Drug inventory management method and system based on PDA (Personal Digital Assistant)

Abstract

The invention provides a PDA (Personal Digital Assistant)-based medicine inventory management method and system. According to the method, a three-dimensional shelf topological structure is constructed, a medicine validity period attenuation rate is modeled into a space-time gradient field, and shelf unit codes are generated; and calculating shelf coordinate offset based on the temperature and humidity fluctuation coefficient, establishing a virtual turnover sequence in the digital twin warehouse, generating a multi-layer topological navigation grid by matching space-time constraints and emergency channel passing parameters, and triggering grid reconstruction based on adjacent coupling parameters to avoid shelf displacement conflicts. Finally, the optimized navigation path is converted into an executable instruction set carrying temperature and humidity compensation parameters, PDA equipment is driven to project a path guiding signal, the positioning precision is verified in real time through the goods shelf coordinate offset, and a closed-loop feedback mechanism is formed. According to the technical scheme provided by the invention, high-precision navigation path generation and closed-loop verification under dynamic disturbance of the storage environment are realized, and the medicine turnover efficiency is remarkably improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of drug inventory management, and in particular, to a drug inventory management method and system based on PDA. Background Art

[0002] With the surging demand for the intelligent upgrade of logistics warehousing and the refined management of drug expiration dates, the warehousing system needs to monitor the temperature and humidity fluctuations in real time and map them to the physical deformation of the shelves to avoid shelf displacement and collision caused by environmental disturbances; it is necessary to synchronize the shelf coordinate offset, the gradient decay of the expiration date (such as the non-linear decay model of drug expiration date with temperature and humidity changes), and the emergency channel parameters in the digital twin to achieve dynamic path reconstruction; it is required to compare the PDA navigation signal with the coordinate offset of the digital twin in real time, and the positioning error needs to be controlled within ±3 cm to ensure the picking accuracy.

[0003] Existing solutions include using a fixed three-dimensional shelf model (such as the lifting and traversing shelf of Jiangsu Huicheng Optoelectronics), relying on preset parameter updates, and being unable to dynamically respond to micron-level deformations caused by temperature and humidity; temperature and humidity monitoring (such as LoRa wireless sensor network) and path planning (such as AGV navigation algorithm) belong to independent modules, lacking a coupling mechanism for shelf unit attribute coding and spatio-temporal constraints; existing expiration date predictions are mostly based on the assumption of a constant environment, and do not embed a gradient field model to quantify the non-linear impact of temperature and humidity fluctuations on the expiration date.

[0004] Existing solutions have the following defects: the update period of the traditional solution is up to the minute level, unable to capture the instantaneous offset of the shelf caused by sudden heat convection, and the collision risk increases by 42%; the path planning does not integrate the gradient field decay rate of the expiration date (such as the exponential relationship between the drug expiration date loss rate and temperature and humidity), resulting in a mismatch in the turnover priority of high-value goods and a 28% loss in warehousing efficiency; the PDA navigation relies on a single RFID positioning (accuracy ±10 cm), without dynamically compensating in combination with the shelf coordinate offset, and the deviation rate of the actual picking path from the digital twin exceeds the limit by 15%. Summary of the Invention

[0005] The embodiments of the present application provide a drug inventory management method and system based on PDA to solve the problem of low drug turnover efficiency in the prior art.

[0006] In a first aspect, the embodiments of the present application provide a drug inventory management method based on PDA, including:

[0007] Construct a three-dimensional shelf topological structure and embed an expiration date gradient field, and generate a dynamic attribute coding shelf unit carrying a temperature and humidity fluctuation coefficient and a neighboring coupling parameter in combination with the temperature and humidity data stream, where the neighboring coupling parameter represents the spatial interference relationship between adjacent shelf units;

[0008] Generate a shelf coordinate offset based on the temperature and humidity fluctuation coefficient, establish a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin, and form a spatiotemporal constraint relationship between the virtual turnover sequence and the validity period gradient field;

[0009] Dynamically matching the spatiotemporal constraint relationship with the emergency channel traffic parameters to generate a navigation grid, triggering grid topology reconstruction through the proximity coupling parameters and generating a navigation path with conflict markers;

[0010] The navigation path is converted into an executable instruction set to drive the PDA device to project a path guidance signal. The path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop.

[0011] Optionally, the generating of the shelf coordinate offset based on the temperature and humidity fluctuation coefficient, establishing a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin, wherein the virtual turnover sequence forms a spatiotemporal constraint relationship with the validity period gradient field, includes:

[0012] Establish a dynamic environmental response model of the temperature and humidity fluctuation coefficient and the three-dimensional shelf topology structure, generate a shelf coordinate offset by real-time correlation calculation between the physical position of the shelf unit and the temperature and humidity data stream, and drive the dynamic correction of the spatial coordinates of the shelf unit in the warehouse digital twin;

[0013] The shelf coordinate offset and the decay rate of the validity period gradient field are temporally and spatially superimposed to generate a priority weight distribution of the shelf unit, wherein the priority weight distribution quantitatively sorts the spatial interference intensity of the adjacent shelf units based on the proximity coupling parameter to form a scheduling logic of a virtual turnover sequence;

[0014] Dynamically adapting the scheduling logic of the virtual turnover sequence to the geometric constraints of the emergency channel passage parameters, dynamically adjusting the scheduling order of the shelf units through the interference intensity threshold triggered by the proximity coupling parameter, and forming a path rule compatible with the emergency channel passage parameters;

[0015] The path rule is parameter-bound with the scheduling logic of the virtual turnover sequence, and the spatial coordinates of the shelf units of the warehouse digital twin are dynamically updated to form a spatiotemporal constraint relationship between the virtual turnover sequence and the spatiotemporal attenuation rate of the validity period gradient field.

[0016] Optionally, dynamically matching the spatiotemporal constraint relationship with emergency channel traffic parameters to generate a navigation grid, triggering grid topology reconstruction through the proximity coupling parameters and generating a navigation path with conflict marks includes:

[0017] Dynamically matching the spatiotemporal constraint relationship with the geometric topological features of the emergency passage parameters, generating a multi-layer topological navigation grid based on the spatial interference intensity of the proximity coupling parameters, wherein the hierarchical distribution of the multi-layer topological navigation grid forms a dynamic binding relationship with the priority of the emergency passage parameters;

[0018] Calculating the path safety gap of the multi-layer topological navigation grid by using the interference intensity threshold triggered by the proximity coupling parameter to generate a path branch structure carrying a conflict mark, wherein the conflict mark forms a dynamic association constraint with the shelf coordinate offset;

[0019] The path branch structure carrying the conflict mark is parameter-superimposed with the spatiotemporal decay rate of the validity period gradient field to generate a path weight distribution based on the conflict propagation coefficient, wherein the dimension of the path weight distribution forms a strict mapping relationship with the hierarchical distribution of the multi-layer topology navigation grid;

[0020] The path weight distribution and the shelf unit space coordinates of the warehouse digital twin are subjected to closed-loop dynamic verification. Through the shelf coordinate offset correction process triggered by the conflict propagation coefficient, the conflict mark of the path branch structure and the spatiotemporal attenuation rate of the validity period gradient field form a rigid synchronization constraint, thereby generating a navigation path with conflict marks.

[0021] Optionally, the interference intensity threshold triggered by the proximity coupling parameter is used to calculate the path safety gap of the multi-layer topology navigation grid to generate a path branch structure carrying a conflict mark, wherein the conflict mark forms a dynamic association constraint with the shelf coordinate offset, including:

[0022] Based on the interference intensity threshold triggered by the proximity coupling parameter, dynamically calculate the path safety gap of the multi-layer topology navigation grid, wherein the dynamic calculation process of the path safety gap integrates the shelf coordinate offset and the physical spacing constraint of the shelf unit to generate an initial distribution of conflict markers;

[0023] The dynamic calculation result is associated with the dynamic change of the shelf coordinate offset in real time, and the path branch break probability calculation triggered by the interference intensity threshold is used to generate a path branch structure carrying a conflict mark, wherein the break probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit;

[0024] The fracture probability of the path branch structure is superimposed on the spatiotemporal decay rate of the effective period gradient field to generate a conflict propagation intensity coefficient, wherein the distribution dimension of the conflict propagation intensity coefficient forms a strict mapping relationship with the hierarchical division of the multi-layer topology navigation grid;

[0025] Dynamically bind the conflict propagation intensity coefficient to the spatial coordinates of the shelf units of the warehouse digital twin, trigger the remapping of conflict markers through the fracture probability threshold of the path branch structure, and generate a path branch structure carrying conflict markers, where the conflict markers form a dynamic association constraint with the shelf coordinate offset.

[0026] Optionally, the real-time association of the dynamic calculation result with the dynamic change of the shelf coordinate offset, and the calculation of the fracture probability of the path branch triggered by the interference intensity threshold, generate a path branch structure carrying conflict markers, and the fracture probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit, including:

[0027] The real-time association of the dynamic calculation result with the dynamic change of the shelf coordinate offset is carried out through the dynamic coupling parameter triggered by the interference intensity threshold, and the dynamic coupling parameter generates an initial fracture probability distribution based on the physical displacement amplitude of the shelf unit and the geometric topological spacing constraint of the path branch structure;

[0028] Based on the initial fracture probability distribution, calculate the fracture probability quantization value through the interference intensity ratio of the displacement amplitude of the shelf coordinate offset to the geometric topological spacing of the path branch structure, and the fracture probability quantization value forms a dynamic mapping constraint relationship with the displacement amplitude of the shelf unit;

[0029] Based on the fracture probability quantization value, dynamically correct the conflict propagation intensity coefficient through the gradient diffusion model of the spatio-temporal attenuation field, and the gradient parameter of the spatio-temporal attenuation field is generated by the spatio-temporal coupling relationship between the drug expiration decay rate and the fracture probability quantization value, and is mapped to the hierarchical division of the multi-layer topological navigation grid;

[0030] Adaptively fuse the conflict propagation intensity coefficient with the dynamic pooling weight parameter, and generate a path branch structure carrying conflict markers through the density distribution of the geometric topological fracture symbols. The dynamic pooling weight parameter is triggered by the dynamic association relationship between the shelf coordinate offset and the terrain stress field coefficient, and the fracture probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit.

[0031] Optionally, based on the fracture probability quantization value, dynamically correct the conflict propagation intensity coefficient through the gradient diffusion model of the spatio-temporal attenuation field, and the gradient parameter of the spatio-temporal attenuation field is generated by the spatio-temporal coupling relationship between the drug expiration decay rate and the fracture probability quantization value, and is mapped to the hierarchical division of the multi-layer topological navigation grid, including:

[0032] Based on the dynamic coupling relationship between the fracture probability quantification value and the crack propagation phase field variable, dynamic crack propagation parameters are generated through the similarity constraint principle, and the crack propagation phase field variable is jointly calibrated by the drug expiration decay rate and the hierarchical division of the multi-layer topological navigation grid;

[0033] The dynamic crack propagation parameters and the equivalent permeability parameters of the fluid-structure coupling phase field model are superimposed in multiple physical fields to generate a conflict propagation intensity correction coefficient, and the equivalent permeability parameter is dynamically calculated through the spatio-temporal coupling relationship between the fluid pressure gradient at the crack tip and the fracture probability quantification value;

[0034] Based on the conflict propagation intensity correction coefficient, the hierarchical division of the multi-layer topological navigation grid is dynamically recalibrated through the similarity constraint of the data-driven probability failure assessment model, and the similarity constraint is jointly triggered by the crack tip restraint state and the geometric topological spacing of the multi-layer topological navigation grid;

[0035] The dynamic recalibration result is adaptively matched with the energy release rate threshold of the crack bifurcation phase field model, and the energy release rate threshold forms a dynamic constraint relationship through the gradient diffusion relationship between the crack propagation phase field variable and the drug expiration decay rate.

[0036] Optionally, the converting the navigation path into an executable instruction set to drive the PDA device to project a path guidance signal, and the path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop, includes:

[0037] Based on the dynamic coupling relationship between the geometric topological spacing of the navigation path and the shelf coordinate offset, a phase encoding sequence of the executable instruction set is generated through dynamic path modulation parameters, and the dynamic path modulation parameters are jointly calibrated by the interference intensity ratio of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude;

[0038] The phase encoding sequence and the frequency domain response parameters of the multi-modal signal projection model are non-linearly superimposed to generate a path guidance signal carrying temperature and humidity compensation, and the frequency domain response parameters are dynamically matched through the geometric characteristics of the projection aperture of the PDA device and the curvature radius of the navigation path;

[0039] Based on the optical diffraction field distribution of the path guidance signal carrying temperature and humidity compensation, the phase distortion of the projection signal is dynamically compensated through a phase field constraint model, and the phase field constraint model is jointly triggered by the shelf coordinate offset and the topological fracture probability of the navigation path to form the initial constraint condition of the positioning verification closed loop;

[0040] The dynamic compensation result is adaptively matched with the closed-loop phase field energy threshold to generate a positioning verification closed loop. The closed-loop phase field energy threshold forms a dynamic balance constraint through the spatiotemporal conduction mechanism of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude.

[0041] In a second aspect, an embodiment of the present application provides a drug inventory management system based on a PDA, comprising:

[0042] The encoding module is used to construct a three-dimensional shelf topology structure and embed the validity period gradient field, and generate a dynamic attribute encoding shelf unit carrying the temperature and humidity fluctuation coefficient and the adjacent coupling parameter in combination with the temperature and humidity data stream. The adjacent coupling parameter represents the spatial interference relationship between adjacent shelf units;

[0043] A generation module, used to generate a shelf coordinate offset based on the temperature and humidity fluctuation coefficient, and establish a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin, wherein the virtual turnover sequence forms a spatiotemporal constraint relationship with the validity period gradient field;

[0044] A matching module, used for dynamically matching the spatiotemporal constraint relationship with the emergency channel traffic parameters to generate a navigation grid, triggering grid topology reconstruction through the proximity coupling parameters and generating a navigation path with conflict marks;

[0045] The conversion module is used to convert the navigation path into an executable instruction set to drive the PDA device to project a path guidance signal, wherein the path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop.

[0046] In a third aspect, an embodiment of the present application provides a computing device, comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a PDA-based drug inventory management method as described in the first aspect.

[0047] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program, which, when executed by a computer, implements a PDA-based drug inventory management method as described in the first aspect.

[0048] In an embodiment of the present application, a three-dimensional shelf topology structure is constructed and an effective period gradient field is embedded. The temperature and humidity data stream is combined to generate a dynamically attributed coded shelf unit carrying a temperature and humidity fluctuation coefficient and a neighboring coupling parameter, wherein the neighboring coupling parameter represents the spatial interference relationship between adjacent shelf units. A shelf coordinate offset is generated based on the temperature and humidity fluctuation coefficient, and a virtual turnover sequence including the shelf coordinate offset is established in the warehouse digital twin. The virtual turnover sequence forms a spatiotemporal constraint relationship with the effective period gradient field. The spatiotemporal constraint relationship is dynamically matched with the emergency channel passage parameters to generate a navigation grid, and the grid topology reconstruction is triggered by the neighboring coupling parameters to generate a navigation path with conflict marks. The navigation path is converted into an executable instruction set to drive the PDA device to project a path guidance signal, and the path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop.

[0049] The technical solution of this application has the following beneficial effects:

[0050] Through the joint modeling of temperature and humidity fluctuation coefficients and neighboring coupling parameters, the shelf unit coding is equipped with the ability to perceive environmental disturbances and improve sensitivity; the spatiotemporal constraint mechanism of the virtual turnover sequence and the expiration date gradient field realizes the coordinated optimization scheduling of drug expiration date and shelf displacement, thereby improving the utilization rate of warehouse resources; the navigation grid reconstruction mechanism based on neighboring coupling parameters can detect and avoid shelf displacement conflicts within 0.5 seconds, thereby improving the response speed of the grid topology method; the path guidance signal carries temperature and humidity compensation parameters and coordinate offsets, and is compared with the digital twin in real time through PDA projection to control the positioning verification error; the attenuation characteristics of the expiration date gradient field, the temperature and humidity fluctuation transmission mechanism, and the emergency channel passage parameters are integrated to reduce the risk of path failure caused by sudden changes in the warehouse environment.

[0051] Furthermore, based on the dynamic response model of temperature and humidity fluctuation coefficient and shelf topology, the shelf coordinate offset is generated in real time and the three-dimensional coordinate correction of the digital twin is driven. By superimposing the attenuation rate of the validity period gradient field and the adjacent coupling parameters, the spatial interference intensity of the shelf unit is quantified to form the priority weight distribution and the scheduling logic of the virtual turnover sequence. The scheduling order is dynamically adjusted to adapt to the geometric constraints of the emergency channel, and finally the spatiotemporal rigid synchronization of the virtual turnover sequence and the validity period gradient field is achieved. Through the dynamic coupling of temperature and humidity fluctuations and shelf displacement, the environmental disturbance is mapped into the coordinate offset correction of the digital twin, so that the scheduling logic of the virtual turnover sequence is rigidly bound to the attenuation rate of the drug validity period. The shelf priority weight distribution quantifies the spatial interference intensity through the adjacent coupling parameters, realizes the dynamic adaptation of the drug turnover path and the emergency channel traffic rules, and improves the scheduling efficiency; at the same time, based on the coordinate offset correction mechanism driven in real time by the temperature and humidity data stream, the positioning error between the digital twin and the actual storage environment is reduced, and the risk of shelf collision caused by environmental fluctuations is reduced.

[0052] These aspects or other aspects of the present application will be more clearly understood in the following description of the embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0054] Figure 1 It is a flowchart of a method for managing drug inventory based on a PDA provided by an embodiment of the present application;

[0055] Figure 2 It is a schematic structural diagram of a system for managing drug inventory based on a PDA provided by an embodiment of the present application;

[0056] Figure 3 It is a schematic structural diagram of a computing device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] In order to enable those skilled in the art to better understand the solution of the present application, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application.

[0058] In some processes described in the specification and claims of the present application and the above-mentioned drawings, there are multiple operations that appear in a specific order. However, it should be clearly understood that these operations can be executed not in the order in which they appear herein or in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations can be executed in sequence or in parallel. It should be noted that the descriptions such as "first" and "second" in this article are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, nor do they limit that "first" and "second" are of different types.

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0060] Figure 1The following is a flowchart of a drug inventory management method based on a PDA provided by an embodiment of the present application. As Figure 1 shown, the method includes:

[0061] 101. Construct a three-dimensional shelf topological structure and embed a validity period gradient field, and generate dynamic attribute-encoded shelf units carrying temperature and humidity fluctuation coefficients and adjacent coupling parameters in combination with temperature and humidity data streams, where the adjacent coupling parameter represents the spatial interference relationship between adjacent shelf units;

[0062] In this step, the three-dimensional shelf topological structure is a three-dimensional model used to describe the spatial layout of shelves in a warehouse, including the position, size, and mutual relationship of each shelf unit

[0063] The validity period gradient field is a virtual field created in the warehouse according to the validity period of drugs, used to guide the storage and retrieval of drugs according to their validity periods, and ensure that drugs with earlier expiration dates are processed first.

[0064] The temperature and humidity data stream is a data sequence of temperature and humidity changes in the warehouse collected in real time through a sensor network, used to monitor the impact of environmental conditions on drug preservation.

[0065] The temperature and humidity fluctuation coefficient is a parameter calculated based on the temperature and humidity data stream, representing the degree of temperature and humidity changes in a specific area, and used to evaluate the environmental stability of this area.

[0066] The adjacent coupling parameter is a parameter describing the spatial interference relationship between adjacent shelf units, reflecting their mutual influence, such as occlusion effects or heat transfer.

[0067] The dynamic attribute-encoded shelf unit combines the validity period gradient field and the temperature and humidity fluctuation coefficient to assign a dynamic code to each shelf unit, and this code is updated with the change of the environment, used to guide the storage and retrieval of drugs.

[0068] In an embodiment of the present application, first, high-precision point cloud data of the warehouse environment is obtained through a laser scanner, and a three-dimensional shelf topological structure is generated using these data. Specifically, the point cloud data is imported into three-dimensional modeling software (such as AutoCAD or Blender) to generate an accurate three-dimensional model. Then, in combination with the data stream collected by the temperature and humidity sensors, the temperature and humidity fluctuation coefficient is calculated using an interpolation algorithm. The adjacent coupling parameter is determined by calculating the distance and angular relationship between adjacent shelf units, representing the spatial interference relationship between them. Then, machine learning algorithms are used to optimize these parameters to generate dynamic attribute-encoded shelf units. Finally, these dynamic attribute encodings are embedded into the three-dimensional shelf topological structure to form dynamic attribute-encoded shelf units carrying temperature and humidity fluctuation coefficients and adjacent coupling parameters.

[0069] Suppose in a large pharmaceutical warehouse, in order to more precisely manage the drug storage environment, the method in step 101 is adopted for data processing and analysis. First, high-precision point cloud data of all the shelves in the warehouse is obtained through a laser scanner and imported into 3D modeling software to generate an accurate 3D model. The specific data is as follows: The total area of the warehouse is 5000 square meters, there are 120 standard shelf units in total, the height of each shelf unit is 3 meters, the width is 1.5 meters, and the depth is 1 meter. Based on this, a 3D shelf topological structure is constructed. Next, using the data stream collected by the temperature and humidity sensors, an interpolation algorithm is used to calculate the temperature and humidity fluctuation coefficient. For example, in the monitoring of a certain day, the average temperature in the warehouse is 22 °C, the relative humidity is 60%, and the temperature and humidity fluctuation coefficient is 0.05. By calculating the distance and angle relationship between adjacent shelf units, the adjacent coupling parameter is obtained as 0.85. Finally, machine learning algorithms are used to optimize these parameters to generate dynamically attributed coded shelf units. Finally, these dynamically attributed codes are embedded into the 3D shelf topological structure to form dynamically attributed coded shelf units carrying the temperature and humidity fluctuation coefficient and the adjacent coupling parameter. These dynamically attributed codes include the temperature and humidity fluctuation coefficient of 0.05 and the adjacent coupling parameter of 0.85, ensuring the stability and safety of the shelf units under different temperature and humidity conditions.

[0070] 102. Generate shelf coordinate offsets based on the temperature and humidity fluctuation coefficient, and establish a virtual turnover sequence containing the shelf coordinate offsets in the warehouse digital twin. The virtual turnover sequence forms a spatio-temporal constraint relationship with the expiration date gradient field.

[0071] In this step, the shelf coordinate offset is the change amount of the position of each shelf unit calculated based on the temperature and humidity fluctuation coefficient, and is used to adjust the shelf position in the virtual turnover sequence.

[0072] The virtual turnover sequence is a virtual path planning in the digital twin containing the shelf coordinate offsets, and is used to simulate the actual turnover situation of the goods in the warehouse.

[0073] The spatio-temporal constraint relationship is to combine the expiration date gradient field and the virtual turnover sequence to form time and space limit conditions, ensuring that drugs are stored and retrieved in the order of expiration date.

[0074] The virtual turnover sequence of the shelf coordinate offsets is a path planning model that combines the expiration date gradient field and the temperature and humidity fluctuation coefficient, and is used to optimize the turnover process of the goods in the warehouse.

[0075] In the embodiment of the present application, the Kalman filter is first used to perform real-time filtering on the temperature and humidity data stream to remove noise interference and generate an accurate temperature and humidity fluctuation coefficient. Then, the shelf coordinate offset is calculated by combining the temperature and humidity fluctuation coefficient through the Bayesian inference method. Specifically, the probability distribution of the temperature and humidity fluctuation coefficient is inferred using the Bayesian formula to obtain the probability distribution of the shelf coordinate offset. Next, a virtual turnover sequence containing the shelf coordinate offset is established in the warehouse digital twin, and the virtual turnover sequence is analyzed using spatiotemporal data mining technology to form a spatiotemporal constraint relationship with the validity period gradient field.

[0076] For example, continuing the above example, in the warehouse digital twin, the Kalman filter is used to perform real-time filtering on the temperature and humidity data stream, and the temperature and humidity fluctuation coefficient is 0.05. The Bayesian inference method is combined with this coefficient to calculate the shelf coordinate offset. The specific data are as follows: In a certain monitoring, the average temperature in the warehouse is 22°C, the relative humidity is 60%, and the temperature and humidity fluctuation coefficient is 0.05. The probability distribution of the temperature and humidity fluctuation coefficient is inferred by the Bayesian formula, and the shelf coordinate offset is 0.3 meters. A virtual turnover sequence containing the shelf coordinate offset is established in the warehouse digital twin, and the virtual turnover sequence is analyzed using spatiotemporal data mining technology to generate a data set that forms a spatiotemporal constraint relationship with the expiration date gradient field. For example, there are 1,000 kinds of medicines in the warehouse, and the expiration date gradient field of each medicine is divided in detail according to its shelf life and storage conditions. These data are optimized by linear regression to ensure the accuracy and effectiveness of the virtual turnover sequence. The final generated virtual turnover sequence includes a shelf coordinate offset of 0.3 meters and forms a spatiotemporal constraint relationship with the expiration date gradient field.

[0077] 103. Dynamically matching the spatiotemporal constraint relationship with the emergency channel traffic parameters to generate a navigation grid, triggering grid topology reconstruction through the proximity coupling parameters and generating a navigation path with conflict marks;

[0078] In this step, the emergency channel traffic parameters are data describing the emergency exit and channel traffic capacity, which are used to plan safe and efficient transportation routes.

[0079] The navigation grid is a two-dimensional or three-dimensional path planning map generated based on virtual turnover sequences and emergency channel passage parameters, which is used to guide logistics activities within the warehouse.

[0080] Mesh topology reconstruction is to automatically adjust the structure of the navigation mesh to adapt to the new spatial layout when the proximity coupling parameters change.

[0081] Conflict markers identify areas in the navigation path where congestion or other problems may occur, so that the path planning can be adjusted in time.

[0082] The navigation path with conflict marking is an optimized path planning that takes into account the influence of emergency channel traffic parameters and adjacent coupling parameters, marks potential problem areas, and provides safe and efficient navigation suggestions.

[0083] In the embodiment of the present application, the emergency channel passage parameters are extracted in real time through laser SLAM technology, and the channel width and obstacle contours are combined with the safety distance threshold to generate a dynamic pass permission range. The navigation grid is constructed using an adaptive constrained Delaunay triangulation algorithm, which encodes the spatiotemporal constraint relationship as a weight function of the grid edge, and simulates the repulsive effect of path conflict through a potential field model. When the neighboring coupling parameters trigger topological reconstruction, a deep reinforcement learning framework is used to train the agent to explore the optimal path between grid nodes. The conflict marker detects the geometric overlap area between the path and the obstacle through a convolutional neural network, and finally outputs a navigation path with conflict markers.

[0084] For example, continuing the above example, the width of the emergency passage and the outline of obstacles in the warehouse are first extracted in real time through laser SLAM technology. The specific data are as follows: In a certain monitoring, the width of the emergency passage in the warehouse is 2 meters, and the outline of the obstacle shows that there are some temporarily stacked goods, occupying 0.5 meters of the passage width. Then, the adaptive constrained Delaunay triangulation algorithm is used to construct the navigation grid, and the spatiotemporal constraint relationship is encoded as a weight function of the grid edge. The potential field model is used to simulate the repulsive effect of path conflict to ensure the safety of path planning. When the neighboring coupling parameters trigger topological reconstruction, the deep reinforcement learning framework is used to train the agent to explore the optimal path between grid nodes. For example, in a path optimization process, the agent successfully found a path that avoids all obstacles and has a total length of 150 meters. Finally, the geometric overlap area between the path and the obstacle is detected by the convolutional neural network, the conflict point is marked, and the navigation path with the conflict mark is output.

[0085] 104. Convert the navigation path into an executable instruction set to drive the PDA device to project a path guidance signal, wherein the path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop.

[0086] In this step, the executable instruction set converts the navigation path into a specific operation guide for the PDA device to read and execute.

[0087] The route guidance signal helps employees find the location of the required medicines and complete the pickup task through the instruction information displayed on the PDA device.

[0088] The positioning verification closed loop ensures the accuracy of the path guidance signal through a real-time feedback mechanism and makes adjustments based on actual conditions.

[0089] The path guidance signal carrying the temperature and humidity fluctuation coefficient and the shelf coordinate offset is a detailed navigation instruction that includes environmental parameters and location information to help employees accurately find the target location.

[0090] For example, continuing with the previous example, first, the executable instruction set is compiled using a hierarchical state machine architecture, and the navigation path is decomposed into an opcode, coordinate increment, and time constraint triple. The specific data is as follows: In a certain path planning, the navigation path contains 10 opcodes, each opcode corresponding to a specific action (such as moving forward, turning, etc.), the coordinate increments are (0.5 meters, 0 meters), (0 meters, 0.3 meters), etc., and the time constraint is that each step action does not exceed 5 seconds. The PDA device uses millimeter-wave radar and UWB positioning technology to track the operator's position in real-time to ensure the accurate positioning of the operator in the warehouse. For example, in an actual operation, the operator's initial position is (0,0), and the target position is (5 meters, 3 meters). The path guidance signal uses phase-encoding modulation technology to embed the temperature and humidity fluctuation coefficient of 0.05 and the shelf coordinate offset of 0.3 meters into the orthogonal frequency division multiplexing subchannel of the carrier signal. The multi-source sensor data is fused through a federated Kalman filter to calculate the residual error between the actual position and the virtual path to ensure the positioning accuracy. The finally generated path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset and is transmitted to the PDA device through a wireless transmission protocol to form a positioning verification closed-loop. The operator holds the PDA device and quickly finds the location of the required medicine according to the accurate instructions and successfully completes the picking task.

[0091] Through the above four steps, the full-process automated management from three-dimensional shelf modeling to dynamic path planning has been successfully achieved. It not only improves the space utilization rate of the warehouse and the safety of medicine storage but also significantly improves work efficiency and service quality. Especially in the field of medicine inventory management, this intelligent solution provides strong support for ensuring medicine quality and the smooth operation of the supply chain. Each step is closely linked to jointly build an efficient and reliable intelligent warehouse management system.

[0092] To further improve the accuracy and real-time performance of medicine warehouse management, especially in dealing with the time and space alignment problems of multi-source data, an optimization scheme is proposed. This scheme generates a quality control report related to the abnormal nodes of the environmental response model by establishing a dynamic environmental response model of the temperature and humidity fluctuation coefficient and the three-dimensional shelf topology structure and combining the expiration date gradient field and the emergency channel passage parameters. This process not only improves the consistency and accuracy of the data but also provides more effective operation suggestions for warehouse managers. In some embodiments, the generating the shelf coordinate offset based on the temperature and humidity fluctuation coefficient and establishing a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin in step 102 includes:

[0093] 1021. Establish a dynamic environmental response model of the temperature and humidity fluctuation coefficient and the three-dimensional shelf topological structure. Generate a shelf coordinate offset through real-time correlation calculation of the physical position of the shelf unit and the temperature and humidity data stream. The shelf coordinate offset drives the dynamic correction of the spatial coordinates of the shelf unit in the warehouse digital twin;

[0094] In step 1021, the temperature and humidity fluctuation coefficient is data describing the degree of temperature and humidity change in a specific area, and is used to evaluate the environmental stability of this area.

[0095] The three-dimensional shelf topological structure is a three-dimensional model used to describe the spatial layout of the shelves in the warehouse, including the position, size and mutual relationship of each shelf unit.

[0096] The dynamic environmental response model is a mathematical model used to describe the relationship between the three-dimensional shelf topological structure and the temperature and humidity data stream, and can adjust the position of the shelf unit according to real-time temperature and humidity changes.

[0097] The shelf coordinate offset is the change amount of the position of each shelf unit calculated based on the temperature and humidity fluctuation coefficient, and is used to adjust its spatial coordinates in the digital twin.

[0098] In the embodiment of the present application, first, a support vector regression (SVR) is used to establish a dynamic environmental response model of the temperature and humidity fluctuation coefficient and the three-dimensional shelf topological structure. Specifically, by real-time collecting the temperature and humidity data stream in the warehouse and using SVR to model these data, a temperature and humidity fluctuation coefficient is generated. Then, the temperature and humidity fluctuation coefficient is subjected to real-time correlation calculation with the physical position of the shelf unit to generate a shelf coordinate offset. To achieve this, a spatial correlation algorithm based on a graph neural network (GNN) is used. This algorithm can effectively capture the spatial relationship between shelf units and adjust their coordinates according to the temperature and humidity fluctuation coefficient. Finally, through this dynamic environmental response model, the spatial coordinates of the shelf unit in the warehouse digital twin are driven to perform dynamic correction.

[0099] 1022. Perform spatio-temporal superposition of the shelf coordinate offset and the decay rate of the expiration date gradient field to generate the priority weight distribution of the shelf unit. The priority weight distribution quantifies and sorts the spatial interference intensity of adjacent shelf units based on the adjacent coupling parameter to form the scheduling logic of the virtual turnover sequence;

[0100] In step 1022, the expiration date gradient field is a virtual field created in the warehouse according to the expiration date of the drugs, used to guide the storage and retrieval of drugs according to their expiration dates, and ensure that the drugs with earlier expiration dates are processed first.

[0101] The decay rate describes how quickly a drug's shelf life changes over time and is used to quantify a drug's shelf life priority.

[0102] The priority weight distribution is a weight value generated according to the shelf coordinate offset and the decay rate of the validity period gradient field, which is used to quantify the priority of each shelf unit.

[0103] The proximity coupling parameter is a parameter that describes the spatial interference relationship between adjacent shelf units and reflects the mutual influence between them.

[0104] In the embodiment of the present application, the shelf coordinate offset and the decay rate of the validity period gradient field are firstly superimposed in time and space to generate the priority weight distribution of the shelf unit. Specifically, the spatio-temporal sequence analysis (STSA) method is used to quantitatively sort the spatial interference intensity of adjacent shelf units in combination with the proximity coupling parameter. The STSA method can effectively capture the changing trend of shelf units in time and space, and adjust the priority weight distribution according to these changes. Through this spatio-temporal superposition, the scheduling logic of the virtual turnover sequence is formed.

[0105] 1023. Dynamically adapt the scheduling logic of the virtual turnover sequence to the geometric constraints of the emergency channel passage parameters, dynamically adjust the scheduling order of the shelf units through the interference intensity threshold triggered by the proximity coupling parameter, and form a path rule compatible with the emergency channel passage parameters;

[0106] In step 1023, the geometric constraints are geometric features that describe the emergency passage parameters, such as width, length, etc., which are used to plan a safe and efficient transportation route.

[0107] The interference intensity threshold is a set threshold. When the proximity coupling parameter exceeds the threshold, the scheduling order of the shelf unit is triggered to be adjusted.

[0108] Path rules are specific operation guidelines generated based on virtual turnover sequence scheduling logic and emergency channel access parameters, which are used to guide logistics activities within the warehouse.

[0109] In the embodiments of the present application, first, the scheduling logic of the virtual turnover sequence is dynamically adapted to the geometric constraints of the emergency passage passing parameters. Specifically, the genetic algorithm (GA) is used to optimize the path rules to make them compatible with the emergency passage passing parameters. The interference intensity threshold triggered by the adjacent coupling parameters is used to dynamically adjust the scheduling order of the shelf units to form path rules compatible with the emergency passage passing parameters. The GA method can find the optimal solution among a large number of possible paths to ensure the safety and efficiency of path planning. Through this dynamic adaptation, path rules compatible with the emergency passage passing parameters are generated to ensure the safety and reliability of the path.

[0110] 1024. Bind the path rules with the scheduling logic of the virtual turnover sequence, and through the dynamic update of the spatial coordinates of the shelf units of the warehouse digital twin, make the virtual turnover sequence form a spatio-temporal constraint relationship with the spatio-temporal decay rate of the validity period gradient field.

[0111] In step 1024, the parameter binding is to bind the path rules with the scheduling logic of the virtual turnover sequence to ensure their consistency and coordination.

[0112] The dynamic update of the spatial coordinates is to make the virtual turnover sequence form a spatio-temporal constraint relationship with the spatio-temporal decay rate of the validity period gradient field through the dynamic update of the spatial coordinates of the shelf units of the warehouse digital twin.

[0113] In the embodiments of the present application, first, the path rules are bound with the scheduling logic of the virtual turnover sequence. Specifically, the particle filter (PF) technology is used to make the virtual turnover sequence form a spatio-temporal constraint relationship with the spatio-temporal decay rate of the validity period gradient field. The PF technology can effectively handle the state estimation problem of non-linear systems to ensure the coordination between the path rules and the scheduling logic of the virtual turnover sequence. In addition, the multi-objective optimization (MOO) technology is introduced to balance multiple objectives (such as safety, efficiency, etc.) between the path rules and the scheduling logic. Through this parameter binding, the virtual turnover sequence forms a spatio-temporal constraint relationship with the spatio-temporal decay rate of the validity period gradient field.

[0114] The following is a specific example:

[0115] In the digital twin system deployed in Area A of the pharmaceutical warehousing center, the temperature fluctuation in the storage area of a batch of cephalosporin injections is collected in real time by the temperature and humidity sensors at the top of the shelves, reaching ±0.8 °C / h (standard deviation 0.35), and the humidity fluctuation coefficient reaches 12% RH / min (relative humidity percentage change rate). Based on this data stream, the support vector regression model calculates that the X-axis offset of the shelf unit B-23F-07 is +15 cm and the Z-axis offset is -8 cm, triggering the dynamic correction of the position of this shelf in the three-dimensional topological structure of the digital twin from the original coordinates (X: 23.5 m, Y: 4.2 m, Z: 2.8 m) to (X: 23.65 m, Y: 4.2 m, Z: 2.72 m). At the same time, the remaining shelf life of this batch of drugs is monitored by the expiration date gradient field to be 182 days, and its decay rate is 0.549% / day (corresponding to the daily increase in priority weight ΔW = 0.2). After spatiotemporal superposition, the priority weight value W of this shelf is generated as 86.7 (full score 100), triggering the coupling parameters of the adjacent shelves B-23F-06 and B-23F-08 to reach 0.85 and 0.78 respectively, forming a virtual turnover sequence with B-23F-07 as the core. When the passage width of the emergency passage C is reduced to 1.2 m due to temporary equipment occupying the road, the genetic algorithm adjusts the turning radius constraint value of the shelf scheduling path from 1.5 m to 0.9 m, and recalculates the interference intensity threshold of the adjacent shelves to below 0.7. Finally, a path rule compatible with the emergency passage passing parameters and including 5 priority shelf units is generated, enabling the virtual turnover sequence to form a spatiotemporal constraint relationship with the spatiotemporal decay rate of the expiration date gradient field.

[0116] Through the implementation of steps 1021 to 1024, the accuracy and real-time performance of pharmaceutical warehouse management have been significantly improved, especially in handling the synchronization of multi-source spatiotemporal data in complex areas (such as high humidity areas). It can identify anomalies in the storage environment earlier, and optimize the scheduling logic based on the comprehensive analysis of the temperature and humidity fluctuation coefficient and the expiration date gradient field to ensure that drugs are stored under the best conditions and follow the expiration date priority principle. This mechanism not only improves work efficiency but also ensures the safety and quality of drugs. In addition, by dynamically adapting to the emergency passage passing parameters, it can quickly adjust the path planning in case of emergencies to ensure the safety of personnel and materials. Overall, this solution has greatly improved the intelligent level of warehouse management and enhanced the ability to respond to emergencies.

[0117] In order to further improve the accuracy and real-time performance of drug warehouse management, especially in dealing with the time and space alignment problem of multi-source data, an optimization scheme is proposed. This scheme generates a navigation grid by dynamically matching the spatiotemporal constraint relationship with the emergency channel passage parameters, and triggers the grid topology reconstruction through the proximity coupling parameters to generate a navigation path with conflict markers. This process not only improves the consistency and accuracy of the data, but also provides more effective operational suggestions for warehouse managers. In some embodiments, the dynamic matching of the spatiotemporal constraint relationship with the emergency channel passage parameters to generate a navigation grid in step 103 includes:

[0118] 1031. Dynamically match the spatiotemporal constraint relationship with the geometric topological features of the emergency passage parameters, generate a multi-layer topological navigation grid based on the spatial interference intensity of the proximity coupling parameters, and form a dynamic binding relationship between the hierarchical distribution of the multi-layer topological navigation grid and the priority of the emergency passage parameters;

[0119] In step 1031, the spatiotemporal constraint relationship is a constraint condition generated based on the temperature and humidity fluctuation coefficient and the expiration date gradient field, which is used to ensure the optimal storage environment for the medicine.

[0120] The geometric topological characteristics of the emergency passage parameters are data describing the geometric shape and topological structure of the emergency passage, such as width, length, etc., which are used to plan safe and efficient transportation routes.

[0121] The spatial interference intensity of the proximity coupling parameter describes the degree of mutual influence between adjacent shelf units and is used to quantify the spatial interference intensity between them.

[0122] The multi-layer topological navigation grid is a hierarchical navigation grid structure, and the hierarchical distribution forms a dynamic binding relationship with the priority of the emergency channel passage parameters.

[0123] In the embodiments of the present application, first, a Variational Autoencoder (VAE) is used to dynamically match the spatio-temporal constraint relationship with the geometric topological features of the emergency passage passing parameters. Specifically, by collecting the geometric topological data in the warehouse in real time and using VAE to model these data, a multi-layer topological navigation grid is generated. To achieve this, a spatial association algorithm based on Tensor Decomposition (TD) is used, which can effectively capture the geometric features of the emergency passage and adjust its hierarchical distribution according to the adjacent coupling parameters. In addition, a method based on Hierarchical Clustering (HC) is introduced to dynamically bind the priorities of the emergency passage passing parameters, ensuring that the navigation grids at different levels form a dynamic binding relationship with the priorities of the emergency passage. Finally, through this dynamic matching, a multi-layer topological navigation grid dynamically bound to the priorities of the emergency passage passing parameters is generated.

[0124] 1032. Calculate the path safety gap for the multi-layer topological navigation grid through the interference intensity threshold triggered by the adjacent coupling parameter, and generate a path branch structure with conflict marks, where the conflict marks form a dynamic association constraint with the shelf coordinate offset;

[0125] In step 1032, the interference intensity threshold is a set threshold. When the adjacent coupling parameter exceeds this threshold, the scheduling order adjustment of the shelf unit is triggered.

[0126] The path safety gap calculation is a technique for evaluating whether a path is safe, ensuring that there are no conflicts in the path.

[0127] The path branch structure with conflict marks includes a path branch structure with conflict marks, which is used to indicate potential conflict points in the path.

[0128] In the embodiments of the present application, first, the path safety gap of the multi-layer topological navigation grid is calculated through the interference intensity threshold triggered by the adjacent coupling parameter, and a path branch structure carrying conflict marks is generated. Specifically, a method based on Graph Embedding (GE) is used to analyze the multi-layer topological navigation grid, and the path safety gap is calculated in combination with the interference intensity threshold. To achieve this, a technique based on Locality-Sensitive Hashing (LSH) is used to quickly find and mark the conflict areas. In addition, a method based on Generative Adversarial Network (GAN) is introduced to optimize the finding and marking process to ensure the accuracy of the path branch structure. Through this path safety gap calculation, a path branch structure carrying conflict marks is generated, and the conflict marks form a dynamic association constraint with the shelf coordinate offset.

[0129] 1033. Superimpose the parameters of the path branch structure carrying conflict marks and the spatio-temporal decay rate of the validity period gradient field to generate a path weight distribution based on the conflict propagation coefficient. The dimension of the path weight distribution forms a strict mapping relationship with the hierarchical distribution of the multi-layer topological navigation grid.

[0130] In step 1033, the conflict propagation coefficient describes the speed and degree of conflict propagation in the path and is used to quantify the safety of the path.

[0131] The path weight distribution is the path weight value generated based on the conflict propagation coefficient and is used to guide path selection.

[0132] The strict mapping relationship means that the dimension of the path weight distribution forms a strict mapping relationship with the hierarchical distribution of the multi-layer topological navigation grid.

[0133] In the embodiments of the present application, first, the path branch structure carrying conflict markers is parameter-overlaid with the spatio-temporal decay rate of the validity period gradient field to generate a path weight distribution based on the conflict propagation coefficient. Specifically, a method based on the Markov Decision Process (MDP) is used to optimize the path branch structure, and the spatio-temporal decay rate of the validity period gradient field is combined to generate a path weight distribution. To achieve this, a technique based on Tensor Manifold Learning (TML) is used to capture the changing trends of the path in time and space, and the path weights are adjusted according to these changes. In addition, a method based on Evolutionary Strategies (ES) is introduced to further optimize the path weight distribution to ensure that its dimension forms a strict mapping relationship with the hierarchical distribution of the multi-layer topological navigation grid. Through this parameter overlay, a path weight distribution based on the conflict propagation coefficient is generated.

[0134] 1034. Perform a closed-loop dynamic verification on the path weight distribution and the spatial coordinates of the shelf units in the warehouse digital twin. Through the shelf coordinate offset correction process triggered by the conflict propagation coefficient, a rigid synchronization constraint is formed between the conflict markers of the path branch structure and the spatio-temporal decay rate of the validity period gradient field, generating a navigation path with conflict markers.

[0135] In step 1034, the closed-loop dynamic verification is a technique for verifying the correctness of the path, ensuring that the path is consistent with the actual environment.

[0136] The rigid synchronization constraint ensures that the conflict markers of the path branch structure are synchronized with the spatio-temporal decay rate of the validity period gradient field.

[0137] The navigation path with conflict markers includes a navigation path with conflict markers, which is used to guide the logistics activities in the warehouse.

[0138] In the embodiments of the present application, a method based on the Extended Kalman Filter (EKF) is used to perform a dynamic verification on the path weight distribution, and the shelf coordinate offset is corrected in combination with the conflict propagation coefficient. To achieve this, a technique based on Bayesian Inference (BI) is used to handle uncertainty and ambiguity problems to ensure the accuracy of path correction. In addition, a method based on Reinforcement Learning (RL) is introduced to optimize the path verification process to ensure a high degree of consistency between the path and the actual environment. Through this closed-loop dynamic verification, a navigation path with conflict markers is generated, ensuring that a rigid synchronization constraint is formed between the conflict markers of the path branch structure and the spatio-temporal decay rate of the validity period gradient field.

[0139] The following is a specific example:

[0140] In the scenario of intelligent pharmaceutical warehousing, assume that a pharmaceutical warehouse needs to urgently process a batch of cold-chain drugs with a remaining shelf life of only 45 days (the shelf-life gradient field is set to α = 0.85), and the temperature and humidity fluctuation coefficients in its storage area are calibrated in real time to the spatio-temporal constraint conditions of ±1.5°C / ±3%RH. When the emergency outbound process is initiated, the warehouse digital twin first collects the current environmental data: the geometric topology characteristics of the emergency passage show that the main passage width is 3.5 meters and the turning radius is 2.8 meters, and the coupling parameter spatial interference intensity of the adjacent shelf units reaches the threshold γ = 0.73 (triggering the adjustment of the shelf scheduling order). Through the variational autoencoder, the shelf coordinate offset (such as the X-axis deviation of ±15 cm and the Y-axis rotation angle θ = 5°) is dynamically matched with the channel topology to generate a multi-level topological navigation grid containing three priority levels - among which the first-level grid maps the fire passage (the passing weight ω 1 = 0.95), the second-level grid corresponds to the dedicated passage for cold-chain drugs (ω 2 = 0.88), and the third-level grid covers the general shelf area (ω 3 = 0.75). In the stage of calculating the path safety gap, it is detected that the coordinate offset Δ = 22 cm of roadway B3 is caused by the abnormal displacement of the stacker, triggering a conflict mark and generating a path branch structure carrying this mark (the conflict propagation coefficient β = 0.68). After superimposing the spatio-temporal decay rate of the shelf-life gradient field (the daily decay rate δ = 0.018), the path weight distribution shows that the optimal path is: starting point → the first-level grid node A7 (weight 0.93) → the second-level grid node D12 (weight 0.79) → avoiding the conflict mark area C5 (the weight drops to 0.62). Finally, the AGV navigation path is closed-loop verified through the Kalman filter, and the shelf coordinate offset is corrected to Δ' = 8 cm, so that the conflict mark and the shelf-life decay form a rigid synchronous constraint, and the navigation path with the conflict mark is output. This navigation path is real-time displayed on the digital twin platform as a green dynamic trajectory line passing through the 1-3-5 level grids, and its path safety index is increased to 92.7%, ensuring that the drugs complete the full-process outbound operation within 12 minutes and 35 seconds.

[0141] Through the implementation of steps 1031 to 1034, the accuracy and real-time performance of pharmaceutical warehouse management have been significantly improved, especially in handling the synchronization of multi-source spatio-temporal data in complex areas (such as high-humidity areas). It can identify anomalies in the storage environment earlier, and optimize the scheduling logic based on the comprehensive analysis of the temperature and humidity fluctuation coefficient and the expiration date gradient field to ensure that drugs are stored under the best conditions and follow the expiration date priority principle. In addition, by dynamically adapting the access parameters of the emergency channel, the system can quickly adjust the path planning in case of emergencies to ensure the safety of personnel and materials. Overall, this solution has greatly improved the intelligent level of warehouse management, enhanced the ability to respond to emergencies, and ensured a high degree of consistency between the path and the actual environment through closed-loop dynamic verification, thus improving the overall efficiency and safety of warehouse operations.

[0142] To further improve the accuracy and real-time performance of pharmaceutical warehouse management, especially in dealing with the temporal and spatial alignment of multi-source data, an optimization solution is proposed. This solution calculates the path safety gap for the multi-layer topological navigation grid through the interference intensity threshold triggered by the proximity coupling parameter, generates a path branch structure with conflict markers, and ensures that the conflict markers form a dynamic association constraint with the shelf coordinate offset. This process not only improves the consistency and accuracy of the data but also provides more effective operation suggestions for warehouse managers. In some embodiments, the step of calculating the path safety gap for the multi-layer topological navigation grid through the interference intensity threshold triggered by the proximity coupling parameter and generating a path branch structure with conflict markers in step 1032 includes:

[0143] 201. Dynamically calculate the path safety gap of the multi-layer topological navigation grid based on the interference intensity threshold triggered by the proximity coupling parameter. The dynamic calculation process of the path safety gap integrates the shelf coordinate offset and the physical spacing constraint of the shelf unit to generate an initial distribution of conflict markers;

[0144] In step 201, the interference intensity threshold is a set threshold. When the proximity coupling parameter exceeds this threshold, it triggers an adjustment of the scheduling order of the shelf units.

[0145] The path safety gap is a technology used to evaluate whether a path is safe to ensure that no conflicts occur in the path.

[0146] The initial distribution of conflict markers is the initial conflict marker distribution generated based on the calculation result of the path safety gap, which is used to indicate potential conflict points in the path.

[0147] The physical spacing constraint describes the physical distance limit between shelf units, which is used to ensure the safety of the path.

[0148] In the embodiments of the present application, first, a Diffusion Tensor Imaging (DTI) technique is used to process the path safety gaps in the multi-layer topological navigation grid. Specifically, by collecting the geometric and physical data in the warehouse in real time and using DTI to model these data, path safety gaps are generated. To achieve this, a spatial correlation algorithm based on Principal Component Analysis (PCA) is used, which can effectively capture the spatial relationship between the shelf units and adjust their path safety gaps according to the adjacent coupling parameters. In addition, a method based on Random Matrix Theory (RMT) is introduced to dynamically calculate the interference intensity threshold to ensure the accuracy of the path safety gaps. Finally, through this dynamic calculation process, the shelf coordinate offset is fused with the physical spacing constraint of the shelf units to generate the initial distribution of the conflict markers.

[0149] 202. Real-time associate the dynamic calculation result with the dynamic change of the shelf coordinate offset, and generate a path branch structure carrying conflict markers through the calculation of the path branch break probability triggered by the interference intensity threshold. The break probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit.

[0150] In step 202, the path branch break probability describes the possibility of the path branch breaking and is used to quantify the safety of the path.

[0151] The displacement amplitude describes the distance that the shelf unit moves and is used to evaluate the safety of the path.

[0152] The positive correlation constraint means that the break probability of the path branch structure is directly proportional to the displacement amplitude of the shelf unit.

[0153] In the embodiments of the present application, first, the dynamic calculation result is associated with the dynamic change of the shelf coordinate offset in real time, and a path branch structure carrying a conflict mark is generated through the calculation of the path branch break probability triggered by the interference intensity threshold. Specifically, a method based on the Hidden Markov Model (HMM) is used to model the path branch break probability, and the path branch break probability is calculated in combination with the interference intensity threshold. To achieve this, a technique based on Kernel Density Estimation (KDE) is used to quickly find and mark the break area. In addition, a method based on Simulated Annealing (SA) is introduced to optimize the generation process of the break probability to ensure the accuracy of the path branch structure. Through this calculation of the path branch break probability, a path branch structure carrying a conflict mark is generated, and the break probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit.

[0154] 203. Perform a superposition operation on the break probability of the path branch structure and the spatio-temporal attenuation rate of the validity period gradient field to generate a conflict propagation intensity coefficient, and the distribution dimension of the conflict propagation intensity coefficient forms a strict mapping relationship with the hierarchical division of the multi-layer topological navigation grid;

[0155] In step 203, the conflict propagation intensity coefficient describes the speed and degree of conflict propagation in the path and is used to quantify the safety of the path.

[0156] The strict mapping relationship means that the distribution dimension of the conflict propagation intensity coefficient forms a strict mapping relationship with the hierarchical division of the multi-layer topological navigation grid.

[0157] In the embodiments of the present application, first, a superposition operation is performed on the break probability of the path branch structure and the spatio-temporal attenuation rate of the validity period gradient field to generate a conflict propagation intensity coefficient. Specifically, a method based on the Recursive Neural Network (RvNN) is used to optimize the path branch structure, and the conflict propagation intensity coefficient is generated in combination with the spatio-temporal attenuation rate of the validity period gradient field. To achieve this, a technique based on Manifold Learning (ML) is used to capture the change trend of the path in time and space, and the conflict propagation intensity coefficient is adjusted according to these changes. In addition, a method based on Particle Swarm Optimization (PSO) is introduced to further optimize the distribution of the conflict propagation intensity coefficient to ensure that its dimension forms a strict mapping relationship with the hierarchical division of the multi-layer topological navigation grid.

[0158] 204. Dynamically bind the conflict propagation intensity coefficient to the spatial coordinates of the shelf units of the warehouse digital twin, trigger the conflict marker remapping through the fracture probability threshold of the path branch structure, and generate a path branch structure carrying conflict markers, where the conflict markers form a dynamic association constraint with the shelf coordinate offset.

[0159] In step 204, dynamic binding: Dynamically bind the conflict propagation intensity coefficient to the spatial coordinates of the shelf units of the warehouse digital twin to ensure that the path is consistent with the actual environment.

[0160] The fracture probability threshold is a set threshold. When the fracture probability of the path branch structure exceeds this threshold, it triggers the conflict marker remapping.

[0161] Conflict marker remapping means reassigning conflict markers according to the fracture probability threshold to ensure the safety of the path.

[0162] In the embodiments of this application, first, the conflict propagation intensity coefficient is dynamically bound to the spatial coordinates of the shelf units of the warehouse digital twin, and the conflict marker remapping is triggered through the fracture probability threshold of the path branch structure to generate a path branch structure carrying conflict markers. Specifically, a method based on Adaptive Filtering (AF) is used to dynamically verify the conflict propagation intensity coefficient, and the conflict markers are corrected in combination with the fracture probability threshold. To achieve this, a technique based on Bayesian Networks (BN) is used to handle uncertainty and ambiguity problems to ensure the accuracy of the conflict markers. In addition, a method based on Reinforcement Learning (RL) is introduced to optimize the conflict marker remapping process to ensure a high degree of consistency between the path and the actual environment. Through this dynamic binding, a path branch structure carrying conflict markers is generated, and the conflict markers form a dynamic association constraint with the shelf coordinate offset.

[0163] The following is a specific example:

[0164] In the digital twin scenario of pharmaceutical warehousing, assume that a navigation grid topology with 1200 groups of movable shelves is deployed in an automated stereoscopic warehouse, and the physical spacing constraint of the shelf unit is set to 1.5 meters. When the outbound task of a certain batch of drugs triggers a shelf coordinate offset (X-axis offset Δ = 0.2 meters, Y-axis offset Δ = 0.15 meters), the modeling based on DTI technology shows that the adjacent coupling parameter reaches 0.83, exceeding the preset interference intensity threshold of 0.8. At this time, the path safety gap is calculated by the PCA algorithm, and it is found that the path gap in the No. 3 roadway area is compressed from 1.8 meters to 1.28 meters (lower than the safety threshold of 1.3 meters). Combining the shelf displacement amplitude of 0.35 meters with the RMT dynamic calibration result, an initial distribution map with 5 conflict markers is generated, and the marker density is 2.8 per 100 square meters. Subsequently, the state transition probability matrix of the hidden Markov model (the number of states N = 3, the transition probability matrix A = [[0.6, 0.3, 0.1], [0.2, 0.5, 0.3], [0.1, 0.4, 0.5]]) is used to model the path branch fracture probability. When the displacement amplitude of the B12 shelf is detected to reach 0.5 meters, the corresponding path branch fracture probability rises to 70% (positive correlation coefficient r = 0.92), and 12 high-risk fracture paths are identified at the 2-4 levels of the navigation grid through kernel density estimation. After superimposing the spatio-temporal decay rate of the validity period gradient field (5% per hour), the conflict propagation intensity coefficient reaches a peak of 0.78 at the third level, forming a strict mapping with the four-layer topology structure (correlation coefficient ρ = 0.99). Finally, the conflict markers in the D7 area are remapped through the Bayesian network. When the fracture probability exceeds the threshold of 0.75, the original 3 conflict markers are expanded to 7, the marker density is increased to 3.2 per 100 square meters, and a dynamic association is formed with the real-time coordinate offset of the shelf (synchronization delay ≤ 50 ms), generating a path branch structure with conflict markers to ensure that the spatial coordinate deviation between the AGV transportation path and the digital twin body is controlled within the range of ±0.05 meters.

[0165] Through the implementation of steps 201 to 204, the accuracy and real-time performance of drug warehouse management are significantly improved, especially in processing multi-source spatiotemporal data synchronization in complex areas (such as high humidity areas). It can identify anomalies in the storage environment earlier, and optimize the scheduling logic based on the comprehensive analysis of the temperature and humidity fluctuation coefficient and the expiration date gradient field to ensure that the drugs are stored under the best conditions and follow the expiration date priority principle. In addition, by dynamically adapting the emergency channel traffic parameters, the path planning can be quickly adjusted in an emergency to ensure the safety of personnel and materials. Overall, the solution has greatly improved the intelligence level of warehouse management, enhanced the ability to respond to emergencies, and ensured a high degree of consistency between the path and the actual environment through closed-loop dynamic verification, thereby improving the overall efficiency and safety of warehouse operations. At the same time, by real-time association of the displacement amplitude of the shelf unit with the probability of path branch breakage, the safety and reliability of the path are ensured, further improving the level of refinement of warehouse management.

[0166] In order to further improve the accuracy and real-time performance of drug warehouse management, especially in dealing with the time and space alignment problem of multi-source data, an optimization scheme is proposed. This scheme associates the dynamic calculation results with the dynamic changes of the shelf coordinate offset in real time, calculates the path branch break probability triggered by the interference intensity threshold, generates a path branch structure carrying a conflict mark, and ensures that the break probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit. This process not only improves the consistency and accuracy of the data, but also provides more effective operational suggestions for warehouse managers. In some embodiments, the step 202 associates the dynamic calculation results with the dynamic changes of the shelf coordinate offset in real time, calculates the path branch break probability triggered by the interference intensity threshold, and generates a path branch structure carrying a conflict mark, including:

[0167] 301. Real-time association of the dynamic calculation result with the dynamic change of the shelf coordinate offset through the dynamic coupling parameter triggered by the interference intensity threshold, wherein the dynamic coupling parameter generates an initial distribution of fracture probability based on the physical displacement amplitude of the shelf unit and the geometric topological spacing constraint of the path branch structure;

[0168] In step 301, the dynamic coupling parameters are data generated based on the physical displacement amplitude of the shelf unit and the geometric topological spacing constraints of the path branch structure, and are used to evaluate the safety of the path.

[0169] The initial fracture probability distribution is an initial fracture probability distribution generated based on the dynamic coupling parameters and is used to indicate potential fracture points in the path.

[0170] The interference intensity threshold is a set threshold, and when the threshold is exceeded, the scheduling order adjustment is triggered.

[0171] In the embodiments of the present application, first, a method based on a Convolutional Neural Network (CNN) is used to process the path safety gaps in a multi-layer topological navigation grid. Specifically, by collecting the geometric and physical data in the warehouse in real time and using the CNN to model this data, path safety gaps are generated. To achieve this, a spatial association algorithm based on a Self-Organizing Map (SOM) is used, which can effectively capture the spatial relationships between the shelf units and adjust their path safety gaps according to the proximity coupling parameters. In addition, a method based on Extreme Value Theory (EVT) is introduced to dynamically calculate the interference intensity threshold to ensure the accuracy of the path safety gaps. Finally, through this dynamic coupling parameter, the geometric topological spacing constraint of the shelf coordinate offset and the path branch structure is fused to generate the initial distribution of the fracture probability.

[0172] 302. Based on the initial distribution of the fracture probability, calculate the fracture probability quantization value through the interference intensity ratio of the displacement amplitude of the shelf coordinate offset to the geometric topological spacing of the path branch structure, and the fracture probability quantization value forms a dynamic mapping constraint relationship with the displacement amplitude of the shelf unit;

[0173] In step 302, the fracture probability quantization value is the data calculated based on the interference intensity ratio of the displacement amplitude of the shelf coordinate offset to the geometric topological spacing of the path branch structure, and is used to quantify the safety of the path.

[0174] The dynamic mapping constraint relationship is the dynamic mapping relationship between the fracture probability quantization value and the displacement amplitude of the shelf unit, ensuring the safety of the path.

[0175] In the embodiments of the present application, first, based on the initial distribution of the fracture probability, calculate the fracture probability quantization value through the interference intensity ratio of the displacement amplitude of the shelf coordinate offset to the geometric topological spacing of the path branch structure. Specifically, a method based on Bayesian Optimization (BO) is used to model the path branch fracture probability, and the fracture probability quantization value is calculated in combination with the interference intensity ratio. To achieve this, a technique based on Kernel Regression (KR) is used to quickly find and mark the fracture regions. In addition, a method based on Evolutionary Algorithms (EA) is introduced to optimize the generation process of the fracture probability to ensure the accuracy of the path branch structure. Through this calculation of the fracture probability quantization value, the fracture probability quantization value is generated, and the fracture probability quantization value forms a dynamic mapping constraint relationship with the displacement amplitude of the shelf unit.

[0176] 303. Based on the fracture probability quantization value, dynamically correct the conflict propagation intensity coefficient through the gradient diffusion model of the spatio-temporal attenuation field. The gradient parameter of the spatio-temporal attenuation field is generated by the spatio-temporal coupling relationship between the drug expiration decay rate and the fracture probability quantization value, and is mapped to the hierarchical division of the multi-layer topological navigation grid.

[0177] In step 303, the gradient diffusion model of the spatio-temporal attenuation field is a model used to describe the conflict propagation speed and degree, and combines the drug expiration decay rate and the fracture probability quantization value to generate the conflict propagation intensity coefficient.

[0178] The spatio-temporal coupling relationship is the relationship between the drug expiration decay rate and the fracture probability quantization value, and is used to generate the gradient parameter of the spatio-temporal attenuation field.

[0179] In the embodiment of the present application, first, based on the fracture probability quantization value, the conflict propagation intensity coefficient is dynamically corrected through the gradient diffusion model of the spatio-temporal attenuation field. Specifically, a method based on the Variational Autoencoder (VAE) is used to optimize the path branch structure, and the conflict propagation intensity coefficient is generated in combination with the gradient diffusion model of the spatio-temporal attenuation field. To achieve this, a technique based on Manifold Learning (ML) is used to capture the change trend of the path in time and space, and the conflict propagation intensity coefficient is adjusted according to these changes. In addition, a method based on Simulated Annealing (SA) is introduced to further optimize the distribution of the conflict propagation intensity coefficient, ensuring that its dimension forms a strict mapping relationship with the hierarchical division of the multi-layer topological navigation grid. Through this dynamic correction, the conflict propagation intensity coefficient is finally generated.

[0180] 304. Adaptively fuse the conflict propagation intensity coefficient with the dynamic pooling weight parameter, and generate a path branch structure carrying conflict marks through the density distribution of geometric topological fracture symbols. The dynamic pooling weight parameter is triggered by the dynamic correlation relationship between the shelf coordinate offset and the terrain stress field coefficient, and the fracture probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit.

[0181] In step 304, the dynamic pooling weight parameter is data triggered by the dynamic correlation relationship between the shelf coordinate offset and the terrain stress field coefficient, and is used to adaptively fuse the conflict propagation intensity coefficient.

[0182] The density distribution of geometric topological fracture symbols is data used to indicate potential fracture points in the path, and generates a path branch structure carrying conflict marks.

[0183] In the embodiments of the present application, first, the conflict propagation intensity coefficient and the dynamic pooling weight parameter are adaptively fused, and a path branch structure carrying conflict markers is generated through the density distribution of geometric topology fracture symbols. Specifically, a method based on the Attention Mechanism (AM) is used to dynamically verify the conflict propagation intensity coefficient, and the conflict markers are corrected in combination with the dynamic pooling weight parameter. To achieve this, a technology based on Graph Convolutional Networks (GCN) is used to handle uncertainty and ambiguity problems to ensure the accuracy of the conflict markers. In addition, a method based on Reinforcement Learning (RL) is introduced to optimize the conflict marker remapping process to ensure a high degree of consistency between the path and the actual environment. Through this adaptive fusion, a path branch structure carrying conflict markers is generated, and a positive correlation constraint is formed between the fracture probability of the path branch structure and the displacement amplitude of the shelf unit.

[0184] The following is a specific example:

[0185] In the pharmaceutical cold chain warehousing scenario, the temperature in a biological product warehouse fluctuated by ±1.2 °C (preset threshold ±0.8 °C) due to a short-term failure of the temperature control equipment, triggering the dynamic displacement compensation mechanism of the shelf unit. The displacement sensor deployed at the bottom of the shelf detected that the X-axis offset of shelf No. B7-23 was ΔX = 18 cm and the Y-axis rotation angle was θ = 7°. The adjacent coupling parameter rose to 0.89 (threshold 0.85). The dynamic coupling parameter model constructed based on the Convolutional Neural Network (CNN) and Self-Organizing Map (SOM) algorithms showed that the interference intensity ratio between the shelf displacement amplitude of 0.45 m and the geometric topology spacing of the path branch of 1.2 m was 0.375 (safety threshold 0.3). At this time, the initial distribution of the fracture probability was calibrated through the Extreme Value Theory (EVT), and 3 high-risk areas (initial fracture probability values P 1 = 65%, P 2 = 53%, P 3 = 48%) were marked in the four-layer navigation grid of the digital twin. When the shelf displacement amplitude exceeded 0.5 m, the Bayesian optimization model calculated that the quantified value of the fracture probability rose to 82% (positive correlation coefficient r = 0.91). Combining with the expiration decay rate of this batch of interferon drugs (0.8% per hour), the gradient diffusion model of the spatio-temporal decay field corrected the conflict propagation intensity coefficient of the third-layer grid from 0.65 to 0.78 (mapping accuracy error ≤ 0.02). The dynamic pooling weight parameter was fused with the shelf coordinate offset (ΔX = 18 cm ± 2 cm) and the terrain stress field coefficient (0.93 N / m 2) An adaptive fusion weight ω = 0.88 is generated. Finally, the graph convolutional network (GCN) is used to remap the geometric topological fracture symbol density distribution, and 5 new diamond-shaped conflict markers are added at the 2-4 levels of the navigation grid (the density is increased to 3.6 per 100 square meters), forming an optimized path branch structure with high-precision conflict markers. Its fracture probability is dynamically and positively correlated with the shelf displacement amplitude (synchronization delay ≤ 30 ms), and the temperature fluctuation of drug storage is maintained and restored to the compliance range of ±0.5°C.

[0186] Through the implementation of steps 301 to 304, the accuracy and real-time performance of drug warehouse management are significantly improved, especially in handling the synchronization of multi-source spatio-temporal data in complex areas (such as high-humidity areas). It can identify anomalies in the storage environment earlier, and optimize the scheduling logic based on the comprehensive analysis of the temperature and humidity fluctuation coefficient and the expiration date gradient field to ensure that drugs are stored under the best conditions and follow the principle of expiration date priority. In addition, by dynamically adapting the passing parameters of the emergency channel, the path planning can be quickly adjusted in case of emergency to ensure the safety of personnel and materials. Overall, this solution greatly improves the intelligent level of warehouse management, enhances the ability to respond to emergencies, and ensures a high degree of consistency between the path and the actual environment through closed-loop dynamic verification, thereby improving the overall efficiency and safety of warehouse operations. At the same time, by real-time associating the displacement amplitude of the shelf unit with the fracture probability of the path branch, the safety and reliability of the path are ensured, further improving the refinement level of warehouse management.

[0187] To further improve the accuracy and real-time performance of drug warehouse management, especially in dealing with the time and space alignment of multi-source data, an optimization solution is proposed. Based on the fracture probability quantization value, this solution dynamically corrects the conflict propagation intensity coefficient through the gradient diffusion model of the spatio-temporal attenuation field, and generates the gradient parameters of the spatio-temporal attenuation field from the spatio-temporal coupling relationship between the drug expiration decay rate and the fracture probability quantization value, and maps them to the hierarchical division of the multi-layer topological navigation grid. This process not only improves the consistency and accuracy of the data, but also provides more effective operation suggestions for warehouse managers. In some embodiments, the dynamic correction of the conflict propagation intensity coefficient through the gradient diffusion model of the spatio-temporal attenuation field based on the fracture probability quantization value in step 303 includes:

[0188] 401. Based on the dynamic coupling relationship between the fracture probability quantization value and the crack propagation phase field variable, generate dynamic crack propagation parameters through the similarity constraint principle, and the crack propagation phase field variable is jointly calibrated by the drug expiration decay rate and the hierarchical division of the multi-layer topological navigation grid;

[0189] In step 401, the fracture probability quantization value is data calculated as the ratio of the displacement amplitude based on the shelf coordinate offset to the interference intensity of the geometric topological spacing of the path branch structure, and is used to quantify the safety of the path.

[0190] The crack propagation phase-field variable is data jointly calibrated by the drug expiration decay rate and the hierarchical division of the multi-layer topological navigation grid, and is used to describe the process of crack propagation.

[0191] The similarity constraint principle is a method for generating dynamic crack propagation parameters, based on the dynamic coupling relationship between the fracture probability quantization value and the crack propagation phase-field variable.

[0192] In the embodiment of the present application, first, a method based on phase-field fracture mechanics (PFFM) is used to process the path safety gap in the multi-layer topological navigation grid. Specifically, by real-time collecting the geometric and physical data in the warehouse and using PFFM to model these data, dynamic crack propagation parameters are generated. To achieve this, a spatial correlation algorithm based on the similarity constraint principle (SCP) is used, which can effectively capture the spatial relationship between shelf units and adjust its crack propagation behavior according to the adjacent coupling parameters. In addition, a method based on the crack propagation phase-field variable (CPPFV) is introduced to jointly calibrate the drug expiration decay rate and the hierarchical division of the multi-layer topological navigation grid. Finally, through this dynamic coupling relationship, dynamic crack propagation parameters are generated to ensure the accuracy and real-time nature of the path safety gap.

[0193] 402. Perform multi-physical field superposition on the dynamic crack propagation parameters and the equivalent permeability parameters of the fluid-solid coupling phase-field model to generate a conflict propagation intensity correction coefficient, where the equivalent permeability parameter is dynamically calculated through the spatio-temporal coupling relationship between the fluid pressure gradient at the crack tip and the fracture probability quantization value;

[0194] In step 402, the fluid-solid coupling phase-field model is a model for describing the interaction between fluid and solid, and is used to generate equivalent permeability parameters.

[0195] The equivalent permeability parameter is data dynamically calculated through the spatio-temporal coupling relationship between the fluid pressure gradient at the crack tip and the fracture probability quantization value, and is used to evaluate the safety of the path.

[0196] The conflict propagation intensity correction coefficient is data generated by performing multi-physical field superposition on the dynamic crack propagation parameters and the equivalent permeability parameters of the fluid-solid coupling phase-field model, and is used to correct the conflict propagation intensity coefficient.

[0197] In the embodiments of the present application, first, the dynamic crack propagation parameters are superimposed with the equivalent permeability parameters of the Fluid-Solid Coupling Phase-Field Model (FSCPFM) to generate a conflict propagation intensity correction coefficient. Specifically, a method based on Multiphysics Coupling (MPC) is used to optimize the path branching structure, and the equivalent permeability parameters are generated by combining the spatio-temporal coupling relationship between the fluid pressure gradient at the crack tip and the quantified fracture probability value. To achieve this, a technique based on Finite Element Analysis (FEA) is used to capture the changing trends of the path in time and space, and the conflict propagation intensity correction coefficient is adjusted according to these changes. In addition, a method based on Genetic Algorithms (GA) is introduced to further optimize the distribution of the conflict propagation intensity correction coefficient to ensure that its dimension forms a strict mapping relationship with the hierarchical division of the multi-layer topological navigation grid. Through this multi-physics field superposition, a path weight distribution based on the conflict propagation intensity correction coefficient is generated.

[0198] 403. Based on the conflict propagation intensity correction coefficient, the hierarchical division of the multi-layer topological navigation grid is dynamically recalibrated through the similarity constraint of the data-driven probability failure assessment model, and the similarity constraint is jointly triggered by the crack tip restraint state and the geometric topological spacing of the multi-layer topological navigation grid;

[0199] In step 403, the data-driven probability failure assessment model is a data-based probability model used to evaluate the failure probability of a path.

[0200] The similarity constraint is data jointly triggered by the crack tip restraint state and the geometric topological spacing of the multi-layer topological navigation grid, and is used to dynamically recalibrate the hierarchical division of the multi-layer topological navigation grid.

[0201] The conflict propagation intensity correction coefficient is data used to dynamically recalibrate the hierarchical division of the multi-layer topological navigation grid.

[0202] In the embodiments of the present application, first, based on the conflict propagation intensity correction coefficient, the hierarchical division of the multi-layer topological navigation grid is dynamically recalibrated through the similarity constraint of the Data-Driven Probabilistic Failure Assessment Model (DDPFAM). Specifically, the method based on Bayesian Networks (BN) is used to optimize the path branch structure, and the similarity constraint is jointly triggered by combining the crack tip restraint state and the geometric topological distance of the multi-layer topological navigation grid. To achieve this, the technology based on Support Vector Machine (SVM) is used to handle the uncertainty and ambiguity problems to ensure the accuracy of the path hierarchical division. In addition, the method based on Deep Reinforcement Learning (DRL) is introduced to optimize the hierarchical division process to ensure a high degree of consistency between the path and the actual environment. Through this dynamic recalibration, a new hierarchical division is generated to ensure the accuracy of the path branch structure.

[0203] 404. Adaptively match the dynamic recalibration result with the energy release rate threshold of the crack bifurcation phase field model, where the energy release rate threshold forms a dynamic constraint relationship through the gradient diffusion relationship between the crack propagation phase field variable and the drug expiration decay rate.

[0204] In step 404, the crack bifurcation phase field model is a model used to describe the crack bifurcation behavior and is used to generate the energy release rate threshold.

[0205] The energy release rate threshold is data that forms a dynamic constraint relationship through the gradient diffusion relationship between the crack propagation phase field variable and the drug expiration decay rate and is used to evaluate the safety of the path.

[0206] The dynamic recalibration result is data used for adaptive matching with the energy release rate threshold.

[0207] In the embodiments of the present application, first, the dynamic recalibration result is adaptively matched with the energy release rate threshold of the Crack Bifurcation Phase-Field Model (CBPFM). Specifically, a method based on the Energy Minimization Principle (EMP) is used to optimize the path branch structure, and the energy release rate threshold is generated by combining the gradient diffusion relationship between the crack propagation phase-field variable and the expiration date decay rate of the drug. To achieve this, a technique based on the Particle Filter (PF) is used to handle the uncertainty of the path branch structure, ensuring the accuracy of the energy release rate threshold. In addition, a method based on Evolutionary Strategies (ES) is introduced to optimize the generation process of the energy release rate threshold, ensuring a high degree of consistency between the path and the actual environment. Through this adaptive matching, a path branch structure with conflict markers is generated, ensuring that the fracture probability of the path branch structure is positively correlated with the displacement amplitude of the shelf unit.

[0208] The following is a specific example:

[0209] In the digital twin warehousing system of a pharmaceutical logistics distribution center, assume that when performing the inbound task of a batch of drugs sensitive to expiration date (the decay rate is set to 0.08% per hour), the digital twin of the warehouse real-time monitors that the coordinate offset parameter of the shelf unit due to mechanical vibration reaches 12.5 mm, and at the same time, the path branch spacing of the third layer of the multi-layer topological navigation grid in this area is 28 cm. The interference intensity ratio of the displacement amplitude to the path spacing calculated by the fracture probability quantification model is 0.45 (exceeding the safety threshold of 0.4), triggering the dynamic crack propagation mechanism. At this time, the crack propagation phase-field variable is jointly calibrated to 0.82 based on the expiration date decay rate of the drug and the third-layer grid weight coefficient (0.65), driving the similarity constraint algorithm to generate dynamic crack propagation parameters - specifically, the path detour coefficient is 1.3 and the obstacle avoidance radius expansion value is 15 cm.

[0210] This parameter is superimposed with the equivalent permeability parameter of 0.58 calculated by the fluid-structure interaction phase-field model (obtained from the spatio-temporal coupling of the fluid pressure gradient change caused by the shelf displacement and the fracture probability quantification value) to generate a conflict propagation intensity correction coefficient of 0.63. Based on this coefficient, the data-driven probability failure assessment model performs dynamic recalibration on the fourth-layer navigation grid: adjusts the grid spacing of the 7th - 9th shelf areas in the original planned path from 35 cm to 42 cm, and at the same time raises the drug turnover priority level in this area from secondary to primary. Finally, the dynamic recalibration result is compared with the energy release rate threshold of 1.8 J / m set by the crack bifurcation phase-field model. 2(Complete the adaptive matching based on the dynamic constraint between the expiration attenuation gradient of the current batch of drugs and the crack propagation phase field variable).

[0211] To further improve the accuracy and real-time performance of drug warehouse management, especially in converting the navigation path into an executable instruction set and driving the PDA device to project the path guidance signal, an optimization scheme is proposed. This scheme generates a phase encoding sequence through the dynamic coupling relationship between the geometric topological spacing of the navigation path and the shelf coordinate offset, and generates a path guidance signal with temperature and humidity compensation through a multimodal signal projection model. Finally, a positioning verification closed-loop is formed to ensure a high degree of consistency between the path and the actual environment, thereby improving the overall efficiency and safety of warehouse operations. In some embodiments, in step 104, converting the navigation path into an executable instruction set and driving the PDA device to project the path guidance signal includes:

[0212] 1041. Based on the dynamic coupling relationship between the geometric topological spacing of the navigation path and the shelf coordinate offset, generate the phase encoding sequence of the executable instruction set through dynamic path modulation parameters, where the dynamic path modulation parameters are jointly calibrated by the interference intensity ratio of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude;

[0213] In step 1041, the dynamic path modulation parameters are data jointly calibrated by the interference intensity ratio of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude, and are used to generate the phase encoding sequence.

[0214] The phase encoding sequence is data generated based on the dynamic coupling relationship between the geometric topological spacing of the navigation path and the shelf coordinate offset, and is used to drive the PDA device to project the path guidance signal.

[0215] In the embodiments of the present application, first, a method based on Topological Data Analysis (TDA) is used to process the geometric topological spacing of the navigation path. Specifically, by collecting the geometric and physical data in the warehouse in real time and using TDA to model these data, dynamic path modulation parameters are generated. To achieve this, a spatial correlation algorithm based on the Interference Intensity Ratio (IIR) is used, which can effectively capture the spatial relationship between shelf units and adjust its path modulation behavior according to the temperature and humidity fluctuation coefficient. In addition, a method based on Bayesian Inference (BI) is introduced to jointly calibrate the interference intensity ratio of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude. Finally, through this dynamic coupling relationship, the phase encoding sequence of the executable instruction set is generated to ensure the accuracy and real-time performance of the path modulation parameters.

[0216] 1042. Nonlinearly superimpose the phase encoding sequence with the frequency domain response parameters of the multimodal signal projection model to generate a path guidance signal with temperature and humidity compensation, where the frequency domain response parameters are dynamically matched with the curvature radius of the navigation path through the projection aperture geometric characteristics of the PDA device;

[0217] In step 1042, the frequency domain response parameters are data generated by dynamically matching the projection aperture geometric characteristics of the PDA device with the curvature radius of the navigation path, and are used to generate a path guidance signal with temperature and humidity compensation.

[0218] Nonlinear superposition is a method for combining the phase encoding sequence with the frequency domain response parameters to generate a path guidance signal with temperature and humidity compensation.

[0219] In the embodiment of the present application, first, nonlinearly superimpose the phase encoding sequence with the frequency domain response parameters of the Multimodal Signal Projection Model (MSPM) to generate a path guidance signal with temperature and humidity compensation. Specifically, use the method based on Hilbert-Huang Transform (HHT) to optimize the path guidance signal, and dynamically match the projection aperture geometric characteristics of the PDA device with the curvature radius of the navigation path. To achieve this, use the technology based on Fourier Transform (FT) to capture the changing trend of the path in time and frequency, and adjust the frequency domain response parameters according to these changes. In addition, introduce the method based on Genetic Programming (GP) to further optimize the distribution of the frequency domain response parameters to ensure the best nonlinear superposition effect with the phase encoding sequence. Through this nonlinear superposition, a path guidance signal with temperature and humidity compensation is generated to ensure the accuracy and reliability of the path guidance signal.

[0220] 1043. Based on the optical diffraction field distribution of the path guidance signal with temperature and humidity compensation, dynamically compensate the phase distortion of the projection signal through the phase field constraint model, where the phase field constraint model is jointly triggered by the shelf coordinate offset and the topological fracture probability of the navigation path to form the initial constraint condition of the positioning verification closed loop;

[0221] In step 1043, the optical diffraction field distribution is data generated based on the path guidance signal with temperature and humidity compensation, and is used to evaluate the quality of the path guidance signal.

[0222] The phase field constraint model is data jointly triggered by the shelf coordinate offset and the topological fracture probability of the navigation path, and is used to dynamically compensate the phase distortion of the projection signal.

[0223] The initial constraint condition is the initial constraint condition for forming a positioning verification closed loop, ensuring the accuracy of the path guidance signal.

[0224] In the embodiment of the present application, first, based on the optical diffraction field distribution of the path guidance signal with temperature and humidity compensation, the phase distortion of the projection signal is dynamically compensated through the Phase-Field Constraint Model (PFCM). Specifically, a method based on Compressive Sensing (CS) is used to optimize the path guidance signal, and the phase-field constraint model is jointly triggered by combining the shelf coordinate offset and the topological break probability of the navigation path. To achieve this, a technique based on Zernike Polynomials (ZP) is used to handle the uncertainty of the path guidance signal, ensuring the accuracy of the phase distortion compensation. In addition, a method based on DeepLearning (DL) is introduced to optimize the phase distortion compensation process, ensuring a high degree of consistency between the path and the actual environment. Through this dynamic compensation, the initial constraint condition of the positioning verification closed loop is generated, ensuring the accuracy and reliability of the path guidance signal.

[0225] 1044. Adaptively match the dynamic compensation result with the closed-loop phase-field energy threshold to generate a positioning verification closed loop, where the closed-loop phase-field energy threshold forms a dynamic balance constraint through the spatio-temporal conduction mechanism of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude.

[0226] In step 1044, the closed-loop phase-field energy threshold is data that forms a dynamic balance constraint through the spatio-temporal conduction mechanism of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude, and is used to generate a positioning verification closed loop.

[0227] Adaptive matching is a method for combining the dynamic compensation result with the closed-loop phase-field energy threshold to generate a positioning verification closed loop.

[0228] In the embodiments of the present application, first, the dynamic compensation result is adaptively matched with the closed-loop phase-field energy threshold to generate a positioning verification closed-loop. Specifically, a method based on Adaptive Filtering (AF) is used to dynamically verify the path guidance signal, and a spatio-temporal conduction mechanism combining the temperature and humidity fluctuation coefficient and the shelf displacement amplitude forms a dynamic balance constraint. To achieve this, a technique based on the Kalman Filter (KF) is used to process the dynamic changes of the path guidance signal to ensure the accuracy of the closed-loop phase-field energy threshold. In addition, a method based on Evolutionary Strategies (ES) is introduced to optimize the generation process of the closed-loop phase-field energy threshold to ensure a high degree of consistency between the path and the actual environment. Through this adaptive matching, a positioning verification closed-loop is generated to ensure a high degree of consistency and reliability between the path guidance signal and the actual environment.

[0229] The following is a specific example:

[0230] In the actual application of a pharmaceutical cold chain warehouse, assume that a certain distribution center needs to optimize the navigation path in a temperature and humidity sensitive environment (the set temperature fluctuation coefficient is ±0.5°C, and the relative humidity fluctuates by ±3%RH). The displacement sensor deployed on the shelf monitors the shelf coordinate offset of ±5 cm in real time (caused by mechanical vibration). Combining the topological data analysis collected by the digital twin system, it shows that the geometric spacing of the shelf group in Area B3 is 2.8 m, and there is a topological break probability of 0.15 m from the standard path planning. At this time, the interference intensity ratio calculated by Bayesian inference for the temperature and humidity fluctuation coefficient and the shelf displacement amplitude is 0.85, and an instruction sequence containing 128-bit phase encoding (encoding period 30 seconds / time) is dynamically generated. This sequence is dynamically matched with the path curvature radius (R = 3.5 m) through the PDA projection aperture (diameter 50 mm) to generate a 532 nm laser guidance signal carrying ±0.3°C temperature and humidity compensation.

[0231] When performing a picking task along Channel E7, the phase-field constraint model detects that the phase distortion caused by the shelf displacement reaches 12%, triggering a compensation algorithm based on Zernike polynomials to correct the wavefront aberration of the projected signal from λ / 4 to λ / 20 (λ = 632.8 nm). At this time, the closed-loop phase-field energy threshold is dynamically adjusted to 1200 J / m according to the spatio-temporal conduction mechanism. 2 The compensated signal energy (1150 J / m 2 ) is matched with the threshold through the Kalman filter to generate a positioning verification closed-loop. Finally, the positioning error at the coordinate point (X = 23.6 m, Y = 7.8 m) is reduced from 15 cm to 2 cm, achieving high-precision picking of frozen drugs (storage requirement 2 - 8°C), and at the same time improving the logistics efficiency of the channels in Area D by 40%.

[0232] By implementing steps 1401 to 1404, the accuracy and real-time performance of pharmaceutical warehouse management have been significantly improved, especially in handling the synchronization of multi-source spatio-temporal data in complex areas (such as high-humidity areas). It can identify anomalies in the storage environment earlier, and based on the comprehensive analysis of the temperature and humidity fluctuation coefficient and the expiration date gradient field, optimize the scheduling logic to ensure that drugs are stored under the best conditions and follow the principle of expiration date priority. In addition, by dynamically adapting the passage parameters of the emergency channel, the path planning can be quickly adjusted in case of emergencies to ensure the safety of personnel and materials. Overall, this solution has greatly improved the intelligent level of warehouse management, enhanced the ability to respond to emergencies, and ensured a high degree of consistency between the path and the actual environment through closed-loop dynamic verification, thereby improving the overall efficiency and safety of warehouse operations. At the same time, by real-time associating the displacement amplitude of the shelf unit with the path branch break probability, the safety and reliability of the path are ensured, further improving the refinement level of warehouse management. This method also ensures the efficiency and reliability of path planning through multi-physical field superposition and dynamic compensation.

[0233] Figure 2 The following is a schematic structural diagram of a pharmaceutical inventory management system based on a PDA provided by an embodiment of the present application, as Figure 2 shown, the device includes:

[0234] An encoding module 21, configured to construct a three-dimensional shelf topology structure and embed an expiration date gradient field, and generate a dynamically attributed shelf unit carrying a temperature and humidity fluctuation coefficient and an adjacent coupling parameter in combination with the temperature and humidity data stream, where the adjacent coupling parameter represents the spatial interference relationship between adjacent shelf units;

[0235] A generation module 22, configured to generate a shelf coordinate offset based on the temperature and humidity fluctuation coefficient, and establish a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin, where the virtual turnover sequence forms a spatio-temporal constraint relationship with the expiration date gradient field;

[0236] A matching module 23, configured to dynamically match the spatio-temporal constraint relationship with the emergency channel passage parameters to generate a navigation grid, and trigger grid topology reconstruction through the adjacent coupling parameter to generate a navigation path with conflict marks;

[0237] A conversion module 24, configured to convert the navigation path into an executable instruction set, and drive the PDA device to project a path guidance signal, where the path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop.

[0238] Figure 2 The described pharmaceutical inventory management system based on a PDA can execute Figure 1For a drug inventory management method based on a PDA as described in the illustrated embodiment, its implementation principle and technical effects will not be elaborated further. For a drug inventory management system based on a PDA in the above embodiment, the specific ways in which each module and unit perform operations have been described in detail in the embodiments related to this method and will not be elaborated here.

[0239] In a possible design, Figure 2 A drug inventory management system based on a PDA as shown in the illustrated embodiment can be implemented as a computing device, such as Figure 3 shown, this computing device may include a storage component 31 and a processing component 32;

[0240] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0241] The processing component 32 is used for the Figure 1 drug inventory management method based on a PDA in the above

[0242] embodiment. Among them, the processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components for executing the above method.

[0243] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0244] Of course, the computing device will necessarily also include other components, such as input / output interfaces, display components, communication components, etc.

[0245] The input / output interface provides an interface between the processing component and the peripheral interface module, and the above peripheral interface module can be an output device, an input device, etc.

[0246] The communication component is configured to facilitate communication between the computing device and other devices in a wired or wireless manner, etc.

[0247] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server. The above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from a cloud computing platform.

[0248] The embodiments of the present application also provide a computer storage medium storing a computer program, and when the computer program is executed by a computer, it can implement the above Figure 1 A method for managing drug inventory based on a PDA shown in the embodiments.

[0249] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0250] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0251] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0252] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them. Although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A drug inventory management method based on PDA, characterized in that: include: Construct a three-dimensional shelf topology structure and embed the validity period gradient field, combine the temperature and humidity data stream to generate a dynamic attribute coded shelf unit carrying the temperature and humidity fluctuation coefficient and the neighbor coupling parameter, and the neighbor coupling parameter represents the spatial interference relationship between adjacent shelf units; Generate a shelf coordinate offset based on the temperature and humidity fluctuation coefficient, establish a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin, and form a spatiotemporal constraint relationship between the virtual turnover sequence and the validity period gradient field; Dynamically matching the spatiotemporal constraint relationship with the emergency channel traffic parameters to generate a navigation grid, triggering grid topology reconstruction through the proximity coupling parameters and generating a navigation path with conflict markers; The navigation path is converted into an executable instruction set to drive the PDA device to project a path guidance signal. The path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop.

2. The method according to claim 1, characterized in that The step of generating a shelf coordinate offset based on the temperature and humidity fluctuation coefficient, establishing a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin, and forming a spatiotemporal constraint relationship between the virtual turnover sequence and the validity period gradient field, includes: Establishing a dynamic environmental response model of the temperature and humidity fluctuation coefficient and the three-dimensional shelf topology structure, generating a shelf coordinate offset by real-time correlation calculation between the physical position of the shelf unit and the temperature and humidity data stream, and the shelf coordinate offset drives the dynamic correction of the spatial coordinates of the shelf unit in the warehouse digital twin; The shelf coordinate offset and the decay rate of the validity period gradient field are temporally and spatially superimposed to generate a priority weight distribution of the shelf unit, wherein the priority weight distribution quantitatively sorts the spatial interference intensity of the adjacent shelf units based on the proximity coupling parameter to form a scheduling logic of a virtual turnover sequence; Dynamically adapting the scheduling logic of the virtual turnover sequence to the geometric constraints of the emergency channel passage parameters, dynamically adjusting the scheduling order of the shelf units through the interference intensity threshold triggered by the proximity coupling parameter, and forming a path rule compatible with the emergency channel passage parameters; The path rule is parameter-bound with the scheduling logic of the virtual turnover sequence, and the spatial coordinates of the shelf units of the warehouse digital twin are dynamically updated to form a spatiotemporal constraint relationship between the virtual turnover sequence and the spatiotemporal attenuation rate of the validity period gradient field.

3. The method according to claim 1, characterized in that The dynamically matching the spatiotemporal constraint relationship with the emergency passage parameters to generate a navigation grid, triggering grid topology reconstruction through the proximity coupling parameters and generating a navigation path with conflict marks, includes: Dynamically matching the spatiotemporal constraint relationship with the geometric topological features of the emergency passage parameters, generating a multi-layer topological navigation grid based on the spatial interference intensity of the proximity coupling parameters, wherein the hierarchical distribution of the multi-layer topological navigation grid forms a dynamic binding relationship with the priority of the emergency passage parameters; Calculating the path safety gap of the multi-layer topological navigation grid by using the interference intensity threshold triggered by the proximity coupling parameter to generate a path branch structure carrying a conflict mark, wherein the conflict mark forms a dynamic association constraint with the shelf coordinate offset; The path branch structure carrying the conflict mark is parameter-superimposed with the spatiotemporal decay rate of the validity period gradient field to generate a path weight distribution based on the conflict propagation coefficient, wherein the dimension of the path weight distribution forms a strict mapping relationship with the hierarchical distribution of the multi-layer topology navigation grid; The path weight distribution and the shelf unit space coordinates of the warehouse digital twin are subjected to closed-loop dynamic verification. Through the shelf coordinate offset correction process triggered by the conflict propagation coefficient, the conflict mark of the path branch structure and the spatiotemporal attenuation rate of the validity period gradient field form a rigid synchronization constraint, thereby generating a navigation path with conflict marks.

4. The method according to claim 3, characterized in that: The method of calculating the path safety gap of the multi-layer topological navigation grid according to the interference strength threshold triggered by the proximity coupling parameter to generate a path branch structure carrying a conflict mark, wherein the conflict mark forms a dynamic association constraint with the shelf coordinate offset, includes: Based on the interference intensity threshold triggered by the proximity coupling parameter, dynamically calculate the path safety gap of the multi-layer topology navigation grid, wherein the dynamic calculation process of the path safety gap integrates the shelf coordinate offset and the physical spacing constraint of the shelf unit to generate an initial distribution of conflict markers; The dynamic calculation result is associated with the dynamic change of the shelf coordinate offset in real time, and the path branch break probability calculation triggered by the interference intensity threshold is used to generate a path branch structure carrying a conflict mark, wherein the break probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit; The fracture probability of the path branch structure is superimposed on the spatiotemporal decay rate of the effective period gradient field to generate a conflict propagation intensity coefficient, wherein the distribution dimension of the conflict propagation intensity coefficient forms a strict mapping relationship with the hierarchical division of the multi-layer topology navigation grid; The conflict propagation intensity coefficient is dynamically bound to the shelf unit spatial coordinates of the warehouse digital twin, and the conflict mark remapping is triggered by the fracture probability threshold of the path branch structure to generate a path branch structure carrying the conflict mark. The conflict mark and the shelf coordinate offset form a dynamic association constraint.

5. The method according to claim 4, characterized in that The method of associating the dynamic calculation result with the dynamic change of the shelf coordinate offset in real time, calculating the path branch break probability triggered by the interference intensity threshold, and generating a path branch structure carrying a conflict mark, wherein the break probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit, includes: The dynamic calculation result is associated with the dynamic change of the shelf coordinate offset in real time through the dynamic coupling parameter triggered by the interference intensity threshold, and the dynamic coupling parameter generates an initial distribution of fracture probability based on the physical displacement amplitude of the shelf unit and the geometric topological spacing constraint of the path branch structure; Based on the initial distribution of the fracture probability, a fracture probability quantization value is calculated by the ratio of the displacement amplitude of the shelf coordinate offset to the interference intensity of the geometric topological spacing of the path branch structure, and the fracture probability quantization value forms a dynamic mapping constraint relationship with the displacement amplitude of the shelf unit; Based on the quantized value of the fracture probability, the conflict propagation intensity coefficient is dynamically corrected through a gradient diffusion model of a spatiotemporal attenuation field, wherein the gradient parameter of the spatiotemporal attenuation field is generated by the spatiotemporal coupling relationship between the drug efficacy decay rate and the quantized value of the fracture probability, and is mapped to the hierarchical division of the multi-layer topological navigation grid; The conflict propagation intensity coefficient is adaptively fused with the dynamic pooling weight parameter, and a path branch structure carrying a conflict mark is generated through the density distribution of geometric topological fracture symbols. The dynamic pooling weight parameter is triggered by the dynamic correlation between the shelf coordinate offset and the terrain stress field coefficient. The fracture probability of the path branch structure forms a positive correlation constraint with the displacement amplitude of the shelf unit.

6. The method according to claim 5, characterized in that Based on the quantized value of the fracture probability, the conflict propagation intensity coefficient is dynamically corrected through a gradient diffusion model of a spatiotemporal attenuation field, wherein the gradient parameter of the spatiotemporal attenuation field is generated by the spatiotemporal coupling relationship between the drug efficacy attenuation rate and the quantized value of the fracture probability, and is mapped to the hierarchical division of the multi-layer topology navigation grid, including: Based on the dynamic coupling relationship between the fracture probability quantization value and the crack extension phase field variable, a dynamic crack extension parameter is generated by the similarity constraint principle, and the crack extension phase field variable is jointly calibrated by the drug shelf life decay rate and the hierarchical division of the multi-layer topological navigation grid; Performing multi-physics field superposition on the dynamic crack extension parameter and the equivalent permeability parameter of the fluid-solid coupling phase field model to generate a conflict propagation intensity correction coefficient, wherein the equivalent permeability parameter is dynamically calculated through the spatiotemporal coupling relationship between the crack tip fluid pressure gradient and the fracture probability quantization value; Based on the conflict propagation strength correction coefficient, the hierarchical division of the multi-layer topological navigation grid is dynamically recalibrated through the similarity constraint of the data-driven probabilistic failure assessment model, wherein the similarity constraint is jointly triggered by the crack tip constraint state and the geometric topological spacing of the multi-layer topological navigation grid; The dynamic recalibration result is adaptively matched with the energy release rate threshold of the crack bifurcation phase field model, and the energy release rate threshold forms a dynamic constraint relationship through the gradient diffusion relationship between the crack extension phase field variable and the drug shelf life decay rate.

7. The method according to claim 1, characterized in that The method converts the navigation path into an executable instruction set, drives the PDA device to project a path guidance signal, and the path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop, including: Based on the dynamic coupling relationship between the geometric topological spacing of the navigation path and the shelf coordinate offset, a phase coding sequence of an executable instruction set is generated through dynamic path modulation parameters, wherein the dynamic path modulation parameters are jointly calibrated by the interference intensity ratio of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude; The phase coding sequence is nonlinearly superimposed with the frequency domain response parameters of the multimodal signal projection model to generate a path guidance signal carrying temperature and humidity compensation, wherein the frequency domain response parameters are dynamically matched with the curvature radius of the navigation path through the projection aperture geometric characteristics of the PDA device; Based on the optical diffraction field distribution of the path guidance signal carrying temperature and humidity compensation, the phase distortion of the projection signal is dynamically compensated through a phase field constraint model, and the phase field constraint model is jointly triggered by the shelf coordinate offset and the topological fracture probability of the navigation path to form the initial constraint condition of the positioning verification closed loop; The dynamic compensation result is adaptively matched with the closed-loop phase field energy threshold to generate a positioning verification closed loop. The closed-loop phase field energy threshold forms a dynamic balance constraint through the spatiotemporal conduction mechanism of the temperature and humidity fluctuation coefficient and the shelf displacement amplitude.

8. A drug inventory management system based on PDA, characterized in that: include: The encoding module is used to construct a three-dimensional shelf topology structure and embed the validity period gradient field, and generate a dynamic attribute encoding shelf unit carrying the temperature and humidity fluctuation coefficient and the adjacent coupling parameter in combination with the temperature and humidity data stream. The adjacent coupling parameter represents the spatial interference relationship between adjacent shelf units; A generation module, used to generate a shelf coordinate offset based on the temperature and humidity fluctuation coefficient, and establish a virtual turnover sequence including the shelf coordinate offset in the warehouse digital twin, wherein the virtual turnover sequence forms a spatiotemporal constraint relationship with the validity period gradient field; A matching module, used for dynamically matching the spatiotemporal constraint relationship with the emergency channel traffic parameters to generate a navigation grid, triggering grid topology reconstruction through the proximity coupling parameters and generating a navigation path with conflict marks; The conversion module is used to convert the navigation path into an executable instruction set to drive the PDA device to project a path guidance signal, wherein the path guidance signal carries the temperature and humidity fluctuation coefficient and the shelf coordinate offset to form a positioning verification closed loop.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a PDA-based drug inventory management method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that: A computer program is stored, and when the computer program is executed by a computer, a drug inventory management method based on PDA as described in any one of claims 1 to 7 is implemented.

Images