SUMO-based highway reconstruction and extension simulation scene construction and evacuation rescue decision method and system

Abstract

The invention discloses an SUMO-based highway reconstruction and extension simulation scene construction and evacuation rescue decision method and system, and the method comprises the following steps: 1, building a dynamic scheme combination decision model based on rescue and evacuation route schemes during the highway reconstruction and extension period in combination with traffic flow operation parameters; step 2, adopting SUMO simulation software to carry out traffic simulation during the highway reconstruction and extension period, and simulating the traffic operation state of the reconstruction and extension road section under the traffic accident state; and 3, realizing a plurality of rescue and evacuation routes in reconstruction and extension simulation, collecting simulation data, performing regression fitting on the decision model, counting simulation results and verifying the effectiveness of the decision model. The system comprises a model building module, a traffic simulation module and a result verification module. According to the method, the scheme decision model is constructed, the appropriate rescue and evacuation scheme in the current traffic state is judged according to the input traffic flow parameters, rapid rescue and evacuation of accidents in the reconstruction and extension period are achieved, and the normal traffic flow operation state is effectively recovered.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic simulation, and in particular to a SUMO-based highway reconstruction and expansion simulation scenario construction and an evacuation and rescue decision-making method and system. Background Art

[0002] Recently, the level of intelligent transportation in my country has continued to improve, the deep integration of the Internet and transportation has accelerated, and intelligent transportation has become a key component of the development of smart cities in China. In order to reduce the time and cost of testing, avoid the risks of on-site testing, and improve the efficiency of testing and optimization, digital twin technology has been widely used in intelligent transportation systems. Traffic simulation technology based on digital twins has developed rapidly. In addition, with the continuous and in-depth application of simulation technology in the field of transportation, by constructing complex simulation scenarios and conducting simulation analysis, a decision-making basis is provided for actual traffic management and control solutions. The present invention takes the highway reconstruction and expansion road as the scenario, studies the rescue and evacuation plan decision under this scenario, combines the two-dimensional traffic simulation software SUMO for scenario construction and data analysis, and is committed to developing a rescue and evacuation decision-making method during highway reconstruction and expansion.

[0003] In a large amount of previous research materials, there has not been a relatively systematic study on rescue and evacuation during the reconstruction and expansion of expressways, and there is a lack of decision-making basis for plans in different scenarios. The plans proposed in some studies have not been verified by scientific means, and the effectiveness of the proposed plans cannot be guaranteed. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a SUMO-based highway reconstruction and expansion simulation scenario construction and its evacuation and rescue decision-making method and system, by constructing a scheme decision model, according to the input traffic flow parameters, to judge the appropriate rescue and evacuation scheme under the current traffic conditions, to achieve rapid rescue and evacuation of accidents during reconstruction and expansion, and effectively restore the normal traffic flow operation state.

[0005] In order to solve the above technical problems, the present invention provides a SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method, comprising the following steps:

[0006] Step 1: Based on the rescue and evacuation route plan during the highway reconstruction and expansion period, a dynamic plan combination decision model is established in combination with traffic flow operation parameters;

[0007] Step 2: Use SUMO simulation software to perform traffic simulation during the reconstruction and expansion of the expressway, and simulate the traffic operation status of the reconstruction and expansion section under the traffic accident state;

[0008] Step 3: Implement multiple rescue and evacuation routes in the reconstruction and expansion simulation, collect simulation data and perform regression fitting on the decision model, statistically analyze the simulation results and verify the effectiveness of the decision model.

[0009] Furthermore, in step 1, based on the rescue and evacuation route plan during the highway reconstruction and expansion period, a dynamic plan combination decision model is established in combination with traffic flow operation parameters, which specifically includes the following steps:

[0010] Step 11: Analyze the conventional rescue routes in accident scenarios during the reconstruction and expansion period of the expressway, and determine feasible rescue routes based on the actual conditions of the road section, so that rescue vehicles can enter the expressway through the nearest interchanges upstream and downstream of the expressway and arrive at the accident site;

[0011] Step 12: Analyze the conventional vehicle evacuation routes after the rescue operation in the accident scene, and determine feasible evacuation routes based on the actual rescue situation, so that stranded social vehicles can quickly leave the accident scene;

[0012] Step 13: Establish a decision regression model. When an accident occurs and blocks all traffic in one direction, calculate the rescue time and road restoration time required for different evacuation routes and rescue routes based on the traffic status indicators at the time of the accident, and determine the optimal matching plan for the rescue route and evacuation route.

[0013] Furthermore, in step 13, the decision regression model is established as follows:

[0014] v ij =β ij,0 +β ij.1 X ij,2 +β ij,3 X ij,4 +...+β ij,n X ij,n +ε

[0015]

[0016] Among them, X ij,1 ,X ij,2 ,...,X ij,n is the independent variable X ij The n eigencomponent matrices of ; and are predicted values, all calculated by the regression model; v ij and c ij are real values, which are obtained from a large amount of simulation data in the early stage; and is the average of all true values; β ij,0 ,β ij,1 ,...,β ij,n and γ ij,0 ,γij,1 ,...,γ ij,n is a model parameter vector, each component of which is a parameter of each combination of evacuation route and rescue route, which is the optimization object; the subscript ij represents the i-th evacuation plan combined with the j-th rescue plan.

[0017] Furthermore, in step 2, the traffic simulation during the reconstruction and expansion of the expressway is performed using SUMO simulation software, and the simulation of the traffic operation state of the reconstruction and expansion section under the traffic accident state specifically includes the following steps:

[0018] Step 21, build a SUMO simulation environment, intercept the highway section based on the map database and import it into SUMO, use the netedit interface of SUMO to modify the attributes of the simulation section according to the actual situation of the reconstruction and expansion area, close the emergency lane and add the expansion road, collect relevant data of the highway reconstruction and expansion area, including traffic flow and vehicle speed, and build a basic simulation environment;

[0019] Step 22: According to the emergency rescue and accident situation, determine the number of emergency rescue points and rescue vehicles, add the information of the emergency rescue points and rescue vehicles to the simulation environment, apply the path search algorithm to search for the shortest path from the emergency rescue point to the evacuation point in the road network, and evaluate the searched path, considering the road congestion and vehicle speed factors, and calculate the feasibility and efficiency of the path;

[0020] Step 23: Preheat the simulation scene, set the accident point location, simulate the accident and implement relevant control measures;

[0021] Step 24. According to the characteristics of the highway reconstruction and expansion area, in order to maintain consistency with normal traffic conditions, special simulation roads and special driving routes are built, and reverse emergency lanes are set on the inner side of the simulation roads. The opening conditions of the lanes are set to that there are no vehicles on the corresponding forward lanes, that is, all vehicles have been evacuated.

[0022] Furthermore, in step 3, multiple rescue and evacuation routes are implemented in the reconstruction and expansion simulation, simulation data is collected and regression fitting is performed on the decision model, and the simulation results are statistically analyzed and the effectiveness of the decision model is verified. Specifically, the following steps are included:

[0023] Step 31: By giving the evacuation routes of the congested vehicles, the vehicles at the accident site are evacuated, and the upstream and downstream interchange traffic is controlled to simulate the evacuation process. By giving the number of rescue vehicles and the driving routes, the emergency rescue process is simulated;

[0024] Step 32: Wait for the road to be cleared and end the simulation, and simultaneously collect simulation data, including emergency rescue time, evacuation time, and vehicle congestion; analyze and evaluate the simulation results according to three scenarios: two-way traffic in the opposite lane, U-turn on the central median, and two-way brief release;

[0025] Step 33: Clean the simulation data, process missing values ​​and outliers, and perform normalization and dimensionality reduction processing;

[0026] Step 34: Fit the regression model according to the obtained simulation data, use the least square method to establish a linear regression prediction model, obtain multiple groups of rescue time and road traffic restoration time under different conditions, and obtain the best linear regression function;

[0027] Step 35: Evaluate the established regression model and verify the effectiveness of the decision model in a simulation environment. Under a given accident scenario, collect the accident scenario road traffic operation indicators, input the decision model to obtain the solution decision results, test it in the simulation scenario, and compare and verify it with the total time indicators of other solution combinations.

[0028] Furthermore, in step 31, by giving the evacuation route of the congested vehicles, the vehicles at the accident site are evacuated, and the upstream and downstream interchange traffic is controlled to simulate the evacuation process. By giving the number of rescue vehicles and the driving route, the simulation of the emergency rescue process specifically includes the following steps:

[0029] Step 311, determine the initial conditions of the route, set the rescue vehicle and the emergency evacuation point, define the rescue vehicle ID and the location of the emergency evacuation point in route.rou.xml; initialize the route information, configure the initial route for each rescue vehicle, and these routes are based on the pre-set traffic network and feasible driving paths;

[0030] Step 312: Start the simulation using the Traci interface, and execute each simulation step repeatedly by looping until the set conditions are met or the rescue / evacuation is completed, and dynamically adjust the route of the rescue vehicle according to the real-time traffic conditions and road network status during the simulation process;

[0031] Step 313: modify the vehicle driving settings to avoid the vehicle from detouring due to encountering a congested area or an accident scene;

[0032] Step 314: By monitoring the arrival of the rescue vehicle at the key road node, it is determined whether the route needs to be adjusted. If the rescue vehicle is close to the accident site, it is necessary to adjust the further driving path according to the set rescue route, that is, to select one of upward forward rescue, downward forward rescue and downward reverse rescue for driving;

[0033] Step 315: When the simulation runs to the set step length, the evacuation simulation begins, and the vehicles on the edge before the accident section drive according to the evacuation route;

[0034] Step 316: The rescue vehicle drives away from the accident edge, and the vehicle drives on the accident edge, and the simulation runs normally until the step length ends.

[0035] Accordingly, a SUMO-based highway reconstruction and expansion simulation scenario construction and its evacuation and rescue decision-making system include: a model building module, a traffic simulation module and a result verification module; the model building module is based on the rescue and evacuation route plan during the highway reconstruction and expansion, and combines the traffic flow operation parameters to establish a dynamic plan combination decision model; the traffic simulation module uses SUMO simulation software to perform traffic simulation during the highway reconstruction and expansion, and simulates the traffic operation status of the reconstruction and expansion section under the traffic accident state; the result verification module realizes multiple rescue and evacuation routes in the reconstruction and expansion simulation, collects simulation data and performs regression fitting on the decision model, and statistically analyzes the simulation results and verifies the effectiveness of the decision model.

[0036] The beneficial effects of the present invention are as follows: the present invention focuses on the rescue and evacuation decisions in the scenario of highway reconstruction and expansion, constructs a simulation scenario of highway reconstruction and expansion, and fits a regression decision model based on the traffic flow parameters in the simulation scenario, effectively finding the optimal rescue and evacuation plan in the accident scenario, helping the traffic flow to resume normal operation, and simulates and verifies the plan and decision, illustrating the effectiveness of the decision-making method; the present invention focuses on a dynamic system framework with high compatibility, and can provide rescue and evacuation decisions for reconstruction and expansion accident scenarios under various conditions, effectively improving rescue efficiency and reducing the impact of accidents on traffic. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is an illustration of the rescue route of the present invention.

[0038] Figure 2 It is an evacuation route illustration diagram of the present invention.

[0039] Figure 3 A demonstration diagram is built for the simulation environment of the present invention. DETAILED DESCRIPTION

[0040] A SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method includes the following steps:

[0041] Step 1: Based on the rescue and evacuation route plan during the highway reconstruction and expansion period, a dynamic scheme combination decision model is established in combination with traffic flow operation parameters, which specifically includes the following steps:

[0042] Step 1.1: Combined construction of highway rescue routes and evacuation routes;

[0043] Step 1: Common highway rescue routes such as Figure 1 As shown in the rescue route diagram, it is generally divided into upward forward rescue, downward forward rescue and downward reverse rescue, so that the rescue vehicle enters the expressway through the nearest interchange upstream and downstream of the expressway to reach the accident scene;

[0044] Step 2: Common highway evacuation routes such as Figure 2 As shown in the (Evacuation Route Illustration), it is generally divided into alternate traffic at different time periods, U-turns and lane-by-lane traffic, so that stranded social vehicles can leave the accident scene;

[0045] Step 3: Determine a combination of feasible rescue routes and evacuation routes based on the actual conditions of the road section where the accident occurred.

[0046] Step 1.2: Establishment of scheme combination decision model;

[0047] The regression function uses the least squares method to build a linear regression prediction model. The least squares method is a mathematical optimization technique that is widely used in data fitting.

[0048] Step 1: Regression model input. The input variables are:

[0049] ① Traffic status variables: upstream traffic flow, downstream traffic flow, and two-way average speed. These variables are variables that can be directly observed or calculated in traffic scenarios, and can reflect the current traffic status through traffic calculations.

[0050] ② Distance variable: distance L between points AP1 , L P1X , L XP2 , L P2B Distance and vehicle speed are the most important variables for obtaining rescue time and recovery time.

[0051] Step 2: Regression model output. The output variables are:

[0052] ① Rescue time matrix V=(v ij ) 4×3 .v ij It represents the rescue time required for the i-th evacuation route and the j-th rescue route.

[0053] ②Road restoration time matrix C=(c ij ) 4×3 .c ij It represents the road restoration time required for the i-th evacuation route and the j-th rescue route.

[0054] Step 3: Perform feature engineering on the input data, including data cleaning, variable selection and normalization. Then, these features X ij As an independent variable, rescue time v ij and road recovery time c ij They were used as dependent variables to establish a linear regression model.

[0055] Step 4: The regression function uses the least squares method to establish a linear regression prediction model. The least squares method is a mathematical optimization technique that is widely used in data fitting.

[0056] v ij =β ij,0 +β ij.1 X ij,2 +β ij,3 X ij,4 +...+β ij,n X ij,n +ε

[0057]

[0058] Among them, X ij,1 ,X ij,2 ,...,X ij,n is the independent variable X ij The n eigencomponent matrices of ; and are predicted values, all calculated by the regression model; v ij and c ij are real values, which are obtained from a large amount of simulation data in the early stage; and is the average of all true values; β ij,0 ,β ij,1 ,...,β ij,n and γ ij,0 ,γ ij,1 ,...,γ ij,n is a model parameter vector, each component of which is the parameter of each combination of evacuation route and rescue route, as the optimization object. The subscript ij represents the i-th evacuation plan combined with the j-th rescue plan.

[0059] Step 1.3: Evaluate the effectiveness of decision route selection.

[0060] The specific process of this step is based on the above steps 1.1 and 1.2:

[0061] Step 1: Calculate the rescue time. The rescue time is the time from the occurrence of the accident to the arrival of the rescue at the accident site. During this period of time, the number of evacuation vehicles and rescue vehicles are intertwined. The rescue time can reflect the impact of the combination of different evacuation routes and rescue routes on the rescue effect;

[0062] Step 2: Calculate the road recovery time. The road recovery time is the time from the occurrence of the accident to the complete restoration of two-way road traffic to normal, which can reflect the overall impact of different evacuation routes and rescue routes on traffic;

[0063] Step 3: Based on the rescue time and road restoration time, calculate the total time from the occurrence of the accident to the completion of evacuation. The plan combination with the shortest total time is the optimal rescue and evacuation plan.

[0064] Step 2: Based on step 1, SUMO simulation software is used to simulate the traffic during the highway reconstruction and expansion period, simulating the traffic operation status of the reconstruction and expansion section under the traffic accident state, which is specifically achieved by the following steps:

[0065] Step 2.1: Build SUMO simulation environment;

[0066] Step 1: Based on the map database, extract the highway sections and import them into SUMO. Use SUMO's netedit interface to modify the section attributes according to the actual situation of the reconstruction and expansion area, close the emergency lane and add the expansion road, collect relevant data of the highway reconstruction and expansion area, including traffic flow and vehicle speed, and build a basic simulation environment.

[0067] Step 2: Accident location setting. Here, the accident section AB to be studied is divided into five sections: A-P1-X-P2-B. Among them, X is the accident point, A and B are the nearest interchange entrances and exits upstream and downstream of the accident point, and P1 and P2 are the nearest emergency turning points on the central median strip upstream and downstream of the accident point.

[0068] Step 2.2: Traffic accident scene construction

[0069] Step 1: Select the specific highway section edge where the accident occurred, determine its edge_id, and set relevant parameters and conditions according to the type of accident to be simulated (such as vehicle collision).

[0070] Step 2: Edit the net.xml file. In the SUMO road network file, mark or modify the road section where the accident occurred. You can do this by setting specific road section attributes, such as lowering the speed limit.

[0071] Step 3: Set the impact of the accident, use the additional file add.xml to define the specific impact of the accident. Set the "variable speed flag" to temporarily change the speed limit, and use the "close lane" command to simulate the situation where the lane is blocked due to an accident.

[0072] Step 4: Configure vehicle behavior and preset detour routes for vehicles that may be affected by the accident by modifying the rou.xml file.

[0073] Step 2.3: Modeling emergency rescue needs;

[0074] Step 1: Input the rescue point location, rescue vehicle information and preset routes into the simulation software. Adjust the simulation settings such as time, weather and traffic flow to simulate different rescue scenarios, and determine the number of emergency rescue points and rescue vehicles.

[0075] Step 2: Select a suitable path search algorithm based on the network structure, such as the Dijkstra algorithm for unweighted graphs, and the A* algorithm for path search with heuristic information. Perform path search for each emergency rescue point to find the shortest path to each evacuation point.

[0076] Step 2.4: Vehicle route construction

[0077] Add roads according to the special requirements of rescue and evacuation routes. In the process of building the simulated road network, set up reverse emergency lanes on the inner side of the simulated roads, and set the opening condition of the lanes to be that there are no vehicles on the corresponding forward lanes. Considering the realization of the rescue route, set up a road at the temporary isolation belt and set a certain speed limit to simulate the actual engineering rescue situation. Figure 3 shown.

[0078] Step 3: Implement multiple rescue and evacuation routes in the reconstruction and expansion simulation, collect simulation data and perform regression fitting on the decision model, statistically analyze the simulation results and verify the effectiveness of the decision model.

[0079] The implementation of step 3 requires that the scheme combination in step 1 and the accident scenario and rescue requirements in step 2 are determined, and then simulation and effectiveness verification are carried out. The specific steps are as follows:

[0080] Step 3.1: Rescue and evacuation route simulation implementation;

[0081] Step 1: Determine the initial conditions of the route, set the rescue vehicle and the emergency evacuation point. Define the rescue vehicle ID and the location of the emergency evacuation point in route.rou.xml. Initialize the route information and configure the initial route for each rescue vehicle. These routes are based on the pre-set traffic network and feasible driving paths.

[0082] Step 2: Use the Traci interface to start the simulation, and execute each simulation step repeatedly until the set conditions are met or the rescue / evacuation is completed. During the simulation, the rescue vehicle's route is dynamically adjusted according to the real-time traffic conditions and road network status.

[0083] Step 3: Modify the vehicle's driving settings to avoid the vehicle having to detour due to congested areas or accident scenes.

[0084] Step 4: By monitoring the rescue vehicle's arrival at key road nodes, determine whether the route needs to be adjusted. If the rescue vehicle is close to the accident scene, it is necessary to adjust the further driving path according to the set rescue route, that is, choose one of the upward forward rescue, downward forward rescue and downward reverse rescue.

[0085] Step 5: When the simulation runs to the set step length, the evacuation simulation begins, and the vehicles on the edge before the accident section drive according to the evacuation route.

[0086] Step 6: After the rescue vehicle leaves the accident edge, the vehicle can drive on the accident edge and the simulation runs normally until the step ends.

[0087] Step 3.2: Collect and analyze simulation results;

[0088] The detectors in the simulation software are used to obtain simulation data, mainly the Multi-Entry-Exit Detectors (E3) detector in SUMO. The output values ​​of the E3 detector, nVehEntered and meanSpeed, are selected as the two-way traffic flow and average speed, respectively. The accident location and upstream and downstream interchange locations are derived from the road network data, as shown in Table 1.

[0089] Table 1 Detector output and parameter meanings

[0090]

[0091] Step 3.3: Simulation data statistics

[0092] The simulation data were statistically analyzed and the experiment was initially divided into 9 groups. Each group was set with different combinations of evacuation routes and rescue routes. The distance data of A-P1, P2-B, and P1-X were set. At the same time, the road traffic volume was changed, and the average speed of social vehicles, forward traffic density, rescue time, and evacuation time indicators were statistically analyzed, as shown in the following table.

[0093]

[0094] Step 3.4: Regression model establishment

[0095] The regression model is fitted according to the obtained simulation data, and the least square method is used to establish a linear regression prediction model. Multiple groups of rescue time and road restoration time are obtained under different conditions, and the best linear regression function is obtained.

[0096] Step 3.5: Verify the validity of the decision function

[0097] The established regression model is evaluated and the effectiveness of the decision model is verified in a simulation environment. Under a given accident scenario, the accident scenario road operation indicators are collected and input into the decision model to obtain the decision results of the scheme. The results are tested in the simulation scenario and compared with the total time indicators of other scheme combinations for verification.

[0098] Accordingly, a SUMO-based highway reconstruction and expansion simulation scenario construction and its evacuation and rescue decision-making system include: a model building module, a traffic simulation module and a result verification module; the model building module is based on the rescue and evacuation route plan during the highway reconstruction and expansion, and combines the traffic flow operation parameters to establish a dynamic plan combination decision model; the traffic simulation module uses SUMO simulation software to perform traffic simulation during the highway reconstruction and expansion, and simulates the traffic operation status of the reconstruction and expansion section under the traffic accident state; the result verification module realizes multiple rescue and evacuation routes in the reconstruction and expansion simulation, collects simulation data and performs regression fitting on the decision model, and statistically analyzes the simulation results and verifies the effectiveness of the decision model.

[0099] The above shows and describes the basic principles and main processes of the present invention, focusing on the rescue and evacuation decisions in the scenario of highway reconstruction and expansion, constructing a simulation scenario of highway reconstruction and expansion, and fitting a regression decision model based on the traffic flow parameters in the simulation scenario, effectively finding the optimal rescue and evacuation plan in the accident scenario, and helping the traffic flow to return to normal operation. The plan and decision were simulated and verified to illustrate the effectiveness of the decision-making method. The present invention focuses on a dynamic system framework with high compatibility. It can provide rescue and evacuation decisions for reconstruction and expansion accident scenarios in various states, effectively improving rescue efficiency and reducing the impact of accidents on traffic.

Claims

1. A SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method, characterized in that: The steps include: Step 1: Based on the rescue and evacuation route plan during the highway reconstruction and expansion period, a dynamic plan combination decision model is established in combination with traffic flow operation parameters; Step 2: Use SUMO simulation software to perform traffic simulation during the highway reconstruction and expansion period, and simulate the traffic operation status of the reconstruction and expansion section under the traffic accident state; Step 3: Implement multiple rescue and evacuation routes in the reconstruction and expansion simulation, collect simulation data and perform regression fitting on the decision model, statistically analyze the simulation results and verify the effectiveness of the decision model.

2. The SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method as claimed in claim 1 is characterized in that: In step 1, based on the rescue and evacuation route plan during the highway reconstruction and expansion period, a dynamic scheme combination decision model is established in combination with traffic flow operation parameters, which specifically includes the following steps: Step 11: Analyze the conventional rescue routes in accident scenarios during the reconstruction and expansion period of the expressway, and determine feasible rescue routes based on the actual conditions of the road section, so that rescue vehicles can enter the expressway through the nearest interchanges upstream and downstream of the expressway and arrive at the accident site; Step 12: Analyze the conventional vehicle evacuation routes after the rescue operation in the accident scene, and determine feasible evacuation routes based on the actual rescue situation, so that stranded social vehicles can quickly leave the accident scene; Step 13: Establish a decision regression model. When an accident occurs and blocks all traffic in one direction, calculate the rescue time and road restoration time required for different evacuation routes and rescue routes based on the traffic status indicators at the time of the accident, and determine the optimal matching plan for the rescue route and evacuation route.

3. The SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method as claimed in claim 2 is characterized in that: In step 13, the decision regression model is established as follows: v ij =b ij,0 +b ij.1 X ij,2 +b ij,3 X ij,4 +...+b ij,n X ij,n +e Among them, X ij,1 ,X ij,2 ,...,X ij,n is the independent variable X ij The n eigencomponent matrices of ; and are predicted values, all calculated by the regression model; v ij and c ij are real values, which are obtained from a large amount of simulation data in the early stage; and is the average of all true values; β ij,0 ,β ij,1 ,...,β ij,n and γ ij,0 ,γ ij,1 ,...,γ ij,n is a model parameter vector, each component of which is a parameter of each combination of evacuation route and rescue route, which is the optimization object; the subscript ij represents the i-th evacuation plan combined with the j-th rescue plan.

4. The SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method as claimed in claim 1 is characterized in that: In step 2, SUMO simulation software is used to perform traffic simulation during the reconstruction and expansion of the expressway. The simulation of the traffic operation status of the reconstruction and expansion section under the traffic accident state specifically includes the following steps: Step 21, build a SUMO simulation environment, intercept the highway section based on the map database and import it into SUMO, use the netedit interface of SUMO to modify the attributes of the simulation section according to the actual situation of the reconstruction and expansion area, close the emergency lane and add the expansion road, collect relevant data of the highway reconstruction and expansion area, including traffic flow and vehicle speed, and build a basic simulation environment; Step 22: According to the emergency rescue and accident situation, determine the number of emergency rescue points and rescue vehicles, add the information of the emergency rescue points and rescue vehicles to the simulation environment, apply the path search algorithm to search for the shortest path from the emergency rescue point to the evacuation point in the road network, and evaluate the searched path, considering the road congestion and vehicle speed factors, and calculate the feasibility and efficiency of the path; Step 23: Preheat the simulation scene, set the accident point location, simulate the accident and implement relevant control measures; Step 24. According to the characteristics of the highway reconstruction and expansion area, in order to maintain consistency with normal traffic conditions, a special simulation road and a special driving route are built, and a reverse emergency lane is set on the inner side of the simulation road. The opening condition of the lane is set to that there is no vehicle on the corresponding forward lane, that is, all vehicles have been evacuated.

5. The SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method as claimed in claim 1 is characterized in that: In step 3, multiple rescue and evacuation routes are implemented in the reconstruction and expansion simulation, simulation data is collected and regression fitting is performed on the decision model, and the simulation results are statistically analyzed and the effectiveness of the decision model is verified. The specific steps include the following: Step 31: By giving the evacuation routes of the congested vehicles, the vehicles at the accident site are evacuated, and the upstream and downstream interchange traffic is controlled to simulate the evacuation process. By giving the number of rescue vehicles and the driving routes, the emergency rescue process is simulated; Step 32: Wait for the road to be cleared and end the simulation, and simultaneously collect simulation data, including emergency rescue time, evacuation time, and vehicle congestion; The simulation results were analyzed and evaluated based on three scenarios: two-way traffic in the opposite lane, U-turn on the central median, and two-way brief release. Step 33: Clean the simulation data, process missing values ​​and outliers, and perform normalization and dimensionality reduction processing; Step 34: Fit the regression model according to the obtained simulation data, use the least square method to establish a linear regression prediction model, obtain multiple groups of rescue time and road traffic restoration time under different conditions, and obtain the best linear regression function; Step 35: Evaluate the established regression model and verify the effectiveness of the decision model in a simulation environment. Under a given accident scenario, collect the accident scenario road traffic operation indicators, input the decision model to obtain the solution decision results, test it in the simulation scenario, and compare and verify it with the total time indicators of other solution combinations.

6. The SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method as claimed in claim 5 is characterized in that: In step 31, by giving the evacuation route of the congested vehicles, the vehicles at the accident point are evacuated, and the upstream and downstream interchange traffic is controlled to simulate the evacuation process. By giving the number of rescue vehicles and the driving route, the simulation of the emergency rescue process specifically includes the following steps: Step 311, determine the initial conditions of the route, set the rescue vehicle and the emergency evacuation point, define the rescue vehicle ID and the location of the emergency evacuation point in route.rou.xml; initialize the route information, configure the initial route for each rescue vehicle, and these routes are based on the pre-set traffic network and feasible driving paths; Step 312: Start the simulation using the Traci interface, and execute each simulation step repeatedly by looping until the set conditions are met or the rescue / evacuation is completed, and dynamically adjust the route of the rescue vehicle according to the real-time traffic conditions and road network status during the simulation process; Step 313: modify the vehicle driving settings to avoid the vehicle from detouring due to encountering a congested area or an accident scene; Step 314: By monitoring the arrival of the rescue vehicle at the key road node, it is determined whether the route needs to be adjusted. If the rescue vehicle is close to the accident site, it is necessary to adjust the further driving path according to the set rescue route, that is, to select one of upward forward rescue, downward forward rescue and downward reverse rescue for driving; Step 315: When the simulation runs to the set step length, the evacuation simulation begins, and the vehicles on the edge before the accident section drive according to the evacuation route; Step 316: The rescue vehicle drives away from the accident edge, and the vehicle drives on the accident edge, and the simulation runs normally until the step length ends.

7. A system based on the SUMO-based highway reconstruction and expansion simulation scenario construction and evacuation and rescue decision-making method as claimed in claim 1, characterized in that: include: Model building module, traffic simulation module and result verification module; The model building module is based on the rescue and evacuation route plans during the highway reconstruction and expansion period, and combines the traffic flow operation parameters to establish a dynamic plan combination decision model; the traffic simulation module uses SUMO simulation software to perform traffic simulation during the highway reconstruction and expansion period, simulating the traffic operation status of the reconstruction and expansion section under the condition of traffic accidents; the result verification module realizes multiple rescue and evacuation routes in the reconstruction and expansion simulation, collects simulation data and performs regression fitting on the decision model, statistically analyzes the simulation results and verifies the effectiveness of the decision model.

Images