Mountain area expressway emergency rescue path planning method and system

Abstract

The invention provides a mountain area highway emergency rescue path planning method and system. The method comprises the following steps: collecting traffic flow data, meteorological data and field state data of a disaster site; generating a dynamic risk map according to the meteorological data, the field state data and the traffic circulation data; determining rescue constraint conditions of the rescue bodies in the emergency rescue process according to the dynamic position information of the rescue bodies near the disaster site and the rescue body types of the rescue bodies; performing hierarchical scheduling on each rescue body based on the dynamic risk map and all rescue constraint conditions to obtain a space-time conflict matrix; and performing multi-target real-time path planning on each rescue body according to the space-time conflict matrix, and generating a space-time conflict-free collaborative rescue path set. By adopting the scheme of the invention, efficient cooperation of multiple rescue bodies can be realized in the multi-disaster-chain coupling evolution environment of the highways in the mountainous area so as to generate a global optimal rescue path without conflicts in time and space.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traffic control, and more particularly, to a mountainous highway emergency rescue path planning method and system. BACKGROUND

[0002] The mountainous highway is prone to geological disasters such as landslides and debris flows under extreme weather (such as heavy rain and earthquakes) due to complex terrain and fragile geological conditions, which may also cause secondary disasters such as traffic congestion and vehicle chain accidents. For example, the Ya'an section of the Beijing-Kunming Expressway was once disrupted for 5 hours due to landslides caused by heavy rain, and the golden rescue period was further delayed due to the conflict between rescue vehicles and stranded vehicles. In this scenario, it is necessary to integrate disaster site data, traffic flow status and rescue resource information in real time to achieve dynamic risk assessment and efficient rescue coordination to reduce disaster losses.

[0003] The traditional emergency rescue mode relies on manual patrol and single-point monitoring, which is difficult to grasp the disaster evolution situation in real time, and the rescue resource scheduling lacks overall coordination, often causing path conflicts, response delays and other problems. In the existing intelligent rescue system, disaster risk assessment relies on static models (such as single weather warning or geological disaster threshold judgment), which cannot quantify the dynamic risk of the "geology-weather-accident" chain evolution (such as the Guizhou rainstorm model only focuses on rainfall). Moreover, rescue scheduling often uses single-objective optimization (such as the shortest path), without considering the differentiated constraints of heterogeneous rescue bodies such as fire trucks, ambulances and engineering vehicles (such as the tolerance of injured personnel to bumps and the weight limit of bridges), resulting in insufficient path feasibility. At the same time, the path coordination planning lacks a time-space conflict resolution mechanism, and multiple vehicles may cause secondary delays when converging. Therefore, how to achieve efficient coordination of multiple rescue bodies in the multi-disaster chain coupling evolution environment of mountainous highways to generate a globally optimal rescue path without time-space conflicts has become a difficult problem in the industry. SUMMARY

[0004] The present application provides a mountainous highway emergency rescue path planning method and system, which can achieve efficient coordination of multiple rescue bodies in the multi-disaster chain coupling evolution environment of mountainous highways to generate a globally optimal rescue path without time-space conflicts.

[0005] In a first aspect, the present application provides a mountainous highway emergency rescue path planning method, comprising: locating a disaster site on a mountainous highway, and then collecting traffic flow data, weather data and site status data of the disaster site; determine a meteorological grid map of a region near the disaster site according to the meteorological data, determine a spatio-temporal evolution trend of geological disasters and secondary accidents at the disaster site based on the field state data and the traffic flow data through a pre-constructed disaster chain coupling deduction model, perform grid algebra superposition on the meteorological grid map and the spatio-temporal evolution trend to generate a dynamic risk map near the disaster site; obtain dynamic position information of each rescue body near the disaster site, and then determine rescue constraint conditions of the each rescue body in the emergency rescue process according to all the dynamic position information and rescue body types of the each rescue body; perform hierarchical scheduling on the each rescue body based on the dynamic risk map and all the rescue constraint conditions, and then obtain a spatio-temporal conflict matrix of all the rescue bodies in the emergency rescue process; perform multi-objective real-time path planning on the each rescue body according to the spatio-temporal conflict matrix to generate a set of spatio-temporally conflict-free cooperative rescue paths.

[0006] In some embodiments, the determining of the meteorological grid map of the region near the disaster site according to the meteorological data specifically includes: perform grid division on the region near the disaster site to obtain a climate analysis region with the disaster site as a center; convert the meteorological data into grid values and fill the grid values into the climate analysis region to generate an initial meteorological grid map; synchronously update the grid values according to the initial meteorological grid map and current time meteorological data to form the meteorological grid map of the region near the disaster site.

[0007] In some embodiments, the determining of the spatio-temporal evolution trend of geological disasters and secondary accidents at the disaster site based on the field state data and the traffic flow data through the pre-constructed disaster chain coupling deduction model specifically includes: extract a ground surface deformation feature near the disaster site from the field state data; construct a terrain-traffic flow change map of the disaster site based on the ground surface deformation feature and the traffic flow data; perform spatio-temporal deduction on the terrain-traffic flow correlation map through the pre-constructed disaster chain coupling deduction model to generate a spatio-temporal evolution probability field of the disaster chain at the disaster site; combine historical disaster accident records to dynamically calibrate the spatio-temporal evolution probability field to obtain the spatio-temporal evolution trend of geological disasters and secondary accidents at the disaster site.

[0008] In some embodiments, the grid algebra superposition on the meteorological grid map and the spatio-temporal evolution trend to generate the dynamic risk map near the disaster site specifically includes: aligning the meteorological grid map with the spatio-temporal evolution trend through a spatio-temporal grid aligner to obtain a spatio-temporal coupling tensor field of multi-modal data fusion; performing grid algebra operation on the spatio-temporal coupling tensor field to obtain all risk grid values in the spatio-temporal coupling tensor field of disaster sites; generating a dynamic risk map near the disaster site according to all the risk grid values.

[0009] In some embodiments, determining the rescue constraint condition of each rescue body in the emergency rescue process according to all the dynamic position information and the rescue body type of each rescue body specifically comprises: extracting the position information of the current time point in the dynamic position information of the rescue body according to the state label of the rescue body for each rescue body; determining the inherent constraint of the rescue body according to the rescue body type of the rescue body; calculating the time window constraint of the rescue body to reach the disaster site based on the position information; performing constraint fusion on the time constraint and the inherent constraint through a dynamic constraint adaptation algorithm to obtain the rescue constraint condition of the rescue body in the emergency rescue process, and then obtaining the rescue constraint condition of each rescue body in the emergency rescue process.

[0010] In some embodiments, performing hierarchical scheduling on each rescue body based on the dynamic risk map and all the rescue constraint conditions to obtain a spatio-temporal conflict matrix of all rescue bodies in the emergency rescue process specifically comprises: constructing a three-dimensional scheduling decision space of all rescue bodies through the dynamic risk map and all the rescue constraint conditions; determining the comprehensive rescue path of each rescue body from the location to the disaster site based on the three-dimensional scheduling decision space; scheduling each rescue body according to all the comprehensive rescue paths; performing spatio-temporal detection on the spatio-temporal convergence points of the comprehensive rescue paths of all rescue bodies after scheduling to generate a spatio-temporal conflict matrix of all rescue bodies in the emergency rescue process.

[0011] In some embodiments, performing multi-objective real-time path planning on each rescue body according to the spatio-temporal conflict matrix to generate a set of spatio-temporal conflict-free cooperative rescue paths specifically comprises: identifying a set of rescue bodies whose rescue paths are affected according to the spatio-temporal conflict matrix; performing rescue path re-planning on each rescue body in the set of rescue bodies to obtain an optimized spatio-temporal matrix that eliminates conflicts; generating a set of spatio-temporal conflict-free cooperative rescue paths of all rescue bodies based on the optimized spatio-temporal matrix.

[0012] In a second aspect, the application provides a mountainous highway emergency rescue path planning system, which comprises: The acquisition module is configured to locate a disaster site on the mountainous highway, and then acquire traffic flow data, meteorological data and field state data of the disaster site. The processing module is configured to determine a meteorological grid map of a region near the disaster site according to the meteorological data, determine a time-space evolution trend of geological disasters and secondary accidents at the disaster site based on the field state data and the traffic flow data through a pre-constructed disaster chain coupling deduction model, and generate a dynamic risk map near the disaster site by performing grid algebra superposition on the meteorological grid map and the time-space evolution trend. The processing module is configured to acquire dynamic position information of each rescue body near the disaster site, and then determine rescue constraint conditions of the each rescue body in the emergency rescue process according to all the dynamic position information and rescue body types of the each rescue body. The processing module is configured to perform hierarchical scheduling on the each rescue body based on the dynamic risk map and all the rescue constraint conditions, and then obtain a time-space conflict matrix of all the rescue bodies in the emergency rescue process. The execution module is configured to perform multi-objective real-time path planning on the each rescue body according to the time-space conflict matrix, and generate a set of time-space conflict-free cooperative rescue paths.

[0013] In a third aspect, the application provides a computer device, which comprises a memory and a processor, the memory stores a code, and the processor is configured to acquire the code and perform the mountainous highway emergency rescue path planning method described above.

[0014] In a fourth aspect, the application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the mountainous highway emergency rescue path planning method described above.

[0015] The technical scheme provided by the embodiments of the application has the following beneficial effects: The mountainous highway emergency rescue path planning method and system provided by the application first locates a disaster site on a mountainous highway, and then collects traffic flow data, meteorological data and field state data of the disaster site; a meteorological grid map of a region near the disaster site is determined according to the meteorological data, a spatio-temporal evolution trend of geological disasters and secondary accidents at the disaster site is determined based on the field state data and the traffic flow data through a pre-constructed disaster chain coupling deduction model, the meteorological grid map and the spatio-temporal evolution trend are subjected to grid algebra superposition, and a dynamic risk map near the disaster site is generated; dynamic position information of each rescue body near the disaster site is obtained, and then rescue constraint conditions of the each rescue body in an emergency rescue process are determined according to all the dynamic position information and rescue body types of the each rescue body; the each rescue body is subjected to hierarchical scheduling based on the dynamic risk map and all the rescue constraint conditions, and then a spatio-temporal conflict matrix of all the rescue bodies in the emergency rescue process is obtained; each rescue body is subjected to multi-objective real-time path planning according to the spatio-temporal conflict matrix, and a set of spatio-temporal conflict-free cooperative rescue paths is generated.

[0016] It can be seen that the application carries out multi-objective real-time path planning for each rescue body according to the space-time conflict matrix to generate a set of space-time conflict-free cooperative rescue paths; first, determining a dynamic risk map can obtain a visual atlas showing the risk distribution near the disaster site at different space-time and updating in real time, and the determination of the dynamic risk map can be through the grid algebra superposition of the meteorological grid map and the space-time evolution trend of the disaster chain, and the real-time meteorological data (such as rainfall, wind speed), the on-site deformation characteristics (such as slope cracks, water accumulation area) and the traffic flow state (such as vehicle density, speed) are converted into a calculable space-time continuous risk field, solving the dynamic fusion problem of multi-source heterogeneous data, providing an intuitive risk reference for rescue path planning, helping to avoid high-risk areas and improving rescue safety; then, determining the rescue constraint condition can obtain all the constraint conditions of the rescue body in the rescue ability and the rescue time when executing the rescue task, and the determination of the rescue constraint condition can be through the integration of the rescue body type and the dynamic position information, setting the targeted time constraint for different rescue bodies (such as binding a "30-minute golden rescue time window" for an ambulance), and through the clear ability boundary (such as the maximum load of 50kg of a UAV, the endurance of 30 minutes) and the task target (such as the ambulance is responsible for the transfer of the wounded, and the engineering vehicle is responsible for the road obstacle removal) of each rescue body, a quantitative basis is provided for hierarchical scheduling, and it is ensured that the planned path is consistent with the actual ability and time requirement of the rescue body; finally, determining the space-time conflict matrix can obtain a matrix showing the conflict intensity of all rescue bodies at the space-time convergence point of the rescue path after the rescue scheduling, and the determination of the space-time conflict matrix can identify the potential conflict of the rescue body in the space-time dimension, so as to provide a conflict resolution target point for subsequent multi-objective path planning by accurately positioning the high conflict area and time period, and through the three functions of efficient detection, quantitative evaluation and linkage adaptation, it becomes the key hub connecting dynamic risk assessment and multi-objective path planning, not only solving the technical pain points of "inefficient conflict detection and lack of basis for cooperative decision-making" in the background technology, but also through the deep integration with the risk map and the rescue constraint, building a closed-loop system of "real-time conflict sensing-intelligent conflict resolution-global optimization cooperation", making the mountainous highway emergency rescue from "experience-driven" to "data-driven", and providing core technical support for realizing "space-time conflict-free cooperative rescue path"; in summary, based on the above scheme, the efficient cooperation of multiple rescue bodies can be realized in the mountainous highway multi-disaster chain coupling evolution environment to generate a globally optimal rescue path without space-time conflict. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is an exemplary flowchart of a mountainous highway emergency rescue path planning method according to some embodiments of the application; Figure 2 is an operation flowchart for determining a space-time evolution trend according to some embodiments of the application; Figure 3is an exemplary flowchart of determining a rescue constraint condition according to some embodiments of the present application; Figure 4 is a structural schematic diagram of a mountainous highway emergency rescue path planning system according to some embodiments of the present application; Figure 5 is an internal structural diagram of a computer device for implementing a mountainous highway emergency rescue path planning method according to some embodiments of the present application. DETAILED DESCRIPTION

[0018] In order to better understand the technical solutions of the present application, the technical solutions of the present application will be described in detail below in combination with the drawings of the specification and specific embodiments.

[0019] Reference Figure 1 The figure is an exemplary flowchart of a mountainous highway emergency rescue path planning method according to some embodiments of the present application, which mainly includes the following steps: In step 101, the disaster site on the mountainous highway is located, and then the traffic flow data, meteorological data and on-site state data of the disaster site are collected.

[0020] It should be noted that in the present application, the traffic flow data refers to a data set composed of vehicle flow, vehicle density and vehicle speed information near the disaster site (such as within a diameter distance of 10 kilometers), which can provide accurate traffic state input for disaster chain coupling deduction, thereby improving the spatio-temporal accuracy and reliability of geological disaster and secondary accident prediction; the meteorological data refers to a data set composed of temperature, precipitation and wind speed information near the disaster site (such as within a diameter distance of 10 kilometers), which can be used to construct a meteorological grid map near the disaster site and superimposed with disaster evolution probability, thereby dynamically reflecting the influence of meteorological conditions on risk distribution; the on-site state data is image data showing the surface deformation characteristics (such as slope fissure, landslide precursor, etc.) near the disaster site (such as within a diameter distance of 10 kilometers), which can reflect the geological state change of the disaster site and enhance the real perception ability of disaster evolution trend deduction.

[0021] In specific implementation, locating the disaster site on a mountain highway and then collecting traffic flow data, meteorological data and on-site status data at the disaster site can be achieved in the following ways, namely: first, the precise coordinates (i.e., longitude and latitude) of the disaster site on a mountain highway can be obtained through the Beidou high-precision positioning system, and the precise coordinates (i.e., longitude and latitude) can be matched to the pre-built highway vector grid to achieve precise positioning at the kilometer pile level; then, the vehicle traffic data (including timestamp, license plate, speed, lane, etc.) near the disaster site (e.g., within a diameter of 10 kilometers) can be obtained through the pre-installed roadside induction coils and electronic toll collection system card interface on the highway, and the Kalman filter algorithm is used to denoise and smooth the vehicle traffic data, and then the real-time vehicle flow, vehicle density and vehicle speed near the disaster site are calculated at preset time intervals (e.g., 1 second), and all vehicle flow, vehicle density and vehicle speed are recorded. The collection of vehicle density and vehicle speed information is used as traffic flow data; then, the observation interface of the National Meteorological Administration's automatic weather station near the disaster site can be called to obtain the temperature, precipitation and wind speed information of the disaster site at preset time intervals (such as 1 second), and the collection of all temperature, precipitation and wind speed information is used as meteorological data; finally, a cluster of multi-rotor drones can be dispatched to take oblique photography of the surface near the disaster site (such as within a diameter of 10 kilometers), and the captured high-definition optical images are used as on-site status data; among them, the drones are equipped with cameras and multispectral sensors, and can fly according to preset routes (such as grid coverage) to collect high-definition optical images with a resolution of 0.1 meters; at the same time, the sliding time window technology can be used to control the quality of the meteorological data to eliminate outliers (such as wind speed exceeding the sensor range), and the meteorological data can be interpolated to a resolution of 1 second through a linear interpolation algorithm to ensure that the meteorological data is synchronized with the traffic flow data.

[0022] In step 102, a meteorological grid map of the area near the disaster site is determined based on the meteorological data, and the spatiotemporal evolution trend of geological disasters and secondary accidents occurring at the disaster site is determined based on the on-site status data and the traffic flow data through a pre-constructed disaster chain coupling deduction model. The meteorological grid map and the spatiotemporal evolution trend are grid algebraically superimposed to generate a dynamic risk map near the disaster site.

[0023] In some embodiments, determining a meteorological grid map of an area near a disaster site based on the meteorological data may be achieved by the following steps: The area near the disaster site is divided into grids to obtain a climate analysis area centered on the disaster site; Converting the meteorological data into grid values ​​and filling them into the climate analysis area to generate an initial meteorological grid map; The grid values ​​are dynamically updated according to the current meteorological data synchronized with the initial meteorological grid map to form a meteorological grid map of the area near the disaster site.

[0024] It should be noted that in the present application, the weather grid map is a real-time weather map based on the initial weather grid map to update the grid value of the latest weather data in real time, which can reflect the changes of meteorological elements near the disaster site in real time, provide real-time weather parameters for generating dynamic risk maps, ensure that the risk assessment is consistent with the latest weather conditions, and improve the timeliness and safety of the rescue path.

[0025] In specific implementation, the climate analysis area centered on the disaster site can be implemented in the following manner, that is, the center point coordinates of the disaster site can be obtained through Beidou high-precision positioning, and a square analysis area with a side length of 10 kilometers is constructed with the center point coordinates as the origin, and then the square analysis area is taken as the climate analysis area centered on the disaster site; wherein, the climate analysis area is a spatial range centered on the disaster site, which is an analysis boundary for meteorological data, can provide a structured storage carrier for meteorological data, and ensure that the meteorological analysis covers the core area affected by the disaster.

[0026] In specific implementation, the meteorological data is converted into grid values and filled into the climate analysis area to generate the initial weather grid map, which can be implemented in the following manner, that is, first, the square analysis area can be divided into a plurality of 10m x 10m square grid units by using an equidistant grid division algorithm, and each grid unit is assigned a unique identifier to generate a three-dimensional matrix containing a plurality of addressable grids; then, the meteorological risk index of each grid unit can be calculated according to the earliest recorded temperature, precipitation and wind speed information in the meteorological data through the geostatistical analysis tool in the geographic information system (such as Arc geographic information system), and all meteorological risk indexes are filled into the corresponding grid units as grid values to generate the initial weather grid map; wherein, the initial weather grid map is an initial weather map showing the initial spatial distribution of meteorological data in the climate analysis area, which can preliminarily quantify the spatial distribution of meteorological elements, provide a benchmark layer for dynamic updating of meteorological data, and ensure data continuity; the geostatistical analysis tool can use Kriging method or inverse distance weighting algorithm to fuse multi-source meteorological data into a unified meteorological risk index.

[0027] In a specific implementation, the dynamic updating of the grid value of the current time meteorological data according to the initial meteorological grid map can be achieved in the following manner: a meteorological data subscription mechanism based on a message queue telemetry transport protocol can be established, meteorological station data updates can be received once every preset time interval (e.g., 1 second), and a Kalman filtering algorithm can be used by a grid calculator of a geographic information system to predict the trend of the change in the grid value and perform incremental updating of the grid value. When the change in the grid value exceeds a preset change threshold (e.g., 0.5), a ripple diffusion updating algorithm is triggered, and a 3*3 neighborhood smoothing process is performed with the change point as the center, thereby obtaining a real-time refreshed meteorological grid map of the area near the disaster site.

[0028] In some embodiments, reference is made to Figure 2 The figure is an operation flow chart for determining the spatio-temporal evolution trend according to some embodiments of the present application. In the present application, the spatio-temporal evolution trend of geological disasters and secondary accidents at the disaster site is determined based on the field state data and the traffic flow data by using a pre-constructed disaster chain coupling deduction model. The following steps can be used to achieve this: The ground surface deformation features near the disaster site are extracted from the field state data. A terrain-traffic flow change map of the disaster site is constructed based on the ground surface deformation features and the traffic flow data. The terrain-traffic flow correlation map is spatio-temporally deduced by using a pre-constructed disaster chain coupling deduction model, and a spatio-temporal evolution probability field of the disaster chain at the disaster site is generated. The spatio-temporal evolution probability field is dynamically calibrated in combination with historical disaster accident records, and a spatio-temporal evolution trend of geological disasters and secondary accidents at the disaster site is obtained.

[0029] It should be noted that, in the present application, the ground surface deformation features are quantitative features that reflect the disaster conditions of the ground surface at the disaster site (e.g., the slope angle, the amount of roadbed settlement, and the depth of accumulated water, etc.). These ground surface deformation features accurately depict the physical state of the disaster site, providing basic parameters for constructing the terrain-traffic flow correlation map and ensuring that the subsequent deduction is consistent with the actual situation on site.

[0030] In a specific implementation, the extraction of the ground deformation features near the disaster site from the field state data can be achieved in the following manner: a pre-trained deep learning-based target detection algorithm (such as YOLOv8) can be used to automatically identify the ground deformation features (such as slope cracks, water accumulation areas, etc.) near the disaster site according to the field state data; wherein the target detection algorithm can be learned on a historical disaster data set, for example: set the batch size to 16, the learning rate to 0.001, and train the optimizer for 100 rounds, then cut the input field state data into 640x640 pixel sub-blocks and normalize them, then call the trained model to convert the crack contour pixel length to actual length by calculation, calculate the crack expansion rate by comparing the results at different time points, perform morphological dilation operation on the water accumulation area to calculate the submergence depth, and finally generate the ground deformation features with geographic coordinates.

[0031] In a specific implementation, the construction of the terrain-traffic variation graph of the disaster site based on the ground deformation features and the traffic flow data can be achieved in the following manner: a three-layer node graph can be constructed based on the ground deformation features and the traffic flow data, for example: 8 key monitoring points (such as the top of the slope, the waist of the slope, etc.) can be set for the slope layer nodes, 5 road section units (such as the 200-meter road section before and after the disaster point) can be set for the roadbed layer nodes, and 3 vehicle clusters (such as congested sections and smooth sections) can be set for the traffic layer nodes; the correlation within the layers can be represented by an adjacency matrix (such as the displacement conduction coefficient between slope nodes), and the correlation between layers can be represented by a coupling matrix (such as the weight of the influence of slope displacement on roadbed stability is 0.6, and the weight of the influence of roadbed damage on traffic speed is 0.8); then the traffic flow data can be converted into traffic layer node attributes (such as congestion index) to generate the terrain-traffic correlation graph of the disaster site; wherein the terrain-traffic correlation graph is a structured graph that reflects the dynamic influence relationship between various elements on the scene of the disaster site, and it contains node attributes and inter-layer coupling weights, and provides a logical framework for spatiotemporal reasoning by quantifying the inter-layer correlation (such as the influence weight of slope on roadbed), making the multi-source data form an organic whole and supporting disaster chain conduction analysis.

[0032] In a specific implementation, the time-space evolution trend of the terrain-vehicle flow correlation graph is deduced by using a pre-constructed disaster chain coupling deduction model to generate a time-space evolution probability field of the disaster chain at the disaster location. This can be achieved in the following manner: an existing time-space deduction model (such as a Python-based cellular automaton-multi-agent system coupling model) can be loaded as the disaster chain coupling deduction model to deduce the time-space evolution probability field of the disaster chain at the disaster location from the terrain-vehicle flow correlation graph; wherein the initial weight of the terrain-vehicle flow correlation graph (such as 0.6) can be input at the initial time, and the disaster chain coupling deduction model can be deduced at a time step of 5 minutes, the state parameters of each node in the terrain-vehicle flow change graph can be calculated by using inter-layer coupling rules (such as the probability of roadbed damage increases by 5% for every 1° increase in slope angle; the probability of vehicle rear-end collision increases by 8% for every 1-level increase in roadbed damage level), and 1000 possible evolution paths can be generated by using Monte Carlo simulation, so as to statistically obtain the probability values of the location represented by each grid unit near the disaster location for geological disasters (such as landslides) and secondary accidents (such as consecutive rear-end collisions) in a future time period (such as 1 hour), thereby forming a time-space evolution probability field; the time-space evolution probability field is a probability time sequence showing the possibility of geological disasters and secondary accidents near the disaster location, which reflects the time-space development possibility of the disaster chain and provides a quantitative basis for subsequent calibration, and the probability distribution can be used to clearly determine the high-risk area and time period, thereby improving the refinement degree of the prediction.

[0033] It should be noted that, in the present application, the time-space evolution trend is the dynamic trend of predicting the time, location and type of geological disasters and secondary accidents near the disaster location, which presents the complete evolution process of the disaster chain, includes the disaster type and occurrence probability at different times and different locations in a future time period (such as 1 hour) and is presented in a three-dimensional dynamic graph, which can provide a disaster evolution basis for a dynamic risk map, ensure that the risk assessment is updated in real time as the disaster develops, and enhance the foresight of rescue dispatching; in a specific implementation, the time-space evolution trend of geological disasters and secondary accidents at the disaster location can be obtained by dynamically calibrating the time-space evolution probability field in combination with historical disaster accident records in the following manner: a calibration sample library can be constructed by using pre-collected historical disaster accident records (including slope parameters, traffic flow states, actual disaster types and occurrence times, etc.) of mountainous areas in the past 10 years, and the coupling weight of the disaster chain coupling deduction model can be adjusted by using the least squares method (such as the historical data shows that the influence weight of the roadbed on the vehicle flow is actually 0.7, and the model parameters are corrected) to correct the deviation of the time-space evolution probability field (such as the prediction probability of a certain grid unit is 20%, the actual probability of a similar scenario in history is 15%, and the calibration is 17%), and at the same time, the calibration coefficient is updated every predetermined update time interval (such as 5 minutes) by using the latest traffic flow data, weather data and on-site state data, and finally the calibrated time-space evolution trend is output.

[0034] In some embodiments, the dynamic risk map near the disaster site is generated by performing grid algebraic superposition on the weather grid map and the spatio-temporal evolution trend, which can be achieved by the following steps: aligning the weather grid map and the spatio-temporal evolution trend in space-time through a spatio-temporal grid aligner to obtain a spatio-temporal coupling tensor field of multi-modal data fusion; performing grid algebraic operations on the spatio-temporal coupling tensor field to obtain all risk grid values in the spatio-temporal coupling tensor field of the disaster site; generating a dynamic risk map near the disaster site according to all risk grid values.

[0035] In a specific implementation, the spatio-temporal coupling tensor field of multi-modal data fusion can be obtained by aligning the weather grid map and the spatio-temporal evolution trend in space-time through a spatio-temporal grid aligner, which can be achieved in the following manner: first, the coordinate systems of the weather grid map and the spatio-temporal evolution trend can be unified based on the latitude and longitude of the disaster site using a spatio-temporal grid aligner of a geographic information system, and then the time step of the spatio-temporal evolution trend is adjusted to be the same as that of the weather grid map through linear interpolation (such as bilinear interpolation), so as to ensure that the time granularity of the two is consistent, and the edge grid of the weather grid map and the spatio-temporal evolution trend is resampled to eliminate the resolution deviation, thereby generating the spatio-temporal coupling tensor field of multi-modal data fusion; wherein the spatio-temporal coupling tensor field is a multi-modal data set containing the latitude and longitude of the disaster site, weather values at different time points, and disaster evolution probabilities. The spatio-temporal coupling tensor field is formed after the weather grid map and the spatio-temporal evolution trend are aligned in space-time, and can unify the spatio-temporal reference of multi-source data, thereby providing a consistent data carrier for subsequent grid algebraic operations and ensuring the spatio-temporal matching of weather and disaster evolution data, thereby improving the fusion accuracy.

[0036] In a specific implementation, the grid algebra operation on the spatiotemporal coupling tensor field can be implemented in the following manner: a grid calculator tool in a geographic information system is called to perform the grid algebra operation on the spatiotemporal coupling tensor field, to obtain the risk grid value of each grid cell in the spatiotemporal coupling tensor field of the disaster site at the current time step. For example, the proportion of meteorological factors (such as rainfall and wind speed) can be set to 0.3, and the proportion of spatiotemporal evolution trends (such as geological disaster probability and secondary accident probability) can be set to 0.7. The operation formula is: the risk grid value is equal to the meteorological grid value multiplied by 0.3, plus the evolution probability value multiplied by 0.7. Then, all the operation results are normalized by using an existing normalization method (such as the maximum-minimum standardization), to map them to the interval of 0-10, eliminate the dimensional difference, and obtain all the risk grid values in the risk grid value spatiotemporal coupling tensor field of the disaster site. The risk grid value is a grid value quantifying the comprehensive risk intensity of a single grid cell in the spatiotemporal coupling tensor field, which can provide a quantitative basis for risk classification, convert risk assessment from qualitative to quantitative, and enhance the accuracy.

[0037] In a specific implementation, the dynamic risk map near the disaster site can be generated in the following manner: all the risk grid values of the spatiotemporal coupling tensor field are classified, for example, the risk grid values can be classified into five risk levels, 0-2 is low risk, 3-4 is lower risk, 5-6 is medium risk, 7-8 is higher risk, and 9-10 is high risk. Then, the classification results are visualized and rendered by using a symbol system tool in a geographic information system, a time axis control is added to realize dynamic playing (such as updating one frame every 5 minutes), and a dynamic risk map near the disaster site is obtained.

[0038] It should be noted that, in this application, the dynamic risk map is a visual map showing the distribution of risks at different times and spaces near the disaster site and being updated in real time. The dynamic risk map supports zooming in, querying the risk level of any grid cell, and corresponding time, can provide an intuitive risk reference for rescue path planning, help to avoid high-risk areas, and improve rescue safety.

[0039] In step 103, the dynamic position information of each rescue body near the disaster site is obtained, and then the rescue constraint conditions of each rescue body in the emergency rescue process are determined according to all the dynamic position information and the rescue body type of each rescue body.

[0040] In a specific implementation, the dynamic position information of each rescue body near the disaster site can be obtained in the following manner: a Beidou dual-mode positioning terminal uniformly provided for the rescue body (such as an ambulance, a drone, an engineering vehicle, etc.) receives satellite signals to generate real-time data of the rescue body near the disaster site (such as within 10 kilometers in straight-line distance), such as the latitude and longitude coordinates, the moving speed, and the remaining endurance, thereby generating the position information (including the latitude and longitude coordinates, the moving speed, and the remaining endurance) of each rescue body with a timestamp, and then encapsulating all the position information in the format of "rescue body label + position information + state label" using the message queue telemetry transport protocol, and transmitting the encapsulated position information to a distributed database in real time to obtain the dynamic position information of each rescue body near the disaster site. In this process, all rescue bodies are stored in the distributed database according to the rescue body type (such as an ambulance, a drone, an engineering vehicle, etc.), and the state label of the rescue body is synchronously marked (such as 1 for idle and ∞ for busy), the distributed database automatically updates the stored data every 30 seconds, and the latest position information is pushed to the rescue dispatch platform in real time through an interface to ensure that the obtained dynamic position information is consistent with the current state of the rescue body. The dynamic position information is data information reflecting the real-time position, the moving speed, and the state (such as idle / busy) of the rescue body (such as an ambulance, a drone, an engineering vehicle, etc.) changing over time, which can track the spatial movement of the rescue body in real time, provide a position reference (such as a time window from the disaster site) for determining the rescue constraint condition, provide real-time spatial data for hierarchical dispatch and path planning, ensure that the rescue body dispatch is consistent with its dynamic position, reduce the time-space conflict, and improve the efficiency of collaborative rescue.

[0041] In some embodiments, reference is made to Figure 3 FIG. 1 is an exemplary flowchart for determining a rescue constraint condition according to some embodiments of the present application, and the determination of the rescue constraint condition of each rescue body in the emergency rescue process according to all the dynamic position information and the rescue body type of each rescue body can be implemented in the following steps: In step 1031, for each rescue body, the position information at the current time point is extracted from the dynamic position information of the rescue body according to the state label of the rescue body; In step 1032, the inherent constraint of the rescue body is determined according to the rescue body type of the rescue body; In step 1033, the time window constraint of the rescue body to reach the disaster site is calculated based on the position information; In step 1034, the time constraint and the inherent constraint are fused by a dynamic constraint adaptation algorithm to obtain the rescue constraint condition of the rescue body in the emergency rescue process, and thus the rescue constraint condition of each rescue body in the emergency rescue process is obtained.

[0042] In a specific implementation, the position information of the rescue body at the current time point can be extracted from the dynamic position information of the rescue body according to the state label of the rescue body, that is, when the state label of the rescue body is in the idle state, the position information of the rescue body at the current time point is extracted from the dynamic position information of the rescue body. The position information is real-time data reflecting the current spatial position of the rescue body (such as longitude and latitude, moving speed, and remaining endurance), and the spatial relationship between the rescue body and the disaster point can be determined through the position information, which provides basic data for calculating the straight-line distance and generating the time window constraint, and ensures the accuracy of the spatial benchmark of the time window calculation.

[0043] In a specific implementation, the inherent constraint of the rescue body can be determined according to the rescue body type of the rescue body, that is, the physical constraint (such as the maximum range of 100 kilometers and the load of 50 kg for a drone, and the maximum water capacity of 8 tons and the endurance time of 4 hours for a fire truck) and the functional constraint (such as the medical transfer of an ambulance and the operation of an engineering vehicle) of the rescue body can be extracted from the existing rescue body capability characteristic library (containing 12 types of ambulances, fire trucks, drones, etc.) according to the rescue body type of the rescue body, and the set of the physical constraint and the functional constraint is taken as the inherent constraint of the rescue body. The inherent constraint is a fixed limit determined by the rescue body type (such as the load of a drone and the number of people in an ambulance), which defines the capability boundary of the rescue body and avoids scheduling tasks beyond the capability of the rescue body (such as transporting materials beyond the load of a drone), ensuring the feasibility of rescue.

[0044] In a specific implementation, the time window constraint of the rescue body reaching the disaster site can be calculated based on the position information, that is, the accurate coordinates (longitude and latitude) of the disaster site can be obtained, and the straight-line distance between the rescue body and the disaster point can be calculated according to the accurate coordinates of the disaster site and the longitude and latitude in the position information of the rescue body using the spherical distance formula. Then, the expected arrival time of the rescue body to the disaster site is calculated according to the moving speed in the position information of the rescue body, and the time window constraint is generated in combination with the emergency rescue golden time (such as 30 minutes for life rescue), for example, the earliest arrival time is calculated by "distance divided by average speed in moving speed", the latest arrival time is calculated by "distance divided by minimum speed in moving speed", and if the latest arrival time exceeds the emergency rescue golden time, the latest arrival time is forced to be set as the emergency rescue golden time. The time window constraint is the earliest and latest time range of the rescue body reaching the disaster point, which limits the time boundary of the rescue body for rescue, ensures the arrival of the rescue body within the effective time (neither delays nor waits too long), and balances the rescue efficiency and resource utilization.

[0045] It should be noted that in the present application, the rescue constraint condition is all the constraints on the rescue ability and rescue time when the rescue body executes the rescue task, which can provide a clear boundary for path planning (such as avoiding high-risk areas beyond the rescue body's ability), and ensure that the planned path meets the actual ability and time requirements of the rescue body; in specific implementation, the rescue constraint condition of the rescue body in the emergency rescue process can be obtained by constraint fusion of the time constraint and the inherent constraint through a dynamic constraint adaptation algorithm, which can be implemented in the following way, that is, the existing dynamic constraint adaptation algorithm (such as the improved Lagrange relaxation algorithm) can be used to fuse the time constraint and the inherent constraint to obtain the rescue constraint condition of the rescue body in the emergency rescue process; wherein the objective function of the dynamic constraint adaptation algorithm can be set as: min(time window violation degree x time constraint weight + inherent constraint violation degree x inherent constraint weight), the time constraint weight and the inherent constraint weight can be dynamically adjusted according to the disaster level, and the time window constraint and the inherent constraint are conflict resolved through iterative optimization: if the time window requirement is too tight to cause the UAV to run out of power, the time window is extended; if the load constraint conflicts with the time window, the life-saving task is prioritized, and finally a rescue constraint vector containing the time constraint factor and the inherent constraint factor is generated as the rescue constraint condition.

[0046] In step 104, each rescue body is scheduled based on the dynamic risk map and all rescue constraint conditions, and then a space-time conflict matrix of all rescue bodies in the emergency rescue process is obtained.

[0047] In some embodiments, the space-time conflict matrix of all rescue bodies in the emergency rescue process can be obtained by scheduling each rescue body based on the dynamic risk map and all rescue constraint conditions in the following steps: A three-dimensional scheduling decision space of all rescue bodies is constructed based on the dynamic risk map and all rescue constraint conditions; Based on the three-dimensional scheduling decision space, a comprehensive rescue path of each rescue body from the location to the disaster site is determined; According to all the comprehensive rescue paths, the scheduling echelon is divided to schedule each rescue body; The space-time intersection points of the comprehensive rescue paths of all scheduled rescue bodies are detected in space-time, and a space-time conflict matrix of all rescue bodies in the emergency rescue process is generated.

[0048] In a specific implementation, the three-dimensional scheduling decision space of all rescue bodies can be constructed by the dynamic risk map and all rescue constraints in the following manner: a three-dimensional coordinate system is established based on the dynamic risk map with the disaster site as the origin (e.g., X represents longitude and Y represents latitude), and all risk grid values in the dynamic risk map are mapped to the Z axis of the three-dimensional coordinate system; then, the comprehensive rescue weight vector of each grid unit is generated by improving the weight calculation logic of the Floyd algorithm, thereby obtaining the three-dimensional scheduling decision space of all rescue bodies; the comprehensive rescue weight vector is composed of the comprehensive rescue weights of all rescue bodies, and the comprehensive rescue weight of each rescue body can be calculated according to the calculation formula of the weight calculation logic, i.e., comprehensive rescue weight of rescue body = (Euclidean distance from grid unit to disaster site + risk grid value of grid unit × time window constraint + time window constraint × inherent constraint factor) × type weight of rescue body (e.g., 1.5 for life rescue); the three-dimensional scheduling decision space is a stereoscopic decision model providing a multi-dimensional quantitative benchmark for path planning of rescue bodies, which converts risk, time and capacity constraints into calculable spatial weight parameters, so that the path planning meets safety, timeliness and feasibility at the same time, and avoids scheduling deviation caused by a single factor.

[0049] In a specific implementation, the comprehensive rescue path of each rescue body from the current location to the disaster site can be determined based on the three-dimensional scheduling decision space in the following manner: for each rescue body, the optimal path search algorithm (e.g., improved A-star algorithm) is used to search for the optimal path to the disaster site according to the location coordinates of the rescue body at the current time in the three-dimensional scheduling decision space, thereby obtaining the comprehensive rescue path of the rescue body from the current location to the disaster site; the heuristic function of the optimal path search algorithm can be set as f(n) = actual cost (i.e., comprehensive rescue weight accumulation) from the current location to the current grid unit + estimated cost (i.e., risk weighted straight-line distance) from the current grid unit to the destination, and different movement rules are set for different types of rescue bodies (e.g., a UAV can pass through an area with a risk grid value ≤ 7, and a ground rescue body can only pass through an area with a risk grid value ≤ 5); the optimal path search algorithm dynamically avoids high-risk grids in the path search process based on the obstacle avoidance strategy, thereby obtaining the comprehensive rescue path of the rescue body from the current location to the disaster site; the comprehensive rescue path is the optimal driving route of the rescue body from the current location to the disaster site, which contains path point coordinates, estimated arrival time and risk grid value, and takes into account risk avoidance, time window constraints and rescue body capacity limitations; different movement rules are set for different types of rescue bodies to ensure that each path meets the inherent constraints (e.g., load and endurance) and meets the requirements of dynamic risk avoidance, thereby improving the reliability of rescue task execution.

[0050] In a specific implementation, the dispatching of each rescue body according to all the integrated rescue paths can be implemented in the following manner, that is, three levels of dispatch teams can be divided according to rescue priorities (life rescue > material transportation > road repair), the first level of dispatch team can be a UAV and an ambulance, the second level of dispatch team can be a fire engine and an engineering vehicle, and the third level of dispatch team can be a transport vehicle. A hierarchical strategy based on dynamic programming (for example, the first level of team needs to start dispatching within 5 minutes after the disaster occurs, the second level of team needs to start dispatching within 10 minutes, and the third level of team needs to start dispatching within 20 minutes) is used to dispatch rescue according to the integrated rescue paths of the rescue bodies in each dispatch team.

[0051] It should be noted that in the present application, the space-time conflict matrix is a matrix showing the conflict intensity of all rescue bodies at the space-time convergence points of the rescue paths after rescue dispatching. The space-time conflict matrix can identify the potential conflicts of the rescue bodies in the space-time dimension, so as to provide conflict resolution targets for subsequent multi-objective path planning by accurately positioning the high-conflict areas and time periods, and avoid the efficiency loss or safety risk caused by the aggregation of rescue bodies in the same space-time area. In a specific implementation, the space-time detection of the space-time convergence points of the integrated rescue paths of all rescue bodies after dispatching, and the generation of the space-time conflict matrix of all rescue bodies in the emergency rescue process can be implemented in the following manner, that is, the time difference (for example, less than 5 minutes) and space difference (for example, less than 100 meters) of the intersection points of all integrated rescue paths can be detected by using an existing space-time conflict detector (for example, a space-time conflict detection algorithm based on a quadtree index), and a quadtree space index is used to accelerate the detection process, that is, the integrated rescue paths are divided into multiple sub-regions according to the space range, and only the paths in adjacent regions are compared in detail to generate the space-time conflict matrix of all rescue bodies in the emergency rescue process. Wherein, the matrix element Cij represents the conflict intensity of the rescue bodies i and j at the space-time convergence point.

[0052] In step 105, multi-objective real-time path planning is performed on each rescue body according to the space-time conflict matrix, and a set of space-time conflict-free cooperative rescue paths is generated.

[0053] In some embodiments, the multi-objective real-time path planning on each rescue body according to the space-time conflict matrix, and the generation of a set of space-time conflict-free cooperative rescue paths can be implemented in the following steps: According to the space-time conflict matrix, a set of rescue bodies whose rescue paths are affected is identified; The rescue paths of each rescue body in the set of rescue bodies are re-planned to obtain an optimized space-time matrix that eliminates conflicts; Based on the optimized space-time matrix, a set of space-time conflict-free cooperative rescue paths of all rescue bodies is generated.

[0054] In a specific implementation, the identification of the rescue body set affected by the rescue path according to the space-time conflict matrix can be implemented in the following manner: threshold filtering can be performed on the space-time conflict matrix, rescue bodies with a conflict intensity greater than a preset conflict intensity threshold (e.g., 30) are extracted, and a union-find set algorithm is used to merge these rescue bodies into conflict-connected components, each conflict-connected component representing a group of mutually conflicting rescue bodies, and the set of all conflict-connected components is taken as the rescue body set affected by the rescue path. The rescue body set is a group of rescue bodies affected by path conflicts in the rescue process. The rescue body set clearly identifies the objects that need to be adjusted in priority. By focusing on conflict-related rescue bodies, indiscriminate adjustment can be avoided, the path optimization efficiency is improved, and a targeted target for precise obstacle avoidance is provided.

[0055] In a specific implementation, the rescue path re-planning for each rescue body in the rescue body set to obtain the optimized space-time matrix that eliminates conflicts can be implemented in the following manner: an existing path planning optimization (e.g., an improved fast exploration random tree algorithm) can be used to re-plan the paths of rescue bodies in different conflict-connected components in the rescue body set, and an optimized space-time matrix that eliminates conflicts is generated. The fast exploration random tree algorithm can construct a search tree in a three-dimensional space, assign a dedicated height layer (e.g., 300 meters for UAV 1 and 350 meters for UAV 2) to each rescue body and set a vertical safety interval (e.g., 50 meters), and then add a conflict time constraint to the path cost function: if rescue body i and rescue body j conflict at time t, the new path of i must satisfy that the arrival time at the conflict point is not equal to t±Δt (e.g., 5 minutes) and a conflict penalty term. Iterative optimization is performed until all paths satisfy the space-time separation constraint (e.g., the spatial interval is greater than or equal to 150 meters and the time interval is greater than or equal to 8 minutes), thereby generating an optimized space-time matrix that eliminates conflicts. The optimized space-time matrix records the path parameters of each rescue body after re-planning the paths of conflict rescue bodies. The optimized space-time matrix stores the path basic data without conflicts. The space-time separation constraint verified by quantification provides a reliable benchmark for generating the final collaborative path, and ensures that subsequent paths do not have secondary conflicts.

[0056] In a specific implementation, the generation of the set of collaborative rescue paths without conflicts in space-time for all rescue bodies based on the optimized space-time matrix can be implemented in the following manner: the path parameters (e.g., longitude, latitude, height, and time) in the optimized space-time matrix are input into a three-dimensional path rendering engine to generate a set of collaborative rescue paths with a time dimension; a space-time consistency verification algorithm can be used to verify the path feasibility: it checks whether each rescue path satisfies its own rescue constraint condition (e.g., load capacity, endurance), and confirms whether all rescue paths satisfy the space-time separation constraint.

[0057] It should be noted that in the present application, the cooperative rescue path set is the sum of all rescue body paths in the rescue process that reach the disaster site without time and space conflicts. The cooperative rescue path set can guide the efficient cooperative action of the rescue bodies, avoid congestion at the time and space convergence point by integrating the rescue paths of all rescue bodies and ensuring time and space consistency, and thus improve the overall rescue response speed and execution safety.

[0058] In addition, another aspect of the present application, in some embodiments, the present application provides a mountainous highway emergency rescue path planning system, referring to Figure 4 The figure is a structural schematic diagram of a mountainous highway emergency rescue path planning system according to some embodiments of the present application. The mountainous highway emergency rescue path planning system 400 includes a collection module 401, a processing module 402, and an execution module 403, which are described as follows: The collection module 401 is mainly used for positioning the disaster site on the mountainous highway, and then collecting the traffic flow data, meteorological data, and on-site state data of the disaster site. The processing module 402 is mainly used for determining the meteorological grid map of the area near the disaster site according to the meteorological data, determining the time and space evolution trend of geological disasters and secondary accidents at the disaster site based on the on-site state data and the traffic flow data through a pre-constructed disaster chain coupling deduction model, and generating a dynamic risk map near the disaster site by grid algebra superposition of the meteorological grid map and the time and space evolution trend. It should be noted that the processing module 402 is also used to obtain the dynamic position information of each rescue body near the disaster site, and then determine the rescue constraint condition of each rescue body in the emergency rescue process according to all the dynamic position information and the rescue body type of each rescue body. In addition, it should be noted that the processing module 402 is also used for hierarchical scheduling of each rescue body based on the dynamic risk map and all rescue constraint conditions, and then obtaining the time and space conflict matrix of all rescue bodies in the emergency rescue process. The execution module 403 is mainly used for multi-objective real-time path planning of each rescue body according to the time and space conflict matrix, and generating a cooperative rescue path set without time and space conflicts.

[0059] The above-mentioned modules in the mountainous highway emergency rescue path planning system can be realized by software, hardware, and their combinations. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to the above-mentioned modules.

[0060] In addition, in an embodiment, the present application provides a computer device, which can be a server, and an internal structure diagram of the computer device can be as shown in Figure 5 The computer device includes a processor, a memory and a network interface connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The database of the computer device is configured to store data of the mountainous highway emergency rescue path planning method. The network interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a mountainous highway emergency rescue path planning method.

[0061] Those skilled in the art can understand that Figure 5 The structure shown in the above embodiment is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0062] In an embodiment, a computer device is also provided, which includes a memory and a processor. The memory stores a computer program. The processor executes the computer program to implement the steps in the above-mentioned mountainous highway emergency rescue path planning method embodiments.

[0063] In an embodiment, a computer readable storage medium is provided, which stores a computer program. The computer program is executed by a processor to implement the steps in the above-mentioned mountainous highway emergency rescue path planning method embodiments.

[0064] In an embodiment, a computer program product or computer program is provided, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium. The processor executes the computer instructions to cause the computer device to perform the steps in the above-mentioned mountainous highway emergency rescue path planning method embodiments.

[0065] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0066] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, but as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0067] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be pointed out that for those skilled in the art, without departing from the concept of the present application, some modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for planning emergency rescue routes on mountainous highways, characterized in that: The steps include: Locate the disaster site on a mountain highway and collect traffic flow data, meteorological data, and on-site status data at the disaster site; Determining a meteorological grid map of an area near the disaster site based on the meteorological data, determining the spatiotemporal evolution of geological hazards and secondary accidents occurring at the disaster site using a pre-constructed disaster chain coupling deduction model based on the on-site status data and the traffic flow data, performing grid algebraic superposition on the meteorological grid map and the spatiotemporal evolution to generate a dynamic risk map near the disaster site; Obtaining dynamic position information of each rescue body near the disaster site, and then determining rescue constraints of each rescue body during the emergency rescue process based on all dynamic position information and the rescue body type of each rescue body; Based on the dynamic risk map and all rescue constraints, each rescuer is hierarchically scheduled, thereby obtaining a spatiotemporal conflict matrix of all rescuers in the emergency rescue process; Multi-objective real-time path planning is performed on each rescue object according to the time-space conflict matrix to generate a set of collaborative rescue paths without time-space conflict.

2. The method according to claim 1, wherein Determining a meteorological grid map of an area near a disaster site based on the meteorological data specifically includes: The area near the disaster site is divided into grids to obtain a climate analysis area centered on the disaster site; Converting the meteorological data into grid values ​​and filling them into the climate analysis area to generate an initial meteorological grid map; The grid values ​​are dynamically updated according to the current meteorological data synchronized with the initial meteorological grid map to form a meteorological grid map of the area near the disaster site.

3. The method according to claim 1, wherein Determining the spatiotemporal evolution of geological disasters and secondary accidents at the disaster site based on the on-site status data and the traffic flow data through a pre-built disaster chain coupling deduction model specifically includes: extracting surface deformation features near the disaster site from the on-site status data; Constructing a terrain-traffic flow change map of the disaster site based on the surface deformation characteristics and the traffic flow data; The terrain-vehicle flow correlation map is subjected to spatiotemporal deduction by a pre-built disaster chain coupling deduction model to generate a spatiotemporal evolution probability field of the disaster chain at the disaster site; The spatiotemporal evolution probability field is dynamically calibrated in combination with historical disaster accident records to obtain the spatiotemporal evolution trend of geological disasters and secondary accidents at the disaster site.

4. The method according to claim 1, wherein Performing grid algebraic superposition on the meteorological grid map and the spatiotemporal evolution situation to generate a dynamic risk map near the disaster site specifically includes: The meteorological grid map is spatiotemporally aligned with the spatiotemporal evolution situation by a spatiotemporal grid aligner to obtain a spatiotemporal coupling tensor field of multimodal data fusion; Performing grid algebraic operations on the space-time coupling tensor field to obtain all risk grid values ​​in the space-time coupling tensor field at the disaster location; Generate a dynamic risk map near the disaster site based on all risk grid values.

5. The method according to claim 1, wherein The rescue constraint conditions of each rescuer in the emergency rescue process are determined based on all dynamic position information and the rescuer type of each rescuer, specifically including: For each rescuer, the current time point position information is extracted from the dynamic position information of the rescuer according to the state label of the rescuer; Determine the inherent constraints of the rescue body according to the rescue body type of the rescue body; Calculating a time window constraint for the rescuer to arrive at the disaster site based on the location information; The time constraint and the inherent constraint are constrained and integrated by a dynamic constraint adaptation algorithm to obtain the rescue constraint conditions of the rescue body in the emergency rescue process, and then obtain the rescue constraint conditions of each rescue body in the emergency rescue process.

6. The method according to claim 1, wherein Based on the dynamic risk map and all rescue constraints, each rescuer is hierarchically scheduled, and then the spatiotemporal conflict matrix of all rescuers in the emergency rescue process is obtained, which specifically includes: Constructing a three-dimensional scheduling decision space for all rescue bodies through the dynamic risk map and all rescue constraints; Determining a comprehensive rescue path for each rescue body from its location to the disaster site based on the three-dimensional scheduling decision space; Divide the dispatch echelons according to all comprehensive rescue routes and dispatch each rescue entity; The spatiotemporal convergence point of the comprehensive rescue paths of all rescue bodies after dispatch is detected in time and space to generate the spatiotemporal conflict matrix of all rescue bodies in the emergency rescue process.

7. The method according to claim 1, wherein Performing multi-objective real-time path planning for each rescuer based on the spatiotemporal conflict matrix to generate a spatiotemporal conflict-free collaborative rescue path set specifically includes: Identifying a set of rescue bodies whose rescue paths are affected according to the spatiotemporal conflict matrix; Replanning the rescue path for each rescuer in the rescuer set to obtain an optimized space-time matrix that eliminates conflicts; A set of collaborative rescue paths without spatial and temporal conflicts for all rescue bodies is generated based on the optimized spatial and temporal matrix.

8. A mountain highway emergency rescue path planning system, characterized in that: The system includes: The acquisition module is used to locate the disaster site on the mountain highway and collect traffic flow data, meteorological data and on-site status data at the disaster site; a processing module for determining a meteorological grid map of an area near the disaster site based on the meteorological data, determining the spatiotemporal evolution of geological hazards and secondary accidents occurring at the disaster site using a pre-built disaster chain coupling deduction model based on the on-site status data and the traffic flow data, and performing grid algebraic superposition on the meteorological grid map and the spatiotemporal evolution to generate a dynamic risk map near the disaster site; The processing module is used to obtain dynamic position information of each rescue body near the disaster site, and then determine the rescue constraint conditions of each rescue body during the emergency rescue process based on all the dynamic position information and the rescue body type of each rescue body; The processing module is used to perform hierarchical scheduling of each rescuer based on the dynamic risk map and all rescue constraints, thereby obtaining a spatiotemporal conflict matrix of all rescuers during the emergency rescue process; The execution module is used to perform multi-objective real-time path planning for each rescue body according to the time-space conflict matrix to generate a set of collaborative rescue paths without time-space conflict.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the mountain highway emergency rescue path planning method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the mountain highway emergency rescue path planning method according to any one of claims 1 to 7 are implemented.

Images