains unchanged. The preset blocking penalty value and the preset passage resistance penalty value herein are preferably configured according to the warehouse operation strategy during the system deployment phase: the blocking penalty value can be uniformly set as a very large weight that is unacceptable to the graph search algorithm, for expressing absolute impassability; the passage resistance penalty value can be configured into several levels according to the restriction level, for example, mild restriction, moderate restriction, and severe restriction correspond to different additional cost values and different speed limit parameters respectively. In this way, the output is not only the conclusion of whether passage is possible, but also the execution information of at what cost to pass and whether speed limit is required, so that the topology judgment result can directly serve the equipment control side.

[0064] For specific reference, take the passage segment edge EDGE_101 as an example, its standard physical width W is 3.0 meters, and the safe passage width threshold W_safe of the on-site AGV equipment is 1.8 meters. This 1.8-meter threshold can be comprehensively determined by the vehicle body width in the equipment file and the safety clearance requirements on both sides, and is used as the standard safe passage condition for the main AGVs in this area. An ideal passage rectangular space Poly_A with a width of 3.0 meters is constructed based on the center line of EDGE_101. The intersection of the contour polygon Poly_B of the temporary stacking object OBJ_001 and Poly_A is calculated, and the overlapping area is 0.85 square meters (greater than zero). The deepest intrusion point of the overlapping area in the direction perpendicular to the center line of EDGE_101 is calculated to obtain the maximum intrusion depth D=0.8 meters, and the remaining effective passage width W_r=3.0-0.8=2.2 meters is calculated. The comparison finds that W_safe (1.8 meters) ≤ W_r (2.2 meters) < W (3.0 meters), which falls into the restricted passage interval.

[0065] In the dynamic constraint layer, mark EDGE_101 as restricted passage, assign the preset passage resistance penalty value (for example, the base cost is 10, the additional penalty value is 15, and the total cost is updated to 25) to EDGE_101, and meanwhile attach an execution control parameter of speed limit 0.8m / s to the attribute of this edge. This speed limit parameter can be pre-configured by the warehouse operation strategy, for example, it is specified that within the restricted passage section, the AGV must pass at a safe speed lower than the default cruising speed to reduce the risk of scratches and collisions. Assume that the maximum intrusion depth D calculated for another passage segment edge EDGE_102 is 1.5 meters, and its remaining effective passage width W_r=3.0-1.5=1.5 meters < W_safe (1.8 meters), then directly assign the preset blocking penalty value INF (infinity) to EDGE_102, and mark it as impassable. If there is a candidate edge whose overlapping area after intersection with OBJ_001 is zero, disassociate this edge from the mapping relationship, and it will no longer enter the subsequent path calculation, so as to avoid invalid constraints remaining in the dynamic temporary passage constraint layer for a long time.

[0066] Through this specific embodiment, the waste of passage capacity caused by misjudging slight intrusion as absolute blocking is avoided, and the safety hazard of equipment forcing through the site and causing scratches and collisions due to space compression misjudgment is completely eliminated, so that the passage cost in the layered topology view can truly and accurately reflect the bearable capacity of the on-site physical space.

[0067] Further, the step of performing connectivity check on the target warehouse-out point on the layered topology view includes: Extracting a local subgraph including all alternative connected paths from the start point of the warehouse-out task to the target warehouse-out point from the layered topology view; Superimposing the preset blocking penalty value or preset passage resistance penalty value of each passage segment edge in the dynamic temporary passage constraint layer onto the corresponding passage segment edge in the local subgraph; Taking the passage segment edge superimposed with the preset blocking penalty value as a disconnected edge in the connectivity check, taking the passage segment edge superimposed with the preset passage resistance penalty value as a high-cost passable edge in the connectivity check, taking the passage segment edge not attached with a passage cost as a default passable edge in the connectivity check, and searching for a connected path from the start point of the warehouse-out task to the target warehouse-out point on the local subgraph; If a connected path is found, it is determined that the target warehouse-out point passes the connectivity check, otherwise it is determined that the target warehouse-out point does not pass the connectivity check.

[0068] Specifically, firstly, based on the starting point of the outbound task (e.g., the current docking point of the equipment or the outbound platform) and the target outbound storage location node, a local subgraph containing all feasible topological branches between the two points is extracted from the global hierarchical topology view using a bidirectional search or a graph extraction algorithm based on geometric distance envelope, forming a restricted topology slice oriented towards the current outbound task. The starting point and target outbound point can be directly derived from the task order generated by the warehouse management system, where the task order at least includes fields such as task identifier, starting node identifier, target storage location node identifier, task priority, and equipment type.

[0069] To ensure that the local subgraph neither misses feasible paths nor excessively expands the search range, the following extraction method can be adopted: First, a geometric envelope region is determined based on the starting point and the target exit point. Then, the nodes and edges within this region are used as the initial candidate set. Subsequently, several layers of adjacency relationships are extended outward along the basic spatial topology until all possible bypass branches are covered. Then, only for the channel segments within this local subgraph, the corresponding cost parameters are extracted from the dynamic temporary passage constraint layer and superimposed. On this local subgraph, edges superimposed with blocking penalty values ​​(infinite) are considered logically disconnected edges (untraversable) by the graph search algorithm; edges superimposed with passage resistance penalty values ​​are considered high-cost traversable edges; edges unaffected by the ground stack retain their original distance or time weight and are used as default traversable edges.

[0070] If the storage node corresponding to the target outbound point has been marked as unworkable in the previous processing, it can be determined that the node is inaccessible before executing the path search, thus directly outputting the result of failing the connectivity check and avoiding meaningless pathfinding. Based on this, a shortest path optimization algorithm based on a priority queue (such as Dijkstra's algorithm or the heuristic A* algorithm) is used to search for a path with the minimum cumulative cost from the starting point to the target outbound point within the local subgraph. If the algorithm successfully converges and returns a continuous sequence of edges with a finite total cost, it indicates that although there is a temporary ground disturbance locally, there is still a connected channel that can avoid the blocking area or has an acceptable cost, and the target outbound point is determined to have passed the connectivity check. If the algorithm returns that the path does not exist or the total cost is infinite after traversing all possible branches of the local subgraph, it indicates that all alternative paths to the target point are completely blocked by the ground disturbance, and the target outbound point is clearly determined to have failed the connectivity check. By slicing, overlaying, and then pathfinding, the impact of dynamic constraints can be limited to the graph that is truly relevant to the current task, thus balancing real-time performance and judgment accuracy.

[0071] This specific implementation plan maps the graded access costs directly to disconnected edges, high-cost edges, and default access edges, enabling the graph pathfinding algorithm to strictly follow the physical constraints of the site. It can adaptively avoid dead ends completely blocked by the ground pile and make a global optimal trade-off between unobstructed paths and locally restricted paths.

[0072] Furthermore, the steps for determining the execution path or task status of the outbound task based on the connectivity verification results include: When the target outbound point is determined to have passed the connectivity check, the searched connectivity path is smoothed and semantically transformed to generate the execution path instruction. When the target outbound location fails the connectivity check, the task status of the outbound task is marked as suspended, and it is checked whether there are other alternative storage location nodes that store the same material information and are topologically connected.

[0073] Specifically, when the target outbound point is determined to be connected, the connected path searched by the algorithm initially appears as a sequence of geometric centerlines of discrete topological nodes and channel segments. To adapt this path to the actual dynamic steering constraints and narrow space passage requirements of on-site handling equipment (such as AGVs, AMRs, or forklifts), curvature continuity path smoothing processing is performed on the discrete path sequence (e.g., using B-spline curve fitting or Bézier curve optimization algorithms) to eliminate abrupt curvature changes at the corners of the broken lines. Subsequently, semantic transformation is performed to fuse the smoothed geometric trajectory with the dynamic constraint attributes of each channel segment edge in the layered topology view, encapsulating and generating an execution path instruction that includes segmented speed limit labels, obstacle avoidance sensitivity parameters, intersection right-of-way signs, and action instruction sets (such as straight driving, deceleration turning, and warehouse front-end positioning and retrieval). This instruction is then sent to the corresponding handling equipment's on-board controller via the Industrial Internet, driving the equipment to strictly complete the operation according to the procedures.

[0074] Conversely, if the verification determines that the target outbound point is unreachable, the blind issuance of the outbound task is immediately blocked, its lifecycle status is changed from pending execution to suspended, and adaptive inventory scheduling and substitution strategies are triggered in the control center. The material code, specifications, batch number, and other material attributes associated with the current outbound task are read. The inventory database of the warehouse management system is searched for all candidate fixed storage location nodes in the current storage area that contain the same material and are in an available state. Subsequently, in the current hierarchical topology view, a rapid reachability and connectivity assessment is initiated for each of these candidate storage location nodes. The storage location with the lowest topological connectivity and the lowest overall passage cost is selected as the alternative storage location. The target point of the outbound task is redirected to this alternative storage location, and the path smoothing and instruction generation process is re-entered. If the query confirms that there is no alternative connected storage location on-site, the task is kept in a suspended state, and a site cleanup collaboration work order containing the blocked passage identifier and occupancy outline is pushed to the monitoring dashboard or manual terminal.

[0075] Through the above technical solutions, when the verification is successful, smoothing and semantic conversion ensure the kinematic feasibility and on-site safety of the issued instructions. When the verification fails, task suspension and active search for alternative storage locations avoid the negative impact of direct task error interruption on the overall outbound cycle time, thus maximizing the continuity and throughput efficiency of warehouse outbound operations.

[0076] Furthermore, the steps for restoring the basic spatial topology to its original connectivity state before the dynamic temporary access constraint layer was superimposed, after the effective lifecycle of the temporary topology constraint object ends or according to the task status of the outbound task, include: When the effective lifecycle of a temporary topology constraint object ends or when the outbound task status is completed, all temporary topology constraint objects are destroyed, and the passage cost attached to the channel segment edge in the basic space topology is revoked, so that the basic space topology is restored to the original connected state before the dynamic temporary passage constraint layer was superimposed.

[0077] Specifically, the lifecycle of a temporary topology constraint object can be triggered by various events: for example, an industrial internet edge node confirms through continuous multi-frame monitoring video analysis or lidar rescanning that obstacles in the corresponding physical space have completely disappeared; or the warehouse control system receives a completion receipt for the relocation work order for the temporary stack; or the object's preset maximum effective lifetime (TTL) countdown reaches zero and verification confirms no physical existence. When the above lifecycle end signal is detected, or when the outbound task associated with the affected area has been confirmed to be safely completed and the on-site occupation has been synchronously eliminated, the corresponding temporary topology constraint object instance is unregistered in the memory management area and its occupied memory space is reclaimed. At the same time, all binding records between the object and the affected channel segment edge and storage location node are synchronously deleted in the mapping relationship table. With the clearing of the mapping association, the dynamic temporary access constraint layer automatically cancels the previously injected blocking penalty value, passage resistance penalty value, and speed limit and other extended attributes to the corresponding channel segment edge. Since the underlying basic spatial topology is saved in a read-only and decoupled form throughout the entire operation, after the upper-level constraints are removed, the edges and nodes of each channel segment in the underlying topology graph are immediately and seamlessly restored to their original standard physical width, rated passage speed and default weight. No complex rollback queries or reconstruction operations need to be performed on the basic graph database throughout the entire process.

[0078] The above technical solution enables dynamic constraints to be added and removed as needed, completely avoiding the problem of residual topology data state caused by high-frequency temporary occupation, ensuring the stability of the basic spatial topology at the physical storage level, and ensuring that the warehouse location management system always maintains the topology state consistent with the site.

[0079] Secondly, see Figure 2This application also proposes a warehouse location management system based on the Industrial Internet, including: Information acquisition module 210 is used to acquire the basic spatial topology of the storage area and the temporary stacking occupancy information in the area where the target outbound point is located through the Industrial Internet. The object instantiation module 220 is used to instantiate the temporary heap occupancy information into a temporary topological constraint object that is independent of the basic spatial topology, and to establish a mapping relationship between the temporary topological constraint object and the basic spatial topology. The view construction module 230 is used to overlay a dynamic temporary access constraint layer on the basic spatial topology based on the mapping relationship, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view. The connectivity verification module 240 is used to perform connectivity verification on the target outbound point on the hierarchical topology view by adding passage costs to the channel segment edges in the basic spatial topology based on the dynamic temporary passage constraint layer. The task decision and topology restoration module 250 is used to determine the execution path or task status of the outbound task based on the connectivity verification result, and restore the basic spatial topology to the original connectivity state before the dynamic temporary access constraint layer was superimposed after the effective life cycle of the temporary topology constraint object ends or according to the task status of the outbound task.

[0080] The above technical solutions solve the problems of equipment obstruction and task deadlock caused by the mismatch between dynamic temporary stacking and static topology, and significantly improve the reliability of industrial internet warehousing and logistics scheduling.

[0081] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A warehouse location management method based on the Industrial Internet, characterized in that, include: The Industrial Internet is used to obtain the basic spatial topology of the warehouse area and the temporary storage occupancy information within the area where the target outbound point is located. The temporary heap occupancy information is instantiated into a temporary topological constraint object that is independent of the basic spatial topology, and a mapping relationship between the temporary topological constraint object and the basic spatial topology is established. Based on the mapping relationship, a dynamic temporary access constraint layer is superimposed on the basic spatial topology, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view. Based on the dynamic temporary access constraint layer, an access cost is added to the channel segment edges in the basic spatial topology, and a connectivity check is performed on the target outbound point on the hierarchical topology view. The execution path or task status of the outbound task is determined based on the connectivity verification result. After the effective lifecycle of the temporary topology constraint object ends or based on the task status of the outbound task, the basic spatial topology is restored to the original connectivity state before the dynamic temporary access constraint layer is superimposed.

2. The warehouse location management method based on the Industrial Internet according to claim 1, characterized in that, The steps for obtaining the basic spatial topology of the storage area and the temporary storage occupancy information within the target storage location area through the Industrial Internet include: The industrial internet platform obtains the basic spatial topology structure, which is organized in the form of a graph and includes storage location nodes, channel intersection nodes, platform handover nodes, and channel and alleyway segments connecting each node. Acquire monitoring video streams, LiDAR scanning data, and temporary cargo release event logs through industrial internet edge computing nodes; Spatial fusion verification is performed on the monitoring video stream, the lidar scanning data, and the temporary cargo release event log to identify temporary occupancy awareness events. The temporary occupancy awareness events and the coordinates of the physical contour polygons of the temporary occupancy awareness events are used as the temporary land storage occupancy information.

3. The warehouse location management method based on the Industrial Internet according to claim 2, characterized in that, The step of instantiating the temporary heap occupancy information into a temporary topological constraint object independent of the basic spatial topology includes: Create a temporary topological constraint object for recording the coordinates of the physical contour polygon and its lifecycle state of the temporary occupancy-aware event, wherein the temporary topological constraint object is stored independently of the underlying spatial topology structure.

4. The warehouse location management method based on the Industrial Internet according to claim 3, characterized in that, The step of establishing the mapping relationship between the temporary topological constraint object and the basic spatial topology includes: Using the geometric center of the physical contour polygon coordinates recorded by the temporary topological constraint object as the center and a preset search radius as the center, a spatial range search is performed in the basic spatial topology structure to filter out all topological nodes and topological edges that fall within the preset search radius as potentially affected topological objects. A mapping relationship is established between the temporary topological constraint object and the basic spatial topology. The temporary topological constraint object is associated with the channel segments and storage nodes in the selected potentially affected topological objects and recorded in the mapping relationship.

5. A warehouse location management method based on the Industrial Internet according to claim 4, characterized in that, The step of superimposing a dynamic temporary access constraint layer onto the basic spatial topology based on the mapping relationship, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view, includes: Based on the mapping relationship, a dynamic temporary access constraint layer is constructed, which includes the temporary topology constraint object and the channel segment edges and storage location nodes in the potentially affected topology object; Using the basic spatial topology as the bottom layer and the dynamic temporary access constraint layer as the upper layer to carry all the temporary topology constraint objects, a hierarchical topology view is formed in which the basic spatial topology and the dynamic temporary access constraint layer are logically superimposed but physically separated in storage.

6. The warehouse location management method based on the Industrial Internet according to claim 5, characterized in that, The step of adding passage costs to the channel segment edges in the basic spatial topology based on the dynamic temporary passage constraint layer includes: For each channel segment edge in the dynamic temporary access constraint layer, extract the standard physical width and safe access width threshold of the channel segment edge in the basic spatial topology; The ideal passage space area of ​​the channel segment edge is constructed based on the standard physical width and the safe passage width threshold. The overlap area between the temporary topological constraint object and the ideal passage space region is calculated based on the physical contour polygon coordinates recorded by the temporary topological constraint object. If the area of ​​the overlapping region is zero, mark the channel segment edge as having no effect and remove it from the mapping relationship; If the area of ​​the overlapping region is greater than zero, calculate the maximum intrusion depth of the overlapping region in the direction perpendicular to the centerline of the channel segment, and calculate the remaining effective passage width of the channel segment after being temporarily occupied and squeezed based on the difference between the standard physical width and the maximum intrusion depth. When the remaining effective passage width is less than the safe passage width threshold, a preset blocking penalty value is added to the edge of the passage segment; When the remaining effective passage width is greater than or equal to the safe passage width threshold and less than the standard physical width, a preset passage resistance penalty value is added to the edge of the passage segment; When the remaining effective passage width is greater than or equal to the standard physical width, no additional passage cost is imposed on the edge of the channel segment.

7. A warehouse location management method based on the Industrial Internet according to claim 6, characterized in that, The step of performing connectivity verification on the target outbound point on the hierarchical topology view includes: Extract a local subgraph from the hierarchical topology view, including all alternative connected paths from the starting point of the outbound task to the target outbound point. The preset blocking penalty value or preset passage resistance penalty value of each channel segment edge in the dynamic temporary passage constraint layer is superimposed on the corresponding channel segment edge in the local subgraph. The channel segment edge with the preset blocking penalty value is used as the disconnected edge in the connectivity check, the channel segment edge with the preset passage resistance penalty value is used as the high-cost passable edge in the connectivity check, and the channel segment edge without the attached passage cost is used as the default passable edge in the connectivity check. A connected path from the starting point of the outbound task to the target outbound point is searched on the local subgraph. If the connected path is found, the target outbound point is determined to have passed the connectivity check; otherwise, the target outbound point is determined to have failed the connectivity check.

8. A warehouse location management method based on the Industrial Internet according to claim 7, characterized in that, The step of determining the execution path or task status of the outbound task based on the connectivity verification result includes: When the target outbound point is determined to pass the connectivity check, the searched connectivity path is smoothed and semantically transformed to generate an execution path instruction. When it is determined that the target outbound point fails the connectivity verification, the task status of the outbound task is marked as suspended, and it is queried whether there are other alternative storage location nodes that store the same material information and are topologically connected.

9. A warehouse location management method based on the Industrial Internet according to claim 1, characterized in that, The step of restoring the basic spatial topology to its original connectivity state before the dynamic temporary access constraint layer was superimposed after the effective lifecycle of the temporary topology constraint object ends or according to the task status of the outbound task includes: When the effective lifecycle of the temporary topology constraint object ends or when the task status of the outbound task is completed, all the temporary topology constraint objects are destroyed, and the passage cost attached to the channel segment edge in the basic spatial topology is revoked, so that the basic spatial topology is restored to the original connected state before the dynamic temporary passage constraint layer was superimposed.

10. A warehouse location management system based on the Industrial Internet, used to execute the warehouse location management method based on the Industrial Internet as described in any one of claims 1 to 9, characterized in that, The system includes: The information acquisition module is used to acquire the basic spatial topology of the storage area and the temporary storage occupancy information within the area where the target outbound point is located through the Industrial Internet. The object instantiation module is used to instantiate the temporary heap occupancy information into a temporary topological constraint object that is independent of the basic spatial topology, and to establish a mapping relationship between the temporary topological constraint object and the basic spatial topology. A view construction module is used to overlay a dynamic temporary access constraint layer on the basic spatial topology based on the mapping relationship, so that the basic spatial topology and the dynamic temporary access constraint layer together constitute a hierarchical topology view; The connectivity verification module is used to add passage costs to the channel segments in the basic spatial topology based on the dynamic temporary passage constraint layer, and to perform connectivity verification on the target outbound point on the hierarchical topology view. The task decision and topology restoration module is used to determine the execution path or task status of the outbound task based on the connectivity verification result, and restore the basic spatial topology to the original connectivity state before the dynamic temporary access constraint layer was superimposed, after the effective life cycle of the temporary topology constraint object ends or according to the task status of the outbound task.

Images