High-speed rescue material scheduling method and system based on internet of things

Abstract

The application relates to a high-speed rescue material dispatching method and system based on the Internet of Things, relates to the technical field of emergency management, and comprises the following steps: dispatching a UAV equipped with a multi-modal sensing unit to fly to a target area for dynamic patrol, receiving a distress signal sent by the target area through an Internet of Things terminal and a patrol signal returned by the UAV, and generating an estimated value of the number of trapped people; acquiring real-time disaster prediction information of the target area; based on the estimated value of the number of trapped people, the real-time disaster prediction information and a preset material demand model, a dynamic material dispatching scheme is generated, and the scheme is issued to a rescue execution unit through an Internet of Things dispatching platform. Therefore, the response speed and timeliness of material dispatching are improved.

Description

Technical Field

[0001] This invention relates to the field of emergency management technology, and in particular to a method and system for high-speed dispatch of emergency supplies based on the Internet of Things. Background Technology

[0002] In the event of natural disasters or major accidents on mountainous highways, rapid and accurate dispatch of rescue supplies is crucial for saving lives and minimizing losses. Traditional methods for counting the number of people trapped rely on a single data source (such as distress signals or drone surveillance images), which are susceptible to environmental noise, sensor errors, and human interference, leading to significant deviations in the statistical results. Current technology lacks a solution capable of dynamically fusing multi-source data, real-time error correction, and generating high-precision estimates. Summary of the Invention

[0003] This invention addresses the technical problems existing in the prior art by providing a method and system for high-speed emergency material dispatching based on the Internet of Things.

[0004] The technical solution of the present invention to solve the above-mentioned technical problems is as follows:

[0005] This invention provides a high-speed emergency supplies dispatching method based on the Internet of Things, the method comprising:

[0006] A drone equipped with a multimodal sensing unit is dispatched to the target area to conduct dynamic patrols. The drone receives distress signals sent by the target area through an Internet of Things terminal and patrol signals returned by the drone, and generates an estimate of the number of people trapped.

[0007] Obtain real-time disaster prediction information for the target area;

[0008] Based on the estimated number of people trapped, the real-time disaster prediction information, and the preset material demand model, a dynamic material dispatch plan is generated and distributed to the rescue execution unit through the Internet of Things dispatch platform.

[0009] Optionally, the dispatch of a drone equipped with a multimodal sensing unit to the target area for dynamic patrol, receiving distress signals sent by the target area through an IoT terminal and patrol signals returned by the drone, and generating an estimate of the number of people trapped, including:

[0010] The drone returns image data of the target area;

[0011] The image data is divided into explicit and implicit environments;

[0012] Obtain distress signals sent in the explicit environment to determine the first explicit number of people;

[0013] Obtain the patrol signals returned from the explicit environment to get the second explicit number of people;

[0014] Obtain distress signals sent and patrol signals returned from the hidden environment to determine the number of people in the hidden environment;

[0015] An estimated number of people trapped is generated based on the first visible number, the second visible number, and the hidden number.

[0016] Optionally, obtaining the distress signals sent in the explicit environment to determine the first explicit number of people includes:

[0017] Based on the distress signals sent in the explicit environment, construct text semantic vectors and user attribute vectors;

[0018] By fusing the text semantic vector and the user attribute vector, a multimodal feature vector is obtained;

[0019] The multimodal feature vector is input into the fully connected layer to obtain the first explicit number of people.

[0020] Optionally, obtaining the patrol signal returned from the explicit environment to obtain the second explicit number of people includes:

[0021] A pre-trained deep learning model is used to perform target detection on image data in the explicit environment to identify the location of potential trapped persons.

[0022] By combining infrared thermal imaging data, it can be verified whether the potential trapped person has human thermal characteristics;

[0023] Using millimeter-wave radar point cloud data, it was confirmed whether the potential trapped person had micro-motion characteristics related to vital signs;

[0024] Output a preliminary set of visible individuals, where each potential trapped individual includes the location, thermal characteristics, and micromotion characteristics.

[0025] Dynamic behavioral analysis is performed on each potential victim in the initial explicit population set 𝑃, and the explicit behavior confidence level is calculated;

[0026] The number of potential victims whose confidence level for the explicit behavior is greater than a preset threshold is counted to obtain the second explicit number.

[0027] Optionally, obtaining the number of people in the concealed environment by acquiring the distress signals sent and the patrol signals returned in the concealed environment includes:

[0028] Based on the distress signals sent in the hidden environment, a signal feature vector is constructed;

[0029] Based on the patrol signals returned from the hidden environment, an environmental feature vector is constructed;

[0030] The signal feature vector and the environment feature vector are input into the association network to calculate the association probability between the signal and the environment;

[0031] Potential hidden regions are generated based on the association probabilities;

[0032] Signal-trajectory matching is performed on the potential hidden regions to obtain the number of hidden individuals.

[0033] Optionally, the association network is a graph neural network; the signal feature vector and the environment feature vector are input into the association network to calculate the association probability between the signal and the environment, using the formula: P ASSOC =GNN(F sig ,F env ), where F sig F represents the signal eigenvector. env This represents the environmental feature vector.

[0034] Optionally, the step of performing signal-trajectory matching on the potential hidden region to obtain the number of hidden individuals includes:

[0035] DBSCAN clustering is performed on the infrared heat sources in the potential hidden regions to exclude isolated points;

[0036] The millimeter-wave radar trajectory of the clustered points is analyzed by LSTM network to identify motion data that conforms to human behavior patterns.

[0037] Align the timestamp of the distress signal with the time window of the predicted trajectory and calculate the matching degree;

[0038] By combining DBSCAN clustering results, motion data, and matching degree, an estimate of the number of hidden individuals is generated.

[0039] A dynamic correction module is introduced to generate a correction coefficient 𝛾 based on the historical hidden population error model and real-time environmental parameters;

[0040] The hidden population estimate is obtained by correcting the correction coefficient 𝛾.

[0041] Optionally, aligning the timestamp of the distress signal with the time window of the predicted trajectory and calculating the matching degree includes:

[0042]

[0043] in, This represents the distress signal at time t. This represents the predicted trajectory at time t.

[0044] Optionally, generating a dynamic material dispatching plan based on the estimated number of trapped people, the real-time disaster prediction information, and a preset material demand model includes:

[0045] Based on the estimated number of people trapped and the preset material demand model, the initial material demand of the target area is calculated;

[0046] A disaster correction factor is introduced based on the real-time disaster prediction information;

[0047] The initial material requirements are corrected using the disaster correction factor.

[0048] The revised initial material requirements are input into the multi-objective optimization scheduling model to generate a dynamic material scheduling scheme.

[0049] This invention also provides a high-speed emergency relief supplies dispatch system based on the Internet of Things, the system comprising:

[0050] The number of people estimation module is used to dispatch drones equipped with multimodal sensing units to the target area for dynamic patrol, receive distress signals sent by the target area through IoT terminals and patrol signals returned by the drones, and generate an estimated number of people trapped.

[0051] The disaster prediction module is used to acquire real-time disaster prediction information for the target area;

[0052] The material dispatching module is used to generate a dynamic material dispatching plan based on the estimated number of trapped people, the real-time disaster prediction information, and the preset material demand model, and then distribute the plan to the rescue execution unit through the Internet of Things dispatching platform.

[0053] In addition, to achieve the above objectives, the present invention also proposes an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing the high-speed emergency material dispatching method based on the Internet of Things as described above.

[0054] Furthermore, to achieve the above objectives, the present invention also proposes a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements the high-speed emergency material dispatching method based on the Internet of Things as described above.

[0055] The beneficial effects of this invention are:

[0056] (1) This invention dispatches a drone equipped with a multimodal sensing unit to the target area for dynamic patrol, receives distress signals sent by the target area through an IoT terminal and patrol signals returned by the drone, and generates an estimated number of trapped people; obtains real-time disaster prediction information of the target area; and generates a dynamic material dispatch plan based on the estimated number of trapped people, the real-time disaster prediction information, and a preset material demand model, and distributes it to the rescue execution unit through the IoT dispatch platform. This improves the response speed and timeliness of material dispatch.

[0057] (2) To improve the accuracy of the estimation of trapped personnel, the image data collected by the UAV is divided into visible and hidden environments. The distress signals sent from the visible environment are obtained to determine the first visible number of people; the patrol signals returned from the visible environment are obtained to determine the second visible number of people; the distress signals sent from the hidden environment and the returned patrol signals are obtained to determine the hidden number of people; based on the first visible number of people, the second visible number of people, and the hidden number of people, an estimated number of trapped personnel is generated. This reduces the misjudgment rate caused by errors in a single data source and improves the efficiency of disaster relief.

[0058] (3) The present invention further combines multimodal perception, dynamic target recognition and confidence verification mechanism to ensure high-precision extraction of the second explicit number of people; and improves the positioning and counting accuracy of the implicit number of people through signal-environment association modeling and spatiotemporal behavior prediction. Attached Figure Description

[0059] Figure 1 A scenario diagram illustrating the IoT-based high-speed emergency supplies dispatching method provided by this invention;

[0060] Figure 2 A flowchart of the high-speed emergency supplies dispatching method based on the Internet of Things provided by the present invention;

[0061] Figure 3 A flowchart for generating an estimate of the number of people trapped, provided by the present invention;

[0062] Figure 4 A schematic diagram of the structure of the Internet of Things-based high-speed emergency material dispatching system provided by the present invention;

[0063] Figure 5 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0064] Figure 6 This is a schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention. Detailed Implementation

[0065] 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.

[0066] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0067] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0068] Please see Figure 1 , Figure 1 This is a scenario diagram illustrating the high-speed emergency supplies dispatching method based on the Internet of Things provided by this invention. (See diagram below.) Figure 1 As shown, the terminal and server are connected via a network, such as a wired or wireless network. The terminal can include, but is not limited to, portable devices such as smartwatches, mobile phones, and tablets with network platform applications installed, as well as fixed terminals such as computers, kiosks, and advertising machines. The server provides users with various business services, including service push servers and user recommendation servers.

[0069] It should be noted that, Figure 1 The scenario diagram of the IoT-based high-speed emergency material dispatching method shown is merely an example. The terminals, servers, and application scenarios described in the embodiments of the present invention are for the purpose of more clearly illustrating the technical solutions of the embodiments of the present invention, and do not constitute a limitation on the technical solutions provided by the embodiments of the present invention. As those skilled in the art will know, with the evolution of the system and the emergence of new business scenarios, the technical solutions provided by the embodiments of the present invention are also applicable to similar technical problems.

[0070] The terminal can be used for:

[0071] A drone equipped with a multimodal sensing unit is dispatched to the target area to conduct dynamic patrols. The drone receives distress signals sent by the target area through an Internet of Things terminal and patrol signals returned by the drone, and generates an estimate of the number of people trapped.

[0072] Obtain real-time disaster prediction information for the target area;

[0073] Based on the estimated number of people trapped, the real-time disaster prediction information, and the preset material demand model, a dynamic material dispatch plan is generated and distributed to the rescue execution unit through the Internet of Things dispatch platform.

[0074] Please see Figure 2 The present invention provides a flowchart of a high-speed emergency supplies dispatching method based on the Internet of Things, comprising the following steps:

[0075] Step S201: Dispatch a drone equipped with a multimodal sensing unit to the target area for dynamic patrol, receive distress signals sent by the target area through an IoT terminal and patrol signals returned by the drone, and generate an estimated number of people trapped.

[0076] The drone is equipped with a multimodal sensing unit (visual, infrared, millimeter-wave radar, and lidar). The IoT terminal integrates an anti-interference communication module and a signal strength detection module.

[0077] In one implementation, please refer to Figure 3 The provided flowchart for generating an estimate of the number of people trapped, step S201 may specifically include the following steps:

[0078] Step S2011: The UAV returns image data of the target area.

[0079] Step S2012: Divide the image data into explicit environment and implicit environment.

[0080] Specifically, the explicit and implicit environments are divided using semantic segmentation algorithms based on the high-resolution image data returned by the drone.

[0081] The visible environment refers to areas in UAV imagery that are directly obstructed or minimally obstructed, including clearly visible trapped individuals (e.g., waving, calling for help, obvious injuries) or their belongings (e.g., distress signs, scattered items). The hidden environment refers to areas with obstructions (e.g., ruins, smoke), low light, or signal interference, where trapped individuals are not directly identified but may present potential signs.

[0082] Step S2013: Obtain the distress signal sent in the explicit environment to obtain the first explicit number of people.

[0083] Specifically, based on the distress signals sent in the explicit environment, a text semantic vector and a user attribute vector are constructed; the text semantic vector and the user attribute vector are fused to obtain a multimodal feature vector; the multimodal feature vector is input into a fully connected layer to obtain the first explicit number of people.

[0084] This can be achieved by encoding the distress signal text using pre-trained language models (such as BERT and RoBERTa) to extract contextual semantic features. For example, the input text might be: "The flood has come, the garage is flooded, and my family of three is trapped!", and the output semantic vector V would be: t ∈R 768 If the distress signal is voice data, it is converted into text and then encoded.

[0085] Optionally, the user attribute vector includes user registration information (such as household size, age, and gender), device information (such as mobile phone model and GPS accuracy), and historical behavior data (such as whether there have been false alarms). The text semantic vector and the user attribute vector are fused using an attention mechanism or a gated fusion unit to obtain a multimodal feature vector.

[0086] Step S2014: Obtain the patrol signal returned from the explicit environment to obtain the second explicit number of people.

[0087] Specifically, a pre-trained deep learning model is used to perform target detection on image data in the explicit environment to identify the location of potential trapped individuals; infrared thermal imaging data is combined to verify whether the potential trapped individuals have human thermal characteristics (temperature range 36–37.5°C); millimeter-wave radar point cloud data is used to confirm whether the potential trapped individuals have micro-motion characteristics related to vital signs (such as small displacements caused by breathing and heartbeat); a preliminary explicit number set X is output, where each potential trapped individual includes the location, thermal characteristics, and micro-motion characteristics; dynamic behavior analysis is performed on each potential trapped individual in the preliminary explicit number set X to calculate the explicit behavior confidence score; potential trapped individuals with explicit behavior confidence scores greater than a preset threshold are counted to obtain a second explicit number of individuals.

[0088] The pre-trained deep learning model can be YOLOv8 or Faster R-CNN.

[0089] The confidence score of overt behavior is calculated by weighting and summing the visual salience score (such as whether the person is in a distress posture or associated with a distress sign), the thermal imaging consistency score (whether the heat source is stable or conforms to the human body contour) and the motion feature score (whether there are periodic micro-movements or whether the person has been still for too long). The weight of each parameter is obtained by training with historical data.

[0090] Furthermore, redundancy detection and deduplication can be performed on the second visible number of people, including: ID tracking of targets in consecutive frames of images (such as using SORT or DeepSORT algorithms) to avoid duplicate counting; if the same target appears continuously in multiple frames and its position is stable, it is counted as 1 person; if the target moves rapidly or splits (such as multiple people gathering), it is split or merged by combining spatial density analysis.

[0091] The aforementioned second explicit number calculation method integrates visual, thermal imaging, radar, and audio signals, significantly reducing the false detection rate; and distinguishes real trapped individuals from static objects through explicit behavior confidence scoring.

[0092] Step S2015: Obtain the distress signals sent from the hidden environment and the patrol signals returned, and obtain the number of hidden personnel.

[0093] Specifically, a signal feature vector is constructed based on the distress signal sent in the hidden environment; an environmental feature vector is constructed based on the patrol signal returned in the hidden environment; the signal feature vector and the environmental feature vector are input into an association network to calculate the association probability between the signal and the environment; potential hidden regions are generated according to the association probability; signal-trajectory matching is performed on the potential hidden regions to obtain the number of hidden individuals.

[0094] Specifically, it can receive distress signals sent from hidden environments, extract signal propagation features (such as signal strength attenuation curves, multipath reflection paths, and timestamps) and behavioral pattern features (such as signal transmission frequency, duration, and similarity to historical signals), and construct a signal feature vector F. sig .

[0095] Specifically, patrol signals can be acquired using multimodal sensing units (infrared, millimeter-wave radar, and lidar) on UAVs, environmental data blocks can be divided, and a three-dimensional spatial grid can be constructed. For each grid, the following operations are performed: heat source distribution modeling: heat source intensity and distribution density are detected using infrared sensors; motion trajectory prediction: the motion trajectory of the trapped individual is constructed using millimeter-wave radar data, and future location is predicted using Kalman filtering; environmental occlusion modeling: occlusion rate and spatial connectivity are generated using lidar point cloud data, thereby constructing an environmental feature vector F. env .

[0096] Optionally, if the association network is a graph neural network, then the association probability formula is: P ASSOC =GNN(F sig ,F env ), where F sig F represents the signal eigenvector. env This represents the environmental feature vector.

[0097] Optionally, the step of performing signal-trajectory matching on the potential hidden area to obtain the number of hidden individuals includes: performing DBSCAN clustering on the infrared heat sources of the potential hidden area to exclude isolated points (such as animal heat sources); analyzing the millimeter-wave radar trajectories of the clustered points using an LSTM network to identify motion data that conforms to human behavioral patterns (such as periodic movement or stationary states); aligning the timestamp of the distress signal with the time window of the predicted trajectory, and calculating the matching degree using the formula: This represents the distress signal at time t. The predicted trajectory at time t is represented; by combining DBSCAN clustering results, motion data, and matching degree, an estimate of the number of hidden persons is generated; a dynamic correction module is introduced to generate correction coefficients based on historical hidden person error models and real-time environmental parameters. ; through correction coefficients The estimated number of hidden individuals is corrected to obtain the total number of hidden individuals.

[0098] Among them, the correction coefficient The calculation formula is: Where O represents the occlusion change rate, N represents the historical hidden population deviation rate, and k1 and k2 are preset coefficients, further calculated using the formula N corr =N imp · The number of hidden individuals is obtained, where N imp This represents an estimated number of people in the hidden population.

[0099] Step S2016: Generate an estimated number of trapped individuals based on the first visible number of individuals, the second visible number of individuals, and the hidden number of individuals.

[0100] Specifically, an adaptive weighting method is used to fuse the first explicit number of people and the second explicit number of people to obtain the final explicit number of people. The final explicit number of people is then summed with the implicit number of people to obtain the estimated number of people trapped, N. est .

[0101] Step S202: Obtain real-time disaster prediction information for the target area.

[0102] Real-time disaster prediction information includes meteorological data (such as rainfall and wind speed), geological disaster risk levels, and the probability of secondary disasters (such as landslides and floods). Meteorological and geological monitoring data can be obtained through the National Meteorological Administration, local meteorological stations, and hydrological monitoring stations of the Ministry of Water Resources. This meteorological and geological monitoring data is then input into the disaster prediction model to output real-time disaster prediction information.

[0103] Step S203: Based on the estimated number of trapped people, the real-time disaster prediction information, and the preset material demand model, a dynamic material dispatch plan is generated and distributed to the rescue execution unit through the Internet of Things dispatch platform.

[0104] Specifically, based on the material demand mapping relationship constructed from historical disaster data, key material types (such as medical kits, drinking water, and tents) and their priorities corresponding to each unit of people are defined, and a material demand model is constructed. Based on the estimated number of trapped people and the preset material demand model, the initial material demand of the target area is calculated; a disaster correction factor is introduced through the real-time disaster prediction information; the initial material demand is corrected using the disaster correction factor; and the corrected initial material demand is input into a multi-objective optimization scheduling model to generate a dynamic material scheduling scheme.

[0105] The formula for calculating the initial material requirements of the target area is as follows: , where w i f represents the priority weight of material type i. i This is the demand function for this type of material. The formula for calculating the disaster correction factor α is: Where R is the disaster intensity (e.g., rainfall), S is the secondary disaster risk score, and k3 and k4 are preset coefficients, thus obtaining the corrected initial material requirements. =D init ·α.

[0106] Furthermore, to minimize overall costs, including transportation costs, time-delay penalties, and resource waste, a multi-objective optimization scheduling model is constructed:

[0107]

[0108] in, This represents the unit transportation cost (related to distance and road conditions). This indicates the quantity of goods k transported from warehouse i to target area j. Indicates the arrival time of the supplies. λ1 and λ2 represent the expected arrival time of the supplies, and λ1 and λ2 represent the weighting coefficients.

[0109] The constraints of the multi-objective optimization scheduling model include inventory capacity, transportation capacity, timeliness requirements, and route feasibility.

[0110] This improves the response speed and timeliness of material dispatch.

[0111] Please see Figure 4 , Figure 4 A schematic diagram of the structure of the Internet of Things-based high-speed emergency material dispatching system provided by the present invention.

[0112] like Figure 4 As shown in the embodiment of the present invention, a high-speed emergency supplies dispatching system based on the Internet of Things is proposed. The system includes:

[0113] The number of people estimation module 301 is used to dispatch a drone equipped with a multimodal perception unit to fly to the target area for dynamic patrol, receive the distress signal sent by the target area through the Internet of Things terminal and the patrol signal returned by the drone, and generate an estimated number of people trapped.

[0114] Disaster prediction module 302 is used to acquire real-time disaster prediction information for the target area;

[0115] The material dispatching module 303 is used to generate a dynamic material dispatching plan based on the estimated number of trapped people, the real-time disaster prediction information, and the preset material demand model, and then distribute the plan to the rescue execution unit through the Internet of Things dispatching platform.

[0116] Please see Figure 5 , Figure 5 This is a schematic diagram illustrating an embodiment of the electronic device provided in this invention. For example... Figure 5 As shown, an embodiment of the present invention provides an electronic device 400, including a memory 410, a processor 420, and a computer program 411 stored in the memory 410 and executable on the processor 420. When the processor 420 executes the computer program 411, it performs the following steps:

[0117] An application is installed on the electronic device 400. The application dispatches a drone equipped with a multimodal sensing unit to the target area for dynamic patrol. It receives distress signals sent by the target area through an IoT terminal and patrol signals returned by the drone, and generates an estimated number of trapped people. It also obtains real-time disaster prediction information for the target area. Based on the estimated number of trapped people, the real-time disaster prediction information, and a preset material demand model, it generates a dynamic material dispatch plan and distributes it to the rescue execution unit through the IoT dispatch platform.

[0118] Please see Figure 6 , Figure 6 This is a schematic diagram illustrating an embodiment of a computer-readable storage medium provided by an embodiment of the present invention. For example... Figure 6 As shown, this embodiment provides a computer-readable storage medium 500 on which a computer program 411 is stored. When the computer program 411 is executed by a processor, it performs the following steps:

[0119] The computer-readable storage medium 500 includes an application program that dispatches a drone equipped with a multimodal sensing unit to the target area for dynamic patrol, receives distress signals sent by the target area through an IoT terminal and patrol signals returned by the drone, generates an estimated number of trapped people, obtains real-time disaster prediction information for the target area, and generates a dynamic material dispatch plan based on the estimated number of trapped people, the real-time disaster prediction information, and a preset material demand model, and distributes it to the rescue execution unit through the IoT dispatch platform.

[0120] It should be noted that the descriptions of each embodiment in the above embodiments have different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0121] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0122] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0123] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0124] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0125] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0126] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A high-speed rescue material scheduling method based on the Internet of Things, characterized in that, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: input the signal feature vector and the environment feature vector into an association network, and calculate the association probability of the signal and the environment; wherein the association network is a graph neural network; the formula for inputting the signal feature vector and the environment feature vector into the association network and calculating the association probability of the signal and the environment is: P ASSOC =GNN(F sig ,F env ), wherein F sig represents the signal feature vector, and F env represents the environment feature vector; The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps:

2. The high-speed emergency supplies dispatching method based on the Internet of Things according to claim 1, characterized in that, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the ； wherein, represents a distress signal at time t, represents a predicted trajectory at time t. 3.The Internet of Things based high-speed rescue material scheduling method according to claim 1, characterized in that, The dynamic material scheduling scheme is generated based on the trapped person number estimation value, the real-time disaster prediction information, and a preset material demand model, and includes: Based on the trapped person number estimation value and the preset material demand model, the initial material demand of the target area is calculated; Through the real-time disaster prediction information, a disaster correction factor is introduced; The initial material demand is corrected by using the disaster correction factor; The corrected initial material demand is input into a multi-objective optimization scheduling model to generate a dynamic material scheduling scheme.

4. The high-speed rescue material scheduling system based on the Internet of Things, characterized in that, The system includes: The number estimation module is used to dispatch a drone equipped with a multi-modal perception unit to fly to the target area for dynamic patrol, receive the distress signals sent by the target area through the Internet of Things terminal and the patrol signals returned by the drone, and generate a trapped person number estimation value; including: The environment division unit is used to return image data of the target area by the drone; the image data is divided into explicit environment and implicit environment; The signal processing unit is used to obtain the distress signals sent in the explicit environment to obtain the first explicit person number; including: Based on the distress signals sent in the explicit environment, a text semantic vector and a user attribute vector are constructed; The text semantic vector and the user attribute vector are fused to obtain a multi-modal feature vector; The multi-modal feature vector is input into a full connection layer to obtain the first explicit person number; The patrol signals returned in the explicit environment are obtained to obtain the second explicit person number; including: A pre-trained deep learning model is used for target detection on the image data in the explicit environment to identify the location of potential trapped persons; Infrared thermal imaging data is combined to verify whether the potential trapped persons have human thermal characteristics; Through millimeter wave radar point cloud data, it is confirmed whether the potential trapped persons have micro-motion characteristics related to vital signs; An initial explicit person number set P is output, each potential trapped person including the location, the thermal characteristics and the micro-motion characteristics; Dynamic behavior analysis is performed on each potential trapped person in the initial explicit person number set P to calculate an explicit behavior confidence; The potential trapped persons with the explicit behavior confidence greater than a preset threshold are counted to obtain the second explicit person number; The distress signals and the returned patrol signals in the implicit environment are obtained to obtain an implicit person number; including: Based on the distress signals sent in the implicit environment, a signal feature vector is constructed; Based on the returned patrol signals in the implicit environment, an environment feature vector is constructed; input the signal feature vector and the environment feature vector into an association network, and calculate the association probability of the signal and the environment; wherein the association network is a graph neural network; the formula for inputting the signal feature vector and the environment feature vector into the association network and calculating the association probability of the signal and the environment is: P ASSOC =GNN(F sig ,F env ), wherein F sig represents the signal feature vector, and F env represents the environment feature vector; According to the correlation probability, a potential implicit area is generated; Signal-trajectory matching is performed on the potential implicit area to obtain an implicit person number; the implicit person number includes: DBSCAN clustering is performed on the infrared heat sources of the potential implicit area to exclude isolated points; The millimeter wave radar trajectory of the clustered points is analyzed through an LSTM network to identify motion data conforming to human behavior patterns; The timestamps of the distress signals are aligned with the time window of the predicted trajectory to calculate a matching degree; The DBSCAN clustering result, the motion data and the matching degree are integrated to generate an implicit person number estimation value; A dynamic correction module is introduced to generate a correction coefficient Y according to a historical implicit person number error model and real-time environment parameters; The hidden number of people is obtained by correcting the hidden number of people estimation value by a correction coefficient Y; An estimated number of trapped people is generated according to the first number of people, the second number of people and the hidden number of people; A disaster prediction module is configured to acquire real-time disaster prediction information of the target area; A material scheduling module is configured to generate a dynamic material scheduling scheme based on the estimated number of trapped people, the real-time disaster prediction information and a preset material demand model, and deliver the dynamic material scheduling scheme to a rescue execution unit through an Internet of Things scheduling platform.

Images