Highway accident rescue resource dynamic scheduling and path optimization method

Abstract

The invention discloses an expressway accident rescue resource dynamic scheduling and path optimization method, which comprises the following steps: acquiring expressway road network GIS data, dynamic interference data and static parameters, and performing cleaning and standardization to form an initial data set; constructing an expressway directed graph, quantifying a congestion coefficient and an ice and snow influence factor, and dividing four types of rescue scenes including no interference, congestion interference, ice and snow interference and composite interference; establishing a sub-scene real-time vehicle speed calculation model by taking the minimization of the total driving time of rescue resources as a target, and setting a single rescue point allocation constraint, a path-allocation association constraint and a binary variable constraint; multi-scene road section weights are calculated in real time, rescue points are screened, the shortest distance of nodes is updated through relaxation iteration, and a rescue path is rebuilt through reverse tracing; and performing multi-dimensional analysis on the rescue path, and optimizing dynamic parameters according to road network characteristics and historical data. The technical problems of inaccurate dynamic road condition recognition, poor road condition topology adaptability and incomplete scene coverage of a traditional method are solved.

Description

Technical Field

[0001] This invention relates to the field of highway emergency rescue and route planning technology, and in particular to a method for dynamic scheduling and route optimization of highway accident rescue resources. Background Technology

[0002] As a core transportation hub in regional economic cycles, the efficiency of emergency rescue on highways directly determines the rate of secondary accidents, the probability of saving lives, and the speed of road network restoration. Current highway rescue technology systems suffer from several core technological deficiencies, making it difficult to meet complex practical needs. Specifically, traditional rescue routes rely on static road network parameters and lack a quantitative correlation between dynamic indicators such as traffic congestion coefficients and snow / ice impact factors and real-time vehicle speeds. When road sections are closed due to snow or ice, or when congestion worsens, calculating routes based on normal traffic parameters fails to find the optimal solution and may even generate invalid routes containing impassable sections. Furthermore, during peak congestion periods, the inability to dynamically adjust travel times for road sections severely impacts decision-making accuracy.

[0003] Meanwhile, traditional methods lack adaptability to road network topology and fail to fully consider the hierarchical characteristics of highway node functions—toll stations are the only entry point for rescue vehicles, and intermediate intersections only serve as road segment connections without external entry points or U-turn lanes. Traditional algorithms perform indiscriminate iterative calculations on all nodes, and perform "starting point-accident point" path solving on intermediate intersection nodes that are not rescue points, resulting in a large proportion of invalid calculations; it is also prone to path-accident point matching deviations. Existing technologies mostly focus on single interference scenarios and do not cover "congestion + snow" combined interference scenarios. In such combined scenarios, the decrease in road surface adhesion and traffic saturation create a "1+1>2" interference amplification effect. For example, moderate congestion reduces vehicle speed to 60% of normal levels, and after light snow, vehicle speed further decreases to 48%, extending rescue time by 108.3%. Existing solutions cannot adapt to such complex constraints, leading to the failure of rescue solutions. Traditional shortest path algorithms have two major drawbacks: First, they assume that the road segment weights (travel times) are fixed and do not update dynamic parameters during iteration, making it impossible to avoid road closures. Second, they do not prune based on the road network topology and continue to iterate on rescue points that are too far from the accident point (the path contains more than 5 road segments) or are topologically inaccessible, which increases the computation time by more than 40% and is prone to outputting non-optimal rescue point allocation schemes.

[0004] Patent document CN202510886203.3 discloses a method and system for intelligent emergency management of highways. It integrates multi-source heterogeneous data to construct a three-dimensional situational awareness database, enabling precise location and prediction of traffic congestion. Based on a Bayesian network, it constructs a virtual emergency scenario on a simulation platform to evaluate the traffic management effectiveness indicators of different control plans, analyze accident impacts and the feasibility of rescue routes. It establishes an optimization objective function to solve for the optimal resource scheduling scheme and distributes the results to intelligent interactive devices in real time. Through device feedback data, it dynamically optimizes the Bayesian network and deep reinforcement learning model, achieving adaptive adjustment of the scheme. Existing technologies cannot solve the core contradictions of inaccurate dynamic road condition recognition, poor road network adaptability, incomplete scenario coverage, and low algorithm efficiency, making it difficult to meet the needs of efficient highway rescue under changing road conditions. Therefore, there is an urgent need to propose a method for dynamic scheduling and route optimization of highway accident rescue resources. Summary of the Invention

[0005] The main objective of this invention is to propose a method for dynamic scheduling and route optimization of highway accident rescue resources, aiming to solve the technical problems of inaccurate dynamic road condition identification, poor road condition topology adaptability, and incomplete scene coverage in traditional methods.

[0006] To achieve the above objectives, the present invention provides a method for dynamic scheduling and route optimization of highway accident rescue resources, wherein the method includes the following steps:

[0007] S1. Data Acquisition and Preprocessing: Collect highway network GIS data, dynamic interference data, and static parameters, and form an initial dataset after cleaning and standardization.

[0008] S2. Road network topology abstraction and dynamic interference modeling: construct a directed graph of highways, quantify congestion coefficient and ice and snow impact factor, and classify four types of rescue scenarios: no interference, congestion interference, ice and snow interference, and combined interference.

[0009] S3. Multi-scenario rescue optimization model construction: With the goal of minimizing the total travel time of rescue resources, a real-time vehicle speed calculation model for different scenarios is established, and single rescue point allocation constraints, path-allocation association constraints, and binary variable constraints are set.

[0010] S4. Improve the Bellman-Ford algorithm to solve the problem, calculate the road segment weights in real time for multiple scenarios, filter rescue points, update the shortest distance of nodes through relaxation iteration, and reconstruct the rescue path by tracing back.

[0011] S5. Result evaluation and parameter optimization: Conduct multi-dimensional analysis of rescue routes and optimize dynamic parameters based on road network characteristics and historical data.

[0012] In one preferred embodiment, the directed graph of the highway in step S2 is as follows:

[0013]

[0014] in, For a directed graph of a highway, The node set includes the toll station node set. and the set of nodes at the middle intersection , ,and , This is a set of road segments, which includes the main road segments. and intersection sections .

[0015] One preferred embodiment is that step S2 quantifies the congestion coefficient, specifically: the congestion coefficient Let be the traffic congestion level at time t on road segment k, and let be the congestion level values: when there is no congestion, =1.0; during mild congestion, =0.8; During moderate congestion, =0.6, during severe congestion, ≤0.4.

[0016] One preferred embodiment is that step S2 quantifies the ice and snow impact factors, specifically: the ice and snow impact factors Let t represent the degree of damage to road surface conditions at section k caused by icy and snowy weather. Values ​​are assigned based on the icy and snow severity level: When there is no icy or snowy weather... =1.0; During Minor Snow, =0.8; During heavy snow, =0.6; during road closures, =0.

[0017] In one preferred embodiment, the objective function of the multi-scenario rescue optimization model in step S3 is:

[0018]

[0019] in, Let k be the actual length of road segment k. Let be the actual vehicle speed at time t on road segment k. From fire station to rescue point The shortest time, For the time t from the rescue point along Direction towards the accident site Use 1 when allocating resources, and 0 otherwise. For lane direction, As a rescue point, This is the location of the accident.

[0020] One preferred embodiment is that step S3 establishes a scenario-specific real-time vehicle speed calculation model, specifically as follows:

[0021]

[0022] Among them, scenario 1 is no interference, scenario 2 is congestion interference, scenario 3 is ice and snow interference, and scenario 4 is combined interference. Let be the actual vehicle speed at time t on road segment k. The maximum speed limit for this section of road The minimum speed limit for rescue vehicles in the emergency lane. Let be the congestion coefficient at time t on road segment k. Let be the ice and snow impact factor at time t on road segment k.

[0023] In one preferred embodiment, step S3 sets single rescue point allocation constraints, path-allocation association constraints, and binary variable constraints, specifically as follows:

[0024] The single rescue point allocation constraint is:

[0025]

[0026] The path-allocation association constraint is:

[0027]

[0028] The binary variable constraint is:

[0029]

[0030]

[0031] in, As a rescue point, Gather at the rescue point. For lane direction, For the time t from the rescue point along Direction towards the accident site Use 1 when allocating resources, and 0 otherwise. As the accident site, For the set of accident points, For the time t from the rescue point to the accident site of The value is 1 for road segment k in the direction of travel, and 0 otherwise. , This is a collection of highway network segments.

[0032] One preferred embodiment, the multi-scenario road segment weights, are specifically as follows:

[0033]

[0034] in, As for the weight of road segments, Let k be the actual length of road segment k.

[0035] One preferred embodiment is that step S4, screening rescue points, specifically involves:

[0036] Rescue points were selected through initial pruning of connected components and fine pruning of topological distance.

[0037] The initial pruning of connected components involves converting the directed graph into an undirected graph, decomposing it into several connected components using depth-first search, and retaining only the fault points. The rescue points within the component form the initial pruning candidate set, and topologically unreachable rescue points are removed.

[0038] The topological distance pruning is performed by using breadth-first search to calculate the topological distance from each rescue point to the accident point in the initial pruning candidate set, setting a threshold, and retaining rescue points whose topological distance is less than or equal to the threshold to form the final candidate set.

[0039] One preferred embodiment is that step S4 involves iteratively updating the shortest distance of the nodes and reconstructing the rescue path by tracing back, specifically as follows:

[0040] Initialization, based on the point of failure Using the path tracing starting point, the final candidate set after topological pruning... and all nodes of the highway network Set up accident point The initial shortest path cost to itself , accident point To all other nodes Initial shortest path cost All nodes The predecessor node is an empty set, that is... ;

[0041] Relaxation iteration, iteration Next, traverse the route. ,like Then update and ;in, It is a set of highway network nodes. It is a collection of highway network segments. The starting point of the road segment. The end point of the road segment, For the first Round-by-round iterative computation from rescue point to node The shortest path cost, The weight is real-time, representing the travel time of the road segment. For the first Round-by-round iterative computation from rescue point to node The shortest path cost;

[0042] Path reconstruction, from the accident point By tracing back the predecessor node, the forward path can be obtained after reversing the order. ;

[0043] Optimal solution selection involves calculating the total travel time of each rescue point in the final candidate set and selecting the rescue point and route corresponding to the minimum value.

[0044] The above-mentioned technical solution of the present invention includes the following steps: data collection and preprocessing, collecting highway network GIS data, dynamic interference data, and static parameters, and forming an initial dataset after cleaning and standardization; road network topology abstraction and dynamic interference modeling, constructing a directed graph of the highway, quantifying congestion coefficients and snow and ice impact factors, and classifying four types of rescue scenarios: no interference, congestion interference, snow and ice interference, and combined interference; construction of a multi-scenario rescue optimization model, aiming to minimize the total travel time of rescue resources, establishing a scenario-specific real-time vehicle speed calculation model, and setting single rescue point allocation constraints, path-allocation association constraints, and binary variable constraints; improvement of the Bellman-Ford algorithm for solving, calculating the weights of road segments in multiple scenarios in real time, screening rescue points, updating the shortest distance of nodes through relaxation iteration, and reconstructing the rescue path by reverse tracing; result evaluation and parameter optimization, performing multi-dimensional analysis of the rescue path, and optimizing dynamic parameters based on road network characteristics and historical data. The present invention solves the technical problems of inaccurate dynamic road condition identification, poor road condition topology adaptability, and incomplete scenario coverage in traditional methods. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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 the structures shown in these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of a method for dynamic scheduling and route optimization of highway accident rescue resources according to an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of scenario 1 of the present invention;

[0048] Figure 3 This is a schematic diagram of scenario 2 in embodiment 2 of the present invention;

[0049] Figure 4 This is a schematic diagram of scenario 3 in embodiment 3 of the present invention;

[0050] Figure 5 This is a schematic diagram of scenario 4 in embodiment 4 of the present invention;

[0051] Figure 6 This is a schematic diagram of the road network topology abstraction, taking the Changchun Ring Expressway and its connecting branch roads as an example in an embodiment of the present invention;

[0052] Figure 7 This is an original schematic diagram of the initial pruning of the road network topology abstract diagram of the Changchun Ring Expressway and its connecting branches in an embodiment of the present invention.

[0053] Figure 8 This is a schematic diagram of the initial pruning in an embodiment of the present invention;

[0054] Figure 9 This is a schematic diagram of the branches after fine pruning according to an embodiment of the present invention;

[0055] Figure 10 This is a schematic diagram illustrating the historical congestion situation of the Changchun Ring Expressway and its connecting branch roads according to an embodiment of the present invention;

[0056] Figure 11 This is a schematic diagram of the historical ice and snow conditions of the Changchun Ring Expressway and its connecting branch roads, according to an embodiment of the present invention.

[0057] Figure 12 This is a schematic diagram of the nodes where the rescue station changes due to road closures caused by ice and snow, considering only the rescue time of rescue vehicles on highways in this embodiment of the invention;

[0058] Figure 13 This is a schematic diagram of nodes where the rescue station changes due to congestion, considering only the rescue time of rescue vehicles on highways in this embodiment of the invention;

[0059] Figure 14 This is a schematic diagram illustrating the changes in rescue stations caused by road closures due to ice and snow, according to an embodiment of the present invention.

[0060] Figure 15 This is a schematic diagram of nodes where the rescue station changes due to congestion, according to an embodiment of the present invention.

[0061] The realization of the objectives, functional features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0062] 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 a part of the embodiments of the present invention, and not all of them. 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.

[0063] Furthermore, the technical solutions of the various embodiments of the present invention can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention.

[0064] See Figure 1 According to one aspect of the present invention, the present invention provides a method for dynamic scheduling and route optimization of highway accident rescue resources, wherein the method includes the following steps:

[0065] S1. Data Acquisition and Preprocessing: Collect highway network GIS data, dynamic interference data, and static parameters, and form an initial dataset after cleaning and standardization.

[0066] S2. Road network topology abstraction and dynamic interference modeling: construct a directed graph of highways, quantify congestion coefficient and ice and snow impact factor, and classify four types of rescue scenarios: no interference, congestion interference, ice and snow interference, and combined interference.

[0067] S3. Multi-scenario rescue optimization model construction: With the goal of minimizing the total travel time of rescue resources, a real-time vehicle speed calculation model for different scenarios is established, and single rescue point allocation constraints, path-allocation association constraints, and binary variable constraints are set.

[0068] S4. Improve the Bellman-Ford algorithm to solve the problem, calculate the road segment weights in real time for multiple scenarios, filter rescue points, update the shortest distance of nodes through relaxation iteration, and reconstruct the rescue path by tracing back.

[0069] S5. Result evaluation and parameter optimization: Conduct multi-dimensional analysis of rescue routes and optimize dynamic parameters based on road network characteristics and historical data.

[0070] Specifically, in this embodiment, the directed graph of the highway in step S2 is as follows:

[0071]

[0072] in, For a directed graph of a highway, The node set includes the toll station node set. and the set of nodes at the middle intersection , That is, all hubs with road segment connection functions in the expressway network are included in the node set, and toll station node set As the starting point for rescue vehicles to enter the highway network, it is recorded as the rescue point set. intermediate intersection node set If it only serves as a road segment connection, its ramp design must meet the requirement of "one-way traffic, no U-turn lanes"; This is a set of road segments, which includes the main road segments. and intersection sections Main road sections This refers to continuous lane sections of the main highway, connecting tollbooth nodes and intermediate nodes, as well as intersection sections. This refers to the ramp section at the intermediate intersection, which only connects intermediate nodes; both types of road sections indicate adjacent node pairs. , Each road segment corresponds to only one direction of travel on the left or right lane, and the lanes are physically separated by the central guardrail. Rescue vehicles can only travel in the current lane and cannot cross lanes to change lanes.

[0073] accident point All located on the road section The middle position is denoted as Its distance from the starting point of the road section The distance is denoted as ;

[0074] From the rescue point to the accident site path , refers to an ordered set of continuous road segments along a fixed lane direction, and its revised and clear definition and expression are as follows:

[0075]

[0076] in, For lane direction, This indicates that the route must travel entirely within the left or right lane and cannot cross lanes. This is the rescue point, i.e., the starting point of the route. For a node to be an intermediate node in a path, the end point of the preceding segment must equal the start point of the following segment, such as... The End yes The starting point ensures a continuous and unbroken path; accident point Located on the road section Between, path length For rescue point arrive The sum of the lengths of the road segments plus to the accident site distance Specifically:

[0077]

[0078] in, For path length, For road section The actual map length, Accident point To the starting point of the road segment The distance.

[0079] Specifically, in this embodiment, step S2 quantifies the congestion coefficient, which specifically involves: the congestion coefficient... Let be the traffic congestion level at time t on road segment k, and let be the congestion level values: when there is no congestion, =1.0; during mild congestion, =0.8; During moderate congestion, =0.6, during severe congestion, ≤0.4.

[0080] Specifically, in this embodiment, see Figures 2-5 Step S2, quantifying the ice and snow impact factors, specifically involves: the ice and snow impact factors Let t represent the degree of damage to road surface conditions at section k caused by icy and snowy weather. Values ​​are assigned based on the icy and snow severity level: When there is no icy or snowy weather... =1.0; During Minor Snow, =0.8; During heavy snow, =0.6; during road closures, =0, When the value is 0, the road segment is completely impassable, and the travel time approaches infinity.

[0081] Table 1. Quantitative Table of the Impact of Dynamic Interference Factors on Rescue Efficiency and Regional Economy

[0082]

[0083] Rescue scenario classification:

[0084] Scenario 1 (No Interference): No Congestion =1.0), no ice or snow ( =1.0);

[0085] Scenario 2 (Congestion Interference): There is congestion ( <1.0), no ice or snow ( =1.0);

[0086] Scenario 3 (Ice and Snow Interference): No Congestion =1.0), with ice and snow ( ≤0.8);

[0087] Scenario 4 (Combined Interference): Congestion exists ( <1.0), with ice and snow ( ≤0.8);

[0088] Table 2 Classification of Highway Accident Rescue Scenarios

[0089]

[0090] Specifically, in this embodiment, the core objective is to minimize the total travel time of rescue resources. A multi-scenario rescue optimization model is constructed by combining road network topology constraints and dynamic interference parameters. The model assumes: sufficient rescue resource reserves: each toll station rescue point corresponds to a dedicated fire station, and its rescue vehicles, personnel, and equipment reserves meet the needs of a single accident point; stable static attributes of the road network: within the rescue decision-making cycle, static attributes such as road segment length, node connection relationships, and toll station open status remain unchanged, while only dynamic parameters... Updated; Rescue point service capabilities are consistent: the equipment configuration, personnel skills, and emergency response procedures of the fire stations corresponding to each toll station are identical, and the preparation time from the fire station to the toll station is the same. (Time to enter the highway from the fire station) is a known fixed value; driving rules are compliant: rescue vehicles strictly follow the one-way traffic rules of highways and are not allowed to cross lanes, only moving between adjacent nodes and not deviating from the road network.

[0091] Specifically, in this embodiment, the objective function of the multi-scenario rescue optimization model in step S3 is:

[0092]

[0093] in, Let k be the actual length of road segment k. Let be the actual vehicle speed at time t on road segment k. From fire station to rescue point The shortest time, For the time t from the rescue point along Direction towards the accident site Use 1 when allocating resources, and 0 otherwise. For lane direction, As a rescue point, This is the location of the accident.

[0094] Specifically, in this embodiment, step S3, establishing a scenario-specific real-time vehicle speed calculation model, specifically involves:

[0095]

[0096] Among them, scenario 1 is no interference, scenario 2 is congestion interference, scenario 3 is ice and snow interference, and scenario 4 is combined interference. Let be the actual vehicle speed at time t on road segment k. The maximum speed limit for this section of road The minimum speed limit for rescue vehicles in the emergency lane. Let be the congestion coefficient at time t on road segment k. Let be the ice and snow impact factor at time t on road segment k.

[0097] Scenario 1: Ignoring ice and snow, and traffic congestion, each rescue vehicle travels at the maximum speed limit on the road segment, i.e. ;

[0098] Scenario 2: Considering traffic congestion, but not ice and snow, in which... ;

[0099] Scenario 3: Considering ice and snow, but not traffic congestion, where... hour, ;

[0100] Scenario 4: Considering ice and snow, and traffic congestion, where, when hour, ;

[0101] In scenarios 3 and 4, when At that time, section k was closed and impassable. , .

[0102] Specifically, in this embodiment, step S3 sets single rescue point allocation constraints, path-allocation association constraints, and binary variable constraints, specifically as follows:

[0103] The single rescue point allocation constraint is:

[0104]

[0105] The path-allocation association constraint is:

[0106]

[0107] The binary variable constraint is:

[0108]

[0109]

[0110] in, As a rescue point, Gather at the rescue point. For lane direction, For the time t from the rescue point along Direction towards the accident site Use 1 when allocating resources, and 0 otherwise. As the accident site, For the set of accident points, For the time t from the rescue point to the accident site of The value is 1 for road segment k in the direction of travel, and 0 otherwise. , This is a collection of highway network segments.

[0111] Specifically, in this embodiment, the constraint is explained as follows: at any given time, resources are allocated from only one rescue point via a specific lane direction at each accident site; this applies to all... , ;

[0112]

[0113] In order to accommodate the actual characteristics of the highway network where the physical separation of the left and right lanes (such as the central guardrail or median strip) prevents rescue vehicles from crossing lanes, Traverse all rescue points and connect each rescue point to the accident site. Sum the allocation variables and limit them to 1. For fixed rescue points At time t, for the accident point The allocation variables are combined, and the sum can only take the values ​​0 and 1. This means that a single rescue point can only allocate resources to one lane direction for a certain accident point at a certain time, or not allocate resources at all. This can avoid unreasonable decisions that waste scheduling resources. If the sum is greater than or equal to 2, it means that at time t, multiple rescue points allocate resources to the accident point, resulting in the repeated investment of rescue vehicles, personnel and other resources. This not only wastes resources, but may also exacerbate traffic congestion due to multiple rescue vehicles entering the accident section at the same time.

[0114] At the same time, the path selection variable is associated with the allocation variable, for all , For road segment indexing, ;

[0115]

[0116] To avoid the contradictory situation of "choosing a route without allocating resources," restrictions should be placed on... (Rescue point) At any moment To reach the accident site without going through this rescue point (Resource allocation), then The value must be 0 to ensure that the route selection is meaningful; to conform to the road network topology logic that "a route consists of specific adjacent road segments", restrictions are imposed if... (Rescue point) At any moment To the accident site (Resource allocation), then The value can be 1 or 0. A value of 1 indicates that the shortest path for rescue vehicles passes through this section of the road. The value 0 indicates that it does not pass through.

[0117] Specifically, in this embodiment, the multi-scenario road segment weights are calculated in real time. The road segment weight is defined as the real-time travel time at time t. Combined with dynamic scene updates, the multi-scenario road segment weights are specifically as follows:

[0118]

[0119] in, As for the weight of road segments, The actual length of road segment k;

[0120] when When a road is closed, the algorithm skips the relaxation iteration of that road segment to avoid invalid paths.

[0121] Specifically, in this embodiment, step S4, screening rescue points, specifically involves:

[0122] Rescue points were selected through initial pruning of connected components and fine pruning of topological distance.

[0123] The initial pruning of connected components involves converting the directed graph into an undirected graph, decomposing it into several connected components (nodes within each component are mutually reachable) using a depth-first search, and retaining only the fault points. The rescue points within the component form the initial pruning candidate set, and topologically unreachable rescue points are removed.

[0124] The topological distance pruning involves calculating the topological distance from each rescue point to the accident point in the initial pruning candidate set using breadth-first search, setting a threshold, and retaining rescue points whose topological distance is less than or equal to the threshold to form the final candidate set. In this invention, the threshold K is 5, but this invention is not strictly limited and can be set as needed. Gradually expand K until the number of candidates is greater than or equal to 2.

[0125] Specifically, in this embodiment, step S4, which involves relaxing and iteratively updating the shortest distance of nodes and reconstructing the rescue path by tracing back, specifically includes:

[0126] Initialization, based on the point of failure Using the path tracing starting point, the final candidate set after topological pruning... and all nodes of the highway network Set up accident point The initial shortest path cost to itself , accident point To all other nodes Initial shortest path cost All nodes The predecessor node is an empty set, that is... ;

[0127] Relaxation iteration, iteration Next, traverse the route. ,like Then update and ;in, It is a set of highway network nodes. It is a collection of highway network segments. The starting point of the road segment. The end point of the road segment, For the first Round-by-round iterative computation from rescue point to node The shortest path cost, The weight is real-time, representing the travel time of the road segment. For the first Round-by-round iterative computation from rescue point to node The shortest path cost;

[0128] Path reconstruction, from the accident point By tracing back the predecessor node, the forward path can be obtained after reversing the order. ;

[0129] Optimal solution selection involves calculating the total travel time of each rescue point in the final candidate set and selecting the rescue point and route corresponding to the minimum value.

[0130] Specifically, in this embodiment, see Figures 6-15 This invention takes a ring expressway and its connecting branches as the research object. The expressway is approximately 90km long, has a closed ring structure, connects 7 external trunk lines, and undertakes the main cross-regional logistics and commuting needs. Its significant characteristics are: frequent snow and ice weather in winter (December to February of the following year), local congestion easily occurs during morning and evening rush hours (7:00-9:00, 17:00-19:00), and often faces the combined interference of "congestion + snow and ice" during winter holidays such as the Spring Festival. The core elements of the abstract road network are constructed by obtaining GIS data of the ring expressway through OpenStreetMap (OSM) and performing topology processing.

[0131] Node layer: 92 nodes in total, including 36 toll station nodes (rescue point set) ) and 56 intermediate intersection nodes ;

[0132] Road segment level: a total of 127 road segments, including 80 main road segments. Intersections with 47 road sections The road sections are numbered according to the "rescue point - intermediate node" rule;

[0133] Congestion coefficient The Baidu Maps API was used to obtain real-time congestion information for the ring expressway, and the congestion coefficient for each road segment was determined; the impact of ice and snow was also considered. Based on local historical data on road closures due to extreme weather in winter, the impact factors of ice and snow on each road section were determined;

[0134] Experimental results:

[0135] When only considering the rescue time of rescue vehicles on highways, the nodes where rescue stations change due to road closures caused by ice and snow are:

[0136] Table 3. Nodes where rescue stations changed due to road closures caused by ice and snow.

[0137]

[0138] When only considering the rescue time of rescue vehicles on highways, the nodes where rescue stations change due to congestion are:

[0139] Table 4. Nodes where rescue stations changed due to congestion

[0140]

[0141] When considering the travel time from a fire station to a highway toll station, the shortest travel time from each toll station to the nearest fire station meeting rescue requirements is obtained via the Baidu Maps API.

[0142] Table 5. Shortest arrival time of the nearest fire station meeting rescue requirements to the highway from each toll station.

[0143]

[0144] The following points have been affected by changes in rescue stations due to road closures caused by ice and snow:

[0145] Table 6. Nodes where rescue stations changed due to road closures caused by ice and snow.

[0146]

[0147] Nodes where rescue stations have changed due to congestion:

[0148] Table 7. Nodes where rescue stations changed due to congestion.

[0149]

[0150] Specifically, in this embodiment, the present invention can realize dynamic and efficient scheduling and optimal route planning of rescue resources after highway accidents. By integrating dynamic interference factors such as traffic congestion and snowy weather in real time, it accurately calculates the travel time of road segments and automatically avoids invalid road segments. By leveraging road network topology layering and pruning strategies, it significantly improves algorithm efficiency and adapts to multiple scenarios such as "no interference, single interference, and compound interference," achieving coordinated optimization of rescue point selection and route planning. Ultimately, it provides a real-time, efficient, and universally applicable solution for highway emergency rescue, road network management, and other fields. It is particularly suitable for complex working conditions such as snowy winters and holiday congestion in northern China, and has significant benefits in shortening rescue time, reducing accident losses, and ensuring smooth road network operation.

[0151] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. All equivalent structural transformations made using the contents of the present invention's specification and drawings under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for dynamic scheduling and route optimization of highway accident rescue resources, characterized in that, Includes the following steps: S1. Data Acquisition and Preprocessing: Collect highway network GIS data, dynamic interference data, and static parameters, and form an initial dataset after cleaning and standardization. S2. Road network topology abstraction and dynamic interference modeling: construct a directed graph of highways, quantify congestion coefficient and ice and snow impact factor, and classify four types of rescue scenarios: no interference, congestion interference, ice and snow interference, and combined interference. S3. Multi-scenario rescue optimization model construction: With the goal of minimizing the total travel time of rescue resources, a real-time vehicle speed calculation model for different scenarios is established, and single rescue point allocation constraints, path-allocation association constraints, and binary variable constraints are set. S4. Improve the Bellman-Ford algorithm to solve the problem, calculate the road segment weights in real time for multiple scenarios, filter rescue points, update the shortest distance of nodes through relaxation iteration, and reconstruct the rescue path by tracing back. S5. Result evaluation and parameter optimization: Conduct multi-dimensional analysis of rescue routes and optimize dynamic parameters based on road network characteristics and historical data.

2. The method for dynamic scheduling and route optimization of highway accident rescue resources according to claim 1, characterized in that, The directed graph of the highway in step S2 is as follows: ； in, For a directed graph of a highway, The node set includes the toll station node set. and the set of nodes at the middle intersection , ,and , This is a set of road segments, which includes the main road segments. and intersection sections .

3. A method for dynamic scheduling and route optimization of highway accident rescue resources according to any one of claims 1-2, characterized in that, Step S2, quantifying the congestion coefficient, specifically involves: the congestion coefficient Let be the traffic congestion level at time t on road segment k, and let be the congestion level values: when there is no congestion, =1.0; during mild congestion, =0.8; During moderate congestion, =0.6, during severe congestion, ≤0.

4.

4. A method for dynamic scheduling and route optimization of highway accident rescue resources according to any one of claims 1-2, characterized in that, Step S2, quantifying the ice and snow impact factors, specifically involves: the ice and snow impact factors Let t represent the degree of damage to road surface conditions at section k caused by icy and snowy weather. Values ​​are assigned based on the icy and snow severity level: When there is no icy or snowy weather... =1.0; During Minor Snow, =0.8; During heavy snow, =0.6; during road closures, =0.

5. A method for dynamic scheduling and route optimization of highway accident rescue resources according to any one of claims 1-2, characterized in that, The objective function of the multi-scenario rescue optimization model in step S3 is: ； in, Let k be the actual length of road segment k. Let be the actual vehicle speed at time t on road segment k. From fire station to rescue point The shortest time, For the time t from the rescue point along Direction towards the accident site Use 1 when allocating resources, and 0 otherwise. For lane direction, As a rescue point, This is the location of the accident.

6. A method for dynamic scheduling and route optimization of highway accident rescue resources according to any one of claims 1-2, characterized in that, Step S3, establishing a scenario-specific real-time vehicle speed calculation model, specifically involves: ； Among them, scenario 1 is no interference, scenario 2 is congestion interference, scenario 3 is ice and snow interference, and scenario 4 is combined interference. Let be the actual vehicle speed at time t on road segment k. The maximum speed limit for this section of road Minimum speed for rescue vehicles in the emergency lane. Let be the congestion coefficient at time t on road segment k. Let be the ice and snow impact factor at time t on road segment k.

7. A method for dynamic scheduling and route optimization of highway accident rescue resources according to any one of claims 1-2, characterized in that, Step S3 sets single rescue point allocation constraints, path-allocation association constraints, and binary variable constraints, specifically as follows: The single rescue point allocation constraint is: ； The path-allocation association constraint is: ； The binary variable constraint is: ； ； in, As a rescue point, Gather at the rescue point. For lane direction, For the time t from the rescue point along Direction towards the accident site Use 1 when allocating resources, and 0 otherwise. As the accident site, For the set of accident points, For the time t from the rescue point to the accident site of The value is 1 for road segment k in the direction of travel, and 0 otherwise. , This is a collection of highway network segments.

8. The method for dynamic scheduling and route optimization of highway accident rescue resources according to claim 6, characterized in that, The multi-scenario road segment weights are specifically as follows: ； in, As for the weight of road segments, Let k be the actual length of road segment k.

9. A method for dynamic scheduling and route optimization of highway accident rescue resources according to claim 6, characterized in that, The step S4, screening rescue points, specifically involves: Rescue points were selected through initial pruning of connected components and fine pruning of topological distance. The initial pruning of connected components involves converting the directed graph into an undirected graph, decomposing it into several connected components using depth-first search, and retaining only the fault points. The rescue points within the component form the initial pruning candidate set, and topologically unreachable rescue points are removed. The topological distance pruning is performed by using breadth-first search to calculate the topological distance from each rescue point to the accident point in the initial pruning candidate set, setting a threshold, and retaining rescue points whose topological distance is less than or equal to the threshold to form the final candidate set.

10. A method for dynamic scheduling and route optimization of highway accident rescue resources according to claim 9, characterized in that, Step S4, which involves relaxing and iteratively updating the shortest distance of nodes and reconstructing the rescue path in reverse, specifically involves: Initialization, based on the point of failure Using the path tracing starting point, the final candidate set after topological pruning... and all nodes of the highway network Set up accident point The initial shortest path cost to itself , accident point To all other nodes Initial shortest path cost All nodes The predecessor node is an empty set, that is... ; Relaxation iteration, iteration Next, traverse the route. ,like Then update and ;in, It is a set of highway network nodes. It is a collection of highway network segments. The starting point of the road segment. The end point of the road segment, For the first Round-by-round iterative computation from rescue point to node The shortest path cost, The weight is real-time, representing the travel time of the road segment. For the first Round-by-round iterative computation from rescue point to node The shortest path cost; Path reconstruction, from the accident point By tracing back the predecessor node, the forward path can be obtained after reversing the order. ; Optimal solution selection involves calculating the total travel time of each rescue point in the final candidate set and selecting the rescue point and route corresponding to the minimum value.

Images