Urban tunnel hidden danger emergency monitoring internet of things large model system, method and medium

Abstract

The application provides an urban tunnel hidden danger emergency monitoring Internet of Things large model system, method and medium. The system comprises an emergency supervision management platform. The method comprises the following steps: obtaining dangerous and hazardous monitoring data based on a positioning device deployed in a target tunnel; determining a plurality of single risk events in the target tunnel through a risk rule table according to the dangerous and hazardous monitoring data, environmental monitoring data and traffic monitoring data; determining a chain risk event in the target tunnel according to the plurality of single risk events and structure information of the target tunnel; generating a first rescue instruction comprising a rescue path and a rescue mode according to the plurality of single risk events and the chain risk event, controlling a corresponding rescue intelligent terminal to navigate according to the rescue path based on the first rescue instruction, and displaying the rescue mode.

Description

Technical Field

[0001] This specification relates to the field of emergency monitoring, and in particular to a large-scale IoT model system, method and medium for emergency monitoring of potential hazards in urban tunnels. Background Technology

[0002] As an important transportation infrastructure, tunnels are enclosed, narrow, and have a complex environment, making them prone to serious casualties and property damage in the event of an accident.

[0003] To improve tunnel safety management, various intelligent systems have been developed. These systems can coordinate and control ventilation, lighting, traffic signals, and broadcasting systems within the tunnel in the event of an accident, minimizing its impact. However, many intelligent systems still suffer from inaccurate positioning, isolated data, and delayed response, which may lead to a lack of foresight in emergency response and an inability to provide precise and visualized command of rescue forces.

[0004] Therefore, it is desirable to provide a large-scale IoT model system, method, and medium for emergency monitoring of potential hazards in urban tunnels, which can improve the accuracy of accident location, enhance the quality of safety monitoring within tunnels, and improve the timeliness, precision, and effectiveness of emergency response. Summary of the Invention

[0005] This specification provides one or more embodiments of an IoT large-scale model system for emergency monitoring of potential hazards in urban tunnels. The system includes an emergency monitoring and management platform configured to: acquire hazardous chemical monitoring data based on a positioning device deployed within the target tunnel; determine multiple single risk events within the target tunnel using a risk rule table based on the hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data; determine cascading risk events within the target tunnel based on the multiple single risk events and the structural information of the target tunnel; generate a first rescue instruction including a rescue path and rescue method based on the multiple single risk events and the cascading risk events; control a corresponding rescue smart terminal to navigate according to the rescue path based on the first rescue instruction; and display the rescue method.

[0006] One embodiment of this specification provides a large-scale IoT model method for emergency monitoring of potential hazards in urban tunnels. The method includes: acquiring hazardous chemical monitoring data based on a positioning device deployed within the target tunnel; determining multiple individual risk events within the target tunnel using a risk rule table based on the hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data; determining cascading risk events within the target tunnel based on the multiple individual risk events and the structural information of the target tunnel; generating a first rescue instruction including a rescue path and rescue method based on the multiple individual risk events and the cascading risk events; controlling a corresponding intelligent rescue terminal to navigate according to the rescue path based on the first rescue instruction, and displaying the rescue method.

[0007] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes a large-scale IoT model method for emergency monitoring of urban tunnel hazards. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is an exemplary system module diagram of an Internet of Things (IoT) large-scale model system for emergency monitoring of urban tunnel hazards, as shown in some embodiments of this specification. Figure 2 This is an exemplary flowchart of an IoT big data model method for emergency monitoring of urban tunnel hazards, as shown in some embodiments of this specification; Figure 3 This is an exemplary schematic diagram of the structure of a prediction model according to some embodiments of this specification. Detailed Implementation

[0010] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0011] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0012] This specification uses flowcharts to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0013] Figure 1 This is an exemplary system module diagram of an Internet of Things (IoT) big data system for emergency monitoring of urban tunnel hazards, as shown in some embodiments of this specification.

[0014] In some embodiments, the Urban Tunnel Hazard Emergency Monitoring IoT Big Data Model System 100 includes: an emergency monitoring user platform 110, an emergency monitoring service platform 120, an emergency monitoring management platform 130, an emergency monitoring sensor network platform 140, and an emergency monitoring object platform 150.

[0015] The IoT big data model refers to an IoT model architecture used to enable the efficient flow of large amounts of data within a system 100. In some embodiments, artificial intelligence models (e.g., ChatGPT, Gemini) can be applied to the IoT model architecture for data sensing and processing.

[0016] Emergency monitoring user platform 110 refers to a platform for interaction with users (such as regulatory personnel). In some embodiments, emergency monitoring user platform 110 includes terminal devices. For example, terminal devices may include mobile devices, tablet computers, consoles, etc. Emergency monitoring user platform 110 can exchange data bidirectionally with emergency monitoring service platform 120.

[0017] Emergency monitoring service platform 120 refers to a platform used for receiving and transmitting data and / or information. In some embodiments, emergency monitoring service platform 120 is configured as a server, processor, or communication module. Emergency monitoring service platform 120 can interact bidirectionally with emergency monitoring user platform 110 and emergency monitoring management platform 130. Communication module refers to a device or software that enables real-time information interaction. For example, communication module can be a mobile phone, video monitor, multimedia computer, etc.

[0018] The emergency monitoring service platform 120 can interact bidirectionally with the data center in the emergency monitoring management platform 130. In some embodiments, the emergency monitoring management platform 130 can obtain monitoring data from the emergency monitoring object platform 150 through the emergency monitoring sensor network platform 140 and store the obtained monitoring data in a database.

[0019] The emergency monitoring and management platform 130 (hereinafter referred to as the management platform) is a comprehensive management platform that manages and coordinates the connections and collaboration between multiple platforms. The emergency monitoring and management platform can be configured as a server or a processor. In some embodiments, the emergency monitoring and management platform 130 communicates with the emergency monitoring object platform 150 through a data center and an emergency monitoring sensor network platform 140.

[0020] In some embodiments, the emergency monitoring and management platform 130 may include a data center. The data center may include a database, a data processing model library, and computing units.

[0021] The database is used to collect, store, and manage data related to emergency monitoring of potential hazards in urban tunnels, such as hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data. Databases can include MySQL, PostgreSQL, InfluxDB, Prometheus, etc. For more information on hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data, please refer to [link to relevant documentation]. Figure 2 and Figure 3 And its related descriptions.

[0022] The data processing model library is used to store pre-trained large-scale data processing models. In some embodiments, the data processing model library may include chatbots, risk identification models, predictive models, etc. For more information on predictive models, please refer to [link to relevant documentation]. Figure 3 And its related descriptions.

[0023] A computing unit is a functional module that performs arithmetic, logical, and other instruction operations. A computing unit may include a processor.

[0024] In some embodiments, the emergency monitoring and management platform 130 is configured to: acquire hazardous chemical monitoring data based on a positioning device deployed in the target tunnel; determine multiple single risk events in the target tunnel based on the hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data, using a risk rule table; determine cascading risk events in the target tunnel based on the multiple single risk events and the structural information of the target tunnel; generate a first rescue instruction including a rescue path and rescue method based on the multiple single risk events and cascading risk events; control the corresponding rescue smart terminal to navigate according to the rescue path based on the first rescue instruction; and display the rescue method.

[0025] For more information about the emergency monitoring and management platform, please refer to [link / reference]. Figure 2 and Figure 3 And its related descriptions.

[0026] The emergency monitoring sensor network platform 140 is used for comprehensive management of sensor information and serves as a communication transmission platform for bidirectional data interaction between the emergency monitoring management platform 130 and the emergency monitoring target platform 150. In some embodiments, the emergency monitoring sensor network platform 140 can be configured as a communication network or gateway. The emergency monitoring sensor network platform 140 is responsible for uploading real-time sensor data collected by the emergency monitoring target platform to the emergency monitoring management platform 130, and for issuing the first rescue command generated by the emergency monitoring management platform 130 to the corresponding emergency monitoring target platform 150 for execution.

[0027] The emergency monitoring platform 150 is a platform for generating monitoring information and executing control information. In some embodiments, the emergency monitoring platform 150 may include various sensors (e.g., gas sensors, smoke concentration sensors, water level sensors), ultra-wideband (UWB) positioning base stations, rescue smart terminals, vehicle-mounted terminals of associated vehicles, associated vehicles for autonomous driving, ventilation devices, and indicating equipment deployed at key nodes of the tunnel. The emergency monitoring platform 150 is responsible for collecting real-time data such as hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data, and uploading them through the emergency monitoring sensor network platform 140; at the same time, it also responds to the first rescue command issued from the emergency monitoring management platform 130 and executes specific emergency operations, such as controlling the corresponding rescue smart terminal to navigate according to the rescue path.

[0028] In some embodiments of this specification, the urban tunnel hazard emergency monitoring IoT big data model system, which integrates UWB precise positioning, chain prediction model and dynamic digital twin platform, can solve the problems of unclear location of emergency events and difficulty in predicting risks in enclosed spaces such as tunnels.

[0029] Figure 2 This is an exemplary flowchart of a large-scale IoT model method for emergency monitoring of urban tunnel hazards, as shown in some embodiments of this specification. Figure 2 As shown, process 200 includes the following steps. In some embodiments, process 200 may be executed by an emergency monitoring and management platform.

[0030] Step 210: Obtain hazardous chemical monitoring data based on the positioning device deployed in the target tunnel.

[0031] A target tunnel refers to a tunnel that requires monitoring and risk management. Hereafter, the target tunnel will also be referred to simply as a tunnel.

[0032] Positioning devices are used to locate target vehicles passing through the target tunnel. For example, positioning devices may include radar. Target vehicles may include vehicles transporting hazardous chemicals (hereinafter referred to as hazardous chemical vehicles).

[0033] In some embodiments, the positioning device is a UWB positioning base station. A UWB positioning base station is a device that uses ultra-wideband radio technology for high-precision positioning.

[0034] In some embodiments, multiple UWB positioning base stations are deployed at preset intervals on the sidewalls or top of the target tunnel. These base stations are interconnected via a wired network to form a UWB positioning network covering the tunnel interior. When a vehicle carrying a UWB tag enters the tunnel, the UWB positioning network provides continuous and accurate location data.

[0035] Hazardous chemical monitoring data refers to the location information of multiple hazardous chemical vehicles acquired within a preset time period. For example, hazardous chemical monitoring data includes the real-time three-dimensional coordinates of hazardous chemical vehicles in a tunnel and their continuously formed driving trajectory data.

[0036] A preset time period refers to a period of time calculated backwards from the current moment. The duration of a preset time period can be set based on actual application or prior experience, such as 3 minutes, 5 minutes, etc.

[0037] Hazardous chemical monitoring data can be obtained in various ways. For example, the management platform can acquire hazardous chemical monitoring data by obtaining the GPS location signal of the hazardous chemical vehicle.

[0038] In some embodiments, hazardous chemical monitoring data is acquired as follows: designated hazardous chemical vehicles and key vehicles (such as buses) are equipped with UWB tags; when a vehicle equipped with a UWB tag enters the target tunnel, the UWB tag on the vehicle will actively emit a UWB signal; after receiving the signal, multiple UWB positioning base stations pre-deployed in the target tunnel use the Time Difference of Arrival (TDOA) algorithm to calculate the real-time three-dimensional coordinates of the UWB tag; by continuously collecting and processing the real-time three-dimensional coordinates, the driving trajectory data of the hazardous chemical vehicle in the tunnel can be generated as hazardous chemical monitoring data.

[0039] Step 220: Based on hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data, identify multiple individual risk events within the target tunnel using a risk rule table.

[0040] Environmental monitoring data refers to data acquired within a preset time period that reflects the environmental conditions inside the tunnel. In some embodiments, environmental monitoring data includes temperature data, carbon monoxide (CO) concentration, nitrogen oxide (NOx) concentration, smoke concentration, and water depth, etc.

[0041] Environmental monitoring data can be acquired by multiple sensors deployed within the target tunnel. For example, temperature data, CO concentration, NOx concentration, and smoke concentration can be obtained in real time by heat-sensitive fire detectors, CO and NOx gas sensors, and smoke concentration sensors deployed on the longitudinal top or side walls of the target tunnel, respectively. Water depth can be monitored in real time by pressure-type water level sensors deployed at the lowest point of the tunnel, sump pits, or drainage pumping stations—locations prone to water accumulation.

[0042] Traffic monitoring data refers to data acquired within a preset time period that reflects the traffic conditions inside a tunnel. Traffic monitoring data includes vehicle image flow, traffic volume, average vehicle speed, and lane occupancy. Traffic monitoring data can be acquired through various methods.

[0043] In some embodiments, the management platform inputs high-definition image streams acquired by high-definition cameras deployed along the tunnel into an image recognition algorithm to obtain traffic monitoring data. Image recognition algorithms include convolutional neural networks (CNNs), object detection algorithms, etc.

[0044] In some embodiments, to compensate for the data loss caused by low visibility (such as smoke or insufficient light) in tunnels caused by high-definition cameras, millimeter-wave radar is also installed in the tunnel to acquire traffic monitoring data. Millimeter-wave radar is unaffected by environmental factors such as light and smoke, and can accurately detect the speed and direction of travel of multiple vehicles in multiple lanes, thereby accurately monitoring traffic data.

[0045] A single risk event refers to a risk event that may occur independently within the target tunnel and is unaffected by other risk events. A single risk event includes both the risk type and risk information.

[0046] Risk type refers to the type of risk event. Risk types include environmental risk events, traffic risk events, and transportation risk events. Environmental risk events include water accumulation, flooding, and fire inside the tunnel. Traffic risk events include vehicle collisions, wrong-way driving, and breakdowns. Transportation risk events include collisions, explosions, or leaks of hazardous chemical vehicles inside the tunnel.

[0047] Risk information refers to specific information related to a risk event. For example, risk information includes the location of the risk event and the degree of risk associated with it.

[0048] In some embodiments, the processor determines multiple individual risk events within the target tunnel based on a risk rule table, using hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data.

[0049] The risk rule table includes the mapping relationship between environmental monitoring data, traffic monitoring data, and hazardous chemical monitoring data and whether various individual risk events (such as environmental risk events, traffic risk events, and transportation risk events) will occur.

[0050] In some embodiments, the risk rule table may include, but is not limited to, the following exemplary rules or mapping relationships.

[0051] For example, the rules for identifying environmental risk events are as follows: when the water depth exceeds a preset water depth threshold, a single risk event is determined to be water accumulation or flooding; when the temperature rise rate of the temperature data exceeds a preset temperature rise threshold, a single risk event is determined to be a fire.

[0052] For example, the rule for identifying transportation risk events is: when the UWB positioning data of a hazardous chemical vehicle overlaps with a traffic event that has been determined to be a "collision" or "breakdown" in time and space, the single risk event is determined to be a collision or breakdown of the hazardous chemical vehicle.

[0053] In some embodiments, the risk rule table is determined based on multiple historical accident data, and the steps are as follows: First, multiple clustering vectors are constructed based on the multiple historical accident data, and the historical accident type corresponding to each clustering vector is determined; based on the historical accident type, the multiple clustering vectors are clustered using a clustering algorithm (e.g., K-Means clustering, DBSCAN, etc.) to determine one or more clusters; for each cluster, the management platform statistically analyzes the monitoring data that significantly differs from the normal state in all clustering vectors, and uses these data as features for constructing the risk rule table. For each type of monitoring data with significant differences, the mean μ and standard deviation σ of all clustering vectors in that cluster are calculated for that type of monitoring data, and the interval μ±σ is calculated. N is defined as the range of characteristic values ​​corresponding to this type of accident in the risk rule table. When real-time monitoring data enters this range of characteristic values, it is determined that the corresponding accident has occurred. N is a positive integer.

[0054] Historical accident data refers to historical data on accidents that occurred within the tunnel. This data includes historical risk types, historical risk information, and hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data from the time of the historical accidents (hereinafter referred to as: historical hazardous chemical data, historical environmental data, and historical traffic data, respectively).

[0055] The specific content of historical accident types and historical risk information is similar to that of risk types and risk information.

[0056] Normal status data can be pre-set by technicians based on experience. The normal status data can differ for each type of historical risk.

[0057] The aforementioned clustering algorithm can group accidents with similar data patterns into one category, thus obtaining multiple clusters, each typically corresponding to a specific accident type. By analyzing the data patterns of historical accident data, a risk rule table can be obtained, leading to more accurate judgment rules.

[0058] In some embodiments, the emergency monitoring and management platform is further configured to update the risk rule table every preset period based on historical hazardous chemical data, historical environmental data, and historical traffic data, using a risk identification model.

[0059] The preset cycle refers to the update cycle of the risk rule table. The duration of the preset cycle can be set based on experience, such as 2 hours, 5 hours, etc.

[0060] In some embodiments, the risk identification model can be a machine learning model. For example, any one or a combination of a deep neural network (DNN) model or other custom model structures.

[0061] The inputs to the risk identification model include historical hazardous chemical data, historical environmental data, and historical traffic data from the previous preset period up to the current time. The output of the risk identification model includes a risk rule table.

[0062] In some embodiments, the risk identification model can be trained using multiple first training samples with a first label. For example, multiple first training samples with a first label can be input into an initial risk identification model. A loss function is constructed using the first label and the results of the initial risk identification model. Based on the loss function, the parameters of the initial risk identification model are iteratively updated using gradient descent or other methods. When a preset condition is met, the model training is complete, and a trained risk identification model is obtained. The preset condition may be that the loss function converges, the number of iterations reaches a threshold, etc.

[0063] In some embodiments, the first training sample may include historical hazardous chemical data, historical environmental data, and historical traffic data from multiple historical preset periods. The first training sample may be obtained based on historical data.

[0064] In some embodiments, the first label may include a risk rule table for the next update period of the historical update period corresponding to the first training sample. The first label may be constructed by obtaining the historical value range (i.e., maximum value to minimum value) of historical monitoring data in the next update period of the historical update period, where there are significant differences between each type of accident and the normal state when an accident actually occurs, and constructing a historical risk rule table based on this as the first label.

[0065] Historical monitoring data includes hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data.

[0066] In some embodiments of this specification, a risk rule table is determined by a trained risk identification model based on historical hazardous chemical data, historical environmental data, and historical traffic data. This makes the risk rule table more realistic, thereby effectively ensuring the accuracy of identifying single risk events.

[0067] In some embodiments, the management platform uses a risk rule table to determine the similarity between the current hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data and the corresponding judgment rules in the risk rule table, and uses the similarity as the risk level of the corresponding risk event.

[0068] There are several ways to calculate similarity. Taking temperature data as an example, if the temperature range corresponding to the preset rule table is A to B, and the current temperature is C: when C is within the range of [A, B], the risk level is 0; when C is greater than B, the similarity (i.e., the risk level) is (CB) / B; when C is less than A, the risk level is (AC) / A.

[0069] The location of a single risk event can be identified based on environmental monitoring data, traffic monitoring data, and hazardous chemical monitoring data.

[0070] Step 230: Based on multiple individual risk events and the structural information of the target tunnel, determine the chain of risk events within the target tunnel.

[0071] Structural information refers to the geometric and physical property data of the target tunnel. For example, structural information includes the tunnel's inner diameter, tunnel height, ventilation system layout, and distribution of fire protection facilities. Structural information can be obtained through design drawings, construction archives, as-built documentation, or on-site measurements and exploration.

[0072] A cascading risk event refers to a secondary or derivative risk event triggered by one or more individual risk events. Compared to individual risk events, cascading risk events can lead to a wider scope of damage and more severe consequences. A cascading risk event includes the type of cascading risk, the predicted time point of risk, and the risk location (hereinafter also referred to as the time of occurrence and location). Types of cascading risks include tunnel structure damage risk, personnel asphyxiation risk, and large-scale leakage and spread of hazardous chemicals. The predicted time point of risk and risk location refer to when and in which specific areas the predicted cascading risk event will affect.

[0073] For example, a single risk event, a fire, could trigger a chain of risk events, including damage to tunnel structures, smoke spreading leading to asphyxiation, and reduced visibility causing secondary traffic accidents. A single risk event, a hazardous chemical leak, could cause widespread pollution and explosions.

[0074] The management platform identifies chain risk events within the target tunnel using multiple methods based on several individual risk events and the structural information of the target tunnel.

[0075] In some embodiments, the management platform may determine a cascading risk event by including the following steps: Cluster vectors are constructed based on multiple historical single risk events recorded at multiple historical moments of the target tunnel and their corresponding historical structural data. The labels of the cluster vectors are the type, time point, and location of the historical risk events that actually occurred within a subsequent period (such as 0.5h, 1h, 2h, etc.) corresponding to the historical moment. Target vectors are constructed based on the current multiple single risk events and current structural data. Multiple cluster vectors and multiple target vectors are clustered to obtain multiple clusters. The cluster in which the target vector is located is denoted as the target cluster. The union of the types, time points, and locations of the actual historical risk events corresponding to all cluster vectors in the target cluster is taken as the multiple chain risk events corresponding to the target vector.

[0076] Step 240: Based on multiple single risk events and chain risk events, generate a first rescue instruction including a rescue route and rescue method. Based on the first rescue instruction, control the corresponding rescue smart terminal to navigate according to the rescue route and display the rescue method.

[0077] A rescue route refers to the planned route taken by rescue personnel from the tunnel entrance or other designated assembly point to the risk location. For example, starting from tunnel entrance A, they may travel along the left lane to station K, or reach the risk location via the emergency exit inside the tunnel.

[0078] Rescue methods refer to the specific rescue actions and technical means taken by rescuers after arriving at a risk location in response to a specific risk event. For example, in the case of a fire, a rescue method could be "rescuers using thermal imaging cameras (TIC) and lighting tools to conduct a 'Z' shaped search along the tunnel sidewall (e.g., using the right-hand rule) to locate trapped personnel," while "using high-pressure water cannons or foam guns to strike and extinguish the flames at their base."

[0079] The management platform can generate the first rescue order in various ways. In some embodiments, the management platform randomly assigns risk events to be handled to each rescue team based on multiple currently identified single risk events and cascading risk events; based on the structural information of the target tunnel, it uses shortest path planning algorithms (such as Floyd, Bellman-Ford, etc.) to determine the optimal rescue path from the current location (or designated entrance) of the rescue personnel to the risk location. A rescue team may be assigned to handle one or more events, and each rescue team includes multiple rescue personnel.

[0080] In some embodiments, the management platform determines the rescue method by querying a preset rescue method table. The preset rescue method table includes standard rescue methods and operating procedures corresponding to various types of risk events (including single risk events and cascading risk events); the management platform retrieves the corresponding rescue method from the preset rescue method table based on the determined risk type.

[0081] The rescue smart terminal corresponding to the rescue command refers to the intelligent device worn or carried by rescue personnel during rescue missions, capable of receiving, displaying, and executing rescue commands. Examples of rescue smart terminals include smart helmets, smartphones, and smartwatches.

[0082] In some embodiments, the management platform can also construct a risk twin model to determine rescue instructions. For further description of this section, see [link to relevant documentation]. Figure 3 .

[0083] In some embodiments of this specification, in tunnels where GPS is not available, sensors such as temperature, gas, and water level are deployed, combined with high-definition cameras and UWB positioning technology, to identify environmental monitoring data, traffic monitoring data, and hazardous chemical monitoring data within the tunnel, providing crucial location data for subsequent risk analysis.

[0084] In some embodiments, process 200 further includes steps 250-270.

[0085] Step 250: Based on the positioning device deployed in the target tunnel, obtain the identification of the hazardous chemical vehicle; based on the hazardous chemical vehicle identification and hazardous chemical monitoring data, identify multiple associated vehicles.

[0086] Hazardous chemical vehicle markings reflect the type and quantity of hazardous chemicals being transported. For example, a hazardous chemical vehicle marking may indicate "flammable liquid, 10 tons" or "highly toxic gas, 5 tons".

[0087] The management platform can directly obtain hazardous chemical vehicle identification from the UWB tags equipped on the vehicles. For example, the UWB tag includes the vehicle ID, type of hazardous chemical, and transport volume.

[0088] Associated vehicles refer to vehicles that may be involved in accidents with or affected by hazardous chemical vehicles. For example, associated vehicles are vehicles within a predetermined distance range from hazardous chemical vehicles.

[0089] The management platform can identify associated vehicles in various ways. For example, it can group vehicles within a preset distance range from hazardous chemical transport vehicles as associated vehicles. This preset distance range is determined based on a first preset table. The first preset table includes preset distance ranges for the potential impact of leaks or fires on different types and quantities of hazardous chemicals. This first preset table is constructed by technical personnel based on prior experience. For example, the hazard distance range corresponding to each type and quantity of hazardous chemical transport is used as the preset distance range.

[0090] Step 260: For each of the multiple associated vehicles, generate a prompt instruction including recommended speed and recommended distance based on the hazardous chemical vehicle identification; control the on-board terminals of the multiple associated vehicles to display the corresponding recommended speed and recommended distance based on the prompt instruction.

[0091] Recommended speed refers to the suggested speed for associated vehicles. Recommended distance refers to the suggested distance between associated vehicles and hazardous chemical trucks, or between associated vehicles themselves. For example, when a hazardous chemical truck is marked "flammable liquid, 10 tons," the system may recommend that associated vehicles travel at a speed of 50 kilometers per hour and maintain a distance of 80 meters from the hazardous chemical truck.

[0092] In some embodiments, the management platform can determine the prompt instruction in multiple ways. For example, the management platform can construct a target vector based on the current hazardous chemical vehicle identifier; and determine the prompt instruction based on the retrieval results of the target vector in a first vector database. The first vector database includes multiple reference vectors and the prompt instruction corresponding to each reference vector.

[0093] The reference vector is constructed based on multiple historical hazardous chemical vehicle identifiers. The label corresponding to the reference vector is determined as follows: among multiple historical vehicle speeds and historical distances corresponding to the historical hazardous chemical vehicle identifiers, the historical speed and historical distance with the lowest number of accidents are used as its corresponding label. The management platform can calculate the vector distance between the target vector and the reference vector, and select the label corresponding to the reference vector with the smallest vector distance as the prompt instruction.

[0094] Step 270: For a vehicle located behind the hazardous chemical vehicle and in autonomous driving mode among multiple associated vehicles, when the distance between the vehicle and the hazardous chemical vehicle is less than the recommended distance, a driving command is generated; the vehicle in autonomous driving mode is controlled to decelerate based on the driving command.

[0095] The management platform can determine the location relationship with the hazardous chemical vehicle by associating the vehicle's location information, and confirm whether it is in autonomous driving mode by the vehicle's own reported operating status.

[0096] In some embodiments, when it is determined that a vehicle following the hazardous materials vehicle is in autonomous driving mode and the distance between them is less than the recommended distance, the management platform generates a driving instruction based on the current speed of the autonomous vehicle and the actual distance between them. The driving instruction may include a target speed after deceleration. The management platform sends the driving instruction to the control system of the autonomous vehicle, controlling the autonomous vehicle to decelerate to the target speed.

[0097] By identifying associated vehicles, selecting the higher-risk vehicles among them, and generating driving instructions, the distance between associated vehicles and hazardous chemical vehicles can be increased, thereby further reducing the probability of chain risk events.

[0098] Figure 3 This is an exemplary schematic diagram of the structure of a prediction model according to some embodiments of this specification.

[0099] In some embodiments, the management platform is further configured to: predict chain risk events within the target tunnel using a prediction model based on multiple individual risk events and structural information; the prediction model is a graph neural network model.

[0100] In some embodiments, such as Figure 3 As shown, prediction model 320 can be a graph neural network (GNN) model.

[0101] In some embodiments, the input to the prediction model includes a risk map 310. Nodes 311 of the risk map 310 are multiple individual risk events, and node features include the risk type and risk information corresponding to each individual risk event.

[0102] In some embodiments, when the distance between two individual risk events is less than a preset distance threshold, an edge 312 is connected between them. In some embodiments, when the overlap rate of monitoring data types corresponding to the risk determination rules of two individual risk events in the risk rule table is greater than a preset overlap rate threshold, an edge 312 is connected between them. The preset distance threshold and the preset overlap rate threshold can be preset based on experience. Edge features include the distance between two individual risk events, the monitoring data types whose risk determination rules of the two individual risk events overlap; or, the time difference between the occurrence times of the two individual risk events. This time difference can be obtained through the timestamp difference of the monitoring data.

[0103] The output of predictive model 320 includes cascading risk events 330.

[0104] In some embodiments, the prediction model 320 can be trained using a second training sample and a second label. The second training sample is constructed from historical risk maps of different tunnels at multiple historical moments. The second label includes the risk type, time of occurrence, and location of the historical risk event that actually occurred within a subsequent period of time corresponding to the historical moment of the second training sample. The duration of the subsequent period can be preset. The training process of the prediction model is similar to that of the risk identification model. For more details on the risk identification model, please refer to [link to relevant documentation]. Figure 2 .

[0105] By using a trained prediction model and risk mapping to predict cascading risk events, the system can more accurately identify potential complex risk scenarios, thereby improving the predictability and response capabilities of tunnel risk management.

[0106] In some embodiments, the management platform is further configured to: generate ventilation control commands including ventilation speed and ventilation direction based on cascading risk events, structural information and ventilation device distribution; and control the corresponding jet fans to run according to the ventilation speed and ventilation direction based on the ventilation control commands to ventilate the target tunnel.

[0107] Ventilation equipment refers to devices installed within a tunnel for ventilation. For example, ventilation equipment includes jet fans. Ventilation equipment distribution refers to the arrangement of ventilation equipment within the tunnel. This distribution can be pre-set by technicians. Ventilation speed refers to the rotational speed of the ventilation equipment. Ventilation direction refers to the orientation of the ventilation equipment. In other words, the ventilation direction determines which direction the ventilation equipment directs the airflow.

[0108] The management platform can determine ventilation control commands in various ways. For example, the platform obtains the ventilation speed of each ventilation device based on a second preset table, which includes ventilation speeds corresponding to cascading risk events. For each ventilation device, the ventilation direction is determined based on its positional relationship with the location of the cascading risk event, as shown in the structural information. For instance, the ventilation direction is the direction that draws air from the location of the cascading risk event to a preset area (such as a pre-defined safe and harmless area). The second preset table is constructed by technical personnel based on prior experience.

[0109] Some embodiments in this specification control ventilation devices based on ventilation control commands, which can reduce the impact of disasters. For example, in the event of a fire, it can reduce smoke and prevent obstruction of vehicle visibility, thereby mitigating the damage caused by the accident.

[0110] In some embodiments, the management platform is further configured to: identify high-risk areas based on multiple single risk events and cascading risk events; identify crowd gathering points based on traffic monitoring data; generate lighting control instructions including display colors and flashing frequencies based on the crowd gathering points and high-risk areas; control lighting devices in the corresponding areas to illuminate with display colors and flash with flashing frequencies based on the lighting control instructions; and generate lighting indication instructions including indicator signs based on the crowd gathering points and high-risk areas; and control corresponding indication devices to display indicator signs based on the lighting indication instructions.

[0111] High-risk areas are areas where risk events are frequently occurring. High-risk areas can be determined by counting the number of single and cascading risk events in each area of ​​the tunnel. For example, the target tunnel can be evenly divided into several areas, and the number of single and cascading risk events in each area can be counted. Areas where the number exceeds a preset threshold are identified as high-risk areas. The preset threshold can be set in advance based on experience.

[0112] A crowd gathering point refers to an area within the target tunnel where people congregate. For example, a crowd gathering point can be an area where the flow of people or vehicles exceeds a certain threshold.

[0113] In some embodiments, crowd gathering points can be determined based on traffic monitoring data. For example, the management platform can use image recognition algorithms to identify areas where traffic flow exceeds a preset traffic flow threshold or where pedestrian flow exceeds a preset pedestrian flow threshold based on high-definition image streams from traffic monitoring data, and then determine these areas as crowd gathering points. The preset traffic flow threshold and preset pedestrian flow threshold can be preset based on experience.

[0114] Lighting control instructions include the display color and flashing frequency of the display devices. These instructions are used to control the display devices within the tunnel to provide warnings or illumination. Display devices include screens and indicator lights within the tunnel. The display color refers to the color shown by the device, such as red for emergency and green for safety. The flashing frequency refers to the frequency at which the display device flashes; a higher flashing frequency indicates a greater emergency.

[0115] The management platform can determine lighting control commands in multiple ways. Based on crowd gathering points and high-risk areas, the platform generates lighting control commands, including display colors and flashing frequencies, using a third preset table. The third preset table includes lighting control commands corresponding to crowd gathering points and high-risk areas. This table can be constructed as follows: high-risk areas that are also crowd gathering points are displayed in red and flash at a first flashing frequency as a warning; other low-risk areas are displayed in green and flash at a second flashing frequency for illumination. The first and second flashing frequencies can be preset based on experience, with the first flashing frequency being greater than the second.

[0116] Lighting guidance instructions are used to control guidance equipment for rescue operations. For example, lighting guidance instructions include directional arrows to guide people to safe areas and rescue signs (such as danger zone signs and rescue route guidance signs) to guide rescuers to carry out relevant rescue operations.

[0117] Lighting directional instructions can be determined as follows: The management platform generates lighting directional signs pointing from high-risk areas / crowd gathering points to safe areas based on the location information of crowd gathering points, high-risk areas, and safe areas. For example, the generated directional arrows point to safe areas, thereby guiding people to those areas.

[0118] By controlling the equipment inside the tunnel with various commands, people can be quickly guided to a safe area, thereby evacuating the flow of people in a timely manner and reducing accident losses.

[0119] In some embodiments, the emergency monitoring and management platform is further configured to: construct a risk twin model based on multiple single risk events, structural information, chain risk events, and rescue location data; send the risk twin model to the emergency monitoring user platform; and receive a second rescue instruction sent by the emergency monitoring user platform.

[0120] Rescue location data refers to the location data of rescue personnel or rescue supplies. In some embodiments, rescue personnel / supplies are equipped with UWB tags, and rescue location data is obtained through deployed UWB positioning base stations. The content of the second rescue instruction is similar to that of the first rescue instruction, except that the second rescue instruction is issued by the rescue commander through the emergency monitoring user platform.

[0121] A digital twin model is a digital twin model that maps real-time data of a physical object into a virtual model through sensors and data acquisition technology, thereby enabling real-time monitoring and management of the physical object.

[0122] A risk twin model is a 3D visualization platform that reproduces a real tunnel at a 1:1 scale. The risk twin model labels the corresponding location within the tunnel with the risk type and information of a single risk event, the risk type and occurrence time of a chain of risk events, and real-time rescue location data.

[0123] As an example, a risk twin model can be constructed as follows: 3D modeling and simulation based on the structural information of the target tunnel (such as using tools like BIM, Unity 3D, or WebGL) and a 3D visualization platform is established; various single risk events, chain risk events, rescue personnel, or materials within the tunnel are labeled and displayed on the 3D visualization platform according to their location and extent in the real tunnel.

[0124] Once the risk twin model is built, the management platform sends it to the emergency monitoring user platform. This allows rescue commanders to visually see the real-time distribution and dynamics of each rescuer, vehicle, risk point, and predicted risk point through the risk twin model on the emergency monitoring user platform. The management platform can receive various secondary rescue instructions (including rescue routes and methods) formulated and issued by rescue commanders based on the risk twin model, or directly generate primary rescue instructions through route planning and other methods based on the actual tunnel conditions reflected in the risk twin model. These rescue instructions are then sent to the rescue personnel's smart terminals through the emergency monitoring object platform, thereby improving the flexibility and reliability of rescue instructions.

[0125] In some embodiments, the risk twin model also includes real-time environmental data.

[0126] In some embodiments, the management platform is further configured to: during a rescue operation, determine a detour area based on a risk twin model; the detour area includes areas with abnormal real-time environmental data and / or areas associated with cascading risk events; determine a rescue path for the rescue personnel based on the detour area, the rescue equipment status of the rescue personnel, and the distance between the rescue personnel and each of the multiple individual risk events; for each rescue path, generate navigation instructions including navigation indicators and navigation colors; and control the indicator devices on the rescue path to display navigation indicators with navigation colors based on the navigation instructions.

[0127] Real-time environmental data refers to environmental monitoring data at various locations at the current moment. For example, real-time environmental data includes real-time temperature data, real-time smoke concentration, etc.

[0128] Detour areas refer to areas that need to be bypassed. Detour areas include areas with abnormal real-time environmental data and / or areas associated with cascading risk events. Areas with abnormal real-time environmental data refer to areas within the tunnel where real-time monitoring data is abnormal; for example, areas with high temperatures or dense smoke.

[0129] The management platform can identify areas where the real-time temperature data reflected by the risk twin model exceeds a preset temperature threshold, and areas where the real-time smoke concentration exceeds a preset concentration threshold, as areas of abnormal real-time environmental data. The preset temperature and concentration thresholds can be pre-set based on experience.

[0130] The region associated with a cascading risk event refers to the region where the cascading risk event occurred or the region that the cascading risk event may endanger.

[0131] In some embodiments, the management platform can, based on a risk twin model, determine the regions associated with a cascading risk event as the regions corresponding to the location of the cascading risk event, and the regions within a preset influence distance range of the location of the cascading risk event. The preset influence distance can be pre-set based on experience.

[0132] The status of rescue equipment refers to the equipment and remaining rescue supplies of each rescue team's personnel.

[0133] The distance between rescuers and each of the multiple individual risk events can be directly determined based on the location of that individual risk event and the rescue positioning data.

[0134] In some embodiments, the management platform can determine the rescue path for rescue personnel in various ways, based on the detour area, the equipment status of rescue personnel, and the distance between rescue personnel and each of the multiple individual risk events. For example, for each group of rescue personnel, the system calculates the priority of each group of rescue personnel for each individual risk event based on their equipment status and distance to each individual risk event. The priority is obtained by a weighted sum of the normalized normalized values ​​of the matching degree between the equipment equipped by the rescue personnel and the risk type of the individual risk event, the remaining rescue supplies, and the distance between the rescue personnel and the individual risk event. Distance is a negative indicator (e.g., normalized using the reciprocal / negative value of the distance), while the matching degree of the risk type and the remaining rescue supplies are positive indicators. Normalization processes include Min-Max normalization, Z-score normalization, etc. The weighting coefficients of the matching degree of the risk type, the remaining rescue supplies, and the distance between the rescue personnel and the individual risk event can be preset based on experience.

[0135] The matching degree between the equipment equipped by rescue personnel and the risk type of a single risk event can be determined by using the overlap rate between the standard rescue equipment and the equipment equipped by rescue personnel, after the standard rescue equipment for a single risk event is determined according to the fourth preset table. The fourth preset table includes the standard rescue equipment corresponding to a single risk event. For example, if the single risk event is a fire, the standard rescue equipment should at least include a fire extinguisher.

[0136] For each rescue team, the management platform can designate the location of the highest-priority single risk event as the target location for that team. Based on shortest path planning algorithms, the platform determines the shortest path to the target location after bypassing multiple detour areas; this shortest path is the rescue route. Shortest path planning algorithms include Single-Source Shortest Path (SSSP) and All-Pairs Shortest Path (APSP), among others.

[0137] Navigation instructions include navigation directions and navigation colors. Navigation directions include the navigation arrows displayed on the rescue route guidance devices and the corresponding text labels. Navigation colors refer to the display color of the navigation arrows. It should be noted that the display color of lighting control instructions refers to the overall color displayed on the display device (i.e., the lighting color); while navigation colors refer to the color of the navigation arrows displayed on the display device.

[0138] Directional arrows are used to indicate the direction to which rescuers should proceed based on the rescue route, while text labels are used to distinguish whether the directional arrow is used for rescue.

[0139] By planning rescue routes for rescue personnel, avoiding detour areas, and using the shortest rescue routes, and matching rescue equipment with the events to be rescued, rescue personnel can carry the appropriate rescue equipment and quickly reach the location of the event, thereby improving rescue effectiveness.

[0140] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0141] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0142] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. An urban tunnel hidden danger emergency monitoring Internet of Things large model system, characterized in that, The system includes an emergency monitoring and management platform, which is configured as follows: Based on the positioning device deployed in the target tunnel, hazardous chemical monitoring data is acquired; the hazardous chemical monitoring data refers to the positioning information of multiple hazardous chemical vehicles acquired within a preset time period; the hazardous chemical monitoring data includes the real-time three-dimensional coordinates of the hazardous chemical vehicles in the tunnel and their continuously formed driving trajectory data; Based on the hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data, multiple individual risk events within the target tunnel are identified using a risk rule table. Based on the multiple individual risk events and the structural information of the target tunnel, the cascading risk events within the target tunnel are determined; Based on the multiple single risk events and the chain of risk events, a first rescue instruction including a rescue route and a rescue method is generated. Based on the first rescue instruction, the corresponding rescue smart terminal is controlled to navigate according to the rescue route and display the rescue method. Based on the positioning device deployed in the target tunnel, the identification of the hazardous chemical vehicle is obtained; Based on the hazardous chemical vehicle identification and the hazardous chemical monitoring data, multiple associated vehicles were identified; For each of the multiple associated vehicles, a prompt instruction including a recommended speed and a recommended distance is generated based on the hazardous chemical vehicle identification. Based on the prompting instructions, the vehicle terminals of the multiple associated vehicles are controlled to display the corresponding recommended speed and recommended distance; as well as For the vehicle located behind the hazardous chemical vehicle and in autonomous driving mode among the multiple associated vehicles, when the distance between the vehicle and the hazardous chemical vehicle is less than the recommended distance, a driving command is generated; based on the driving command, the vehicle in autonomous driving mode is controlled to decelerate.

2. The system of claim 1, wherein, The emergency monitoring and management platform is further configured as follows: Every preset period, the risk rule table is updated based on historical hazardous chemical data, historical environmental data, and historical traffic data using a risk identification model; the risk identification model is a machine learning model.

3. The system of claim 1, wherein, The emergency monitoring and management platform is further configured as follows: Based on the multiple individual risk events and the structural information, a prediction model is used to predict the chain of risk events within the target tunnel; the prediction model is a graph neural network model.

4. The system of claim 3, wherein, The emergency monitoring and management platform is further configured as follows: Based on the chain of risk events, the structural information, and the distribution of ventilation devices, a ventilation control command including ventilation speed and ventilation direction is generated; Based on the ventilation control command, the corresponding jet fan is controlled to run according to the ventilation speed and the ventilation direction to ventilate the target tunnel.

5. The system of claim 3, wherein, The emergency monitoring and management platform is further configured as follows: High-risk areas are identified based on the multiple individual risk events and the chain of risk events. Based on the traffic monitoring data, the points where people gather were identified; Based on the crowd gathering point and the high-risk area, a lighting control command including a display color and a flashing frequency is generated. Based on the lighting control command, the lighting equipment in the corresponding area is controlled to illuminate with the display color and flash with the flashing frequency. Based on the crowd gathering points and the high-risk areas, lighting instruction commands including indicator signs are generated, and corresponding indicator devices are controlled to display the indicator signs based on the lighting instruction commands.

6. The system of claim 3, wherein, The system also includes an emergency monitoring user platform; the emergency monitoring management platform is further configured as follows: A risk twin model is constructed based on the multiple individual risk events, the structural information, the chain of risk events, and the rescue location data; The risk twin model is sent to the emergency monitoring user platform, and a second rescue instruction is received from the emergency monitoring user platform.

7. The system of claim 6, wherein, The risk twin model also includes real-time environmental data; the emergency monitoring and management platform is further configured to: During the rescue operation, detour areas are determined based on the risk twin model; the detour areas include areas where the real-time environmental data is abnormal and / or areas associated with the cascading risk events. The rescue route of the rescue personnel is determined based on the detour area, the rescue equipment status of the rescue personnel, and the distance between the rescue personnel and each of the multiple individual risk events. For each of the rescue routes, generate navigation instructions including navigation indicators and navigation colors; Based on the navigation instructions, control the indicating devices on the rescue route to display the navigation instructions in the navigation colors.

8. A method for large-scale IoT modeling of emergency monitoring of potential hazards in urban tunnels, characterized in that, The method is executed by the emergency monitoring and management platform, and the method includes: Based on the positioning device deployed in the target tunnel, hazardous chemical monitoring data is acquired; the hazardous chemical monitoring data refers to the positioning information of multiple hazardous chemical vehicles acquired within a preset time period; the hazardous chemical monitoring data includes the real-time three-dimensional coordinates of the hazardous chemical vehicles in the tunnel and their continuously formed driving trajectory data; Based on the hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data, multiple individual risk events within the target tunnel are identified using a risk rule table. Based on the multiple individual risk events and the structural information of the target tunnel, the cascading risk events within the target tunnel are determined; Based on the multiple single risk events and the chain of risk events, a first rescue instruction including a rescue route and a rescue method is generated. Based on the first rescue instruction, the corresponding rescue smart terminal is controlled to navigate according to the rescue route and display the rescue method. Based on the positioning device deployed in the target tunnel, the hazardous chemical vehicle identification is obtained; based on the hazardous chemical vehicle identification and the hazardous chemical monitoring data, multiple associated vehicles are identified; For each of the multiple associated vehicles, a prompt instruction including a recommended speed and recommended distance is generated based on the hazardous chemical vehicle identification; based on the prompt instruction, the on-board terminals of the multiple associated vehicles are controlled to display the corresponding recommended speed and recommended distance; and For the vehicle located behind the hazardous chemical vehicle and in autonomous driving mode among the multiple associated vehicles, when the distance between the vehicle and the hazardous chemical vehicle is less than the recommended distance, a driving command is generated; based on the driving command, the vehicle in autonomous driving mode is controlled to decelerate.

9. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions from the storage medium, the computer executes a method for an IoT-based large-scale model for emergency monitoring of potential hazards in urban tunnels, including: Based on the positioning device deployed in the target tunnel, hazardous chemical monitoring data is acquired; the hazardous chemical monitoring data refers to the positioning information of multiple hazardous chemical vehicles acquired within a preset time period; the hazardous chemical monitoring data includes the real-time three-dimensional coordinates of the hazardous chemical vehicles in the tunnel and their continuously formed driving trajectory data; Based on the hazardous chemical monitoring data, environmental monitoring data, and traffic monitoring data, multiple individual risk events within the target tunnel are identified using a risk rule table. Based on the multiple individual risk events and the structural information of the target tunnel, the cascading risk events within the target tunnel are determined; Based on the multiple single risk events and the chain of risk events, a first rescue instruction including a rescue route and a rescue method is generated. Based on the first rescue instruction, the corresponding rescue smart terminal is controlled to navigate according to the rescue route and display the rescue method. Based on the positioning device deployed in the target tunnel, the hazardous chemical vehicle identification is obtained; based on the hazardous chemical vehicle identification and the hazardous chemical monitoring data, multiple associated vehicles are identified; For each of the multiple associated vehicles, a prompt instruction including a recommended speed and recommended distance is generated based on the hazardous chemical vehicle identification; based on the prompt instruction, the on-board terminals of the multiple associated vehicles are controlled to display the corresponding recommended speed and recommended distance; and For the vehicle located behind the hazardous chemical vehicle and in autonomous driving mode among the multiple associated vehicles, when the distance between the vehicle and the hazardous chemical vehicle is less than the recommended distance, a driving command is generated; based on the driving command, the vehicle in autonomous driving mode is controlled to decelerate.

Images