Active flood monitoring method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of flood disaster emergency response, in particular to an active flood monitoring method and device, electronic equipment and a storage medium. Comprising the following steps: on the basis of rainfall forecast data and a preset cloud base database, calculating by using a preset flood model to obtain a space-time risk tensor; constructing an objective function based on the space-time risk tensor, the regional road network and the waterlogging-prone point location distribution, and determining an optimal path of each mobile platform based on the constraint condition and the objective function so as to control each mobile platform to move to the risk monitoring point; determining the ponding area of each risk monitoring point by using a preset target detection algorithm, and determining the ponding depth of each risk monitoring point in combination with a preset digital elevation base map; and according to the ponding depth of each risk monitoring point, a proper monitoring strategy is automatically selected, and flood monitoring is carried out on each risk monitoring point. Therefore, the problems of weak perception capability, limited coverage range, untimely operation response and the like of a single monitoring means in related technologies are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of flood disaster emergency response, and in particular to an active flood monitoring method, device, electronic equipment and storage medium. Background Art

[0002] With climate change and advancing urbanization, urban rainstorms and floods are becoming more frequent, posing a serious threat to life and property. Due to a lack of comprehensive monitoring of the complex urban environment and the evolution of flood disasters, accurate prediction and targeted management of urban flood disasters are impossible. Consequently, redundant deployment, which consumes significant manpower and resources, is necessary to minimize the impact of floods. Therefore, scientific flood monitoring to guide effective disaster prevention and mitigation is of great practical significance.

[0003] Among the related technologies, flood monitoring strategies generally include three categories: flood monitoring based on satellite remote sensing, flood monitoring based on drone platforms, and flood monitoring based on the Internet of Things.

[0004] However, the monitoring methods of the above-mentioned monitoring strategies are mostly based on historical flood disaster experience to determine the monitoring implementation methods. The monitoring methods are single, the coverage is limited, and the operational response is not timely, which urgently needs to be solved. Summary of the Invention

[0005] The present invention provides an active flood monitoring method, device, electronic device and storage medium to solve the problems of weak perception ability, limited coverage and untimely operation response of a single monitoring means in related technologies, thereby improving the efficiency and accuracy of flood disaster prevention.

[0006] To achieve the above objectives, a first embodiment of the present invention provides an active flood monitoring method, comprising the following steps: Obtaining first rainfall forecast data and second rainfall forecast data for the area to be monitored, and calculating a spatiotemporal risk tensor using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and a preset cloud-based database; Based on the spatiotemporal risk tensor, the target area road network, and the historical distribution of flood-prone points, an objective function is constructed, and based on preset constraints and the objective function, an optimal path for each preset mobile platform is determined; based on the optimal path for each preset mobile platform, each preset mobile platform is controlled to move from an initial stop point to a preset risk monitoring point; Based on a preset target detection algorithm, using each preset mobile platform to determine the water accumulation area of ​​each preset risk monitoring point, and based on the water accumulation area of ​​each preset risk monitoring point and a preset digital elevation base map, determine the water accumulation depth of each preset risk monitoring point; According to the water accumulation depth of each preset risk monitoring point, the monitoring strategy of each preset risk monitoring point is determined, and based on the monitoring strategy of each preset risk monitoring point, flood monitoring is performed on each preset risk monitoring point to obtain flood monitoring results.

[0007] According to one embodiment of the present invention, determining the monitoring strategy for each preset risk monitoring point according to the water accumulation depth of each preset risk monitoring point includes: Based on each of the preset risk monitoring points, at least one first target risk monitoring point is selected whose water accumulation depth is less than a preset depth value; Control the preset mobile platform corresponding to at least one first target risk monitoring point to move to the target point corresponding to each first target risk monitoring point, and use a ranging water level meter to measure the water depth of the target point corresponding to each first target risk monitoring point.

[0008] According to an embodiment of the present invention, determining the monitoring strategy for each preset risk monitoring point according to the water accumulation depth of each preset risk monitoring point further includes: Based on each of the preset risk monitoring points, at least one second target risk monitoring point is selected whose water accumulation depth is greater than or equal to a preset depth value; Control the preset mobile platform corresponding to at least one second target risk monitoring point to call a preset bionic robot carrying a pressure water level gauge to move to the target point corresponding to each second target risk monitoring point, and use the pressure water level gauge to measure the water depth at the target point corresponding to each second target risk monitoring point.

[0009] According to one embodiment of the present invention, the objective function is:

[0010] in, To monitor the total cost, is the number of the preset mobile platforms, 、 、 are target weights, is the balance factor, is the path distance cost, For the risk cost, is the energy cost, Cost of duplicate work for multiple machines.

[0011] According to one embodiment of the present invention, before calculating the spatiotemporal risk tensor using the preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and the preset cloud-based base database, the method further includes: Acquire optical image data and point cloud data of the target scene area; Detecting image elements of a target object in the optical image data using the preset target detection algorithm to obtain processed optical image data, geo-registering the point cloud data with the optical image data using a K-nearest neighbor sampling and sparse graph convolution algorithm, and assigning the geographic coordinates of the point cloud data to the target object in the optical image data to obtain registered optical image data and registered point cloud data; Mapping the registered optical image data to an image feature space using a first preset neural network, and mapping the registered point cloud data to a point cloud feature space using a second preset neural network; Using a preset feature mapping network, aligning the image feature space and the point cloud feature space to obtain a target feature vector, and training a target segmentation model based on the target feature vector; The target segmentation model is used to interpret the geometric boundary information of the target object, and the geographic coordinate information and geometric size attributes of the target object are extracted, and the geometric boundary information, the geographic coordinate information and the geometric size attributes are updated to the preset cloud-based database.

[0012] The active flood monitoring method proposed in an embodiment of the present invention calculates a spatiotemporal risk tensor using a preset flood model based on rainfall forecast data and a preset cloud-based database. An objective function is constructed based on the spatiotemporal risk tensor, the regional road network, and the distribution of flood-prone points. The optimal path for each mobile platform is determined based on the constraints and the objective function to control the movement of each mobile platform to a preset risk monitoring point. A preset target detection algorithm is used to determine the accumulated water area at each risk monitoring point, and then, combined with a preset digital elevation base map, the accumulated water depth at each risk monitoring point is determined. Based on the accumulated water depth at each risk monitoring point, an appropriate monitoring strategy is automatically selected to conduct flood monitoring for each risk monitoring point. This solves the problems of weak perception capabilities, limited coverage, and untimely operational responses associated with single monitoring methods in related technologies, thereby improving the efficiency and accuracy of flood disaster prevention.

[0013] To achieve the above objectives, a second embodiment of the present invention provides an active flood monitoring device, comprising: an acquisition module, configured to obtain first rainfall forecast data and second rainfall forecast data for the area to be monitored, and calculate a spatiotemporal risk tensor using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and a preset cloud-based database; a control module, configured to construct an objective function based on the spatiotemporal risk tensor, the target area road network, and the historical distribution of flood-prone points, determine an optimal path for each preset mobile platform based on preset constraints and the objective function, and control each preset mobile platform to move from an initial stop point to a preset risk monitoring point based on the optimal path of each preset mobile platform; a determination module, configured to determine the accumulated water area of ​​each preset risk monitoring point using each preset mobile platform based on a preset target detection algorithm, and determine the accumulated water depth of each preset risk monitoring point based on the accumulated water area of ​​each preset risk monitoring point and a preset digital elevation base map; The monitoring module is used to determine the monitoring strategy of each preset risk monitoring point according to the water accumulation depth of each preset risk monitoring point, and based on the monitoring strategy of each preset risk monitoring point, perform flood monitoring on each preset risk monitoring point to obtain flood monitoring results.

[0014] According to one embodiment of the present invention, the monitoring module is specifically configured to: Based on each of the preset risk monitoring points, at least one first target risk monitoring point is selected whose water accumulation depth is less than a preset depth value; Control the preset mobile platform corresponding to at least one first target risk monitoring point to move to the target point corresponding to each first target risk monitoring point, and use a ranging water level meter to measure the water depth of the target point corresponding to each first target risk monitoring point.

[0015] According to one embodiment of the present invention, the monitoring module is further configured to: Based on each of the preset risk monitoring points, at least one second target risk monitoring point is selected whose water accumulation depth is greater than or equal to a preset depth value; Control the preset mobile platform corresponding to at least one second target risk monitoring point to call a preset bionic robot carrying a pressure water level gauge to move to the target point corresponding to each second target risk monitoring point, and use the pressure water level gauge to measure the water depth at the target point corresponding to each second target risk monitoring point.

[0016] According to one embodiment of the present invention, the objective function is:

[0017] in, To monitor the total cost, is the number of the preset mobile platforms, 、 、 are target weights, is the balance factor, is the path distance cost, For the risk cost, is the energy cost, Cost of duplicate work for multiple machines.

[0018] According to one embodiment of the present invention, before calculating the spatiotemporal risk tensor using the preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and the preset cloud-based base database, the obtaining module is further configured to: Acquire optical image data and point cloud data of the target scene area; Detecting image elements of a target object in the optical image data using the preset target detection algorithm to obtain processed optical image data, geo-registering the point cloud data with the optical image data using a K-nearest neighbor sampling and sparse graph convolution algorithm, and assigning the geographic coordinates of the point cloud data to the target object in the optical image data to obtain registered optical image data and registered point cloud data; Mapping the registered optical image data to an image feature space using a first preset neural network, and mapping the registered point cloud data to a point cloud feature space using a second preset neural network; Using a preset feature mapping network, aligning the image feature space and the point cloud feature space to obtain a target feature vector, and training a target segmentation model based on the target feature vector; The target segmentation model is used to interpret the geometric boundary information of the target object, and the geographic coordinate information and geometric size attributes of the target object are extracted, and the geometric boundary information, the geographic coordinate information and the geometric size attributes are updated to the preset cloud-based database.

[0019] The active flood monitoring device proposed in an embodiment of the present invention calculates a spatiotemporal risk tensor using a preset flood model based on rainfall forecast data and a preset cloud-based database. An objective function is constructed based on the spatiotemporal risk tensor, the road network distribution, and the distribution of flood-prone points. The optimal path for each mobile platform is determined based on the constraints and the objective function to control each mobile platform's movement to a preset risk monitoring point. A preset target detection algorithm is used to determine the accumulated water area at each risk monitoring point, and then, combined with a preset digital elevation base map, the accumulated water depth at each risk monitoring point is determined. Based on the accumulated water depth at each risk monitoring point, an appropriate monitoring strategy is automatically selected to conduct flood monitoring for each risk monitoring point. This solves the problems of weak perception capabilities, limited coverage, and untimely operational responses associated with single monitoring methods in related technologies, thereby improving the efficiency and accuracy of flood disaster prevention.

[0020] To achieve the above-mentioned objectives, the third aspect of the present invention proposes an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the active flood monitoring method as described in the above-mentioned embodiment.

[0021] To achieve the above objectives, a fourth embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the active flood monitoring method as described in the above embodiments.

[0022] To achieve the above objectives, a fifth embodiment of the present invention provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it is used to implement the active flood monitoring method as described in the above embodiment.

[0023] Additional aspects and advantages of the present invention will be set forth in part in the description which follows and, in part, will be obvious from the description which follows, or may be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which: Figure 1 This is a flow chart of an active flood monitoring method provided according to an embodiment of the present invention; Figure 2 is a schematic diagram of a multimodal data processing flow according to an embodiment of the present invention; Figure 3 is a block diagram of a preset mobile platform according to one embodiment of the present invention; Figure 4 is a flowchart of a dynamic path planning algorithm for a preset mobile platform according to one embodiment of the present invention; Figure 5 A block diagram of an active flood monitoring device according to an embodiment of the present invention; Figure 6 A schematic structural diagram of an electronic device provided according to an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present invention, but are not to be construed as limiting the present invention.

[0026] The following describes the active flood monitoring method, device, electronic device and storage medium according to embodiments of the present invention with reference to the accompanying drawings. First, the active flood monitoring method according to embodiments of the present invention will be described with reference to the accompanying drawings.

[0027] Figure 1 4 is a flow chart of an active flood monitoring method according to an embodiment of the present invention.

[0028] For example, Figure 1 As shown, the active flood monitoring method includes the following steps: In step S101, first rainfall forecast data and second rainfall forecast data for the area to be monitored are obtained, and a spatiotemporal risk tensor is calculated using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data and a preset cloud-based database.

[0029] It should be understood that the first rainfall forecast data refers to the 0-2 hour short-term rainfall forecast based on statistical extrapolation of weather radar data; the second rainfall forecast data refers to the 2-6 hour short-term rainfall forecast based on numerical model calculations. The pre-configured cloud-based database is a real-time dynamic database built on a cloud platform. It is used to store and manage the multimodal perception data required for urban flood modeling, including the geometric boundaries of urban infrastructure (such as bridge and culvert boundaries, road boundaries), geographic coordinates (including elevation), and geometric dimension attributes; the latest environmental perception data (such as water depth and risk point information) collected and wirelessly transmitted in real time by sensors on mobile platforms (such as lidar and motion cameras); and static geographic information such as high-resolution digital elevation basemaps and historical distribution of flood-prone areas. The spatiotemporal risk tensor describes the likelihood of flooding and its potential severity within a specific time and space range.

[0030] Specifically, in order to monitor and warn of flood risks in a specific area, the first and second rainfall forecast data for the area can be obtained. These data are usually provided by the meteorological department and contain forecast information on rainfall conditions in the future. Then, combined with the preset cloud-based database, the first and second rainfall forecast data can be used to drive the preset flood model. Through comprehensive analysis and calculation, a spatiotemporal risk tensor reflecting the flooding and traffic obstruction of the road section can be obtained. , the spatiotemporal risk tensor can represent each path node in the next 6 hours At different times The predicted risk magnitude is of great reference value for formulating countermeasures and conducting emergency responses.

[0031]

[0032] in, Path node At the moment The depth of water accumulation can be calculated by the preset flood model; The median value of the critical water depth for the preset mobile platform to stagnate is generally taken as 0.3 meters. That is, when the water depth on the path exceeds 0.3 meters, the preset mobile platform (the mobile platform equipped with the sensor) cannot pass; It is the preset speed attenuation coefficient of the mobile platform, which is generally taken as 4.

[0033] For ease of understanding, the following details how to obtain the preset cloud-based database.

[0034] In some embodiments, before calculating the spatiotemporal risk tensor using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and the preset cloud-based base database, the method further includes: obtaining optical image data and point cloud data of the target scene area; detecting the image elements of the target object in the optical image data using a preset target detection algorithm to obtain processed optical image data; and geo-registering the point cloud data with the optical image data using a K-nearest neighbor sampling and sparse graph convolution algorithm, and assigning the geographic coordinates of the point cloud data to the target object in the optical image data to obtain the registered optical image data. The method comprises the following steps: first, a preset neural network is used to map the registered optical image data to the image feature space, and second, a preset neural network is used to map the registered point cloud data to the point cloud feature space; a preset feature mapping network is used to align the image feature space and the point cloud feature space to obtain the target feature vector, and a target segmentation model is obtained based on the target feature vector training; the target segmentation model is used to interpret the geometric boundary information of the target object, and the geographic coordinate information and geometric size attributes of the target object are extracted, and the geometric boundary information, geographic coordinate information and geometric size attributes are updated to the preset cloud-based base database.

[0035] Specifically, if Figure 2As shown, embodiments of the present invention can utilize relevant technical means to acquire optical image data and point cloud data of a target scene area. For example, optical image data (i.e., high-resolution surface images) can be acquired using RGB (Red, Green, Blue) or near-infrared bands using a sports camera, an optical camera mounted on a drone, or satellite remote sensing equipment. Furthermore, a lidar (LiDAR) can be used for three-dimensional spatial scanning to generate point cloud data with centimeter-level accuracy. After obtaining the optical image data and point cloud data, a preset target detection algorithm (such as the YOLOv5 (You Only Look Once version 5) algorithm) can be utilized. The YOLOv5 network serves as the backbone network for image feature extraction. Through its unique Focus structure (for efficiently extracting initial image features) and CSP (CrossStage Partial Network) module, multi-scale feature fusion of high-resolution optical images is achieved. This process not only improves processing speed but also enhances the network's ability to capture details. As a result, the YOLOv5 network can quickly and automatically detect and annotate target objects in optical image data, namely various underlying features, including bridge and culvert boundaries, road boundaries, stormwater inlets, and river outlets, thereby generating processed optical image data. K-nearest neighbor sampling and sparse graph convolution algorithms can also be used to encode and annotate the geometric features of target objects from point cloud data, thereby generating processed point cloud data. The processed optical image data and processed point cloud data are then geo-referenced. This involves aligning the data from different sensors (optical image and point cloud data) in a geographic coordinate system to accurately locate the target to be analyzed, making the two data types spatially comparable.

[0036] Next, a first preset neural network (such as a convolutional neural network) is used to encode the registered optical image data into the image feature space, and a second preset neural network (such as the neural network shown in the figure) is used to encode the registered point cloud data into the point cloud feature space. Then, a feature mapping network (i.e., a preset feature mapping network) based on a lightweight transformer network (Transformer) is constructed, and the preset feature mapping network is used to align the image feature space and the point cloud feature space to form a feature vector based on a unified latent space representation (i.e., a target feature vector). Thus, a target segmentation model based on a unified latent space representation can be trained based on the target feature vector. This model can quickly interpret the geometric boundaries of the target object and extract information such as the target object's geographic coordinates (including elevation) and geometric size attributes. Finally, through wireless transmission technology, the latest identified information parameters are synchronously updated to the preset cloud-based base database to achieve real-time data update and storage.

[0037] In step S102, an objective function is constructed based on the spatiotemporal risk tensor, the target area road network, and the historical distribution of flood-prone points. Based on the preset constraints and the objective function, the optimal path of each preset mobile platform is determined. Based on the optimal path of each preset mobile platform, each preset mobile platform is controlled to move from the initial stop point to the preset risk monitoring point.

[0038] It's understood that the target area road network refers to the topological structure data of the road network within the monitored area, including road nodes (such as intersections and key facility locations), road section connectivity (whether the road is passable), and dynamic traffic capacity (such as the actual traffic status after waterlogging). The historical distribution of flood-prone locations refers to areas with high incidence of flooding, determined through historical monitoring data or disaster records.

[0039] Among them, Figure 3 As shown, the pre-set mobile platform refers to a type of autonomous monitoring vehicle (e.g., a vehicle-based intelligent monitoring platform) that integrates multimodal sensors and an intelligent decision-making system. In an embodiment of the present invention, the pre-set mobile platform is equipped with monitoring equipment such as lidar, motion cameras, drones, and liquid level gauges (i.e., ranging and pressure water level gauges). Furthermore, the pre-set mobile platform also integrates a small bionic robot and a microcomputer system (including memory, processor, and communication interface). The microcomputer system is pre-installed with map resources such as an autonomous driving electronic map, a high-resolution digital elevation basemap, and the distribution of historical flood-prone areas in the urban area. These resources assist the pre-set mobile platform in dynamic path planning, flood risk identification, terrain analysis, and rapid collection and intelligent processing of water-related element information. Pre-set constraints are set based on the objective function and primarily constrain the access payload, starting and ending locations, and multi-sensor coordination of each pre-set mobile platform. These constraints will be elaborated upon later.

[0040] Specifically, after obtaining the spatiotemporal risk tensor, an objective function can be constructed based on the spatiotemporal risk tensor, the target area road network, and the historical distribution of flood-prone points. This objective function is then calculated based on the path distance cost, risk risk cost, and energy efficiency of multiple pre-set mobile platforms. Combined with pre-set constraints, the dynamic flood optimization swarm algorithm, based on reinforcement learning and the objective function, enables swarm collaboration among the pre-set mobile platforms, thereby obtaining the optimal path for multiple pre-set mobile platforms from their initial stop to the pre-set risk monitoring point. Different pre-set mobile platforms typically correspond to different optimal paths, and each pre-set mobile platform has the same initial stop but different pre-set risk monitoring points.

[0041] It should be noted that before performing path planning, the embodiment of the present invention can also divide the area to be monitored into multiple evenly distributed grids, each grid can represent a node, thereby constructing a grid map of the area to be monitored, making path planning more intuitive.

[0042] Optionally, in some embodiments, the objective function is:

[0043] in, To monitor the total cost, is the number of preset mobile platforms, 、 、 are target weights, is the balance factor, is the path distance cost, For the risk cost, is the energy cost, Cost of duplicate work for multiple machines.

[0044] The path cost can be:

[0045] in, For nodes To Node distance, Indicates the default mobile platform Whether the path can be passed (affected by the connectivity between sections and the risk level of the section) joint control), For mobile platforms Total driving time, is the mission time limit (controlled by the mobile platform's energy storage), is the time penalty coefficient (the degree of impact of the additional path distance cost caused by the mobile platform running out of energy and having to be retrieved by other vehicles, which can be taken as 103 here), is the total number of nodes, 、 For the node.

[0046] The risk cost can be:

[0047]

[0048] in, is the attenuation coefficient, Estimated arrival point The maximum time, For this moment, is the time decay factor. By introducing the time decay factor (It is known that the predicted risk weight closer to the current time The higher the value, the more attention the path planning will pay to the upcoming high-risk sections.

[0049] Energy costs can be:

[0050] in, is the driving power of the vehicle, is the sensing system power, is the battery energy conversion efficiency, For mobile platforms Driving distance.

[0051] The calculation formula is:

[0052] in, is the friction coefficient, is the mass of the vehicle, is the acceleration due to gravity, is the speed of the vehicle, is the ground inclination angle, is the air resistance coefficient, is the air density, is the frontal area of ​​the car.

[0053] The cost of repeated work on multiple machines can be:

[0054] in, For mobile platforms Responsible monitoring area, For mobile platforms Responsible monitoring area, is the overlapping area penalty coefficient (here it can be taken as 2). By introducing this cost, overlapping monitoring between the preset mobile platforms is guided to be minimized, thereby effectively improving the monitoring coverage.

[0055] Optionally, the plurality of preset constraints may include the following: the preset mobile platform is allowed to visit each target location node once; the preset starting location of the mobile platform is located at a given docking center; the preset travel time of the mobile platform satisfies the following formula:

[0056] in, Arrival node for the preset mobile platform time, Arrival node for the preset mobile platform time, For the preset mobile platform at the node The residence time, For the preset mobile platform slave node To Node driving time.

[0057] Furthermore, the embodiment of the present invention adopts a dynamic flood optimization swarm reinforcement learning algorithm to solve the above problem to minimize the total monitoring cost. Optimization problems with the goal, such as Figure 4 As shown, the following steps may be included: using a randomly generated path on a grid map by a preset mobile platform as initialization, iterative updates driven by navigation behavior, dynamically adjusting the weights to calculate the fitness value, and selecting the solution to the objective function based on the fitness value; when the population meets the convergence conditions, outputting the optimal path for multiple preset mobile platforms to move from the initial stop point to the preset risk monitoring point. Convergence conditions include emergency termination when the number of iterations reaches a preset number, the fitness value is less than a preset threshold, or the flood risk value of all paths exceeds a threshold. The preset number or preset threshold can be set according to actual needs and is not specifically limited here.

[0058] At the same time, during the solution process, the dynamic flood optimization swarm reinforcement learning algorithm establishes a navigation mechanism of "global search + local optimization + population evolution" to optimize the path plan: (1) Fine-tune the initial path: Randomly generate a path plan, fine-tune the path, ensure that the initial solution meets the basic constraints, and calculate each fitness value, that is, the total cost of the path.

[0059] (2) Calculate the fitness value: Calculate the total travel cost of each path and select the best solution based on the fitness value; retain the optimal solution and the top 30% of the best solutions as the population basis for the next generation.

[0060] (3) Dynamic multi-objective trade-off: The dynamic flood optimization algorithm can effectively jump out of the local optimal solution, dynamically balance multiple objectives, find the global optimal path, and dynamically adjust the weights according to real-time data. (4) Adaptive mutation: For high-risk or high-energy consumption nodes in the path, they are replaced with adjacent safety grids according to probability.

[0061] It should be noted that in the process of determining the optimal path for each preset mobile platform, the navigation simulation method may include at least one of local search, local optimization and population evolution, among which the global search is: applying large-scale perturbations to each path to break through the local optimum and discover potential high-quality areas; the local optimization is: fine-tuning the path; the population evolution is: selecting the top 30% high-quality paths as parents according to the fitness value ranking, cutting and splicing the two paths at random nodes through cross-recombination, and replacing the high-risk or high-energy consumption nodes in the path with adjacent safety grids according to probability through adaptive mutation.

[0062] It is understandable that compared with the existing particle swarm optimization algorithm and the classic dung beetle algorithm, the dynamic flood optimization group reinforcement learning algorithm has a better monitoring effect on the total cost. The internal union takes into account indicators such as path distance, flood risk, energy efficiency and multi-vehicle coordinated balance, and uses dynamically adjustable weights , realizing the adaptive allocation of real-time priorities of multiple objectives, can effectively solve the limitations of particle swarm optimization (the fixed weight of single-objective optimization makes it difficult to cope with dynamic multi-objectives, such as the real-time balance requirements of distance, risk, and energy consumption, and is prone to falling into local optimality) and classic dung beetle algorithm (based on fixed behavioral rules, such as single-objective optimization logic such as rolling, foraging, and reproduction, lacking multi-objective coordination and real-time decision-making capabilities).

[0063] Furthermore, a dynamic flood optimization swarm reinforcement learning algorithm establishes a navigation mechanism combining "global search + local optimization + population evolution." This algorithm applies large-scale perturbations to mobile paths to escape local optima and probabilistically replaces high-risk or high-energy consumption nodes with adjacent safety grids, achieving adaptive variation. Furthermore, by randomly cutting and splicing possible paths and incorporating flood and waterlogging risk predictions to avoid falling into high-risk areas, it effectively addresses the limitations of the ant colony algorithm (which suffers from low initial search rates and difficulty updating paths in real time during high-risk, dynamically changing flooding scenarios) and the classic dung beetle algorithm (which, under single-objective optimization conditions, uses fixed behavioral rules that cannot cope with unexpected situations such as path failures, such as a sudden increase in water depth at a node).

[0064] In step S103, based on the preset target detection algorithm, each preset mobile platform is used to determine the water accumulation area of ​​each preset risk monitoring point, and based on the water accumulation area of ​​each preset risk monitoring point and the preset digital elevation base map, the water accumulation depth of each preset risk monitoring point is determined.

[0065] Specifically, when the preset mobile platform reaches the preset risk monitoring point, the preset mobile platform can accurately identify the water accumulation area (water accumulation range) of the preset risk monitoring point from the optical image through the preset target detection algorithm, and combine with the preset digital elevation base map (that is, the high-resolution digital elevation base map preset in the microcomputer system of the preset mobile platform) to accurately calculate the water accumulation depth of the preset risk monitoring point.

[0066] In step S104, the monitoring strategy of each preset risk monitoring point is determined according to the water accumulation depth of each preset risk monitoring point, and flood monitoring is performed on each preset risk monitoring point based on the monitoring strategy of each preset risk monitoring point to obtain a flood monitoring result.

[0067] In other words, after determining the water depth at each pre-set risk monitoring point, this depth information can be used to determine the appropriate monitoring strategy for different water depths, ensuring timely and effective risk warning and management. Based on the monitoring strategy corresponding to each pre-set risk monitoring point, flood monitoring can be performed at that pre-set risk monitoring point, resulting in corresponding flood monitoring results.

[0068] As a possible implementation method, in some embodiments, the monitoring strategy for each preset risk monitoring point is determined based on the water accumulation depth at each preset risk monitoring point, including: based on each preset risk monitoring point, screening out at least one first target risk monitoring point where the water accumulation depth at the monitoring point is less than the preset depth value; controlling the preset mobile platform corresponding to at least one first target risk monitoring point to move to the target point corresponding to each first target risk monitoring point, and using a ranging water level meter to measure the water accumulation depth at the target point corresponding to each first target risk monitoring point.

[0069] Among them, the preset depth value can be pre-set by researchers in this field, or obtained through a limited number of experiments, or obtained through a limited number of computer simulations. No specific limitation is made here. Preferably, the embodiment of the present invention sets the preset depth value to 0.15m.

[0070] Specifically, for the first target risk monitoring point among the preset risk monitoring points where the water depth is less than the preset depth value (such as 0.15m), the preset mobile platform corresponding to each first target risk monitoring point can be controlled to travel to the target point corresponding to each first target risk monitoring point (a specific location in the water accumulation area, used to collect water depth information), and a ranging water level meter can be used to further measure the actual water depth.

[0071] As a possible implementation method, in other embodiments, the monitoring strategy of each preset risk monitoring point is determined according to the water accumulation depth of each preset risk monitoring point, and also includes: based on each preset risk monitoring point, screening out at least one second target risk monitoring point whose water accumulation depth is greater than or equal to the preset depth value; controlling the preset mobile platform corresponding to at least one second target risk monitoring point to call a preset bionic robot carrying a pressure water level gauge to move to the target point corresponding to each second target risk monitoring point, and using the pressure water level gauge to measure the water accumulation depth of the target point corresponding to each second target risk monitoring point.

[0072] Specifically, for the second target risk monitoring points among the preset risk monitoring points where the water depth is greater than or equal to the preset depth value (such as 0.15m), the preset mobile platform corresponding to each second target risk monitoring point can be controlled to call the preset bionic robot (that is, a small bionic robot integrated in the preset mobile platform), and the preset bionic robot is organized to carry a pressure water level gauge and move in a semi-submersible mode to the target point corresponding to each second target risk monitoring point. Subsequently, the pressure water level gauge is released to use the pressure water level gauge to measure the water depth at the target point corresponding to each second target risk monitoring point.

[0073] In summary, the active flood monitoring method proposed in the embodiment of the present invention has at least the following beneficial effects: (1) The embodiments of the present invention can realize active perception and targeted monitoring of urban floods, improve the timeliness and support of flood monitoring, and serve the construction of livable, resilient, and smart cities.

[0074] (2) The present invention constructs an objective function based on the travel distance, flood risk cost, and energy efficiency of multiple preset mobile platforms, and sets multiple constraints on the objective function to obtain a true path planning result. Based on the dynamic flood optimization swarm reinforcement learning algorithm, the path planning is performed on multiple preset mobile platforms to obtain the target path for the preset mobile platform to move from the initial stop point to the preset risk monitoring point. The algorithm can effectively escape the local optimal solution, dynamically balance multiple objectives, and find the global optimal path. It can also dynamically adjust the weights based on real-time data, which not only improves the accuracy and efficiency of the optimal path search, but also ensures the practical feasibility and dynamic adaptability of the path plan, ensuring that it can adapt to the complex urban road network environment and has high applicability.

[0075] The active flood monitoring method proposed in an embodiment of the present invention calculates a spatiotemporal risk tensor using a preset flood model based on rainfall forecast data and a preset cloud-based database. An objective function is constructed based on the spatiotemporal risk tensor, the regional road network, and the distribution of flood-prone points. The optimal path for each mobile platform is determined based on the constraints and the objective function to control the movement of each platform to a preset risk monitoring point. A preset target detection algorithm is used to determine the accumulated water area at each risk monitoring point, and then, combined with a preset digital elevation base map, the accumulated water depth at each risk monitoring point is determined. Based on the accumulated water depth at each risk monitoring point, an appropriate monitoring strategy is automatically selected to conduct flood monitoring for each risk monitoring point. This solves the problems of weak perception capabilities, limited coverage, and untimely operational responses associated with single monitoring methods in related technologies, thereby improving the efficiency and accuracy of flood disaster prevention.

[0076] Next, an active flood monitoring device according to an embodiment of the present invention will be described with reference to the accompanying drawings.

[0077] Figure 54 is a block diagram of an active flood monitoring device according to an embodiment of the present invention.

[0078] like Figure 5 As shown, the active flood monitoring device 10 includes: an acquisition module 100 , a control module 200 , a determination module 300 and a monitoring module 400 .

[0079] The acquisition module 100 is configured to obtain first rainfall forecast data and second rainfall forecast data for the area to be monitored, and calculate a spatiotemporal risk tensor using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and a preset cloud-based database. The control module 200 is configured to construct an objective function based on the spatiotemporal risk tensor, the target area road network, and the historical distribution of flood-prone areas, and determine the optimal path for each preset mobile platform based on preset constraints and the objective function. Based on the optimal path for each preset mobile platform, each preset mobile platform is controlled to move from an initial stop point to a preset risk monitoring point. Determination module 300, for determining the waterlogged area of ​​each preset risk monitoring point using each preset mobile platform based on a preset target detection algorithm, and determining the waterlogged depth of each preset risk monitoring point based on the waterlogged area of ​​each preset risk monitoring point and a preset digital elevation base map; The monitoring module 400 is used to determine the monitoring strategy of each preset risk monitoring point according to the water accumulation depth of each preset risk monitoring point, and based on the monitoring strategy of each preset risk monitoring point, perform flood monitoring on each preset risk monitoring point to obtain flood monitoring results.

[0080] Optionally, in some embodiments, the monitoring module 400 is specifically configured to: Based on each preset risk monitoring point, at least one first target risk monitoring point is selected whose water accumulation depth is less than a preset depth value; Control at least one preset mobile platform corresponding to a first target risk monitoring point to move to a target point corresponding to each first target risk monitoring point, and use a ranging water level meter to measure the water depth at the target point corresponding to each first target risk monitoring point.

[0081] Optionally, in some embodiments, the monitoring module 400 is further configured to: Based on each preset risk monitoring point, screening out at least one second target risk monitoring point where the water accumulation depth at the monitoring point is greater than or equal to the preset depth value; Control the preset mobile platform corresponding to at least one second target risk monitoring point to call the preset bionic robot carrying the pressure water level gauge to move to the target point corresponding to each second target risk monitoring point, and use the pressure water level gauge to measure the water depth at the target point corresponding to each second target risk monitoring point.

[0082] Optionally, in some embodiments, the objective function is:

[0083] in, To monitor the total cost, is the number of preset mobile platforms, 、 、 are target weights, is the balance factor, is the path distance cost, For the risk cost, is the energy cost, Cost of duplicate work for multiple machines.

[0084] Optionally, in some embodiments, before calculating the spatiotemporal risk tensor using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and a preset cloud-based base database, the module 100 is further configured to: Acquire optical image data and point cloud data of the target scene area; Using a preset target detection algorithm, the image elements of the target object in the optical image data are detected to obtain processed optical image data. The point cloud data and the optical image data are then geo-referenced using K-nearest neighbor sampling and sparse graph convolution algorithms. The geographic coordinates of the point cloud data are assigned to the target object in the optical image data to obtain registered optical image data and registered point cloud data. Mapping the registered optical image data to an image feature space using a first preset neural network, and mapping the registered point cloud data to a point cloud feature space using a second preset neural network; Using the preset feature mapping network, the image feature space and the point cloud feature space are aligned to obtain the target feature vector, and the target segmentation model is trained based on the target feature vector; The target segmentation model is used to interpret the geometric boundary information of the target object, and the geographic coordinate information and geometric size attributes of the target object are extracted, and the geometric boundary information, geographic coordinate information and geometric size attributes are updated to a preset cloud-based database.

[0085] It should be noted that the above explanations of the embodiment of the active flood monitoring method are also applicable to the active flood monitoring device of this embodiment, and will not be repeated here.

[0086] The active flood monitoring device proposed in an embodiment of the present invention calculates a spatiotemporal risk tensor using a preset flood model based on rainfall forecast data and a preset cloud-based database. An objective function is constructed based on the spatiotemporal risk tensor, the regional road network, and the distribution of flood-prone points. The optimal path for each mobile platform is determined based on the constraints and the objective function to control the movement of each mobile platform to a preset risk monitoring point. A preset target detection algorithm is used to determine the accumulated water area at each risk monitoring point, and then, combined with a preset digital elevation base map, the accumulated water depth at each risk monitoring point is determined. Based on the accumulated water depth at each risk monitoring point, an appropriate monitoring strategy is automatically selected to perform flood monitoring for each risk monitoring point. This solves the problems of weak perception capabilities, limited coverage, and untimely operational responses associated with single monitoring methods in related technologies, thereby improving the efficiency and accuracy of flood disaster prevention.

[0087] Figure 6 This is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device may include: A memory 601 , a processor 602 , and a computer program stored in the memory 601 and executable on the processor 602 .

[0088] When the processor 602 executes the program, the active flood monitoring method provided in the above embodiment is implemented.

[0089] Furthermore, the electronic device further includes: The communication interface 603 is used for communication between the memory 601 and the processor 602 .

[0090] The memory 601 is used to store computer programs that can be run on the processor 602 .

[0091] The memory 601 may include a high-speed RAM (Random Access Memory) memory, and may also include a non-volatile memory, such as at least one disk memory.

[0092] If the memory 601, processor 602, and communication interface 603 are implemented independently, the communication interface 603, memory 601, and processor 602 can be connected to each other via a bus and communicate with each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. Buses can be divided into address buses, data buses, control buses, etc. For ease of representation, Figure 6Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0093] Optionally, in a specific implementation, if the memory 601, the processor 602 and the communication interface 603 are integrated on a chip, the memory 601, the processor 602 and the communication interface 603 can communicate with each other through an internal interface.

[0094] The processor 602 may be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present invention.

[0095] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which implements the above-mentioned active flood monitoring method when executed by a processor.

[0096] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the active flood monitoring method as described above is implemented.

[0097] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

[0098] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0099] Although the embodiments of the present invention have been shown and described above, it will be understood that the above embodiments are illustrative and are not to be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. An active flood monitoring method, characterized in that: The following steps are involved: Obtaining first rainfall forecast data and second rainfall forecast data for the area to be monitored, and calculating a spatiotemporal risk tensor using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and a preset cloud-based database; Based on the spatiotemporal risk tensor, the target area road network, and the historical distribution of flood-prone points, an objective function is constructed, and based on preset constraints and the objective function, an optimal path for each preset mobile platform is determined; based on the optimal path for each preset mobile platform, each preset mobile platform is controlled to move from an initial stop point to a preset risk monitoring point; Based on a preset target detection algorithm, using each preset mobile platform to determine the water accumulation area of ​​each preset risk monitoring point, and based on the water accumulation area of ​​each preset risk monitoring point and a preset digital elevation base map, determine the water accumulation depth of each preset risk monitoring point; According to the water accumulation depth of each preset risk monitoring point, the monitoring strategy of each preset risk monitoring point is determined, and based on the monitoring strategy of each preset risk monitoring point, flood monitoring is performed on each preset risk monitoring point to obtain flood monitoring results.

2. The active flood monitoring method according to claim 1, characterized in that: The step of determining a monitoring strategy for each preset risk monitoring point according to the water accumulation depth of each preset risk monitoring point includes: Based on each of the preset risk monitoring points, at least one first target risk monitoring point is selected whose water accumulation depth is less than a preset depth value; Control the preset mobile platform corresponding to the at least one first target risk monitoring point to move to the target point corresponding to each first target risk monitoring point, and use a ranging water level meter to measure the water depth of the target point corresponding to each first target risk monitoring point.

3. The active flood monitoring method according to claim 1, characterized in that: The step of determining a monitoring strategy for each of the preset risk monitoring points according to the water depth of each of the preset risk monitoring points further includes: Based on each of the preset risk monitoring points, at least one second target risk monitoring point is selected whose water accumulation depth is greater than or equal to a preset depth value; Control the preset mobile platform corresponding to at least one second target risk monitoring point to call a preset bionic robot carrying a pressure water level gauge to move to the target point corresponding to each second target risk monitoring point, and use the pressure water level gauge to measure the water depth at the target point corresponding to each second target risk monitoring point.

4. The active flood monitoring method according to claim 1, characterized in that: The objective function is: in, To monitor the total cost, is the number of the preset mobile platforms, 、 、 are target weights, is the balance factor, is the path distance cost, For the risk cost, is the energy cost, Cost of duplicate work for multiple machines.

5. The active flood monitoring method according to claim 1, characterized in that: Before calculating the spatiotemporal risk tensor using the preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and the preset cloud-based database, the method further includes: Acquire optical image data and point cloud data of the target scene area; Detecting image elements of a target object in the optical image data using the preset target detection algorithm to obtain processed optical image data, geo-registering the point cloud data with the optical image data using a K-nearest neighbor sampling and sparse graph convolution algorithm, and assigning the geographic coordinates of the point cloud data to the target object in the optical image data to obtain registered optical image data and registered point cloud data; Mapping the registered optical image data to an image feature space using a first preset neural network, and mapping the registered point cloud data to a point cloud feature space using a second preset neural network; Using a preset feature mapping network, aligning the image feature space and the point cloud feature space to obtain a target feature vector, and training a target segmentation model based on the target feature vector; The target segmentation model is used to interpret the geometric boundary information of the target object, and the geographic coordinate information and geometric size attributes of the target object are extracted, and the geometric boundary information, the geographic coordinate information and the geometric size attributes are updated to the preset cloud-based database.

6. An active flood monitoring device, characterized in that: include: an acquisition module, configured to obtain first rainfall forecast data and second rainfall forecast data for the area to be monitored, and calculate a spatiotemporal risk tensor using a preset flood model based on the first rainfall forecast data, the second rainfall forecast data, and a preset cloud-based database; a control module, configured to construct an objective function based on the spatiotemporal risk tensor, the target area road network, and the historical distribution of flood-prone points, determine an optimal path for each preset mobile platform based on preset constraints and the objective function, and control each preset mobile platform to move from an initial stop point to a preset risk monitoring point based on the optimal path of each preset mobile platform; a determination module, configured to determine the accumulated water area of ​​each preset risk monitoring point using each preset mobile platform based on a preset target detection algorithm, and determine the accumulated water depth of each preset risk monitoring point based on the accumulated water area of ​​each preset risk monitoring point and a preset digital elevation base map; The monitoring module is used to determine the monitoring strategy of each preset risk monitoring point according to the water accumulation depth of each preset risk monitoring point, and based on the monitoring strategy of each preset risk monitoring point, perform flood monitoring on each preset risk monitoring point to obtain flood monitoring results.

7. The active flood monitoring device according to claim 6, characterized in that: The monitoring module is used to: Based on each of the preset risk monitoring points, at least one first target risk monitoring point is selected whose water accumulation depth is less than a preset depth value; Control the preset mobile platform corresponding to the at least one first target risk monitoring point to move to the target point corresponding to each first target risk monitoring point, and use a ranging water level meter to measure the water depth of the target point corresponding to each first target risk monitoring point.

8. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the active flood monitoring method according to any one of claims 1 to 5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the active flood monitoring method according to any one of claims 1 to 5.

10. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, is used to implement the active flood monitoring method according to any one of claims 1 to 5.

Images