Multi-dimensional dynamic perception integrated cooperative command and dispatch system

Abstract

The invention belongs to the technical field of emergency command and dispatch management evaluation, and particularly discloses and provides a multi-dimensional dynamic perception integrated cooperative command and dispatch system, which comprises a traffic weight updating module for updating the real-time traffic weight of each link of a whole road network according to the real-time vehicle traffic speed and the real-time ponding depth; the ponding trend prediction module determines a ponding prediction area based on the starting point and the ending point of the scheduling task, and predicts the ponding depth change of each link in the area in a future time period through hydrological deduction; the passing time window judgment module is used for judging a time window for keeping a passable state of each link in a future time period in combination with the accumulated water depth change; and the passing plan generation module is used for generating a scheduling path and a time plan by adopting a time-dependent shortest path search algorithm. According to the invention, the problems that the scheduling instruction is disjointed from the dynamic environment and the rescue deployment is in a passive response state are solved, and the accuracy and response efficiency of emergency command are improved.

Description

Technical Field

[0001] This invention belongs to the field of emergency command, dispatch, management and evaluation technology. Specifically, it relates to an integrated collaborative command and dispatch system with multi-dimensional dynamic perception. Background Technology

[0002] When cities experience flooding, disaster information relies primarily on manual reporting and some remote sensing imagery. This results in slow updates and difficulty in real-time monitoring of the overall extent and evolution of the disaster, leading to indiscriminate deployment of rescue forces. Simultaneously, the command center cannot monitor the location, progress, and remaining capacity of these forces in real time, often resulting in dispatch instructions being out of sync with the actual situation on the ground.

[0003] Existing technologies, such as the multi-level linkage combat unit command and dispatch system disclosed in Chinese invention patent application number 202511619494.6, primarily focus on multi-level linkage and visualized command. This system involves constructing an information database, generating electronic maps, identifying anomalies, establishing a topological network, and visualizing the situation, ultimately selecting areas for dispatch.

[0004] Existing technologies, such as the Chinese invention patent application with application number 202410902832.6, disclose a distributed command and dispatch method and system based on a collaborative integrated platform. This method constructs a collaborative integrated platform that allows local centers to access it and sets timeout takeover rules to achieve backup scheduling and collaborative support for cross-regional resources. This solves the problems of untimely cross-regional collaboration and response.

[0005] The aforementioned existing technologies rely on preset rules for command and dispatch, such as selection associations and timeout takeover. This means dispatch instructions are issued solely based on straight-line distances on maps or administrative affiliations, without considering real-time changes in road network conditions due to flooding, landslides, or traffic control. This can lead to dispatched rescue vehicles and teams failing to arrive on time or even becoming stranded en route, thus compromising the final dispatch effectiveness.

[0006] Furthermore, the aforementioned existing technologies rely primarily on manual reporting and periodic remote sensing imagery for understanding disaster situations. The perception of key trends such as the spread of floodwaters and the shift of risk areas remains discrete and delayed. The inability to make short-term predictions of disaster evolution trends based on real-time meteorological, water flow, and geographic data leaves rescue deployments in a reactive state. Summary of the Invention

[0007] In view of this, in order to solve the above problems, an integrated collaborative command and dispatch system with multi-dimensional dynamic perception is proposed.

[0008] The objective of this invention can be achieved through the following technical solution: This invention provides an integrated collaborative command and dispatch system with multi-dimensional dynamic perception. The system includes: a traffic weight update module, which updates the real-time traffic weight of each link in the entire road network based on the real-time vehicle traffic speed and real-time water depth of each link in the target area.

[0009] The flooding trend prediction module receives a scheduling task that includes the starting point, ending point, latest arrival time, and vehicle type. Based on the scheduling task, it determines the flooding prediction area. Using the current flooding depth of each link in the target area as the initial condition, it predicts the change in flooding depth of each link in the flooding prediction area in the future time period through hydrological extrapolation.

[0010] The passage window determination module, in conjunction with the changes in water depth, determines the time window during which each link will remain passable in the future.

[0011] The passage plan generation module, based on the real-time passage weight, the time window of the passability status, and the scheduling task, uses a time-dependent shortest path search algorithm to generate and output the scheduling path and time plan from the starting point to the destination.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) By combining real-time vehicle traffic speed and real-time water depth, the present invention updates the real-time traffic weight of each link in the entire road network, enabling the command center to accurately grasp the real traffic status of each road segment at the current moment, eliminating the information blind spot caused by dynamic changes in road conditions, and providing a reliable data foundation for the implementation of all subsequent scheduling decisions.

[0013] (2) By delineating the water accumulation prediction area, this invention only performs refined hydrological simulations on the task-related areas, which solves the problem of large calculation volume and long update cycle for the whole-area prediction. It also solves the problem of slow information update and fragmentation caused by the existing reliance on manual reporting and periodic remote sensing images, enabling the command center to quickly obtain the future water accumulation evolution trend of key areas and realize the transformation from passive response to forward-looking perception.

[0014] (3) By determining the time window in which each link remains passable in the future, the present invention transforms the complex hydrological prediction results into clear time constraints that can be directly used for route planning, enabling scheduling decisions to predict and avoid links that may fail due to water accumulation during vehicle travel, thereby improving the efficiency of subsequent route generation.

[0015] (4) This invention uses a time-dependent shortest path search algorithm to plan the path by combining real-time passage weight and passable time window dual dynamic constraints. The generated scheduling path can ensure that the rescue force arrives at the target on time along a reliable path throughout the entire process. This fundamentally solves the problem of the scheduling instructions being out of sync with the dynamic environment and the rescue deployment being in a passive response state, thus ensuring the final scheduling effect. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a schematic diagram of the system module structure connection of the present invention;

[0018] Figure 2 This is a schematic diagram of the real-time access weight update process of the present invention;

[0019] Figure 3 This is a schematic diagram of the real-time traffic capacity index calculation process of the present invention. Detailed Implementation

[0020] 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.

[0021] Currently, in urban flood disaster emergency command, dispatch decisions mainly rely on pre-set rules and human experience. Disaster information acquisition depends on manual reporting and periodic remote sensing imagery. Changes in road traffic conditions are difficult to perceive in real time and dynamically update, resulting in a serious disconnect between dispatch instructions and actual road conditions on site, and a lack of direction in the deployment of rescue forces. At the same time, due to the lack of short-term forecasting capabilities for the spread of floodwater, the command center remains in a passive response state and cannot adjust rescue plans in advance according to future developments.

[0022] To address the aforementioned issues, this invention discloses an integrated collaborative command and dispatch system with multi-dimensional dynamic perception, which effectively improves the accuracy and response efficiency of emergency command.

[0023] Please see details. Figure 1 As shown, Figure 1The present invention provides an integrated collaborative command and dispatch system with multi-dimensional dynamic perception, which includes: a passage weight update module, a water accumulation trend prediction module, a passage time window determination module, and a passage plan generation module.

[0024] In the above, the water accumulation trend prediction module is connected to the passage weight update module and the passage time window determination module, respectively, and the passage plan generation module is connected to the passage weight update module, the water accumulation trend prediction module and the passage time window determination module, respectively.

[0025] The traffic weight update module updates the real-time traffic weight of each link in the entire road network based on the real-time vehicle traffic speed and real-time water depth of each link in the target area.

[0026] Among them, the vehicle speed can be collected in real time by microwave radar or geomagnetic sensors deployed on the side of each link, and the water depth can be collected in real time by pressure water level gauges or ultrasonic water level sensors deployed in the pre-set monitoring points on the road surface of each link. The pre-set monitoring points are located at the lowest point of each link, near the storm drain, and historically flood-prone points, with at least 1 to 2 monitoring points deployed on each link.

[0027] In flood emergency command scenarios, when the water is shallow, traffic speed is mainly affected by traffic flow, similar to normal road conditions. However, as the water deepens, vehicles must slow down, and the actual traffic capacity is determined by both speed and water depth. When the water exceeds the safe wading depth for vehicles, the road is completely blocked. Traditional command systems rely on static electronic maps or historical traffic data, and the dispatch routes they generate are essentially based on the assumption that roads are usable under ideal conditions. This often results in rescue vehicles encountering impassable flooded sections en route, forcing them to turn back or detour, delaying crucial rescue time.

[0028] Therefore, this invention updates real-time traffic weights by combining real-time vehicle speed and real-time water depth to quantify the actual traffic conditions in the city, thereby solving the fundamental problem that the command center is unaware of or has difficulty knowing the actual road conditions.

[0029] Specifically, please see Figure 2 As shown, the specific update steps of the real-time traffic weight include: S1, real-time collection of vehicle traffic speed and water depth data for each link.

[0030] S2. Compare the collected real-time water depth with the preset first water depth threshold and second water depth threshold, wherein the second water depth threshold is greater than the first water depth threshold.

[0031] Understandably, the first water accumulation threshold is preferably set as the average height of the curb height of urban roads. For example, the standard curb height of urban roads is usually 10 to 15 centimeters, and the present invention can preferably set it as 12 centimeters as the first water accumulation threshold.

[0032] The second water accumulation threshold is determined based on the minimum ground clearance and safe wading depth of various emergency rescue vehicles. For example, statistical analysis of the technical parameters of common rescue vehicles such as fire trucks, ambulances, and engineering rescue vehicles shows that their minimum ground clearance ranges from 18 cm to 25 cm, and their maximum safe wading depth ranges from 30 cm to 50 cm. To ensure safety redundancy, the second water accumulation threshold is set at 30 cm, meaning that when the water depth reaches this value, ordinary rescue vehicles can no longer pass safely.

[0033] S3. If the real-time water depth of a certain link is lower than the first water depth threshold, mark the link as the first link, obtain the preset benchmark passage speed of the link, and calculate the ratio of the real-time vehicle passage speed to the benchmark passage speed. Use this ratio as the real-time passage weight of the link.

[0034] Understandably, the baseline speed can be set with reference to urban traffic regulations. For example, for urban expressways, the baseline speed is set at 80 kilometers per hour. For urban arterial roads, the baseline speed is set at 60 kilometers per hour. If the design speed is missing, the average speed of the road segment during periods without water accumulation and congestion (such as 2 to 4 a.m.) is taken as the baseline speed.

[0035] S4. If the real-time water depth of a certain link is between the first water depth threshold and the second water depth threshold, then the link is marked as the second link. The real-time traffic capacity index is calculated based on the real-time vehicle speed and the real-time water depth, and used as the real-time traffic weight.

[0036] Please refer to Figure 3 As shown, in this embodiment, for the road marked as the second link, its capacity index is calculated through the following steps: S41, Calculate the initial capacity index: The ratio of the real-time vehicle speed of the second link to the baseline speed is used as the capacity factor. The real-time water depth of the second link is compared with a second preset water depth threshold to obtain the real-time water depth ratio. The difference between 1 and the water depth ratio is taken as the depth influence factor. The capacity index is multiplied by the depth influence factor to obtain the real-time initial capacity index.

[0037] S42. Monitoring Abnormal Water Level Rise: A time series of water level rises is constructed based on real-time water depth, continuously monitoring the change in water depth between adjacent time points. When it is detected that the increase in water depth at the current time exceeds a preset threshold, that time point is marked as a water depth rise node. If no increase in water depth exceeding the preset threshold is detected at the current time, it is determined that there is no abnormal water level rise, and the initial traffic capacity index is directly output as the final traffic capacity index for that time point, without proceeding with subsequent steps. Furthermore, the sampling frequency of water depth changes can be set to once per minute to ensure that minute-level water level changes can be captured.

[0038] It should be noted that the preset rise change threshold can be set with reference to the short-term heavy rainfall intensity specified in the outdoor drainage design standard. The same applies to subsequent rise change thresholds. For example, it can be set to 2 cm per minute. In specific implementation, the rise change threshold can be corrected according to the road longitudinal slope. The larger the longitudinal slope, the faster the confluence velocity, and the higher the threshold should be.

[0039] S43. Calculate the speed decrease: Obtain the vehicle speed within a preset associated time window before and after the water depth rise node. Based on the speed at the beginning of the associated time window (i.e., the time after the water depth rise node is extended forward by the first or second time), calculate the difference between this speed and the lowest speed within the window. Then divide the obtained difference by the speed at the beginning of the window to obtain the maximum decrease in speed within the associated time window.

[0040] S44. Determine and output the final traffic capacity index: If the maximum drop does not exceed the preset speed drop threshold, directly output the initial traffic capacity index. If the maximum drop exceeds the preset speed drop threshold, mark the water depth rise node as a correction node, and use the ratio of the speed drop threshold to the maximum drop as a correction factor. Multiply the initial traffic capacity index by the correction factor to obtain the corrected traffic capacity index and output it.

[0041] Understandably, the speed reduction threshold is set as follows: Average traffic speeds are collected for multiple consecutive time periods on at least three main urban roads under normal conditions (no water accumulation, no accidents). For each road, the rate of change of traffic speed between adjacent time periods is calculated, and the absolute values ​​of the rates of change for all adjacent time periods are statistically analyzed to obtain the set of speed fluctuation amplitudes for that road under normal conditions. The speed fluctuation amplitude sets of all roads are summarized, and their average is taken as the speed reduction threshold. The length of adjacent time periods can be set to 5 minutes, consistent with the time step of hydrological projections.

[0042] S5. If the real-time water depth of a link is greater than or equal to the second water depth threshold, the real-time passage weight of that link will be set to 0, indicating that the link is completely impassable. The real-time passage weight will be updated every 1 to 5 minutes, with a 1-minute update cycle during emergency response.

[0043] The following supplementary explanations are needed in the specific execution of step S44 above: The associated time window is set according to the following rules: calculate the average rate of change of water depth within the preset time before the water depth rises and calculate the standard deviation of vehicle traffic speed within the preset time after the water depth rises. Considering that the confluence time of urban small watersheds is generally 5 to 10 minutes, the process of water accumulation from generation to formation of significant changes is usually concentrated in this period. Therefore, the preset time is preferably set to 5 minutes, and the average rate of change of water depth and the standard deviation of vehicle traffic speed can be calculated using a sliding window method.

[0044] If the average rate of change of water depth is less than or equal to a preset threshold for increase, and the standard deviation of vehicle speed is less than or equal to a preset threshold for speed fluctuation, then the time of occurrence of the water depth increase node is used as the center, extending forward and backward for a first duration. Otherwise, the time is extended forward and backward for a second duration, respectively, thus obtaining a forward window and a backward window. The second duration is longer than the first duration. The sum of the forward window and the backward window is then used as the preset associated time window. The first duration ranges from 2 to 3 minutes, and the second duration ranges from 8 to 10 minutes. Preferably, the first duration is set to 3 minutes and the second duration is set to 8 minutes. The first and second durations can be dynamically adjusted according to road grade and traffic flow, with the upper limit for main roads and the lower limit for secondary roads; the duration is appropriately extended during peak hours and appropriately shortened during off-peak hours.

[0045] It should be understood that the preset traffic speed fluctuation threshold can be derived from the statistical results of the standard deviation of traffic speed on multiple urban main roads under normal conditions of no water accumulation and no accidents. For example, the normal fluctuation range is generally 2 to 4 kilometers per hour, and the preset traffic speed fluctuation threshold of the present invention can preferably be set to 3 kilometers per hour.

[0046] The water accumulation trend prediction module receives a scheduling task that includes the starting point, the ending point, the latest arrival time, and the vehicle type. Based on the scheduling task, it determines the water accumulation prediction area and uses the current water accumulation depth of each link in the target area as the initial condition. Through hydrological extrapolation, it predicts the changes in water accumulation depth of each link in the water accumulation prediction area in the future period.

[0047] In urban flood disaster emergency command scenarios, conducting high-precision hydrological simulations across the entire area presents challenges due to the large computational load and long update cycles, failing to meet the timeliness requirements of real-time dispatch. However, dispatch decisions should primarily focus on areas directly related to the current task, namely candidate routes and surrounding roads that could serve as alternative routes. Furthermore, focusing solely on the candidate routes themselves may overlook the impact of floodwater spread on surrounding roads, resulting in a lack of predictive data to support subsequent route adjustments.

[0048] Based on this, this step delineates a reasonable flooding prediction area that can cover both the roads related to the task and surrounding related roads that may be affected by flooding.

[0049] Specifically, the process for delineating the flooding prediction area is as follows: After receiving the scheduling task, the flooding trend prediction module first determines the flooding prediction area based on the starting point and ending point in the task. Specifically, using the shortest path algorithm or the K-shortest path algorithm, at least one candidate path connecting the starting point and ending point is generated in the entire road network. This embodiment of the invention defaults to using the shortest path algorithm to generate one candidate path; when alternative solutions are needed, the K-shortest path algorithm is used to generate multiple candidate paths, and the K value can be set to 3-5 according to the importance of the task. Subsequently, all links covered by these candidate paths are extracted to form a first link set. Then, a preset buffer range is expanded outward from each link in the first link set as the center, and all links falling within the buffer are included to form the final flooding prediction area. The preset buffer range can be set according to the road level. For example, for urban arterial roads, the preset buffer range is set to extend 200 meters upstream and downstream of the link. For secondary arterial roads and branch roads, the preset buffer range is set to extend 100 meters upstream and downstream. Furthermore, the aforementioned shortest path algorithm or K shortest path algorithm are existing path planning techniques, and their specific execution process will not be elaborated further.

[0050] It should be noted that by delineating the water accumulation prediction area and conducting refined hydrological simulations only for the task-related areas, the problem of large computational load and long update cycle for overall prediction has been solved. At the same time, the problem of slow and fragmented information updates caused by relying on manual reporting and periodic remote sensing images has also been solved. This enables the command center to quickly obtain the future water accumulation evolution trend of key areas and realize the transformation from passive response to proactive perception.

[0051] Because the extent and depth of floodwater are constantly evolving, issuing instructions based solely on whether a road is passable at the current moment could result in vehicles arriving at a road already submerged. Therefore, this step introduces a time dimension to predict future risks. By using the current floodwater depth as an initial condition and integrating real-time rainfall, topographic elevation data, and underground pipe network data, hydrological extrapolation is used to predict changes in floodwater depth at various points within the flood prediction area over future periods.

[0052] Specifically, the execution process of hydrological extrapolation and prediction is as follows: R1. Divide the water accumulation prediction area into grids. This yields individual grid cells, where the grid size can be set to 10 meters × 10 meters. Each grid cell serves as the basic unit for hydrological calculations, recording its center point coordinates and corresponding elevation data.

[0053] R2. Based on the location of each grid cell, extract the topographic elevation data of that cell from the Digital Elevation Model (DEM). Extract the corresponding underground pipe network information from the urban drainage pipe network GIS data, including the location of pipe network nodes, pipe segment connections, pipe diameter, and slope. Simultaneously, establish the correspondence between grid cells and road links, i.e., which link(s) each grid cell belongs to, so that the grid water depth can be mapped back to the links later.

[0054] R3. The current water depth of each link is used as the initial water depth of the corresponding grid cell. That is, according to the correspondence between links and grid cells, the current water depth of the link is assigned to all grid cells it covers. Simultaneously, real-time meteorological data is accessed, and the current rainfall intensity and future short-term rainfall forecast are distributed to each grid cell according to a preset time step, serving as an external water source input. The preset time step can be exemplified as 5 minutes in this invention, and can be dynamically adjusted according to the latest arrival time of the scheduled task; the shorter the latest arrival time, the smaller the time step value.

[0055] R4. Based on the surface elevation difference between grid cells, the D8 algorithm is used to determine the surface water flow direction of each grid cell, i.e., the flow direction to the cell with the lowest elevation among the eight adjacent grid cells. Simultaneously, based on the underground pipe network connection relationship, the path of water accumulation in each grid cell into the pipe network is determined, i.e., which grid cells correspond to pipe network nodes and how water enters the pipe network through storm drains. Based on this, the water depth change of each grid cell is iteratively calculated according to a preset time step. Within each time step, the following sub-steps are executed sequentially: increase the water volume of the grid cell based on the rainfall input.

[0056] Based on the direction of surface flow, water exceeding the storage capacity of a grid cell is distributed to downstream grid cells.

[0057] Based on the path of the accumulated water flowing into the pipe network, the water entering the pipe network is transmitted along the pipeline to the downstream node and discharged from the outlet.

[0058] Update the water depth of each grid cell and pipe network node as the initial value for the next time step. Repeat the above iterations until the simulation of all preset time steps is completed, such as the simulation of the next 3 hours.

[0059] R5. Based on the correspondence between each link and grid cell, extract the water depth of the grid cells covered by the link at the end of each time step. By taking the maximum value of the water depth of these grid cells, obtain the predicted water depth of the link at that time step. Arrange the predicted values ​​of all time steps in chronological order to generate a sequence of water depth changes for the link in future periods.

[0060] During the hydrological extrapolation and prediction process from steps R1 to R5, continuous monitoring of the real-time traffic weight of each link within the current waterlogging prediction area is also included. When a link's traffic weight is detected to be lower than a preset weight threshold at the current moment, that link is marked as a low-weight link. The weight threshold can be configured according to the road level; for example, for main roads, the weight threshold can be set to 0.3, and for secondary roads, the weight threshold can be set to 0.2. Subsequently, the low-weight link and links within its adjacent preset range are expanded into the current waterlogging prediction area to form an updated waterlogging prediction area. Hydrological extrapolation is immediately initiated for the newly added area to update its future waterlogging depth change sequence. The value of the adjacent preset range can be consistent with the preset buffer range, for example, extending 200 meters upstream and downstream of the low-weight link.

[0061] This invention provides a way to dynamically update the water accumulation prediction area based on real-time traffic weights. This ensures that when road conditions suddenly worsen, the relevant areas can be included in the prediction range and the water accumulation data can be updated in a timely manner. Furthermore, by combining hydrological extrapolation with the dynamic updating of the water accumulation prediction area, it provides data support for the determination of subsequent traffic time windows and time-varying path planning, thus ensuring the consistency between dispatch instructions and the dynamic environment.

[0062] The passage window determination module, in conjunction with the changes in water depth, determines the time window during which each link will remain passable in the future.

[0063] Since the water depth sequence obtained from hydrological projection is a set of continuous time and depth data, and the path planning algorithm is constrained by when it is passable and when it is not, a passability window determination module is set up to output the time window in which each link remains passable in the future period, thereby transforming the continuous water depth data into discrete passability time windows, which facilitates the execution of the subsequent path planning algorithm.

[0064] Specifically, after obtaining the sequence of water depth changes for each link in the future time period, the passage window determination module determines the time window for each link to remain passable in the future time period according to the following steps: Based on the vehicle type marked in the scheduling task, it queries the vehicle parameter configuration table for the corresponding wading depth threshold for that vehicle type. For example, for ordinary sedans, the wading depth threshold is 20 cm. For SUVs, the wading depth threshold is 30 cm. For special rescue vehicles such as fire trucks and ambulances, the wading depth threshold is 40 cm.

[0065] For each link, the water depth value at each time step in its water depth change sequence is compared with the obtained vehicle wading depth threshold. For each time step, it is determined whether the water depth at that step exceeds the vehicle wading depth threshold.

[0066] Scan the entire water depth variation sequence to identify periods where the water depth at all consecutive time steps does not exceed the wading depth. Record the start and end times of these periods as the time windows during which the link remains passable for that vehicle type. If a link has multiple discontinuous passable periods, they are recorded as separate time windows. If the water depth at all time steps in the entire water depth variation sequence does not exceed the threshold, the link remains passable throughout the entire prediction period. If the water depth at all time steps exceeds the vehicle wading depth threshold, the link is completely impassable during the prediction period.

[0067] This module transforms complex hydrological forecasts into explicit time constraints that can be directly used for route planning by determining the time window during which each link will remain passable in the future. This enables scheduling decisions to anticipate and avoid links that may fail due to deepening water during vehicle travel, thereby improving the efficiency of subsequent route generation.

[0068] The passage plan generation module, based on the real-time passage weight, the time window of the passability status, and the scheduling task, uses a time-dependent shortest path search algorithm to generate and output the scheduling path and time plan from the starting point to the destination.

[0069] Specifically, a time-dependent shortest path search algorithm is used to generate the scheduling path and time plan. The specific execution steps are as follows: the start and end points in the scheduling task are mapped to corresponding nodes in the entire road network, serving as the starting and target nodes for the path search. Road network nodes typically correspond to road intersections or road endpoints, while links are road segments connecting two nodes.

[0070] Set the earliest arrival time of the starting node to the task start time, i.e., the time when the vehicle departs from the starting point, and add the starting node to the set of nodes to be expanded (usually implemented using a priority queue). At the same time, initialize an extremely large earliest arrival time (such as infinity) for all other nodes, indicating that they have not yet arrived.

[0071] Select the node with the smallest and earliest arrival time from the set of nodes to be expanded as the current node, and traverse all outgoing links starting from the current node.

[0072] For each outgoing link, perform the following sub-step: determine the time to enter the link based on the earliest arrival time of the current node.

[0073] Based on the passage weight at that moment, calculate the passage time required to pass through the link, add the earliest arrival time of the current node to the passage time required to pass through the link, and obtain the candidate arrival time of the downstream node.

[0074] Determine whether the candidate arrival time is within the passable time window for the task vehicle type of the link.

[0075] If the candidate arrival time is within the time window and is less than the earliest arrival time currently recorded by the downstream node, then update the earliest arrival time of the downstream node to the candidate arrival time, record the current link as the predecessor link, and add the downstream node to the set of nodes to be expanded.

[0076] Repeat the above expansion process until the node taken from the set of nodes to be expanded is the target node, which indicates that the shortest feasible path from the starting point to the destination has been found. Finally, based on the preceding link information recorded by the target node, backtrack level by level to obtain the complete scheduling path from the starting point to the destination. At the same time, generate the corresponding time plan based on the earliest arrival time of each node, that is, the estimated time when the vehicle arrives at each key node on the path.

[0077] It should be added that the required travel time through this link is calculated as follows: Obtain the baseline travel speed and link length, denoted as... and At the same time, the passage weight at the moment of entry into this link is recorded as The real-time passage weight is a value between 0 and 1, used to characterize the passage efficiency of the link at the current moment. When the weight is 1, it means that the passage is completely smooth (reaching the baseline speed), and when the weight approaches 0, it means that passage is difficult or completely interrupted.

[0078] Calculate travel time , When the passage weight is 1, the passage time equals the link length divided by the base speed, which is the ideal passage time. When the passage weight decreases, the actual passage speed... The weight is reduced accordingly, and the travel time is extended proportionally. If the weight is 0, it means that the link is completely impassable, and the travel time approaches infinity. The algorithm will automatically exclude this link during path search.

[0079] Traditional shortest path algorithms (such as Dijkstra's algorithm) assume that link travel time is static and constant. However, in flood disaster scenarios, road conditions change dynamically over time. The travel time required for vehicles to enter the same link at different times may vary significantly, and the link may only be passable during specific time periods. Static algorithms cannot handle this time-dependent characteristic, causing the generated path to become invalid due to changes in road conditions during actual execution. Therefore, this invention employs a time-dependent shortest path search algorithm and uses the passable time window as a hard constraint to ensure that the planned path meets the passable conditions throughout its entirety.

[0080] In this embodiment, real-time traffic weights ensure that route planning is based on current road conditions, and the passable time window ensures that the route remains passable throughout the actual time period for vehicle travel. During the expansion process, the algorithm dynamically checks the feasibility of each link, ensuring that the final generated route is not only spatially connected but also temporally continuous and feasible. The dispatch route and time plan output by this algorithm can be directly sent to the rescue vehicle terminal, providing drivers with accurate navigation guidance and estimated arrival times. This solves the problem of traditional dispatch instructions being disconnected from the dynamic environment, ensuring the final dispatch effect.

[0081] The above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described, or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, and all such modifications and additions should fall within the protection scope of the present invention.

Claims

1. A multi-dimensional dynamic perception integrated collaborative command and dispatch system, characterized in that, The system includes: The traffic weight update module updates the real-time traffic weight of each link in the entire road network based on the real-time vehicle traffic speed and real-time water depth of each link in the target area. The water accumulation trend prediction module receives a scheduling task containing the origin, destination, latest arrival time and vehicle type, and determines the water accumulation prediction area based on the scheduling task. Using the current water accumulation depth of each link in the target area as the initial condition, it predicts the change of water accumulation depth of each link in the water accumulation prediction area in the future time period through hydrological extrapolation. The passage window determination module, in conjunction with the changes in water depth, determines the time window during which each link will remain passable in the future period. The passage plan generation module, based on the real-time passage weight, the time window of the passability status, and the scheduling task, uses a time-dependent shortest path search algorithm to generate and output the scheduling path and time plan from the starting point to the destination.

2. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 1, characterized in that: The specific steps for updating the real-time access weight include: If the real-time water depth of a certain link is lower than the preset first water depth threshold, the link is recorded as the first link, the preset benchmark passage speed corresponding to the first link is obtained, the ratio of the real-time vehicle passage speed of the first link to the benchmark passage speed is calculated, and the ratio is used as the real-time passage weight of the first link. If the real-time water depth of a certain link is between a preset first threshold and a second water depth, the link is recorded as the second link. Based on the real-time vehicle speed of the second link and the real-time water depth, a real-time traffic capacity index is calculated and used as the real-time traffic weight of the link, wherein the second water depth is greater than the first water depth. If the real-time water accumulation depth of a certain link is greater than or equal to the preset second water accumulation threshold, then the passage weight of that link is assigned to 0.

3. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 2, characterized in that: The specific calculation process for the real-time traffic capacity index includes: The ratio of the real-time vehicle speed of the second link to the benchmark speed is used as the real-time traffic capacity factor. At the same time, the real-time water depth of the second link is compared with the second preset water depth threshold to obtain the real-time water depth ratio. The difference between 1 and the water depth ratio is taken as the depth influence factor. The real-time traffic capacity factor and the real-time depth influence factor are multiplied to obtain the real-time initial traffic capacity index. Based on real-time water depth, a water depth sequence is constructed and continuously monitored. When it is detected that the increase in water depth at the current moment compared to the previous moment exceeds a preset threshold for increase, the current moment is marked as a water depth increase node. Otherwise, the initial traffic capacity index at that moment is output as the final traffic capacity index at that moment. Obtain the vehicle speed within a preset associated time window before and after the water depth rises, and calculate the maximum decrease in speed within the associated time window. If the maximum decrease does not exceed the preset speed decrease threshold, the initial traffic capacity index of the water depth rise node is directly output as its final traffic capacity index. If the maximum drop exceeds the preset speed drop threshold, the ratio of the speed drop threshold to the maximum drop is used as a correction factor. The initial traffic capacity index of the water depth rise node is multiplied by the correction factor to obtain the corrected traffic capacity index, which is then output.

4. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 3, characterized in that: The associated time window is set according to the following rules: Calculate the average rate of change of water depth within a preset time period before the water depth rises and calculate the standard deviation of vehicle speed within a preset time period after the water depth rises. If the average rate of change of water depth is less than or equal to the preset threshold for increase, and the standard deviation of vehicle speed is less than or equal to the preset threshold for speed fluctuation, then the first duration is extended forward and backward, centered on the moment when the water depth rises. Otherwise, the second duration is extended forward and backward, serving as a forward window and a backward window, with the second duration being greater than the first duration. The sum of the forward window and the backward window is used as the preset associated time window.

5. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 1, characterized in that: The specific method for determining the predicted flooding area is as follows: Based on the starting point and the ending point, at least one candidate path is determined in the entire road network, and all links covered by the candidate path are extracted as a first link set. The first set of links and the links within its adjacent preset range are collectively designated as the water accumulation prediction area.

6. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 1, characterized in that: During the prediction process, the flooding trend prediction module also performs the following dynamic adjustment steps: Obtain the real-time traffic weight of each link within the flood prediction area; When the passage weight of a link is lower than the preset weight threshold at the current moment, the link is marked as a low-weight link; The low-weighted link and its adjacent links within a preset range are expanded into the water accumulation prediction area to form an updated water accumulation prediction area.

7. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 1, characterized in that: The specific steps for performing the hydrological simulation are as follows: The water accumulation prediction area is divided into grids to obtain each grid cell; Based on the location of each grid cell, each cell is assigned terrain elevation data, underground pipeline data, and corresponding links; The current water depth of each link is used as the initial water depth of the corresponding grid cell, and real-time rainfall data is distributed to each grid cell according to a preset time step. The direction of surface water flow is determined based on the surface elevation between grid cells. At the same time, the path of water flowing into the underground pipe network is determined based on the connection relationship of the underground pipe network. The water depth of each grid cell is iteratively calculated according to the preset time step. Based on the water depth of the grid cells covered by each link at each time step, a sequence of water depth changes for that link is generated.

8. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 7, characterized in that: The specific steps for determining the time window for each link to remain passable in the future period are as follows: Based on the vehicle type marked in the scheduling task, obtain the corresponding vehicle wading depth threshold; The water depth at each time step in the water depth change sequence of each link is compared with the wading depth threshold. The time period in which the water depth does not exceed the wading depth threshold for all consecutive time steps is marked as the time window in which the link remains passable for that vehicle type.

9. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 1, characterized in that: The specific steps for generating and outputting the scheduling path and time plan from the starting point to the destination are as follows: The starting point and ending point in the scheduling task are mapped to corresponding nodes in the entire road network, serving as the starting node and the target node, respectively. Starting from the starting node, initialize the earliest arrival time of the starting node to the task start time, and add the starting node to the set of nodes to be expanded; Select the node with the smallest and earliest arrival time from the set of nodes to be expanded as the current node, and traverse all outgoing links starting from the current node; For each outgoing link, perform the following sub-steps: The time to enter the link is determined based on the earliest arrival time of the current node; Based on the passage weight at that moment, calculate the passage time required to pass through the link, add the earliest arrival time of the current node to the passage time required to pass through the link, and obtain the candidate arrival time of the downstream node. Determine whether the candidate arrival time is within the passable status time window of the link for the task vehicle type; If the candidate arrival time is within the time window and is less than the earliest arrival time currently recorded by the downstream node, then update the earliest arrival time of the downstream node to the candidate arrival time, record the current link as the predecessor link, and add the downstream node to the set of nodes to be expanded. Repeat the above sub-steps until the node taken from the set of nodes to be expanded is the target node. Based on the recorded predecessor links, backtrack to obtain the scheduling path from the starting point to the end point and the corresponding time plan.

10. The integrated collaborative command and dispatch system with multi-dimensional dynamic perception as described in claim 9, characterized in that: The calculation method for the required travel time through this link is as follows: Obtain the baseline transit speed and link length of the link, denoted as . and At the same time, the passage weight at the moment of entry into this link is recorded as ; Calculate travel time , .

Images