Emergency resource intelligent linkage method and system combined with 5g and edge computing

Abstract

The application provides a first-aid resource intelligent linkage method and system combined with 5G and edge computing, relates to the technical field of resource linkage, and comprises the following steps: allocating a special transmission channel for a first-aid event through a 5G network slice, and unloading signals to a nearby edge node for real-time analysis. Based on the prediction of the availability of candidate first-aid resources, the event demand characteristics are matched and calculated to generate a scheduling decision. After the decision is issued, the traffic signals are synchronously controlled and the resource positions are monitored in real time, so that dynamic path re-planning is realized. The application improves the real-time performance, accuracy and linkage efficiency of first-aid resource scheduling.

Description

Technical Field

[0001] This invention relates to the field of resource linkage technology, and in particular to an intelligent linkage method and system for emergency medical resources that combines 5G and edge computing. Background Technology

[0002] In the field of emergency medical care, efficient and precise resource allocation is crucial for improving patient survival rates and treatment outcomes. Current technologies typically rely on centralized emergency command platforms. Upon receiving an emergency alarm, the platform, based on the event location, a static resource database, and simple distance calculations, issues dispatch instructions to ambulances and other emergency resources via public mobile communication networks. Simultaneously, some systems attempt to integrate with urban traffic management systems to request priority passage for emergency vehicles on missions. This model constitutes the current standard practice for emergency resource allocation.

[0003] However, the aforementioned conventional practices have significant drawbacks. Firstly, centralized processing and public network transmission inherently involve latency. The reporting of emergency event trigger signals, platform parsing, and the issuance of decision-making instructions all require multi-hop network transmission and centralized computation. In time-sensitive emergency scenarios, precious "golden time" is consumed in data processing and transmission. Secondly, existing dispatch decisions rely on insufficient information dimensions and poor real-time performance. Decisions are typically based on the static registration location of emergency resources or the approximate location of the last reported information, lacking accurate predictions of their actual operational status, real-time traffic conditions, and dynamic availability. This often leads to unreasonable dispatch instructions; for example, dispatched ambulances may be performing other tasks, located in congested areas, or have malfunctioning equipment, preventing timely response and delaying optimal rescue opportunities. Furthermore, traffic signal priority instructions are often triggered independently and with a lag, failing to deeply coordinate with real-time vehicle locations and dynamic routes, resulting in overall operational efficiency that needs improvement. Summary of the Invention

[0004] This invention provides a method and system for intelligent linkage of emergency medical resources that combines 5G and edge computing, which can solve the problems in the prior art.

[0005] A first aspect of this invention provides a method for intelligent linkage of emergency medical resources combining 5G and edge computing, comprising: Acquire an emergency response event trigger signal, which includes the coordinates of the event location; allocate a dedicated transmission channel for the emergency response event trigger signal based on 5G network slicing, and offload the emergency response event trigger signal to an edge computing node that is geographically adjacent to the event location coordinates through the dedicated transmission channel; the edge computing node analyzes the emergency response event trigger signal in real time to obtain a feature set of event handling requirements; Spatiotemporal correlation analysis is performed on the historical trajectory data of each candidate emergency medical resource in the pre-constructed candidate emergency medical resource set to predict the availability of each candidate emergency medical resource within a predetermined time window and generate availability prediction results; Based on the event handling requirement feature set and the availability prediction results, the candidate emergency resources set is matched and calculated according to the resource capacity coverage constraint and the shortest arrival time constraint to generate an emergency resource dispatch decision. Based on the emergency medical resource dispatch decision, the system issues driving route instructions to the dispatched emergency medical resource terminal and issues priority passage instructions to the traffic signal control nodes along the driving route. At the same time, the system collects the location data of the dispatched emergency medical resources in real time through the dedicated transmission channel. If the location data deviates from the driving route, the system triggers route replanning and updates the driving route instructions.

[0006] The emergency response event trigger signal also includes an event type identifier and an injury level identifier; the edge computing node performs real-time analysis of the emergency response event trigger signal to obtain an event handling requirement feature set, including: Based on the coordinates of the event location, the geographical distance between each candidate edge computing node and the event location is calculated, and the current available computing resources of each candidate edge computing node are collected. According to the dual-constraint selection rule of the shortest geographical distance and the available computing resources meeting the requirements of the parsing task, the target edge computing node is determined from each candidate edge computing node, and the emergency event trigger signal is offloaded to the target edge computing node through the dedicated transmission channel. The target edge computing node extracts the event type identifier and the injury level identifier from the emergency event trigger signal, respectively; Based on the event type identifier and the injury level identifier, a joint query is performed in the event-requirement mapping table stored locally on the target edge computing node to obtain the resource type requirement set and response time limit corresponding to the event type identifier and the injury level identifier. The resource type requirement set and the response time limit are then encapsulated into the event handling requirement feature set.

[0007] Perform a joint query on the event-demand mapping table stored locally on the target edge computing node to obtain the set of resource type demands and the upper limit of response time corresponding to the event type identifier and the injury level identifier, including: Each record in the event-demand mapping table consists of four fields: event type identifier, injury level identifier, resource type demand set, and response time limit. The resource type demand set field records the name of each resource type required to complete the handling of the event type and the minimum number of calls for each resource type. Using the event type identifier and the injury level identifier as the joint query key, perform an exact match in the event-demand mapping table; when there is a unique exact match record, directly extract the resource type demand set field value and the response time upper limit field value of the record, and use them as the resource type demand set and the response time upper limit, respectively; When no exact match is found, the event type identifier remains unchanged. The event-demand mapping table is traversed in descending order of severity corresponding to the injury level identifier. The resource type demand set field value and response time upper limit field value of the first matching record with the event type identifier are taken as the resource type demand set and the response time upper limit, respectively.

[0008] Spatiotemporal correlation analysis is performed on the historical trajectory data of each candidate emergency medical resource in a pre-constructed candidate emergency medical resource set to predict the availability of each candidate emergency medical resource within a predetermined time window, generating availability prediction results, including: The historical location sequence and historical task status sequence of each candidate emergency medical resource within a predetermined statistical period are obtained from the emergency medical resource status database; the historical location sequence is divided into several time segments with a preset time granularity; the spatial aggregation area and movement direction features of each candidate emergency medical resource are extracted for the location data in each time segment to obtain the spatiotemporal distribution features of each candidate emergency medical resource. Based on the aforementioned spatiotemporal distribution characteristics, the expected location intervals of each candidate emergency medical resource at each moment within the predetermined time window are extrapolated to obtain the predicted location intervals of each candidate emergency medical resource. Based on the frequency of candidate emergency resources being idle in each time segment of the historical task status sequence, calculate the idle probability value of each candidate emergency resource within the predetermined time window; The availability prediction result is formed by combining the predicted location range of each candidate emergency medical resource with the idle probability value.

[0009] The historical location sequence is divided into several time segments with a preset time granularity. For the location data within each time segment, spatial clustering regions and movement direction features of each candidate emergency medical resource are extracted to obtain the spatiotemporal distribution features of each candidate emergency medical resource, including: For each time segment, all historical location records of each candidate emergency medical resource within that time segment are statistically analyzed, the centroid coordinates of all historical location records are calculated, and the distance between the location record farthest from the centroid coordinates and the centroid coordinates is used as the location dispersion radius. The centroid coordinates and the location dispersion radius are jointly recorded as the spatial clustering area of ​​the candidate emergency medical resource within that time segment. For two adjacent time segments on the time axis, the difference vector obtained by subtracting the centroid coordinates of the spatial clustering region of the previous time segment from the centroid coordinates of the spatial clustering region of the later time segment is used as the movement direction feature of the candidate emergency rescue resource between the two adjacent time segments. Using time segment numbers as the sorting index, the spatial clustering areas of each candidate emergency medical resource within each time segment and the corresponding movement direction features between each adjacent time segment are arranged in chronological order to form the spatiotemporal distribution features of each candidate emergency medical resource.

[0010] Based on the event handling requirement feature set and the availability prediction results, the candidate emergency medical resources set is matched and calculated according to resource capacity coverage constraints and shortest arrival time constraints to generate an emergency medical resource dispatch decision, including: Extract the resource type demand set from the event handling demand feature set, compare the capability type identifier of each candidate emergency resource in the candidate emergency resource set with the resource type demand set item by item, and filter out the candidate emergency resources whose capability type identifier covers at least one resource type demand in the resource type demand set to obtain the capability matching candidate set. The predicted location range of each candidate emergency medical resource is extracted from the availability prediction results. The predicted arrival time of each candidate emergency medical resource is calculated from the centroid coordinates of the predicted location range of each candidate emergency medical resource as the starting point and the location coordinates of the event as the ending point. The capability matching candidate set is sorted in order of predicted arrival time from shortest to longest to obtain the arrival time sorted candidate set. Based on the resource capacity coverage constraint, candidate emergency resources are selected sequentially according to the order of the arrival time sorted candidate set, until the set of capability type identifiers of the selected candidate emergency resources completely covers the set of resource type requirements. The set of identifiers of the selected candidate emergency resources and the corresponding quantity of each resource type are encapsulated into the emergency resource scheduling decision.

[0011] Based on the emergency medical resource dispatch decision, a travel route instruction is issued to the dispatched emergency medical resource terminal, and a priority passage instruction is issued to the traffic signal control nodes along the travel route. Simultaneously, the location data of the dispatched emergency medical resources is collected in real time through the dedicated transmission channel. If the location data deviates from the travel route, route replanning is triggered and the travel route instruction is updated, including: Extract the identification information of the dispatched emergency resources from the emergency resource dispatch decision, obtain the current location coordinates of the dispatched emergency resources, query the road network status data to obtain the shortest driving path between the current location coordinates and the location coordinates of the event occurrence, encapsulate the shortest driving path into a driving path instruction, and send it to the corresponding dispatched emergency resource terminal through the dedicated transmission channel; Extract the identifiers of each traffic signal control node along the driving route, and issue priority passage instructions to the traffic signal control nodes corresponding to each traffic signal control node identifier, so that the dispatched emergency medical resources can obtain priority passage throughout the entire driving process. The location data of the dispatched emergency resources is collected in real time through the dedicated transmission channel, and the deviation distance between the coordinates corresponding to the location data and the driving path is calculated. When the deviation distance exceeds the preset allowable range, the road network status data is re-queried with the coordinates corresponding to the location data as a new starting point, and a new shortest driving path is calculated from the new starting point to the coordinates of the location where the event occurred. The new shortest driving path is encapsulated into an updated driving path instruction and sent to the dispatched emergency resource terminal through the dedicated transmission channel.

[0012] A second aspect of the present invention provides an intelligent emergency resource linkage system combining 5G and edge computing, comprising: A signal acquisition unit is used to acquire an emergency event trigger signal, wherein the emergency event trigger signal includes the coordinates of the location where the event occurred; The signal transmission unit is used to allocate a dedicated transmission channel for the emergency event trigger signal based on 5G network slicing, and to offload the emergency event trigger signal to an edge computing node that is geographically adjacent to the location coordinates of the event through the dedicated transmission channel. The edge computing node then analyzes the emergency event trigger signal in real time to obtain a feature set of event handling requirements. The resource prediction unit is used to perform spatiotemporal correlation analysis on the historical trajectory data of each candidate emergency medical resource in the pre-constructed candidate emergency medical resource set, predict the availability of each candidate emergency medical resource within a predetermined time window, and generate availability prediction results. The resource matching unit is used to match and calculate the candidate emergency resources set according to the event handling requirement feature set and the availability prediction result, in accordance with the resource capacity coverage constraint and the shortest arrival time constraint, and generate an emergency resource scheduling decision. The dispatch execution unit is used to issue driving route instructions to the dispatched emergency medical resource terminals based on the emergency medical resource dispatch decision, and to issue priority passage instructions to the traffic signal control nodes through which the driving route passes. At the same time, it collects the location data of the dispatched emergency medical resources in real time through the dedicated transmission channel. If the location data deviates from the driving route, it triggers route replanning and updates the driving route instructions.

[0013] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0015] This method accurately predicts the availability of each resource within a predetermined time window by analyzing the spatiotemporal correlation of historical trajectory data of candidate emergency resources. This overcomes the lag in traditional static resource status management, enables proactive perception of the dynamic availability status of emergency resources, and improves the predictability and accuracy of dispatch decisions.

[0016] Based on real-time analyzed event demands and predicted resource availability, matching calculations are performed under the dual constraints of resource capacity coverage and minimizing arrival time. This method comprehensively considers the professional and time-sensitive requirements of emergency medical services, generating globally optimized scheduling decisions to minimize emergency response time while meeting the necessary emergency response capabilities.

[0017] During the dispatch decision-making and execution process, instructions are simultaneously issued to emergency medical resource terminals and traffic signal control nodes, achieving integrated coordination between vehicle dispatch and traffic management. Combined with real-time monitoring of emergency medical resource locations via dedicated channels and a route deviation replanning mechanism, a dynamic closed-loop dispatch execution guarantee system is constructed. This ensures that dispatch instructions are reliably executed and can respond to emergencies during execution, guaranteeing that emergency medical resources can quickly and smoothly reach the incident scene along the optimal route. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the intelligent linkage method for emergency medical resources combining 5G and edge computing, as described in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0021] Figure 1 This is a flowchart illustrating the intelligent linkage method for emergency medical resources combining 5G and edge computing, as described in an embodiment of the present invention. Figure 1 As shown, the method includes: Acquire an emergency response event trigger signal, which includes the coordinates of the event location; allocate a dedicated transmission channel for the emergency response event trigger signal based on 5G network slicing, and offload the emergency response event trigger signal to an edge computing node that is geographically adjacent to the event location coordinates through the dedicated transmission channel; the edge computing node analyzes the emergency response event trigger signal in real time to obtain a feature set of event handling requirements; Spatiotemporal correlation analysis is performed on the historical trajectory data of each candidate emergency medical resource in the pre-constructed candidate emergency medical resource set to predict the availability of each candidate emergency medical resource within a predetermined time window and generate availability prediction results; Based on the event handling requirement feature set and the availability prediction results, the candidate emergency resources set is matched and calculated according to the resource capacity coverage constraint and the shortest arrival time constraint to generate an emergency resource dispatch decision. Based on the emergency medical resource dispatch decision, the system issues driving route instructions to the dispatched emergency medical resource terminal and issues priority passage instructions to the traffic signal control nodes along the driving route. At the same time, the system collects the location data of the dispatched emergency medical resources in real time through the dedicated transmission channel. If the location data deviates from the driving route, the system triggers route replanning and updates the driving route instructions.

[0022] In one optional implementation, the emergency event trigger signal further includes an event type identifier and an injury level identifier; The edge computing node performs real-time analysis of the emergency response event trigger signal to obtain an event handling requirement feature set, including: Based on the coordinates of the event location, the geographical distance between each candidate edge computing node and the event location is calculated, and the current available computing resources of each candidate edge computing node are collected. According to the dual-constraint selection rule of the shortest geographical distance and the available computing resources meeting the requirements of the parsing task, the target edge computing node is determined from each candidate edge computing node, and the emergency event trigger signal is offloaded to the target edge computing node through the dedicated transmission channel. The target edge computing node extracts the event type identifier and the injury level identifier from the emergency event trigger signal, respectively; Based on the event type identifier and the injury level identifier, a joint query is performed in the event-requirement mapping table stored locally on the target edge computing node to obtain the resource type requirement set and response time limit corresponding to the event type identifier and the injury level identifier. The resource type requirement set and the response time limit are then encapsulated into the event handling requirement feature set.

[0023] In the actual operation of an emergency incident management system, the information dimensions carried by the emergency incident trigger signal directly affect the accuracy of subsequent handling procedures. Besides the location coordinates of the incident, this signal also includes two key attributes: an event type identifier and an injury level identifier. The event type identifier distinguishes different emergency scenarios such as traffic accidents, cardiac arrest, severe trauma, and poisoning / asphyxiation. It typically uses a predefined coding system, for example, coding traffic accidents as "E01," cardiac arrest as "E02," and severe trauma as "E03." The injury level identifier reflects the patient's urgency level and can be divided into three levels: critical, urgent, and ordinary, corresponding to different resource allocation priorities and response time requirements.

[0024] Once an emergency response signal is transmitted to the system via a dedicated channel, a suitable edge computing node needs to be selected for parsing and processing. Multiple edge computing nodes are typically deployed within a city, distributed across different geographical locations, each undertaking regional computing tasks. To achieve rapid response, the geographical distance between each candidate edge computing node and the event location is first calculated based on the event's coordinates. This distance calculation can employ the Euclidean distance formula. By obtaining the latitude and longitude coordinates of each edge computing node and performing spatial distance calculations with the latitude and longitude coordinates of the event location, the actual geographical distance in kilometers is obtained.

[0025] Another concurrent evaluation task is to collect the current available computing resources of each candidate edge computing node. The computing resources of an edge computing node mainly include the number of processor cores, memory capacity, and storage space. The current available computing resources refer to the amount of resources that are not currently being used; this parameter is reported in real time by the node's internal resource monitoring module. For example, if an edge computing node is configured with an 8-core processor and 16GB of memory, and a current task is using 3 cores and 6GB of memory, then the available computing resources are 5 cores and 10GB of memory. The computing resource requirements for parsing emergency event trigger signals are pre-set based on signal complexity, generally requiring at least 2 processor cores and 4GB of memory.

[0026] After obtaining the geographical distance and available computing resources, a dual-constraint selection rule is used to determine the target edge computing node. This rule sets two necessary conditions: first, the geographical distance must be minimized to ensure signal transmission delay is minimized; second, the available computing resources must meet the requirements of the parsing task to avoid processing failure or delay due to insufficient resources. In practice, all candidate nodes are first sorted in ascending order of geographical distance. Then, starting with the closest node, the available computing resources are checked one by one to see if they meet the requirements. If the nearest node has sufficient resources, it is directly selected as the target edge computing node; if the nearest node has insufficient resources, the next closest nodes are checked sequentially until a node that simultaneously meets both the requirements of proximity and sufficient resources is found. Once the target edge computing node is determined, the emergency event trigger signal is immediately and completely offloaded to that node through a dedicated transmission channel, enabling rapid transfer of the computing task.

[0027] Upon receiving an emergency response signal, the target edge computing node initiates a signal parsing process. This process first identifies the signal's data format and verifies its integrity to ensure it hasn't been damaged during transmission. Then, it proceeds to the key information extraction stage, extracting the event type identifier and injury level identifier from the signal's data structure. These two identifier fields occupy fixed positions within the signal's data packet, and extraction is completed by directly reading the values ​​at the corresponding positions using a field locator. For example, in a specific emergency response signal, the event type identifier field might have a value of "E02," and the injury level identifier field might have a value of "L1," representing cardiac arrest and a critical level, respectively.

[0028] After extracting the event type identifier and injury level identifier, these two identifiers need to be converted into specific event handling requirements. The target edge computing node maintains an event-requirement mapping table in its local storage. This table records the set of resource type requirements and response time limits corresponding to different combinations of event types and injury levels. The mapping table is designed with a multi-dimensional index, using the event type identifier and injury level identifier as the index keys for joint queries. When executing a joint query, the extracted "E02" and "L1" are used as query conditions to retrieve matching record rows from the mapping table.

[0029] The query results typically return a set of resource type requirements, including descriptions of various emergency medical resource needs. Examples include ambulances requiring defibrillators, cardiovascular specialists, and hospitals capable of coronary intervention. Each resource requirement is stored in a structured format, containing attributes such as resource type code, quantity requirements, and technical specifications. The response time cap explicitly defines the maximum allowed time from event triggering to the arrival of emergency resources at the scene. For critical events like cardiac arrest, the response time cap might be set at 8 minutes, meaning resources must be dispatched and arrive at the scene within 8 minutes.

[0030] After obtaining the resource type demand set and response time limit, the target edge computing node encapsulates this information to form a standardized event handling demand feature set. This feature set is organized as a structured data object, containing complete information such as an array of demand types, quantity parameters for each type of resource, technical specification constraints, and time limit parameters. The encapsulated event handling demand feature set serves as the direct input for subsequent resource matching and scheduling decisions, ensuring that the configuration of emergency resources perfectly matches the actual event needs and avoiding resource waste or under-configuration. The entire process from signal reception to demand feature set generation is completed locally on the edge computing node, with processing latency typically controlled within 200 milliseconds, laying a solid foundation for rapid emergency response.

[0031] Perform a joint query on the event-demand mapping table stored locally on the target edge computing node to obtain the set of resource type demands and the upper limit of response time corresponding to the event type identifier and the injury level identifier, including: Each record in the event-demand mapping table consists of four fields: event type identifier, injury level identifier, resource type demand set, and response time limit. The resource type demand set field records the name of each resource type required to complete the handling of the event type and the minimum number of calls for each resource type. Using the event type identifier and the injury level identifier as the joint query key, perform an exact match in the event-demand mapping table; When a unique exact match is found, the resource type requirement set field value and the response time limit field value of that record are directly extracted and used as the resource type requirement set and the response time limit, respectively. When no exact match is found, the event type identifier remains unchanged. The event-demand mapping table is traversed in descending order of severity corresponding to the injury level identifier. The resource type demand set field value and response time upper limit field value of the first matching record with the event type identifier are taken as the resource type demand set and the response time upper limit, respectively.

[0032] Upon receiving the event type identifier and injury level identifier, the target edge computing node needs to retrieve the resource configuration requirements matching the emergency event from its local storage. To achieve rapid response, the target edge computing node pre-establishes and maintains an event-requirement mapping table. This mapping table uses a structured data organization method to associate and store different types of emergency events with the required resource types and response time requirements.

[0033] The event-demand mapping table uses a fixed four-field structure for data organization. The first field is the event type identifier, storing a unique code for event types such as traffic accidents, fires, and hazardous chemical leaks. This identifier typically uses a hierarchical coding system, for example, using two digits to represent the major category, with subsequent digits representing the sub-category. The second field is the injury level identifier, recording the codes for different injury levels such as minor, moderate, severe, and critical. This field establishes a clear correspondence with the severity of the injury; a higher value indicates a more severe injury. The third field is the resource type demand set field. This field records the various resource information required to complete the handling of a specific event type in the form of a structured array or list. Each record contains two sub-elements: the resource type name and the minimum number of resource instances to be called for that resource type. The resource type name includes specific resource categories such as ambulances, medical experts, fire trucks, chemical protection equipment, and traffic control personnel. The minimum number of instances to be called specifies the minimum number of instances of that resource type that need to be dispatched, ensuring that emergency response capabilities meet on-site needs. The fourth field is the response time limit field, which records the maximum allowable time from event reporting to the arrival of the first batch of resources on site. This time is determined based on the urgency of the event and the severity of the injury, and is usually quantified and stored in minutes.

[0034] When obtaining resource requirement information, the exact match query process is executed first. The received event type identifier and injury level identifier are combined to form a joint query key, and a full table scan or index-based fast retrieval is performed in the event-requirement mapping table. The query process compares the event type identifier field and the injury level identifier field simultaneously. Only when the values ​​of both fields of a record are exactly the same as the query key is it considered an exact match. If the query result returns a unique exact match record, the resource type requirement set field value and the response time limit field value are directly read from that record. The resource type requirement set field value is parsed into a set structure containing multiple resource requirement items. Each requirement item explicitly specifies the resource type and quantity requirements. For example, a record of a traffic accident with serious injuries may specify the need for 3 ambulances, 2 trauma surgeons, and a certain number of units of blood products. The response time limit field value is extracted as a specific time constraint parameter, such as 15 minutes or 20 minutes. This parameter will be used for the timeliness evaluation of subsequent resource scheduling plans.

[0035] When a joint query fails to find an exact match, a downgraded matching mechanism is triggered to prevent query failures from disrupting the emergency response. This mechanism is designed based on the inclusive relationship between injury levels; that is, the resource allocation required for a higher injury level can necessarily meet the treatment needs of a lower injury level, but the reverse is not true.

[0036] The downgrade matching process keeps the event type identifier unchanged, only adjusting the injury level identifier. Specifically, it sorts the injuries according to the severity corresponding to the injury level identifier, starting from the most severe level and proceeding towards the least severe level, traversing all records in the event-demand mapping table that match the current event type identifier. The injury severity sorting is usually based on a predefined level system, such as critical-serious-moderate-minor, with a corresponding identifier code of 4-3-2-1.

[0037] During the traversal, for each record whose event type identifier matches the queried event type identifier, the injury level identifier of that record is immediately checked. The first matching record encountered is taken, and the resource type requirement set field value and response time limit field value are extracted from it, and these are returned as the current query result. The first-match principle is used to ensure that the obtained resource configuration plan will not be insufficient relative to the actual injury level, because traversing in descending order of severity ensures that the resource configuration level corresponding to the retrieved record is not lower than the actual requirement.

[0038] The data for the event-demand mapping table is typically developed by emergency management departments based on historical case statistics and expert experience, and is pre-configured offline on various edge computing nodes. The mapping table supports a periodic update mechanism. When emergency response standards change or new event types are added, the updated content is pushed to the edge nodes via incremental synchronization. After receiving the updated data, the edge nodes perform atomic replacement or supplementation operations on their local mapping tables, ensuring that queries are always based on the latest resource demand specifications. The mapping table uses in-memory resident storage to support millisecond-level query response times, while backups are maintained in persistent storage to handle abnormal situations such as node restarts.

[0039] By employing a joint query and degradation matching mechanism through the event-demand mapping table, abstract event descriptions can be quickly transformed into a specific list of resource type requirements and time constraints, providing clear input parameters for subsequent resource availability assessment and scheduling decisions. This mechanism ensures query accuracy while enhancing the system's fault tolerance through a degradation matching strategy, preventing interruptions to emergency response processes due to incomplete mapping table data, and ensuring that effective resource requirement guidance information can be obtained under various event scenarios.

[0040] Spatiotemporal correlation analysis is performed on the historical trajectory data of each candidate emergency medical resource in a pre-constructed candidate emergency medical resource set to predict the availability of each candidate emergency medical resource within a predetermined time window, generating availability prediction results, including: Obtain the historical location sequence and historical task status sequence of each candidate emergency medical resource within a predetermined statistical period from the emergency medical resource status database; The historical location sequence is divided into several time segments with a preset time granularity. For the location data in each time segment, the spatial clustering area and movement direction features of each candidate emergency medical resource are extracted to obtain the spatiotemporal distribution features of each candidate emergency medical resource. Based on the aforementioned spatiotemporal distribution characteristics, the expected location intervals of each candidate emergency medical resource at each moment within the predetermined time window are extrapolated to obtain the predicted location intervals of each candidate emergency medical resource. Based on the frequency of candidate emergency resources being idle in each time segment of the historical task status sequence, calculate the idle probability value of each candidate emergency resource within the predetermined time window; The availability prediction result is formed by combining the predicted location range of each candidate emergency medical resource with the idle probability value.

[0041] Predicting the availability of candidate emergency medical resources requires a comprehensive analysis of their spatiotemporal movement patterns and mission status changes. In practice, a connection is first established with an emergency medical resource status database, which records real-time dynamic information on all emergency vehicles, medical personnel, and emergency equipment within the city. Complete historical data for each candidate emergency medical resource within a predetermined statistical period is retrieved from the database. This period can be set to the past 7 days, 14 days, or 30 days, depending on actual needs, to ensure sufficient data sample size and timeliness. Historical location sequences record the latitude and longitude coordinates of the resource at each moment, with a sampling frequency typically ranging from 30 seconds to 5 minutes; historical mission status sequences indicate whether the resource is idle, dispatched, returned to the hospital, or under maintenance at each moment.

[0042] After acquiring historical data, the location sequence is segmented according to a preset time granularity. The choice of time granularity needs to balance computational accuracy and efficiency, and is usually set to 15 minutes, 30 minutes, or 1 hour. The entire statistical period is divided into several time segments according to the time granularity. For example, with a granularity of 30 minutes, a day is divided into 48 time segments. For each time segment, all location points of the resources within that time period are extracted, and a density clustering algorithm is used to identify the spatial clustering areas of the location points. In specific implementation, the clustering radius can be set from 500 meters to 2000 meters. When the density of location points in a certain area exceeds a threshold, the area is marked as a spatial clustering area. This area usually corresponds to locations where resources frequently stay, such as emergency stations, hospitals, or transportation hubs.

[0043] While extracting spatial clustering regions, the movement direction characteristics of resources within that time segment are calculated. By analyzing the coordinate changes of adjacent locations, the average azimuth and speed of movement are calculated. If a resource remains relatively stationary within a time segment, its primary location is recorded; if a resource exhibits a clear movement trend, its primary movement direction vector is recorded. The spatial clustering regions and movement direction characteristics of different time segments are arranged chronologically to form a spatiotemporal distribution characteristic sequence of resources. This sequence reflects the typical location distribution patterns and movement laws of resources at different time periods.

[0044] When performing extrapolation calculations based on spatiotemporal distribution characteristics, for each moment within a predetermined time window, the spatiotemporal distribution characteristics of the same or similar time periods in historical data are matched. For example, if it is necessary to predict the location of an ambulance between 9:00 AM and 10:00 AM the next morning, the location distribution characteristics of all weekday mornings between 9:00 AM and 10:00 AM are extracted from historical data. By statistically analyzing the distribution range of location data for the same historical time period, the boundary of the region where the resource is most likely to appear at the prediction time is determined. The extrapolation calculation uses a probability density estimation method, projecting historical location points onto a city map grid, calculating the location probability density of each grid cell, and selecting the grid area with a cumulative probability of 85% to 95% as the prediction location interval. This interval is expressed in the form of latitude and longitude coordinate ranges or polygonal regions, covering the geographical area where the resource has a high probability of appearing.

[0045] For resources with clear movement trends, trajectory prediction is incorporated into the extrapolation calculation. Based on historical movement direction characteristics and average speed, combined with the urban road network structure, the possible location range of the resource at the predicted time is estimated. When historical data shows that the resource frequently travels between emergency stations and hospitals along fixed routes during specific periods, the predicted location range is mainly concentrated within the buffer zone along that route. If the resource's movement pattern is random, the predicted location range is correspondingly expanded to cover the main traffic arteries within its service area.

[0046] The idle probability value is calculated based on the historical task state sequence. For each time segment, the number of times the candidate emergency resources were idle in the same historical time segment and the total number of times are counted, and the idle frequency ratio is calculated. For example, if an ambulance was idle 15 times between 9:00 AM and 9:30 AM in the past 20 working days, the idle frequency ratio for that time segment is 75%. Considering the temporal dependence of task states, a Markov state transition mechanism is introduced for correction. When a resource is in a dispatch state in the previous time segment, its probability of becoming idle in the next time segment is affected by the task duration distribution. By statistically analyzing the probability distribution of historical task durations, the conditional probability of a resource returning to idle at the target time under different preceding states is calculated. The idle probability value for each time segment is obtained by weighting and combining the frequency ratio and the conditional probability, with the value ranging from 0 to 1. A higher value indicates higher resource availability.

[0047] To address the impact of unforeseen factors on availability, a time-lapse mechanism is introduced into the calculation. Historical data closer to the prediction time is given higher weight to ensure that the prediction results reflect the latest trends in resource status. When a significant increase or decrease in recent task load is detected, the calculation weight of the idle probability value is dynamically adjusted to make the prediction more closely match the current operational status.

[0048] The final availability prediction result contains two core pieces of information: the predicted location range of each candidate emergency medical resource within a predetermined time window, and the corresponding idle probability value at each time. The predicted location range is expressed using geographic coordinate boundaries or region identifiers, and the idle probability value is presented as a percentage or decimal. These two pieces of information are stored together to form a structured prediction data table, where each record corresponds to the availability status of a resource at a specific time. This prediction result provides a quantitative basis for subsequent resource scheduling decisions. The scheduling algorithm can prioritize resources that are geographically close and have a high probability of availability based on the spatial distance between the resource and the event location, combined with the idle probability, significantly shortening response time and improving the efficiency of emergency medical services.

[0049] For example, the historical location sequence is divided into several time segments with a preset time granularity. For the location data within each time segment, spatial clustering regions and movement direction features of each candidate emergency medical resource are extracted to obtain the spatiotemporal distribution features of each candidate emergency medical resource, including: For each time segment, all historical location records of each candidate emergency medical resource within that time segment are statistically analyzed, the centroid coordinates of all historical location records are calculated, and the distance between the location record farthest from the centroid coordinates and the centroid coordinates is used as the location dispersion radius. The centroid coordinates and the location dispersion radius are jointly recorded as the spatial clustering area of ​​the candidate emergency medical resource within that time segment. For two adjacent time segments on the time axis, the difference vector obtained by subtracting the centroid coordinates of the spatial clustering region of the previous time segment from the centroid coordinates of the spatial clustering region of the later time segment is used as the movement direction feature of the candidate emergency rescue resource between the two adjacent time segments. Using time segment numbers as the sorting index, the spatial clustering areas of each candidate emergency medical resource within each time segment and the corresponding movement direction features between each adjacent time segment are arranged in chronological order to form the spatiotemporal distribution features of each candidate emergency medical resource.

[0050] After obtaining the historical location sequences of each candidate emergency medical resource, in-depth spatiotemporal analysis of this location data is required. The historical location sequences are divided according to a preset time granularity, which can be set to different scales such as 15 minutes, 30 minutes, or 1 hour depending on the actual application scenario. The entire historical time span is then divided into several consecutive time segments according to this granularity, with each time segment covering the same duration to ensure comparability of analyses across different time periods. For example, dividing a 24-hour day into 30-minute segments yields 48 time segments, each numbered sequentially from 0 to 47.

[0051] For a specific time segment, the location records of all candidate emergency medical resources within that segment are collected. Assuming an ambulance generates 12 location records within a 30-minute time segment, statistical analysis is performed on these 12 records to calculate their centroid coordinates. These centroid coordinates represent the center of the ambulance's activity within that time segment, reflecting its primary service area.

[0052] After obtaining the centroid coordinates, the Euclidean distance between each location record and the centroid is calculated. All 12 records are iterated through, and the distance corresponding to the record with the largest distance value is identified. This largest distance is defined as the location dispersion radius. This radius reflects the spatial span of the ambulance's activity range within this time segment; a larger radius indicates a wider activity range, while a smaller radius indicates relatively concentrated activity. The centroid coordinates and location dispersion radius are combined to form a triple. This triple fully describes the spatial clustering area of ​​the candidate emergency medical resources within this time segment, including both central location information and dispersion information.

[0053] Correlation analysis is performed on two adjacent time segments on the time axis to extract the movement trend of candidate emergency medical resources over time. The direction of the centroid displacement vector indicates the direction of movement of the ambulance's activity center between these two time segments, and the magnitude of the vector represents the distance of movement. If the vector points northeast and has a large magnitude, it indicates that the ambulance's service area is shifting towards the northeast with a significant magnitude; if the magnitude is close to zero, it indicates that the ambulance's activity center remains relatively stable within the two time segments. This directional movement characteristic can reveal the dynamic allocation pattern of emergency medical resources; for example, ambulances may concentrate in commercial areas during the morning rush hour and return to residential areas at night.

[0054] For a specific candidate emergency medical resource, after calculating the spatial clustering regions for all time segments and extracting movement direction features between adjacent time segments, this information needs to be integrated in chronological order. A data structure indexed by time segment number is established. For a scenario containing 48 time segments, a spatial clustering region array of length 48 and a movement direction feature array of length 47 are constructed. The k-th element of the spatial clustering region array stores the triplet corresponding to time segment k. The k-th element of the movement direction feature array stores the displacement vector from time segment k to k+1. The two arrays are alternately arranged to form a complete sequence: the spatial clustering region of time segment 0, the directional movement characteristics from time segment 0 to 1, the spatial clustering region of time segment 1, the directional movement characteristics from time segment 1 to 2, and so on until the spatial clustering region of the last time segment. This structured data arranged in chronological order fully depicts the spatiotemporal evolution trajectory of the candidate emergency medical resource over a historical period, including both the spatial distribution state at each moment and the dynamic transitions between states, thus constituting the spatiotemporal distribution characteristics of the emergency medical resource.

[0055] The above processing procedure is executed on all candidate emergency medical resources, and each resource obtains an independent description of its spatiotemporal distribution characteristics. The spatiotemporal distribution characteristics of different resources may vary significantly. Some resources may exhibit strong periodic movement patterns, with their movement directions being highly similar within the same time period; some resources may have relatively fixed activity ranges, with their location radius remaining consistently small; and some resources may exhibit irregular wandering patterns. These differentiated spatiotemporal distribution characteristics provide rich historical behavioral data for subsequent intelligent dispatching decisions, enabling the system to predict the future location trends of each resource based on its historical activity patterns, thereby achieving more accurate resource matching and path planning.

[0056] Based on the event handling requirement feature set and the availability prediction results, the candidate emergency medical resources set is matched and calculated according to resource capacity coverage constraints and shortest arrival time constraints to generate an emergency medical resource dispatch decision, including: Extract the resource type demand set from the event handling demand feature set, compare the capability type identifier of each candidate emergency resource in the candidate emergency resource set with the resource type demand set item by item, and filter out the candidate emergency resources whose capability type identifier covers at least one resource type demand in the resource type demand set to obtain the capability matching candidate set. The predicted location range of each candidate emergency medical resource is extracted from the availability prediction results. The predicted arrival time of each candidate emergency medical resource is calculated from the centroid coordinates of the predicted location range of each candidate emergency medical resource as the starting point and the location coordinates of the event as the ending point. The capability matching candidate set is sorted in order of predicted arrival time from shortest to longest to obtain the arrival time sorted candidate set. Based on the resource capacity coverage constraint, candidate emergency resources are selected sequentially according to the order of the arrival time sorted candidate set, until the set of capability type identifiers of the selected candidate emergency resources completely covers the set of resource type requirements. The set of identifiers of the selected candidate emergency resources and the corresponding quantity of each resource type are encapsulated into the emergency resource scheduling decision.

[0057] After obtaining the event handling requirement feature set and availability prediction results, a precise resource matching calculation process needs to be executed to ensure that the scheduling decision not only meets the event handling capacity requirements but also optimizes the response time. The entire matching calculation process is divided into three progressive stages: capacity adaptation screening, spatiotemporal efficiency assessment, and coverage optimization selection.

[0058] During the capability matching and screening phase, the resource type requirement set is analyzed from the event handling demand characteristics. This set is usually represented in structured data form, containing requirement items for different types of resources such as emergency physicians, nurses, ambulances, and specialized medical equipment. For example, in a serious traffic accident scenario, the resource type requirement set might include specific items such as trauma surgeons, intensive care nurses, monitoring ambulances equipped with ventilators, and portable defibrillators.

[0059] The candidate emergency medical resources set is then traversed to obtain the pre-stored capability type identifier for each resource. These identifiers are coded using a standardized classification system to ensure comparability of capability attributes across different resources. During the item-by-item comparison operation, the capability type identifier of each candidate emergency medical resource is matched against each requirement item in the resource type requirement set.

[0060] The verification process employs a set inclusion relationship determination, specifically checking whether the capability type identifier of a candidate resource corresponds to any item in the demand set. Candidate emergency resources whose capability type identifiers satisfy at least one resource type requirement are included in the capability matching candidate set. This screening mechanism effectively eliminates resource units whose capability attributes completely fail to meet the event requirements, significantly narrowing the data range for subsequent calculations.

[0061] In the spatiotemporal efficiency assessment phase, the predicted location intervals for each member of the capability-matching candidate set are retrieved from the availability prediction results. These predicted location intervals are typically represented by rectangular envelopes or polygonal regions, reflecting the possible location distribution range of candidate emergency resources at the current moment. The centroid coordinates of this interval are calculated as representative points of the resource locations; these centroid coordinates are obtained by weighted averaging of the coordinates of all vertices within the region. Using the calculated centroid coordinates as the starting point and the event occurrence location coordinates as the ending point, a spatial path analysis task is constructed. The calculation of predicted arrival time comprehensively considers straight-line distance, road network structure, real-time traffic flow status, and historical travel speed data.

[0062] In practical implementation, a shortest path algorithm based on road network topology can be used, combined with the average vehicle speed parameter for the current time period, to obtain the estimated travel time from the centroid location to the event location. Different speed correction coefficients are applied for different vehicle types; for example, ambulances have priority, and their predicted arrival time can be reduced by a specific proportion compared to ordinary vehicles. After calculating the predicted arrival time for all members of the capability-matching candidate set, a sorting operation is performed. The sorting is based on the predicted arrival time value from smallest to largest, with the candidate emergency medical resource with the shortest time at the top of the sequence, and resources with longer times placed later, forming a ranked candidate set by arrival time. This ranking result directly reflects the response advantage of each candidate resource in the time dimension.

[0063] During the coverage optimization selection phase, a greedy selection strategy is executed based on resource capability coverage constraints. Starting from the first candidate in the arrival time sorted candidate set, the capability type identifiers of candidate emergency resources are checked sequentially. The capability type identifier of the first candidate resource is added to the selected capability type set, and the resource identifier is recorded in the selected resource list. Subsequently, the capability type identifier of the next candidate emergency resource is extracted, and a union operation is performed with the selected capability type set to determine whether the merged capability type set completely covers the resource type demand set. Complete coverage means that for every demand in the resource type demand set, a corresponding capability identifier can be found in the selected capability type set.

[0064] If complete coverage is not yet achieved, the current candidate resource is added to the selected resource list, and the set of selected capability types is updated. The next candidate resource is then checked. This iterative process continues until the set of selected capability types and the set of resource type requirements achieve a complete match. At this point, the selection process terminates, obtaining the smallest possible resource combination that ensures comprehensive satisfaction of capability requirements while prioritizing resource units with shorter arrival times. Regarding resource quantity constraints, if there are quantity requirements in the resource type requirement set, such as the need for two emergency physicians, the selected quantity of each resource type is accumulated during the selection process until a specified threshold is reached.

[0065] Finally, the unique identifiers of the selected candidate emergency medical resources and the specific quantities corresponding to each resource type are structurally encapsulated to generate an emergency medical resource scheduling decision containing resource allocation schemes. This decision data structure clearly identifies the resource units to be scheduled and their capability types, providing a clear basis for subsequent instruction issuance and task execution.

[0066] The entire matching calculation process achieves multi-objective collaborative optimization of resource scheduling through three progressive screenings: capability adaptation, time optimization, and coverage constraints. Under the premise of meeting all capability requirements for event handling, it minimizes resource arrival time and improves emergency response efficiency for sudden events.

[0067] Based on the emergency medical resource dispatch decision, a travel route instruction is issued to the dispatched emergency medical resource terminal, and a priority passage instruction is issued to the traffic signal control nodes along the travel route. Simultaneously, the location data of the dispatched emergency medical resources is collected in real time through the dedicated transmission channel. If the location data deviates from the travel route, route replanning is triggered and the travel route instruction is updated, including: Extract the identification information of the dispatched emergency resources from the emergency resource dispatch decision, obtain the current location coordinates of the dispatched emergency resources, query the road network status data to obtain the shortest driving path between the current location coordinates and the location coordinates of the event occurrence, encapsulate the shortest driving path into a driving path instruction, and send it to the corresponding dispatched emergency resource terminal through the dedicated transmission channel; Extract the identifiers of each traffic signal control node along the driving route, and issue priority passage instructions to the traffic signal control nodes corresponding to each traffic signal control node identifier, so that the dispatched emergency medical resources can obtain priority passage throughout the entire driving process. The location data of the dispatched emergency medical resources are collected in real time through the dedicated transmission channel, and the deviation distance between the coordinates corresponding to the location data and the driving path is calculated. When the deviation distance exceeds the preset allowable range, the road network status data is re-queried using the coordinates corresponding to the location data as the new starting point. The new shortest driving path from the new starting point to the coordinates of the location where the event occurred is calculated. The new shortest driving path is encapsulated into an updated driving path instruction and sent to the dispatched emergency medical resource terminal through the dedicated transmission channel.

[0068] After the emergency medical resource dispatch decision is generated, the system needs to accurately transmit the dispatch instructions to the dispatched emergency medical resource terminals and coordinate relevant traffic signal facilities to facilitate the passage of emergency medical resources. First, the system extracts the identification information of the dispatched emergency medical resources from the generated emergency medical resource dispatch decision data structure. This identification information includes key information such as the emergency vehicle number, the vehicle terminal equipment identification code, and the driver's contact information. The identification information is then matched with vehicle files in the emergency medical resource database to obtain the current location coordinates of the emergency medical resource. Location coordinates are typically expressed in latitude and longitude format, with an accuracy of six decimal places to ensure positioning accuracy.

[0069] After obtaining the current location coordinates, the road network status data query interface is immediately invoked. The road network status data includes real-time traffic conditions for all roads within the city, covering information such as road flow, congestion levels, construction sections, and temporary control zones. During the query process, the current location coordinates of the dispatched emergency medical resources are used as the starting point, and the location coordinates of the incident are used as the ending point. Route planning calculations are then performed in conjunction with the real-time road network status data. The route planning algorithm comprehensively considers multiple factors such as road distance, travel time, and road condition complexity, employing an improved Dijkstra's algorithm or the A* algorithm to calculate the shortest travel path. The evaluation criterion for the shortest path is not simply the shortest physical distance, but a comprehensive indicator that combines the shortest travel time with optimal road conditions.

[0070] After calculating the shortest driving route, the route information is structured and encapsulated. The encapsulation includes the complete sequence of road segments, the length of each segment, estimated travel time, turn instructions at key intersections, and possible precautions along the route. The route instructions use a standardized data format to ensure accurate parsing by the in-vehicle navigation system. The encapsulated driving route instructions are then transmitted to the corresponding dispatched emergency medical resource terminal via a dedicated transmission channel. This dedicated transmission channel uses an encrypted transmission protocol to ensure the instructions are not tampered with or intercepted during transmission. Upon receiving the driving route instructions, the in-vehicle terminal immediately displays the planned route on the navigation screen and provides real-time navigation prompts to the driver through a voice broadcast system.

[0071] Meanwhile, the system performs a detailed analysis of the shortest travel route, identifying all traffic signal control nodes along the route. Traffic signal control nodes refer to intersections or road sections equipped with intelligent traffic lights; these nodes typically have remote control capabilities and are connected to the city's intelligent traffic management platform. By analyzing the geographic coordinate sequence of the travel route and combining it with the traffic signal control node database, the system matches the identification information of all signal control nodes that emergency medical resources will pass through on their travel route. Each traffic signal control node has a unique device number and network address, and the system establishes a communication connection based on this identification information.

[0072] When issuing priority passage instructions to traffic signal control nodes, the instructions include the vehicle identification of the dispatched emergency medical resources, the estimated arrival time window at the intersection, and the duration of priority passage. Upon receiving the priority passage instruction, the traffic signal control nodes adjust their signal timing schemes within the specified time window, providing a green light for the direction of travel of the emergency vehicle, while appropriately extending the green light duration to ensure the emergency vehicle can pass through the intersection smoothly. The implementation of the priority passage mechanism needs to consider the impact on traffic flow in other directions; therefore, the signal control strategy aims to minimize interference with regular traffic order while ensuring priority passage for emergency medical resources. Through this coordinated control, dispatched emergency medical resources can obtain priority passage throughout their journey from the starting point to the incident location, significantly reducing travel time.

[0073] As emergency medical resources travel along the planned route, the system continuously collects real-time location data via a dedicated transmission channel. The data collection frequency is typically set to once per second to once every two seconds to ensure timely detection of vehicle position changes. The collected location data is recorded in latitude and longitude coordinates, along with a timestamp. The system compares the coordinates corresponding to the real-time location data with the preset travel path and calculates the deviation distance. The deviation distance is calculated using a point-to-curve shortest distance algorithm, projecting the actual location coordinates onto the planned path and measuring the straight-line distance between the projected point and the actual location point.

[0074] The system sets a preset allowable range as the threshold for deviation detection, typically between 50 and 100 meters. When the calculated deviation distance is less than the preset allowable range, it is determined that the emergency medical resources are still traveling normally on the planned route, and the system continues to maintain normal monitoring. When the deviation distance exceeds the preset allowable range, it means that the emergency medical resources have deviated from the original route for some reason, which could be due to the driver actively choosing another route based on the actual situation on site, or it could be due to being forced to change course due to a sudden road situation.

[0075] Upon detecting a deviation, the system immediately triggers a path replanning mechanism. The coordinates corresponding to the currently collected real-time location data are used as the new starting point, while the coordinates of the event's location remain unchanged as the destination. The road network status data query interface is then invoked again. This new road network status data is the latest real-time data, reflecting the current road traffic conditions. Based on the new starting point and the latest road network status, the path planning algorithm is re-executed to calculate the new shortest travel path from the new starting point to the event's location coordinates. The calculation process for the new path is the same as the initial path planning process, also comprehensively considering various influencing factors to ensure the path's optimality.

[0076] After calculation, the new shortest route is encapsulated in the same data format to generate an updated route instruction. This updated instruction is sent to the dispatched emergency medical resource terminal via a dedicated transmission channel. Upon receiving the update instruction, the in-vehicle navigation system immediately replaces the original route with the new planned route and provides a voice prompt to the driver confirming the route has been updated. Simultaneously, the system re-identifies traffic signal control nodes along the new route and issues or updates priority passage instructions to these nodes, ensuring that emergency medical resources also receive priority passage on the new route. The entire route replanning and instruction update process is completed within seconds, maximizing the timeliness of emergency medical missions.

[0077] A second aspect of the present invention provides an intelligent emergency resource linkage system combining 5G and edge computing, comprising: A signal acquisition unit is used to acquire an emergency event trigger signal, wherein the emergency event trigger signal includes the coordinates of the location where the event occurred; The signal transmission unit is used to allocate a dedicated transmission channel for the emergency event trigger signal based on 5G network slicing, and to offload the emergency event trigger signal to an edge computing node that is geographically adjacent to the location coordinates of the event through the dedicated transmission channel. The edge computing node then analyzes the emergency event trigger signal in real time to obtain a feature set of event handling requirements. The resource prediction unit is used to perform spatiotemporal correlation analysis on the historical trajectory data of each candidate emergency medical resource in the pre-constructed candidate emergency medical resource set, predict the availability of each candidate emergency medical resource within a predetermined time window, and generate availability prediction results. The resource matching unit is used to match and calculate the candidate emergency resources set according to the event handling requirement feature set and the availability prediction result, in accordance with the resource capacity coverage constraint and the shortest arrival time constraint, and generate an emergency resource scheduling decision. The dispatch execution unit is used to issue driving route instructions to the dispatched emergency medical resource terminals based on the emergency medical resource dispatch decision, and to issue priority passage instructions to the traffic signal control nodes through which the driving route passes. At the same time, it collects the location data of the dispatched emergency medical resources in real time through the dedicated transmission channel. If the location data deviates from the driving route, it triggers route replanning and updates the driving route instructions.

[0078] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0079] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0080] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0081] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

A method for intelligent linkage of emergency medical resources combining 1.5G and edge computing, characterized in that: include: Acquire an emergency response event trigger signal, wherein the emergency response event trigger signal includes the coordinates of the location where the event occurred; Based on 5G network slicing, a dedicated transmission channel is allocated for the emergency event trigger signal, and the emergency event trigger signal is offloaded to an edge computing node that is geographically adjacent to the location coordinates of the event through the dedicated transmission channel. The edge computing node analyzes the emergency event trigger signal in real time to obtain the event handling requirement feature set. Spatiotemporal correlation analysis is performed on the historical trajectory data of each candidate emergency medical resource in the pre-constructed candidate emergency medical resource set to predict the availability of each candidate emergency medical resource within a predetermined time window and generate availability prediction results; Based on the event handling requirement feature set and the availability prediction results, the candidate emergency resources set is matched and calculated according to the resource capacity coverage constraint and the shortest arrival time constraint to generate an emergency resource dispatch decision. Based on the emergency medical resource dispatch decision, the system issues driving route instructions to the dispatched emergency medical resource terminal and issues priority passage instructions to the traffic signal control nodes along the driving route. At the same time, the system collects the location data of the dispatched emergency medical resources in real time through the dedicated transmission channel. If the location data deviates from the driving route, the system triggers route replanning and updates the driving route instructions.

2. The method according to claim 1, characterized in that, The emergency response trigger signal also includes an event type identifier and an injury level identifier; The edge computing node performs real-time analysis of the emergency response event trigger signal to obtain an event handling requirement feature set, including: Based on the coordinates of the event location, the geographical distance between each candidate edge computing node and the event location is calculated, and the current available computing resources of each candidate edge computing node are collected. According to the dual-constraint selection rule of the shortest geographical distance and the available computing resources meeting the requirements of the parsing task, the target edge computing node is determined from each candidate edge computing node, and the emergency event trigger signal is offloaded to the target edge computing node through the dedicated transmission channel. The target edge computing node extracts the event type identifier and the injury level identifier from the emergency event trigger signal, respectively; Based on the event type identifier and the injury level identifier, a joint query is performed in the event-requirement mapping table stored locally on the target edge computing node to obtain the resource type requirement set and response time limit corresponding to the event type identifier and the injury level identifier. The resource type requirement set and the response time limit are then encapsulated into the event handling requirement feature set.

3. The method according to claim 2, characterized in that, Perform a joint query on the event-demand mapping table stored locally on the target edge computing node to obtain the set of resource type demands and the upper limit of response time corresponding to the event type identifier and the injury level identifier, including: Each record in the event-demand mapping table consists of four fields: event type identifier, injury level identifier, resource type demand set, and response time limit. The resource type demand set field records the name of each resource type required to complete the handling of the event type and the minimum number of calls for each resource type. Using the event type identifier and the injury level identifier as the joint query key, perform an exact match in the event-demand mapping table; when there is a unique exact match record, directly extract the resource type demand set field value and the response time upper limit field value of the record, and use them as the resource type demand set and the response time upper limit, respectively; When no exact match is found, the event type identifier remains unchanged. The event-demand mapping table is traversed in descending order of severity corresponding to the injury level identifier. The resource type demand set field value and response time upper limit field value of the first matching record with the event type identifier are taken as the resource type demand set and the response time upper limit, respectively.

4. The method according to claim 1, characterized in that, Spatiotemporal correlation analysis is performed on the historical trajectory data of each candidate emergency medical resource in a pre-constructed candidate emergency medical resource set to predict the availability of each candidate emergency medical resource within a predetermined time window, generating availability prediction results, including: The historical location sequence and historical task status sequence of each candidate emergency medical resource within a predetermined statistical period are obtained from the emergency medical resource status database; the historical location sequence is divided into several time segments with a preset time granularity; the spatial aggregation area and movement direction features of each candidate emergency medical resource are extracted for the location data in each time segment to obtain the spatiotemporal distribution features of each candidate emergency medical resource. Based on the aforementioned spatiotemporal distribution characteristics, the expected location intervals of each candidate emergency medical resource at each moment within the predetermined time window are extrapolated to obtain the predicted location intervals of each candidate emergency medical resource. Based on the frequency of candidate emergency resources being idle in each time segment of the historical task status sequence, calculate the idle probability value of each candidate emergency resource within the predetermined time window; The availability prediction result is formed by combining the predicted location range of each candidate emergency medical resource with the idle probability value.

5. The method according to claim 4, characterized in that, The historical location sequence is divided into several time segments with a preset time granularity. For the location data within each time segment, spatial clustering regions and movement direction features of each candidate emergency medical resource are extracted to obtain the spatiotemporal distribution features of each candidate emergency medical resource, including: For each time segment, all historical location records of each candidate emergency medical resource within that time segment are statistically analyzed, the centroid coordinates of all historical location records are calculated, and the distance between the location record farthest from the centroid coordinates and the centroid coordinates is used as the location dispersion radius. The centroid coordinates and the location dispersion radius are jointly recorded as the spatial clustering area of ​​the candidate emergency medical resource within that time segment. For two adjacent time segments on the time axis, the difference vector obtained by subtracting the centroid coordinates of the spatial clustering region of the previous time segment from the centroid coordinates of the spatial clustering region of the later time segment is used as the movement direction feature of the candidate emergency rescue resource between the two adjacent time segments. Using time segment numbers as the sorting index, the spatial clustering areas of each candidate emergency medical resource within each time segment and the corresponding movement direction features between each adjacent time segment are arranged in chronological order to form the spatiotemporal distribution features of each candidate emergency medical resource.

6. The method according to claim 1, characterized in that, Based on the event handling requirement feature set and the availability prediction results, the candidate emergency medical resources set is matched and calculated according to resource capacity coverage constraints and shortest arrival time constraints to generate an emergency medical resource dispatch decision, including: Extract the resource type demand set from the event handling demand feature set, compare the capability type identifier of each candidate emergency resource in the candidate emergency resource set with the resource type demand set item by item, and filter out the candidate emergency resources whose capability type identifier covers at least one resource type demand in the resource type demand set to obtain the capability matching candidate set. The predicted location range of each candidate emergency medical resource is extracted from the availability prediction results. The predicted arrival time of each candidate emergency medical resource is calculated from the centroid coordinates of the predicted location range of each candidate emergency medical resource as the starting point and the location coordinates of the event as the ending point. The capability matching candidate set is sorted in order of predicted arrival time from shortest to longest to obtain the arrival time sorted candidate set. Based on the resource capacity coverage constraint, candidate emergency resources are selected sequentially according to the order of the arrival time sorted candidate set, until the set of capability type identifiers of the selected candidate emergency resources completely covers the set of resource type requirements. The set of identifiers of the selected candidate emergency resources and the corresponding quantity of each resource type are encapsulated into the emergency resource scheduling decision.

7. The method according to claim 1, characterized in that, Based on the emergency medical resource dispatch decision, a travel route instruction is issued to the dispatched emergency medical resource terminal, and a priority passage instruction is issued to the traffic signal control nodes along the travel route. Simultaneously, the location data of the dispatched emergency medical resources is collected in real time through the dedicated transmission channel. If the location data deviates from the travel route, route replanning is triggered and the travel route instruction is updated, including: Extract the identification information of the dispatched emergency resources from the emergency resource dispatch decision, obtain the current location coordinates of the dispatched emergency resources, query the road network status data to obtain the shortest driving path between the current location coordinates and the location coordinates of the event occurrence, encapsulate the shortest driving path into a driving path instruction, and send it to the corresponding dispatched emergency resource terminal through the dedicated transmission channel; Extract the identifiers of each traffic signal control node along the driving route, and issue priority passage instructions to the traffic signal control nodes corresponding to each traffic signal control node identifier, so that the dispatched emergency medical resources can obtain priority passage throughout the entire driving process. The location data of the dispatched emergency resources is collected in real time through the dedicated transmission channel, and the deviation distance between the coordinates corresponding to the location data and the driving path is calculated. When the deviation distance exceeds the preset allowable range, the road network status data is re-queried with the coordinates corresponding to the location data as a new starting point, and a new shortest driving path is calculated from the new starting point to the coordinates of the location where the event occurred. The new shortest driving path is encapsulated into an updated driving path instruction and sent to the dispatched emergency resource terminal through the dedicated transmission channel. An intelligent emergency resource linkage system combining 8.5G and edge computing, used to implement the method as described in any one of claims 1-7, characterized in that, include: A signal acquisition unit is used to acquire an emergency event trigger signal, wherein the emergency event trigger signal includes the coordinates of the location where the event occurred; The signal transmission unit is used to allocate a dedicated transmission channel for the emergency event trigger signal based on 5G network slicing, and to offload the emergency event trigger signal to an edge computing node that is geographically adjacent to the location coordinates of the event through the dedicated transmission channel. The edge computing node then analyzes the emergency event trigger signal in real time to obtain a feature set of event handling requirements. The resource prediction unit is used to perform spatiotemporal correlation analysis on the historical trajectory data of each candidate emergency medical resource in the pre-constructed candidate emergency medical resource set, predict the availability of each candidate emergency medical resource within a predetermined time window, and generate availability prediction results. The resource matching unit is used to match and calculate the candidate emergency resources set according to the event handling requirement feature set and the availability prediction result, in accordance with the resource capacity coverage constraint and the shortest arrival time constraint, and generate an emergency resource scheduling decision. The dispatch execution unit is used to issue driving route instructions to the dispatched emergency medical resource terminals based on the emergency medical resource dispatch decision, and to issue priority passage instructions to the traffic signal control nodes through which the driving route passes. At the same time, it collects the location data of the dispatched emergency medical resources in real time through the dedicated transmission channel. If the location data deviates from the driving route, it triggers route replanning and updates the driving route instructions.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images