Dynamic medical warehouse management method and device based on unmanned aerial vehicle, and medium

Abstract

The embodiment of the invention discloses a dynamic medical warehouse management method and device based on an unmanned aerial vehicle, and a medium, and relates to the technical field of warehouse management, and the method comprises the steps: obtaining dynamic device operation data in a dynamic medical warehouse and real-time warehouse temperature data in a plurality of preset warehouse regions, and carrying out the real-time warehouse temperature data based on the real-time warehouse temperature data; predicting the environment temperature so as to determine the single inspection duration corresponding to the unmanned aerial vehicle inspection; according to the dynamic equipment operation data and the static warehouse distribution information, determining a space-time conflict sub-region, planning an inspection path corresponding to unmanned aerial vehicle inspection according to the space-time conflict sub-region and the single inspection duration, and determining an unmanned aerial vehicle inspection path in each preset warehouse region; and inspecting each preset warehouse area according to the unmanned aerial vehicle inspection path, collecting corresponding area inspection information, performing storage risk analysis on the dynamic medical warehouse based on the area inspection information, determining goods risk data, and performing goods management on the dynamic medical warehouse.

Description

Technical Field

[0001] This specification relates to the technical field of warehouse management, and particularly to a dynamic medical warehouse management method, device and medium based on unmanned aerial vehicles (UAVs). Background Art

[0002] In the field of medical supplies warehousing, especially in medical cold chain warehouses that require strict temperature control for vaccines, biological agents, etc., the low-temperature environment (usually -25 ℃ to 4 ℃ ) and the highly dynamic operation scenarios pose double challenges to warehousing management. In the traditional manual inspection mode, operators need to frequently enter and exit the cold storage to perform tasks such as inventory counting and expiration verification of goods. Low-temperature exposure is likely to cause health risks, and the heavy protective equipment greatly reduces work efficiency. On the other hand, the precise management of medical supplies requires real-time knowledge of the storage dynamics. However, the interval period of manual inspections is long, and it is difficult to promptly discover problems such as misplacement of goods in storage locations, which may lead to out-of-control expiration dates of drugs or quality and safety accidents.

[0003] Currently, automated guided vehicles (AGVs) are generally introduced in medical warehouses to realize the outbound transportation of goods, and a liftable pallet system is adopted to optimize the utilization rate of vertical space. For example, drugs with high-frequency access are allocated to the middle-layer shelves. Warehouses that adopt such AGVs or liftable pallets are called dynamic medical warehouses. In a dynamic medical warehouse, while the dynamic equipment improves the operation efficiency, it also forms continuous moving obstacles. During the inspection of other warehouses, UAV inspections can be carried out during the time when the dynamic equipment is not running. However, in a medical warehouse, there is a possibility that dynamic equipment such as drug retrieval is running at any time. In this case, the preset path of the AGV overlaps with the inspection area of the UAV, and the automatic lifting of the pallet causes the real-time change of the shelf structure. Most of the existing UAV inspection technologies are designed for static warehouse environments, and their obstacle avoidance logic relies on lidar or visual SLAM (simultaneous localization and mapping) algorithms. Traditional path planning algorithms (such as A* and RRT) only avoid obstacles based on static maps and do not consider the influence of the movement of AGVs or the lifting of pallets. Sudden obstacles force the UAV to frequently hover emergently, resulting in task interruption. In addition, the battery life of commercial UAVs decays significantly in environments below -10 ℃ which makes it difficult to meet the continuous inspection requirements of medical warehouses.

[0004] Therefore, when a UAV device conducts inspections in a medical cold chain warehouse with dynamic equipment, the inspection process is affected by dynamic obstacles and low-temperature battery life, making it difficult to meet the continuous inspection requirements of medical warehouses, resulting in risks in the storage process of goods in the warehouse. Summary of the Invention

[0005] One or more embodiments of this specification provide a method, device, and medium for dynamic medical warehouse management based on unmanned aerial vehicles (UAVs) to solve the following technical problems: When a UAV conducts inspections in a medical cold chain warehouse with dynamic equipment, the inspection process is affected by dynamic obstacles and low-temperature endurance, making it difficult to meet the continuous inspection requirements of the medical warehouse and resulting in risks during the storage process of goods in the warehouse.

[0006] One or more embodiments of this specification adopt the following technical solutions:

[0007] One or more embodiments of this specification provide a method for dynamic medical warehouse management based on UAVs. The method includes: obtaining dynamic equipment operation data in the dynamic medical warehouse and real-time warehouse temperature data in multiple preset warehouse areas, predicting the environmental temperature of the preset warehouse areas based on the real-time warehouse temperature data to determine the single inspection duration corresponding to the UAV inspection; determining space-time conflict sub-areas through the dynamic equipment operation data and the static warehouse distribution information corresponding to the preset warehouse areas, and planning the inspection path corresponding to the UAV inspection according to the space-time conflict sub-areas and the single inspection duration to determine the UAV inspection path in each preset warehouse area; conducting inspections on each preset warehouse area according to the UAV inspection path, collecting corresponding area inspection information, and performing storage risk analysis on the dynamic medical warehouse based on the area inspection information to determine goods risk data for managing the goods in the dynamic medical warehouse.

[0008] One or more embodiments of this specification provide a device for dynamic medical warehouse management based on UAVs, including:

[0009] At least one processor; and,

[0010] A memory communicatively connected to the at least one processor; wherein,

[0011] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0012] A non-volatile computer storage medium provided by one or more embodiments of this specification stores computer-executable instructions, and the computer-executable instructions are set to: execute the above method.

[0013] The above at least one technical solution adopted in the embodiments of this specification can achieve the following beneficial effects: Through the embodiments of this specification, by obtaining real-time warehouse temperature data and predicting the environmental temperature of a preset warehouse area, the single inspection duration corresponding to the UAV inspection is determined, and according to the battery life of the UAV under different temperature conditions, the inspection duration is reasonably planned to avoid the interruption of the inspection task due to insufficient battery life; in a dynamic medical warehouse, the preset path of the AGV overlaps with the inspection area of the UAV, and the automatic lifting of the pallet causes the real-time change of the shelf structure. Most of the existing UAV inspection technologies are designed for static warehouse environments and are difficult to handle these dynamic obstacles. The embodiments of this specification determine the spatio-temporal conflict sub-areas by obtaining dynamic equipment operation data and combining static warehouse distribution information, and plan the UAV inspection path accordingly, so that the UAV can avoid the activity areas and times of dynamic equipment, effectively avoid conflicts with the movement of the AGV or the lifting of the pallet, reduce the frequent emergency hovering and task interruption caused by sudden obstacles, ensure the smooth progress of the UAV inspection task, and improve the inspection efficiency and safety; the traditional manual inspection mode has problems such as long interval periods and difficulty in timely detecting misplacement of goods locations, and it is difficult to meet the requirements of precise management of medical supplies for real-time grasping of storage dynamics. The embodiments of this specification use the UAV to inspect the preset warehouse area according to the planned inspection path, collect area inspection information, including RFID scan data and goods location coordinate data, and through the storage risk analysis of these information, the goods risk data can be accurately determined, such as expiration risk level, storage location deviation, and temperature zone compliance identification, etc. Based on these data, the goods are managed, and problems such as misplacement of goods locations and out-of-control expiration of drugs can be timely discovered and processed, improving the accuracy and timeliness of medical supplies management and reducing the probability of quality and safety accidents; a variety of technical means such as environmental temperature prediction, dynamic equipment operation data analysis, UAV inspection path planning, and goods risk analysis are comprehensively used to realize the intelligent management of dynamic medical warehouses. From determining the inspection duration, planning the inspection path to conducting storage risk analysis and goods management, the entire process has a high degree of automation, reduces manual intervention, and improves the management efficiency and scientific nature of decision-making; it can obtain dynamic equipment operation data and real-time warehouse temperature data in real time, and adjust the inspection strategy and path planning in a timely manner according to these data. By reasonably planning the UAV inspection path and duration, the damage and maintenance costs of the UAV caused by battery life problems and conflicts with dynamic equipment are reduced; on the other hand, precise goods management can avoid material waste and economic losses caused by problems such as misplacement of goods locations and expiration of drugs, and the automated management method reduces the labor cost required for manual inspection, thus effectively reducing the operating cost of the medical warehouse. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] To more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments recorded in this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings:

[0015] Figure 1 It is a schematic flow chart of a dynamic medical warehouse management method based on an unmanned aerial vehicle provided by an embodiment of this specification;

[0016] Figure 2 It is a schematic structural diagram of a dynamic medical warehouse management device provided by an embodiment of this specification. Detailed implementation manners

[0017] In order to enable those in the technical field to better understand the technical solutions in this specification, the following will clearly and completely describe the technical solutions in the embodiments of this specification in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments of this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this specification.

[0018] An embodiment of this specification provides a dynamic medical warehouse management method based on an unmanned aerial vehicle. It should be noted that the execution subject in the embodiments of this specification can be a server or any device with data processing capabilities. Figure 1 It is a schematic flow chart of a dynamic medical warehouse management method based on an unmanned aerial vehicle provided by an embodiment of this specification, as Figure 1 shown, mainly including the following steps:

[0019] Step S101, obtain the dynamic device operation data in the dynamic medical warehouse and the real-time warehouse temperature data in multiple preset warehouse areas, and based on the real-time warehouse temperature data, predict the environmental temperature of the preset warehouse area to determine the single inspection duration corresponding to the drone inspection.

[0020] In an embodiment of this specification, the real-time warehouse temperature data in the preset warehouse area is collected through temperature sensors set in the medical warehouse. It should be noted that the preset warehouse area here is the inspection area corresponding to the drone device. If the warehouse is a small warehouse, only one drone device can be used to achieve warehouse inspection, then the preset warehouse area here is the dynamic medical warehouse; if multiple drone devices are used for warehouse inspection in the warehouse, the inspection area of the drone is used as the basis for area division, and the inspection area corresponding to the drone is determined as the preset warehouse area.

[0021] A dynamic medical warehouse refers to a warehouse with automated equipment operations inside. For example, automated guided vehicles (AGVs) are used to transport goods out of the warehouse, and a liftable pallet system is adopted to automatically lift the position of the pallets in the shelves, so as to improve the space utilization rate of the shelves in the vertical direction. Data linkage is carried out with the control system corresponding to the above-mentioned dynamic equipment to obtain the operation data of the dynamic equipment in the dynamic medical warehouse. It should be noted that the operation data of the dynamic equipment here can be the data related to the expected operation of the dynamic equipment, that is, the operation plan.

[0022] Different from other types of warehouses, especially in medical cold chain warehouses, the environment is usually at a low temperature. Generally, the warehouse environment is -25 ℃ to 4°C. In a low-temperature environment, the endurance of UAV equipment is significantly reduced. During the process of using UAVs for warehouse management, the single inspection duration of UAVs is related to management efficiency and the continuity of the management process.

[0023] Based on the real-time warehouse temperature data, predict the environmental temperature of the preset warehouse area to determine the single inspection duration corresponding to the UAV inspection, specifically including: obtaining the historical warehouse temperature data within the preset warehouse area, where the preset warehouse area is the inspection task execution area corresponding to the pre-set UAV; training the pre-constructed temperature prediction model based on the historical warehouse temperature data, and determining the predicted temperature time series of the preset warehouse area according to the real-time warehouse temperature data and the temperature prediction model; obtaining the pre-generated UAV endurance mapping data, where the UAV endurance mapping data includes multiple temperature intervals and the benchmark endurance duration of the UAV within each temperature interval; determining the single inspection duration corresponding to the UAV inspection based on the predicted temperature time series and the UAV endurance mapping data.

[0024] In one embodiment of the present specification, historical warehouse temperature data within a preset warehouse area is obtained. The preset warehouse area is the area where the inspection tasks corresponding to the drones are to be executed as pre-set, and the historical temperatures within each area are used for training the prediction model. The temperature data is uploaded at 5-second intervals by temperature sensors deployed in each partition of the warehouse, and the data format can be {area ID, timestamp, temperature value}. It may also include the status of the refrigeration system, such as the start / stop status of the cold storage compressor, the air supply volume, the set temperature, etc. Here, in addition to the above methods, it may also include the record of the opening and closing of the warehouse door, and the opening and closing time can be recorded by a door magnetic sensor. The temperature prediction model here can be a Long Short-Term Memory (LSTM) temperature prediction model. The input layer in the model structure can be set to 60-step historical data (5 seconds / step, a total of 5 minutes), including multiple features such as temperature and refrigeration intensity. The hidden layer includes 2 layers of LSTM, with 128 units in each layer, and Dropout = 0.2. The output layer is the temperature prediction sequence for the next 120 steps (10 minutes). The loss function during the training process can be set to the mean squared error (MSE) and a temperature abrupt change penalty term. An example of the loss function is as follows: Loss = MSE + λ·Σ|ΔT_pred|2, where ΔT_pred is the difference between adjacent predicted temperatures, λ = 0.1, and the Adam optimizer is used with a learning rate of 0.001. The training data is the warehouse operation data for the past 3 months (including summer / winter working conditions).

[0025] It should be noted that here, real-time temperature data should be combined. Relying solely on historical data or periodic sampling may miss instantaneous fluctuations, such as the temperature rise caused by the opening of the warehouse door for 30 seconds, while real-time data can capture such events and avoid prediction deviations. The temperature prediction model (such as LSTM) needs to dynamically correct the weights through real-time data. For example, when it is detected that the temperature suddenly rises by 3 ℃ (warehouse door opening), the model can immediately adjust the prediction curve for the next 10 minutes. If the recovery of the refrigeration system is delayed, the real-time data can trigger an extension of the predicted low-temperature maintenance time.

[0026] In one embodiment of the present specification, the endurance time of the drone at different temperatures is measured through experiments to generate the drone endurance mapping data corresponding to each drone. The drone endurance mapping data includes multiple temperature ranges and the reference endurance duration of the drone within each such temperature range. Based on the predicted temperature time series and the drone endurance mapping data, the single-inspection duration corresponding to the drone inspection is determined.

[0027] Based on the predicted temperature time series and the UAV endurance mapping data, determine the single inspection duration corresponding to the UAV inspection, specifically including: extracting the predicted temperature data corresponding to each predicted timestamp based on the predicted temperature time series, matching the predicted temperature data with the temperature intervals in the UAV endurance mapping data, and determining the temperature intervals that each predicted timestamp falls into; when there are multiple temperature intervals that a predicted timestamp falls into, count the cumulative duration corresponding to each of these temperature intervals, and use the ratio of the cumulative duration to the total duration of the predicted temperature time series to determine the duration correction weight corresponding to the temperature interval; through the duration correction weight, perform a weighted operation on the baseline endurance duration of the UAV in each temperature interval to determine the single inspection duration corresponding to the preset warehouse area.

[0028] In an embodiment of the present specification, through the predicted temperature time series, extract the predicted temperature data corresponding to each predicted timestamp, match the predicted temperature data with the temperature intervals in the UAV endurance mapping data, and for each predicted temperature, find the temperature interval it falls into, so as to determine the temperature interval corresponding to each predicted timestamp. If the predicted temperatures in the predicted temperature time series belong to the same temperature interval, then use the baseline endurance duration corresponding to this temperature interval in the UAV endurance mapping data as the single inspection duration corresponding to the UAV in the preset warehouse area. If the predicted temperatures in the predicted temperature time series belong to different temperature intervals, then count the cumulative duration of each temperature interval that a predicted temperature falls into, that is, in the entire predicted temperature time series, how many timestamps' predicted temperatures fall into a specific temperature interval. Use the ratio of the cumulative duration of each temperature interval that a predicted temperature falls into to the total duration of the predicted temperature time series to determine the duration correction weight corresponding to the temperature interval, that is to say, the duration correction weight here reflects the proportion of each temperature interval in the entire prediction time period. According to the calculated duration correction weight, perform a weighted operation on the baseline endurance duration of the UAV in each temperature interval, multiply the baseline endurance duration of each temperature interval by the corresponding duration correction weight, and then add up all the results to obtain the single inspection duration corresponding to the preset warehouse area. For example, assume that the predicted temperature time series covers the temperature prediction for the next 10 hours, with one data point per hour. Assume that the UAV endurance mapping data contains three temperature intervals: [-20,-10] ℃ 、[-10,0]°C and [0,10] ℃, the corresponding reference endurance times are 20 minutes, 30 minutes, and 40 minutes respectively; through matching, it is found that the temperature in the range of [-20, -10] °C lasts for 3 hours, the temperature in the range of [-10, 0] °C lasts for 4 hours, and the temperature in the range of [0, 10] °C lasts for 3 hours. Then the duration correction weights for these three temperature ranges are 0.3, 0.4, and 0.3 respectively. Through weighted calculation, the single inspection duration = 0.3×20 + 0.4×30 + 0.3×40 = 30 minutes. It should be noted that the specific values and interval divisions in the above example are only for illustration.

[0029] Medical cold chain warehouses are usually in a low-temperature environment. Especially when the temperature is below -10 °C, the battery life of commercial drones will significantly decay. By obtaining the historical warehouse temperature data in the preset warehouse area, training the temperature prediction model, and combining the real-time warehouse temperature data to predict the environmental temperature, and then determining the single inspection duration according to the predicted temperature time series and the drone endurance mapping data, the impact of the low-temperature environment on the drone endurance can be fully considered, avoiding the interruption of the inspection task caused by insufficient endurance, making the determination of the drone inspection duration more reasonable, and thus better meeting the needs of continuous inspection of medical warehouses. For example, during periods of low temperature, the prediction model will give corresponding low-temperature predictions. Then, when determining the inspection duration, considering the decay of the drone endurance at this time, the single inspection duration is appropriately shortened to ensure that the drone can complete the inspection task within the allowable battery power range; obtaining the pre-generated drone endurance mapping data containing multiple temperature ranges and the drone reference endurance time within each temperature range can establish a connection between the temperature and the drone endurance time. Determining the single inspection duration based on the predicted temperature time series and this mapping data enables the inspection plan to be adjusted according to the endurance ability of the drone under different temperature conditions, helping to reasonably arrange the drone inspection tasks, improve the inspection efficiency, and avoid problems such as resource waste or incomplete inspection caused by unreasonable inspection duration arrangements; since the storage of goods in medical warehouses requires strict monitoring, traditional manual inspections have deficiencies such as long interval periods and difficulty in detecting problems in a timely manner. By accurately determining the single inspection duration of the drone, it helps to achieve more effective continuous inspections, can timely detect problems such as misplacement of goods and out-of-control drug expiration dates that may cause quality and safety accidents, reduce the risks during the storage process of goods in the warehouse, and ensure the quality and safety of medical supplies.

[0030] Step S102, determine the spatio-temporal conflict sub-region through the dynamic device operation data and the static warehouse distribution information corresponding to the preset warehouse area, so as to plan the inspection path corresponding to the drone inspection according to the spatio-temporal conflict sub-region and the single inspection duration, and determine the drone inspection path within each preset warehouse area.

[0031] Determine the spatio-temporal conflict sub-region based on the dynamic device operation data and the static warehouse distribution information corresponding to the preset warehouse area, specifically including: obtaining the dynamic device operation data, where the dynamic device operation data includes AGV scheduling task data and pallet lifting task data; extracting spatio-temporal features from the dynamic device operation data to determine the dynamic operation time feature and dynamic operation space feature within the preset warehouse area; determining the target time interval based on the single inspection duration, slicing the target time interval to determine multiple time slices, and constructing the three-dimensional space corresponding to the preset warehouse area based on the static warehouse distribution information to determine the three-dimensional space snapshot corresponding to each time slice; performing operation space positioning in the corresponding three-dimensional space snapshot according to the dynamic operation time feature and the dynamic operation space feature to determine at least one operation area corresponding to each time slice; merging adjacent operation areas belonging to the same time slice to determine the space conflict sub-region corresponding to each time slice, so as to determine the spatio-temporal conflict sub-region.

[0032] In one embodiment of this specification, obtain the AGV scheduling task data and the pallet lifting task data. The AGV scheduling task data can be obtained through the real-time task queue of the AGV central scheduling system (such as the MiR Fleet, Geek+ system), including the task path (coordinate sequence), speed, and task time window (start / end time) of the AGV. The pallet lifting task data can be obtained through the shelf operation log of the warehouse management system (WMS), including the shelf lifting operation instruction, that is, the target height and the lifting time period. It should be noted that both the AGV scheduling task data and the pallet lifting task data here are tasks of planned operations, that is, for the future tasks that have not been scheduled or lifted.

[0033] Extract the spatio-temporal characteristics of dynamic equipment operation data to determine the dynamic operation time characteristics and dynamic operation space characteristics within the preset warehouse area. It should be noted that the AGV time characteristic is the time range during which the AGV occupies a certain spatial coordinate, and the AGV space characteristic is the spatial position corresponding to the AGV path; the shelf time characteristic refers to the time period during which the shelf height changes (such as 14:05:00 - 14:07:30), and the shelf space characteristic is the spatial position corresponding to the pallet where the target moves. The specific feature extraction can be achieved through the following steps. First, extract the time characteristics. Calculate the occupancy time of the AGV on each path segment, such as the coordinate (x, y, z) is occupied by the AGV from t = 10s to t = 15s; obtain the lifting operation time line, mark the start / end time of the shelf height change, and calculate the spatial occupancy duration during the shelf lifting, such as the shelf S1 rises to a height of 3m from t = 30s to t = 60s. Secondly, extract the spatial characteristics. Expand the AGV path into a cylindrical safety area, and the radius of the cylindrical safety area can be set to the AGV width + 0.5m, and dynamically adjust the path coverage range according to the estimated speed of the AGV. Generate a three-dimensional cube area according to the lifted target height. For example, when the height is lifted from 2m to 3m, the z-axis range is 2 - 3m, and the area of z = 2 - 3m is not passable from t = 30s to t = 60s.

[0034] Divide the task time into multiple time slices according to the duration of a single inspection. For example, if the duration of a single inspection is 20 minutes (1200 seconds), then it is sliced into 1200 time slices. Based on the static warehouse distribution information, such as shelf positions, aisles, temperature zones, etc., establish a grid-based three-dimensional map, and the accuracy can be set to 0.5m × 0.5m × 0.5m. Each time slice corresponds to a three-dimensional space snapshot. In the three-dimensional snapshot of each time slice, mark the spatial area occupied by the dynamic equipment according to its spatio-temporal characteristics. For example, in the time slice of t = 50s, AGV-001 occupies the area of (x = 5 - 7m, y = 10 - 12m, z = 0 - 2m) in aisle A. Within the same time slice, merge the areas occupied by dynamically adjacent equipment into a larger conflict sub-region. For example, the shelf S1 occupies (x1, y1, z = 2 - 3m) at t = 50s, and the shelf S2 occupies (x2, y2, z = 2 - 3m). If they are spatially adjacent, they are merged into a conflict area. Determine the spatio-temporal conflict sub-regions in the above manner.

[0035] Through the above technical solution, in a dynamic medical warehouse, the preset path of the AGV overlaps with the inspection area of the UAV. The automatic lifting of the pallet causes real-time changes in the shelf structure, forming continuous moving obstacles. By obtaining the AGV scheduling task data and the pallet lifting task data, and extracting the spatio-temporal features of the dynamic equipment operation data, it is possible to accurately grasp the operation time and spatial range of the dynamic equipment. Based on this information, the spatio-temporal conflict sub-regions are determined, enabling the UAV to avoid the activity areas of the dynamic equipment during inspection, avoiding conflicts with the movement of the AGV or the lifting of the pallet, reducing the situation where the UAV frequently hovers emergently or even interrupts the task due to sudden obstacles, thus effectively coping with the interference of dynamic obstacles to the inspection and ensuring the smooth progress of the UAV inspection task; by slicing the target time interval, constructing a three-dimensional space snapshot corresponding to each time slice, and performing operation space positioning and region merging in the snapshot according to the spatio-temporal features of the dynamic operation, it is possible to carefully divide the spatial conflict sub-regions within each time slice, enabling the UAV inspection path planning to more accurately avoid these conflict regions, without wasting time and energy in the conflict regions for unnecessary obstacle avoidance operations, thereby improving the inspection efficiency, avoiding the risk of spatial collision with the dynamic equipment, greatly enhancing the safety of the UAV inspection, and reducing the possibility of UAV damage; considering the dynamic changes of the dynamic equipment, the spatio-temporal conflict sub-regions are determined through the comprehensive analysis of the dynamic equipment operation data and the static warehouse distribution information, making up for the deficiencies of the existing technology in dealing with the dynamic warehouse environment, and making the UAV inspection technology more suitable for complex environments such as medical cold chain warehouses with dynamic equipment.

[0036] According to the spatio-temporal conflict sub-region and the single inspection duration, the inspection path corresponding to the UAV inspection is planned to determine the UAV inspection path within each preset warehouse area, specifically including: on the basis of the three-dimensional space corresponding to the preset warehouse area, adding a time dimension to construct a four-dimensional search space, where the four-dimensional search space includes multiple search nodes, and each search node includes the node spatial position coordinates and the time dimension identifier; according to the spatio-temporal conflict sub-region and the single inspection duration, path planning is performed within the four-dimensional search space to generate the UAV inspection path within the preset warehouse area.

[0037] In one embodiment of this specification, a time dimension (t) is added to the three-dimensional space corresponding to the preset warehouse area on the basis of the traditional three-dimensional space (x, y, z) to form a four-dimensional search space (x, y, z, t). The four-dimensional search space includes a plurality of search nodes, and each of the search nodes includes node spatial position coordinates and a time dimension identifier; each node represents the position state of the unmanned aerial vehicle at a specific time. For example, the node N(x = 5, y = 10, z = 2, t = 50) represents that the unmanned aerial vehicle is located at the coordinate (5, 10, 2) at t = 50 seconds. According to the spatio-temporal conflict sub-region and the single inspection duration, path planning is performed in the four-dimensional search space to generate the unmanned aerial vehicle inspection path within the preset warehouse area.

[0038] According to the spatio-temporal conflict sub-region and the single inspection duration, path planning is performed in the four-dimensional search space to generate the unmanned aerial vehicle inspection path within the preset warehouse area, which specifically includes: according to the spatio-temporal conflict sub-region and the single inspection duration, setting the spatio-temporal constraint conditions corresponding to the path planning, where the spatio-temporal constraint conditions include that the unmanned aerial vehicle path nodes do not fall into the spatio-temporal range corresponding to the spatio-temporal conflict sub-region, and the planned path duration of the unmanned aerial vehicle is not greater than the single inspection duration; obtaining the pre-set inspection shelf sequence, and determining the target inspection shelf sequence corresponding to the current inspection task in the inspection shelf sequence according to the single inspection duration and the historical inspection efficiency, where the target inspection shelf sequence includes a plurality of sequentially arranged inspection shelves; taking the unmanned aerial vehicle charging node in the preset warehouse area as the starting point and the ending point, and performing path planning in the four-dimensional search space according to the spatio-temporal constraint conditions to generate the unmanned aerial vehicle inspection path corresponding to the target inspection shelf sequence.

[0039] In one embodiment of this specification, according to the spatio-temporal conflict sub-region and the single inspection duration, the spatio-temporal constraint conditions corresponding to the path planning are set. The constraint conditions here include a hard obstacle avoidance rule, that is, the unmanned aerial vehicle path nodes shall not fall into the time-space range of the spatio-temporal conflict sub-region. If the spatio-temporal conflict sub-region occupies passage A (x = 5 - 7, y = 10 - 12, z = 0 - 2) at t = 50 - 55 seconds, the unmanned aerial vehicle cannot enter this area during t = 50 - 55 seconds. In addition to the above hard obstacle avoidance rule, the total task duration does not exceed the single inspection duration.

[0040] Through the task instruction of the warehouse management system, if it is necessary to inspect shelf S1 → S3 → S5, obtain the inspection shelf sequence and the historical inspection efficiency. The historical inspection efficiency is the average inspection time of the shelves recorded in the database. For example, each shelf needs to stay for 30 seconds. According to the single inspection duration and the historical efficiency, calculate the number of shelves that can be completed. The formula example is as follows:

[0041] "Maximum number of shelves = single - time duration / (average inspection time + flight time)". If the single - time duration is 1080 seconds, and each shelf takes 60 seconds (including 30 seconds for inspection + 30 seconds for flight), then at most 18 shelves can be inspected. In the above - mentioned manner, in the sequence of shelves to be inspected, determine the target sequence of shelves to be inspected corresponding to the current inspection task. The target sequence of shelves to be inspected includes multiple shelves to be inspected arranged in order, decomposing the global task into executable sub - goals to avoid global task timeout.

[0042] Taking the drone charging node in the preset warehouse area as the starting point and the ending point, according to the spatio - temporal constraint conditions, use the spatio - temporal A* algorithm to implement the drone inspection path corresponding to the target sequence of shelves to be inspected. Starting from the starting point, expand nodes to adjacent grids, and mark the arrival time for each node. Only allow expanding nodes that are not occupied by the spatio - temporal conflict sub - regions. The four - dimensional path planning process is as follows: First is the initialization search, starting from the charging node (0, 0, 0, t_start), where t_start = 0. Return to the charging node (0, 0, 0, t_end), where t_end ≤ single - time inspection duration. During the node expansion process, the drone can move to adjacent grids (up, down, left, right, forward, backward,

[0043] ±0.5m). For each grid moved, the time increases by n seconds, which is related to the flight speed of the drone. Check whether the expanded node falls into the spatio - temporal conflict sub - region. Design the cost function f(n)=g(n)+h(n), where the actual cost g(n) represents the cumulative cost from the starting point to the current node (distance × 0.5 + temperature risk × 0.2 + time × 0.3), and the heuristic cost h(n) is used to estimate the remaining cost from the current node to the ending point (Euclidean distance + remaining number of shelves × average inspection time). During the expansion process, preferentially expand the node with the smallest f(n) value.

[0044] After that, perform the operation of inserting necessary nodes. Force - insert a stop node at the target shelf coordinates (for example, S1(5, 10, 2) needs to arrive at t = 30 seconds and stay for 30 seconds). Ensure that when accessing the shelf, the area where it is located is not occupied by dynamic equipment. After determining the path, perform path - closing verification and integrity verification. The last node of the path needs to return to the charging node, which must be (0, 0, 0, t_end), and t_end ≤ 1080 seconds. All target shelves must be visited, and the paths between the shelves are continuous.

[0045] Through the above technical solution, by constructing a four-dimensional search space including the time dimension and combining the information of the spatio-temporal conflict sub-regions, the conflict situations of each spatial position in the warehouse at different time points can be accurately identified. When the unmanned aerial vehicle (UAV) plans the inspection path, it can effectively avoid the areas and time periods where the dynamic equipment is operating according to the time dimension identifier of the search node and the node spatial position coordinates, completely solving the problem that the traditional path planning algorithm only avoids obstacles based on the static map and does not consider the impact of dynamic equipment, and greatly reducing the risk of task interruption caused by dynamic obstacles; the medical warehouse requires the UAV to meet the continuous inspection requirements, but the battery life of commercial UAVs decays significantly in low-temperature environments. By planning the path in the four-dimensional search space according to the duration of a single inspection, the limited battery life can be fully utilized. By reasonably arranging the inspection sequence and path, the UAV can traverse the conflict-free areas at different time points within the allowable battery power range, avoiding power waste or inability to complete the inspection task caused by unreasonable paths, ensuring that continuous and efficient inspections can be achieved even under limited low-temperature battery life, and guaranteeing the safety of medical supplies storage; when planning the path in the four-dimensional search space, the dynamic changes in the warehouse space layout and time dimension are fully considered. The inspection route can be flexibly adjusted according to the spatio-temporal conflict sub-regions corresponding to different time slices, and important areas or areas prone to problems can be preferentially inspected, while avoiding waiting or detouring in conflict areas, greatly improving the inspection efficiency; incorporating the time dimension into the path planning consideration and closely combining it with the operation data of dynamic equipment and the static warehouse distribution information enable the UAV inspection path planning to adapt to the changing environment in the warehouse in real time, ensuring the smooth completion of the inspection task in a complex and highly dynamic environment and improving the intelligent management level of the entire medical warehouse.

[0046] Step S103: Inspect each preset warehouse area according to the UAV inspection path, collect the corresponding area inspection information, and based on the area inspection information, conduct a storage risk analysis on the dynamic medical warehouse to determine the goods risk data, so as to manage the goods in the dynamic medical warehouse.

[0047] In an embodiment of this specification, the UAV inspection path is obtained based on the operation plan of the dynamic equipment and the predicted temperature conditions, and has a certain degree of pre-emptiveness. Inspect each preset warehouse area according to the UAV inspection path to collect the corresponding area inspection information. During this process, in order to avoid emergencies, a real-time conflict response mechanism can also be set. If a certain section of the path is blocked by newly detected conflicts, search for the subsequent path from the current node. Compensate for the time deviation caused by obstacle avoidance by adjusting the flight speed (±0.2 m / s).

[0048] Based on the inspection information of this area, conduct a storage risk analysis on this dynamic medical warehouse to determine the goods risk data, specifically including: obtaining the inspection information of this area, where the inspection information of this area includes RFID scan data and goods location coordinate data, and the RFID scan data includes goods batch number, production date, and expiration date; according to the RFID scan data, conduct an expiration risk analysis on the goods to determine the expiration risk level of the goods in this dynamic medical warehouse; through the goods location coordinate data, conduct a goods location risk analysis on the goods to determine the storage location deviation and temperature zone compliance identification corresponding to the goods; based on the expiration risk level of the goods, the storage location deviation, and the temperature zone compliance identification, determine the goods risk data.

[0049] In an embodiment of this specification, the inspection information of the area includes RFID scan data and goods location coordinate data. The RFID scan data contains the goods batch number, production date, and expiration date, uniquely identifying each batch of goods and providing the key time attributes of the goods. The goods location coordinate data clarifies the specific location information of the goods in the warehouse. Based on the production date and expiration date in the RFID scan data, conduct an expiration risk analysis on the goods. By calculating the time difference between the current time and the expiration date and combining the preset risk assessment rules, the expiration risk level of the goods can be determined. For example, goods with a shorter distance to the expiration date may be classified as a high-risk level, while goods with a longer distance to the expiration date are of a low-risk level. The grading method here can be set according to the types of goods existing in different warehouses.

[0050] Conduct a goods location risk analysis on the goods using the goods location coordinate data. By comparing the actual goods location coordinates with the preset correct goods location coordinates, the storage location deviation corresponding to the goods can be determined. If the placement position of the goods deviates greatly from the specified position, it may affect the inventory management efficiency of the warehouse, increase the time cost of finding the goods, and even may lead to problems such as shipping errors. On the other hand, according to the temperature setting of the area where the goods are located and the storage temperature requirements of the goods themselves, determine the temperature zone compliance identification. This can judge whether the goods are stored in a suitable temperature area to ensure that their quality is not affected by temperature. Consider comprehensively the expiration risk level of the goods, the storage location deviation, and the temperature zone compliance identification and other information to determine the goods risk data. Combining the information of these three dimensions can comprehensively reflect the risk status of the goods in the warehouse. By automatically analyzing the RFID scan data and the goods location coordinate data, there is no need for manual frequent entry into the low-temperature environment, reducing the health risks of the operators. And it can monitor the expiration date and storage location of the goods in real time and accurately, timely discover problems such as misplacement of goods locations, making up for the deficiencies of long interval and low efficiency of manual inspections, and improving the timeliness and accuracy of the management of goods in the medical warehouse.

[0051] Inventory management is performed on the dynamic medical warehouse, specifically including: identifying at least one risky item according to the item risk data, where the risk types corresponding to the risky items include at least one of expiration risk, location deviation risk, and temperature zone violation risk; determining the current storage location of the item corresponding to the risky item, and sending the current storage location of the item to the control end corresponding to the dynamic device, so as to perform risk management operations on the risky item through the dynamic device, where the risk management operations include any one or more of goods location movement operations and priority outbound operations.

[0052] In an embodiment of the present specification, all items in the warehouse are screened according to the previously determined item risk data. Since the item risk data comprehensively considers information in multiple dimensions such as expiration risk level, storage location deviation, and temperature zone compliance identification, it can accurately judge which items are at risk. When the risk type corresponding to an item includes at least one of expiration risk (such as approaching the expiration date), location deviation risk (the actual storage location does not match the specified location), and temperature zone violation risk (not stored in the appropriate temperature zone), it is identified as a risky item. Through this step, the items in the warehouse that need special attention and handling can be quickly located, providing a clear target for subsequent management operations. For each identified risky item, determine its current actual storage location in the warehouse. Send the current storage location of the risky item to the control end corresponding to the dynamic device (such as AGV, liftable pallet system, etc.). After receiving the location information, the control end selects appropriate risk management operations according to the specific risk type of the risky item and the management strategy of the warehouse. These operations include goods location movement operations and priority outbound operations. The goods location movement operation is to move the item to the correct location or the appropriate temperature zone when there is a location deviation risk or a temperature zone violation risk for the item, and the priority outbound operation is to give priority to arranging the outbound of the item when there is an expiration risk for the item to avoid losses caused by expiration. Through the automated operation of the dynamic device, the risky items can be processed efficiently, reducing manual intervention and improving management efficiency.

[0053] Through automated identification and management of risky goods, the need for manual intervention is reduced. There is no need for manual workers to frequently enter the low-temperature medical cold chain warehouse to search for and handle risky goods, reducing the health risks of operators due to low-temperature exposure. At the same time, the timeliness and accuracy of handling risky goods are improved, making up for the deficiencies of manual inspections in terms of efficiency and real-time performance. In a dynamic medical warehouse, the operation of AGVs and the automatic lifting of pallets and other dynamic equipment operations increase the complexity of management. By making full use of the automation capabilities of dynamic equipment and according to the location information of risky goods, corresponding operations are carried out by controlling dynamic equipment. This enables efficient management of risky goods in an environment where dynamic equipment is constantly changing, adapting to the highly dynamic operation scenario of a dynamic medical warehouse, ensuring the stability and effectiveness of inventory management, and achieving a data closed-loop for warehouse management data.

[0054] Through the embodiments of this specification, by obtaining real-time warehouse temperature data and predicting the environmental temperature of a preset warehouse area, the single inspection duration corresponding to the UAV inspection is determined. According to the endurance ability of the UAV under different temperature conditions, the inspection duration is reasonably planned to avoid interruption of the inspection task due to insufficient endurance. In a dynamic medical warehouse, the preset path of the AGV overlaps with the inspection area of the UAV, and the automatic lifting of the pallet causes the real-time change of the shelf structure. Most of the existing UAV inspection technologies are designed for static warehouse environments and are difficult to cope with these dynamic obstacles. The embodiments of this specification determine the spatio-temporal conflict sub-areas by obtaining dynamic equipment operation data and combining static warehouse distribution information, and plan the UAV inspection path accordingly, so that the UAV can avoid the activity areas and times of dynamic equipment, effectively avoid conflicts with the AGV movement or pallet lifting, reduce the frequent emergency hovering and task interruption caused by sudden obstacles, ensure the smooth progress of the UAV inspection task, and improve the inspection efficiency and safety. The traditional manual inspection mode has problems such as long interval periods and difficulty in timely detecting misplacement of goods positions, and it is difficult to meet the requirements of precise management of medical supplies for real-time grasping of storage dynamics. The embodiments of this specification use the UAV to inspect a preset warehouse area according to the planned inspection path, collect area inspection information, including RFID scan data and goods position coordinate data, and through the storage risk analysis of this information, the goods risk data can be accurately determined, such as expiration risk level, storage position deviation, and temperature zone compliance identification, etc. Based on these data, the goods are managed, and problems such as misplacement of goods positions and out-of-control drug expiration can be detected and processed in a timely manner, improving the accuracy and timeliness of medical supply management and reducing the probability of quality and safety accidents. A variety of technical means such as environmental temperature prediction, dynamic equipment operation data analysis, UAV inspection path planning, and goods risk analysis are comprehensively used to realize the intelligent management of dynamic medical warehouses. From determining the inspection duration, planning the inspection path to conducting storage risk analysis and goods management, the entire process has a high degree of automation, reduces manual intervention, and improves the management efficiency and scientific nature of decision-making. It can obtain dynamic equipment operation data and real-time warehouse temperature data in real time, and adjust the inspection strategy and path planning in a timely manner according to these data. By reasonably planning the UAV inspection path and duration, the damage and maintenance costs of the UAV caused by endurance problems and conflicts with dynamic equipment are reduced. On the other hand, precise goods management can avoid material waste and economic losses caused by problems such as misplacement of goods positions and expired drugs, and the automated management method reduces the labor costs required for manual inspection, thus effectively reducing the operating costs of medical warehouses.

[0055] The embodiments of this specification also provide a dynamic medical warehouse management device based on a UAV, such as Figure 2As shown, the device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the above method.

[0056] An embodiment of this specification also provides a non-volatile computer storage medium storing computer-executable instructions configured to execute the above method.

[0057] The various embodiments in this specification are described in a progressive manner. For the parts that are the same or similar among the various embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for the embodiments of the device, equipment, and non-volatile computer storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and reference can be made to the corresponding parts of the method embodiments for the relevant content.

[0058] The above describes specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve the desired results. Additionally, the processes depicted in the figures do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The device and medium provided by the embodiments of this specification correspond one-to-one with the method. Therefore, the device and medium also have beneficial technical effects similar to those of the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the device and medium will not be elaborated here.

[0060] Those skilled in the art should understand that the embodiments of this specification can be provided as a method, a system, or a computer program product. Therefore, this specification can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, this specification can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] This specification is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the specification. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices produce means for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0062] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufactured article including instruction means that implement the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0063] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 or means for implementing the functions specified in multiple blocks.

[0064] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and a memory.

[0065] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory, such as read-only memory (ROM) or flash memory (flash RAM). The memory is an example of computer-readable media.

[0066] A computer-readable medium includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can store information accessible by a computing device. As defined herein, a computer-readable medium does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0067] It should also be noted that the term " comprises ” , " includes ” or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the statement " comprises an ……” element does not preclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the said element.

[0068] The above description is only one or more embodiments of this specification and is not intended to limit this specification. For those skilled in the art, various changes and modifications can be made to one or more embodiments of this specification. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of one or more embodiments of this specification shall be included within the scope of the claims of this specification.

Claims

1. A dynamic medical warehouse management method based on drones, characterized in that: The method comprises: Acquire dynamic equipment operation data in a dynamic medical warehouse and real-time warehouse temperature data in multiple preset warehouse areas, and predict the ambient temperature of the preset warehouse area based on the real-time warehouse temperature data to determine the single inspection duration corresponding to the drone inspection; Determine the spatiotemporal conflict sub-area through the dynamic equipment operation data and the static warehouse distribution information corresponding to the preset warehouse area, so as to plan the inspection path corresponding to the drone inspection according to the spatiotemporal conflict sub-area and the single inspection duration, and determine the drone inspection path in each preset warehouse area; According to the drone inspection route, each preset warehouse area is inspected, and corresponding area inspection information is collected. Based on the area inspection information, storage risk analysis is performed on the dynamic medical warehouse, and product risk data is determined to manage products in the dynamic medical warehouse.

2. The method for dynamic medical warehouse management based on drone according to claim 1 is characterized in that: Based on the real-time warehouse temperature data, the ambient temperature of the preset warehouse area is predicted to determine the single inspection duration corresponding to the drone inspection, specifically including: Acquire historical warehouse temperature data in the preset warehouse area, wherein the preset warehouse area is a patrol task execution area corresponding to a preset drone; Training a pre-built temperature prediction model based on the historical warehouse temperature data, and determining a predicted temperature time series of the preset warehouse area according to the real-time warehouse temperature data and the temperature prediction model; Acquire pre-generated drone endurance mapping data, wherein the drone endurance mapping data includes a plurality of temperature intervals and a baseline endurance duration of the drone in each of the temperature intervals; Based on the predicted temperature time series and the drone endurance mapping data, a single inspection duration corresponding to the drone inspection is determined.

3. The method for dynamic medical warehouse management based on drone according to claim 2 is characterized in that: Based on the predicted temperature time series and the drone endurance mapping data, determining a single inspection duration corresponding to the drone inspection specifically includes: Extract the predicted temperature data corresponding to each predicted timestamp based on the predicted temperature time series, match the predicted temperature data with the temperature interval in the drone endurance mapping data, and determine the temperature interval corresponding to each predicted timestamp; When there are multiple falling temperature intervals, the cumulative duration corresponding to each falling temperature interval is counted, and the duration correction weight corresponding to the falling temperature interval is determined by the ratio of the cumulative duration to the total duration of the predicted temperature time series; The reference flight duration of the UAV in each temperature range is weighted by the duration correction weight to determine the single inspection duration corresponding to the preset warehouse area.

4. The method for dynamic medical warehouse management based on drone according to claim 1, characterized in that: Determining the spatiotemporal conflict sub-area by using the dynamic equipment operation data and the static warehouse distribution information corresponding to the preset warehouse area specifically includes: Acquire the dynamic equipment operation data, wherein the dynamic equipment operation data includes AGV scheduling task data and pallet lifting task data; Extracting spatiotemporal features from the dynamic device operation data to determine the dynamic operation time features and dynamic operation space features within the preset warehouse area; Determine a target time interval based on the duration of a single inspection, slice the target time interval to determine a plurality of time slices, and construct a three-dimensional space corresponding to the preset warehouse area based on the static warehouse distribution information to determine a three-dimensional space snapshot corresponding to each of the time slices; According to the dynamic operation time feature and the dynamic operation space feature, performing operation space positioning in the corresponding three-dimensional space snapshot, and determining at least one operation area corresponding to each of the time slices; Adjacent operation areas belonging to the same time slice are merged to determine the spatial conflict sub-area corresponding to each time slice, so as to determine the spatiotemporal conflict sub-area.

5. The method for dynamic medical warehouse management based on drone according to claim 1, characterized in that: Planning the inspection path corresponding to the drone inspection according to the time-space conflict sub-area and the single inspection duration, and determining the drone inspection path in each of the preset warehouse areas, specifically includes: On the basis of the three-dimensional space corresponding to the preset warehouse area, a time dimension is added to construct a four-dimensional search space, wherein the four-dimensional search space includes a plurality of search nodes, each of which includes a node space position coordinate and a time dimension identifier; According to the space-time conflict sub-area and the single inspection duration, path planning is performed in the four-dimensional search space to generate a drone inspection path in the preset warehouse area.

6. The method for dynamic medical warehouse management based on drone according to claim 5 is characterized in that: According to the time-space conflict sub-area and the single inspection duration, path planning is performed in the four-dimensional search space to generate a drone inspection path in the preset warehouse area, specifically including: According to the spatiotemporal conflict sub-area and the duration of a single inspection, the spatiotemporal constraint conditions corresponding to the path planning are set, wherein the spatiotemporal constraint conditions include that the drone path node does not fall into the spatiotemporal range corresponding to the spatiotemporal conflict sub-area, and the planned path time of the drone does not exceed the duration of a single inspection; Obtain a preset inspection shelf sequence, and determine a target inspection shelf sequence corresponding to the current inspection task in the inspection shelf sequence according to the single inspection duration and the historical inspection efficiency, wherein the target inspection shelf sequence includes a plurality of inspection shelves arranged in sequence; Taking the drone charging nodes in the preset warehouse area as the starting point and the end point, path planning is performed in the four-dimensional search space according to the time and space constraints to generate a drone inspection path corresponding to the target inspection shelf sequence.

7. The method for dynamic medical warehouse management based on drone according to claim 1, characterized in that: Based on the regional inspection information, storage risk analysis is performed on the dynamic medical warehouse to determine the product risk data, specifically including: Acquire the regional inspection information, wherein the regional inspection information includes RFID scanning data and cargo location coordinate data, and the RFID scanning data includes cargo batch, production date and expiration date; Performing an expiration risk analysis on the goods according to the RFID scanning data to determine the expiration risk level of the goods in the dynamic medical warehouse; Through the cargo location coordinate data, a cargo location risk analysis is performed on the goods to determine the storage location deviation and temperature zone compliance mark corresponding to the goods; The product risk data is determined based on the expiration risk level of the product, the storage location deviation and the temperature zone compliance mark.

8. The method for dynamic medical warehouse management based on drone according to claim 7, characterized in that: The dynamic medical warehouse is managed for inventory, specifically including: Identify at least one risky product according to the product risk data, wherein the risk type corresponding to the risky product includes at least one of an expiration date risk, a location deviation risk, and a temperature zone violation risk; Determine the current location of the risky goods corresponding to the risky goods, and send the current storage location of the goods to the control end corresponding to the dynamic device, so as to perform risk management operations on the risky goods through the dynamic device, wherein the risk management operations include any one or more of the cargo location movement operations and the priority outbound operations.

9. A dynamic medical warehouse management device based on drones, characterized in that: The device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the method according to any one of claims 1 to 8.

10. A non-volatile computer storage medium storing computer executable instructions, characterized in that: The computer executable instructions are configured to execute the method according to any one of claims 1 to 8.

Images