CN-122022093-A - Dynamic planning method and system for water supplementing path

CN122022093ACN 122022093 ACN122022093 ACN 122022093ACN-122022093-A

Abstract

The invention discloses a dynamic programming method and a dynamic programming system for a water supplementing path, wherein the method comprises the steps of generating N water supplementing paths based on the current position of a fire truck needing water supplementing and the position of the fire truck which can be used for water supplementing based on preset rules, marking all marked targets on the N water supplementing paths, marking the marked targets as targets which potentially influence road smoothness or influence traffic safety, extracting the positions of all the passed targets, calculating the required time of the fire truck passing through the corresponding path targets based on the preset running speed of the fire truck, determining the moment of the fire truck passing through the corresponding path targets based on the current time and the required time of the corresponding passing through path targets, and having self-adaptive reconstruction capability instead of searching for an unoccupied complete path, so that the decision toughness of the system under extreme conditions is ensured, a new feasible scheme can be still output by the system after the N initial paths fail, and uninterrupted execution of the water supplementing task is ensured.

Inventors

Request for anonymity
Request for anonymity
Request for anonymity
Request for anonymity
Request for anonymity
Request for anonymity
Request for anonymity

Assignees

河北数微信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260114

Claims (10)

1. The dynamic programming method for the water supplementing path is characterized by comprising the following steps of: Based on the current position of the fire truck needing water replenishment and the position of the fire hydrant capable of being used for water replenishment, generating N water replenishment paths based on preset rules, marking all marking targets on the N water replenishment paths of the fire truck, wherein the marking targets are targets which potentially influence road smoothness or influence traffic safety; extracting the positions of all the passing targets, calculating the required time of the fire truck passing through the corresponding passing targets based on the preset running speed of the fire truck, and determining the moment of the fire truck passing through the corresponding passing targets based on the current time and the required time of the corresponding passing targets; inputting threat time sequence data acquired correspondingly by the passing targets and corresponding required time into a pre-constructed threat prediction model to obtain cost scores of the fire engine at the moment when the fire engine passes through the corresponding path targets; Judging whether nodes of the route target influence area are removed from the navigation map according to the route target type and the cost score, if so, selecting the node closest to the removed node as a substitute node, reconstructing N water compensating paths, releasing the water compensating paths which can be communicated with the position of the fire hydrant, and reconstructing the running speed of the water compensating paths.
2. The method of claim 1, wherein the marking object type comprises a fixed object and a moving object, the fixed object comprises a roadside high-rise building, the moving object comprises a toxic gas area on a fire scene, and the threat time sequence data comprises fixed time sequence data of the fixed object and moving time sequence data of the moving object.
3. The method for dynamically planning a water make-up path according to claim 2, wherein the method for obtaining the travel speed of the reconstructed water make-up path comprises: and obtaining a fixed target closest to the position of the fire hydrant on each reconstructed water supplementing path, dividing the distance from the current fire truck to the fixed target by the required time for the fire truck to pass through the fixed target, and obtaining the running speed of the reconstructed water supplementing path.
4. The method for dynamically planning a water compensating path according to claim 3, wherein if the route target type is a fixed target and the cost score of the fixed target exceeds the corresponding cost score threshold, the route target influence area node is removed from the navigation map, otherwise, the route target influence area node is not removed.
5. The method for dynamically planning a water compensating path according to claim 3, wherein if the path target type is a moving target and the cost score of the moving target exceeds the corresponding cost score threshold, the path target influence area node is removed from the navigation map, otherwise, the path target influence area node is not removed.
6. The method of claim 2, wherein the threat prediction model comprises a collapse prediction model, and the training method of the collapse prediction model comprises: the method comprises the steps of collecting a historical data set in advance, preprocessing the data including fixed time sequence data and collapse probability and fragment distribution range corresponding to the fixed time sequence data, converting the fixed time sequence data into a plurality of characteristic samples according to preset sliding step length and sliding window length, respectively corresponding the characteristic samples to labels with different appointed time, wherein the labels are the collapse probability and the fragment distribution range, and the characteristic samples and the labels form training data; And training the collapse prediction model by taking training data as input of the collapse prediction model, taking a prediction feature sample and a label of a future appointed time as output, taking a feature sample and an actual label of the future appointed time as a prediction target, taking the maximized prediction accuracy as a training target, and obtaining the collapse prediction model meeting the prediction accuracy, wherein the collapse prediction model is an autoregressive integral moving average model, a time convolution network, a long-term and short-term memory network or a gating circulation unit.
7. The method for dynamically planning a water compensating path according to claim 6, wherein the method for obtaining the cost score of the fixed target comprises: the cost score of the fixed target is a weighted sum of collapse probability and the descaled fragment distribution range.
8. The method according to claim 1, wherein if N water compensating paths are reconstructed, there is no water compensating path communicating with the hydrant position, re-executing the water compensating path based on a preset rule, and generating N water compensating paths.
9. The method for dynamically planning a water compensating path according to claim 1, further comprising constructing a multi-objective path evaluation vector for each water compensating path according to cost scores, water compensating path transit time and fire hydrant water compensating capacity related data of each path objective at the passing time of the fire truck, wherein the multi-objective path evaluation vector at least comprises a path safety index, a transit time index and a water compensating efficiency index; And taking the maximized path safety, the shortest passing time and the optimal water supplementing efficiency as multi-objective optimization targets, calling a preset multi-objective path planning algorithm to solve the N water supplementing paths to obtain a pareto optimal water supplementing path set, and selecting the objective water supplementing path from the pareto optimal water supplementing path set according to a preset decision strategy.
10. A system for dynamic planning of a make-up water path, characterized in that it is configured to implement a method for dynamic planning of a make-up water path according to any one of claims 1 to 9, the system comprising: The path planning module is used for generating N water supplementing paths based on the current position of the fire truck needing water supplementing and the position of the fire hydrant capable of being used for supplementing water and a preset rule, marking all marked targets on the N water supplementing paths of the fire truck as path targets, wherein the marked targets are targets which potentially influence road smoothness or influence traffic safety; The analysis module is used for extracting the positions of all the passing targets, calculating the required time for the fire truck to pass through the corresponding passing targets based on the preset running speed of the fire truck, and determining the moment when the fire truck passes through the corresponding passing targets based on the current time and the required time for the corresponding passing targets; the threat prediction module is used for inputting threat time sequence data acquired correspondingly by the passing target and corresponding required time into a pre-constructed threat prediction model to obtain a cost score of the fire truck at the moment when the fire truck passes through the corresponding passing target; and the rescheduling triggering module judges whether nodes of the route target influence area are removed from the navigation map according to the route target type and the cost score, if so, the node closest to the removed node is selected and used as a substitute node, N water compensating paths are reconstructed, the water compensating paths which can be communicated with the position of the fire hydrant are released, and the running speed of the water compensating paths is reconstructed.

Description

Dynamic planning method and system for water supplementing path Technical Field The invention relates to the technical field of path planning, in particular to a dynamic planning method and system for a water compensating path. Background In fire rescue operations, the continuous combat capability of fire engines is highly dependent on an efficient, reliable uninterrupted water supply, where the planning of the make-up water path from the water source (e.g., hydrant) to the fire front is critical. Currently, the existing path planning technology in the industry, even though systems considering fire threat factors, generally have a core algorithm built on a static or quasi-static assumption that a road network is basically connected, and the systems severely rely on preset electronic navigation map data to abstract the road into a connected graph model and search the shortest path or the optimal path on the basis of the preset electronic navigation map data, for example, a classical Dijkstra algorithm or an a-x algorithm. However, under extremely complex fire scene environments, the conventional path planning technology has the fundamental defects that a planning model based on static environment assumption cannot cope with a rapidly-changing dynamic fire scene, a risk assessment mechanism adopting fixed future time output ignores the space-time difference of fire trucks reaching different positions, so that the fire trucks possibly drive into dangerous death traps when the fire trucks are safe and reach, and meanwhile, when a road network breaks suddenly, the traditional graph searching algorithm falls into a long-time deadlock state and cannot output effective paths in time. These defects severely restrict the water replenishing efficiency of the fire truck and directly threaten the overall effect of fire-fighting and rescue. Therefore, a method and a system for dynamically planning the water compensating path are provided to solve the difficulties in the prior art, which are the problems to be solved by those skilled in the art. Disclosure of Invention The invention aims to provide a dynamic planning method and system for a water compensating path, which solve the defects in the background technology. In order to achieve the above purpose, the invention provides a dynamic planning method for a water compensating path, which comprises the following steps: Based on the current position of the fire truck needing water replenishment and the position of the fire hydrant capable of being used for water replenishment, generating N water replenishment paths based on preset rules, marking all marking targets on the N water replenishment paths of the fire truck, wherein the marking targets are targets which potentially influence road smoothness or influence traffic safety; extracting the positions of all the passing targets, calculating the required time of the fire truck passing through the corresponding passing targets based on the preset running speed of the fire truck, and determining the moment of the fire truck passing through the corresponding passing targets based on the current time and the required time of the corresponding passing targets; inputting threat time sequence data acquired correspondingly by the passing targets and corresponding required time into a pre-constructed threat prediction model to obtain cost scores of the fire engine at the moment when the fire engine passes through the corresponding path targets; Judging whether nodes of the route target influence area are removed from the navigation map according to the route target type and the cost score, if so, selecting the node closest to the removed node as a substitute node, reconstructing N water compensating paths, releasing the water compensating paths which can be communicated with the position of the fire hydrant, and reconstructing the running speed of the water compensating paths. Further, the marking object type includes a fixed object including a road side high-rise building and a moving object including a toxic gas area on a fire scene, and the threat time sequence data includes fixed time sequence data of the fixed object and moving time sequence data of the moving object. Further, the method of obtaining the travel speed of the reconstructed water replenishment path includes: and obtaining a fixed target closest to the position of the fire hydrant on each reconstructed water supplementing path, dividing the distance from the current fire truck to the fixed target by the required time for the fire truck to pass through the fixed target, and obtaining the running speed of the reconstructed water supplementing path. Further, if the route target type is a fixed target and the cost score of the fixed target exceeds the corresponding cost score threshold, the route target influence area node is removed from the navigation map, otherwise, the route target influence area node is not removed. Further, if the path target type is a moving