A highway accident rescue control method and system fusing car-road cooperation and dual-mode traffic flow

Abstract

The application discloses a kind of fusion highway accident rescue control method and system of road cooperation and double mode traffic flow, first to the unified configuration of accident scene parameter and obtain the configured accident scene parameter, then the configured accident scene parameter is verified and generates scene initialization instruction set, then based on scene initialization instruction set, road network is abstracted as directed graph and initializes lane set and section set;The application realizes the function of accurately controlling the process of highway accident rescue by adopting the way of fusion road cooperation and double mode traffic flow, not only improves the scientific level of accident rescue and traffic control, but also produces multidimensional benefits in different groups and application fields, and through the implementation of multi-scheme deduction and quantitative evaluation before accident disposal, the rescue path and lane control strategy can be optimized, the response time is shortened, the congestion diffusion risk is reduced, and the emergency management efficiency and decision-making scientificity are significantly improved.

Description

Technical Field

[0001] This invention relates to the field of highway accident rescue and control technology, specifically to a highway accident rescue and control method and system that integrates vehicle-road cooperation and dual-mode traffic flow. Background Technology

[0002] Currently, existing highway accident rescue and control methods generally suffer from fragmented scenario modeling and incomplete link structures. This results in highway accident rescue and control schemes only being able to model the static lane occupation of accident vehicles or the single-path travel of rescue vehicles. This not only lacks a complete closed-loop depiction of the entire process but also fails to accurately reflect the coupling relationship between accident evolution and handling. Furthermore, existing highway accident rescue and control methods lack reusable mathematical and structured statistical mechanisms for core indicators, leading to strong subjectivity and poor comparability. Most existing methods do not incorporate vehicle-road cooperation conditions into a unified framework, making it impossible to depict the linkage between lane closures, dynamic traffic restrictions, temporary isolation facility deployment and removal infrastructure control measures, and rescue vehicle behavior. This results in insufficient reproduction of the actual control effect. Additionally, the traditional traffic flow generation in existing highway accident rescue and control methods primarily uses a single mode of fixed or random routes, making it difficult to reflect the mixed characteristics of focused traffic flow around the accident site and random traffic flow across the entire area. Therefore, it is necessary to design a highway accident rescue and control method and system that integrates vehicle-road cooperation and dual-mode traffic flow. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of existing technologies and to better and more effectively address the common problems in existing highway accident rescue and control methods, such as fragmented scenario modeling and incomplete links. These problems result in highway accident rescue and control schemes that can only model the static lane occupation of accident vehicles or the single-path travel of rescue vehicles, lacking a complete closed-loop depiction of the entire process and failing to accurately reflect the coupling relationship between accident evolution and handling. Furthermore, existing highway accident rescue and control methods lack reusable mathematical and structured statistical mechanisms for core indicators, leading to strong subjectivity and poor comparability in the schemes. Since most existing highway accident rescue and control methods do not incorporate vehicle-road cooperation conditions into a unified framework, they cannot characterize the relationship between infrastructure-side control measures such as lane closures, dynamic traffic restrictions, and the deployment and removal of temporary isolation facilities, and the behavior of rescue vehicles. The dynamic relationship between traffic flow and traffic flow in existing highway accident rescue and control methods leads to insufficient reproduction of the actual control effect. Furthermore, the traditional traffic flow generation in existing methods primarily uses a single mode of fixed or random routes, making it difficult to reflect the mixed characteristics of focused traffic flow around the accident site and random traffic flow throughout the entire area. This paper proposes a highway accident rescue and control method and system that integrates vehicle-road cooperation and dual-mode traffic flow. This method enables precise control of the highway accident rescue process using integrated vehicle-road cooperation and dual-mode traffic flow, not only improving the scientific level of accident rescue and traffic control but also generating multi-dimensional benefits across different groups and application areas. Moreover, by conducting multi-scheme simulations and quantitative assessments before accident handling, it can optimize rescue routes and lane control strategies, shorten response time, reduce the risk of congestion spread, and significantly improve emergency management efficiency and the scientific nature of decision-making.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A highway accident rescue and control method integrating vehicle-road cooperation and dual-mode traffic flow includes the following steps: Step A: Configure the accident scenario parameters in a unified manner and obtain the configured accident scenario parameters. Then, verify the configured accident scenario parameters and generate a scenario initialization instruction set. Step B: Based on the scene initialization instruction set, the road network is abstracted into a directed graph and the lane set and road segment set are initialized. Then, based on the initialized lane set and road segment set, a method for measuring the straight-line distance between vehicles and a method for measuring the distance along the road travel path are constructed. Step C: Based on the straight-line distance between vehicles, model the concentrated traffic flow affected by the accident near the accident site and generate the focused traffic flow within the accident-affected area. Then, based on the distance measurement along the road travel path, model the regional background traffic and generate random traffic flow across the entire road network. Step D: Based on the focused traffic flow and random traffic flow, the traffic flow intensity margin accumulation compensation mechanism is used to process the case where the expected number of vehicles generated in each step under the distance step length is not an integer, and the processed focused traffic flow and processed random traffic flow are obtained. Then, the probability of the vehicle type of the processed focused traffic flow and processed random traffic flow is set, and the vehicle type is generated based on multivariate distribution sampling. Step E: Construct the accident event triggering conditions and the accident state evolution process, and then, based on the accident event triggering conditions and the accident state evolution process, enter the accident response stage after the accident occurs to generate a rescue vehicle dispatch plan and rescue vehicle driving trajectory. Step F involves performing vehicle-road cooperative management on the road infrastructure side during the rescue process based on the processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan, and rescue vehicle travel trajectory, and obtaining the vehicle-road cooperative management results. Step G: Based on the vehicle-road cooperative management and control results, model the temporary work area at the accident site and generate a spatial model of the temporary work area at the accident site. Then, based on the spatial model of the temporary work area at the accident site, assess the traffic flow impact caused by the temporary work area at the accident site and obtain the traffic flow impact assessment results. Step H involves completing the accident scene rescue task based on the traffic flow impact assessment results and removing the rescue vehicles and accident vehicles from the accident scene. After the rescue vehicles and accident vehicles have left the accident scene, the road control is lifted and traffic operation index data is output, thereby completing the highway accident rescue control operation.

[0005] The aforementioned highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow includes step A, which involves uniformly configuring accident scenario parameters and obtaining the configured accident scenario parameters, then verifying the configured accident scenario parameters and generating a scenario initialization instruction set. The specific steps are as follows. Step A1: Configure the accident scenario parameters uniformly and obtain the configured accident scenario parameters. The accident scenario parameters include the accident road segment identifier, the longitudinal position of the accident point, the accident lane, the accident vehicle triggering conditions, the rescue vehicle departure position, the rescue vehicle delay time, the rescue vehicle operation time, the rescue path constraints, the cone deployment parameters, the vehicle-road cooperative switch, and the traffic flow parameters. Step A2 involves verifying the configured accident scenario parameters and generating a scenario initialization instruction set, as shown in formula (1). (1) in, The result of the parameter set consistency check, and only if After the accident scenario parameters are configured, they pass the verification. It is an indicator function, and the indicator function If the condition is met, the value is 1; otherwise, the indicator function is used. Set to 0; For the first Each scene parameter This represents the total number of scene parameters. and Parameters The minimum and maximum valid values.

[0006] The aforementioned highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow includes step B, which involves abstracting the road network into a directed graph based on the scenario initialization instruction set and initializing the lane set and road segment set. Then, based on the initialized lane set and road segment set, a method for measuring straight-line distance between vehicles and a method for measuring distance along the road travel path are constructed. The specific steps are as follows. Step B1: Based on the scenario initialization instruction set, the road network is abstracted into a directed graph and the lane set and road segment set are initialized, as shown in formula (2). (2) in, It is a directed graph. For the set of all nodes in the road network, It is a set of directed edges; Step B2 involves constructing a method for measuring straight-line distance between vehicles and a method for measuring distance along the road travel path based on the initialized lane set and road segment set. The specific steps are as follows: Step B21: Construct a method for measuring the straight-line distance between vehicles, wherein the method for measuring the straight-line distance between vehicles is used to determine close-range interaction and collision detection, as shown in formula (3). (3) in, For vehicles and vehicles The straight-line distance between vehicles. and vehicles x and y coordinates and vehicles The x and y coordinates; Step B22: Construct a method for measuring the distance along the road travel path. This method is used for calculating the estimated arrival time of rescue vehicles, controlling the length of the cone queue, and propagating congestion waves, as shown in formula (4). (4) in, For vehicles and vehicles The distance of the driving path along the road between them and vehicles and vehicles The section of road where it is located For from vehicles The section of road To the vehicle The section of road The shortest path length, and vehicles and vehicles The longitudinal position within the lane, For vehicles The section of road Its total physical length For path The Middle The path correction length for each turning node.

[0007] The aforementioned highway accident rescue and control method integrating vehicle-road cooperation and dual-mode traffic flow includes step C, which involves modeling the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance metric between vehicles and generating focused traffic flow within the accident-affected area. Then, it involves modeling the regional background traffic based on the distance metric along the road travel path and generating random traffic flow across the entire road network. The specific steps are as follows. Step C1 involves modeling the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance measurement method between vehicles, and generating the focused traffic flow within the accident-affected area, as shown in formula (5). (5) in, Let t be the intensity of the traffic flow. To focus on peak traffic intensity, It is a natural exponential function. The time of the accident. For peak traffic delay time, The standard deviation of the time distribution; Step C2 involves modeling the regional background traffic based on the distance measurement method along the road travel path and generating random traffic flow across the entire road network, as shown in formula (6). (6) in, Let be the intensity of the random traffic flow at time t. The average intensity of random traffic flow. For fluctuation coefficient, The start time, This refers to the traffic flow fluctuation cycle.

[0008] The aforementioned highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow, in step D, uses a traffic flow intensity margin accumulation compensation mechanism to process the case where the expected number of vehicles generated in each step is a non-integer under the discrete step length, based on focused traffic flow and random traffic flow, to obtain the processed focused traffic flow and processed random traffic flow. Then, probabilities are set for the vehicle types of the processed focused traffic flow and processed random traffic flow, and vehicle type is generated based on multinomial distribution sampling. The specific steps are as follows. Step D1: Based on the focused traffic flow and random traffic flow, for the case where the expected number of vehicles generated per step under the distance step length is not an integer, a traffic flow intensity margin accumulation compensation mechanism is used to process and obtain the processed focused traffic flow and processed random traffic flow. The specific traffic flow intensity margin accumulation compensation mechanism is shown in formula (7). (7) in, The actual number of vehicles generated in each step. The desired number of vehicles to generate per simulation step. To round down, To accumulate the margin, The expected number of vehicles to be generated per simulation step. The decimal part, To keep only the fractional part and discard the integer part for any real number, This represents the total number of vehicles generated during the entire cycle. Total duration; Step D2 involves setting probabilities for vehicle types in the processed focused traffic flow and the processed random traffic flow, and generating vehicle models based on multinomial distribution sampling, as shown in formulas (8) and (9). (8) (9) in, To generate vehicle models The probability, For vehicle types The generation weights, This represents the total number of vehicle models. Let m be the weight of the m-th type of vehicle. This is the vehicle category number.

[0009] The aforementioned highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow includes step E, which involves constructing accident event triggering conditions and the accident state evolution process. Based on these conditions, the method then enters the accident response phase after the accident occurs, generating a rescue vehicle dispatch plan and rescue vehicle trajectories. The specific steps are as follows: Step E1: Construct the accident event triggering conditions and the accident state evolution process. The specific steps are as follows: Step E11: Construct the accident event triggering conditions, as shown in formula (10). (10) in, For the accident trigger indication function, and when At that time, an accident was triggered. For the trigger position range, The longitudinal position of the vehicle involved in the accident. This is the starting value at the accident trigger location. The length of the trigger position interval, The speed of the vehicle involved in the accident. To trigger the speed threshold, This is a function of the geometric distance between the accident vehicle and the vehicle involved in the conflict. For vehicles involved in the conflict, Distance is used to determine collisions; Step E12: Construct the accident state evolution process. Specifically, write the accident event information into the system's event timeline, and then trigger the subsequent rescue dispatch, traffic control measures, and indicator statistics window switching. The accident triggering adopts a collision event subscription or callback mechanism for real-time detection. Step E2 involves generating a rescue vehicle dispatch plan and rescue vehicle trajectories based on the accident event triggering conditions and the accident state evolution process after the accident occurs, entering the accident response phase. During the approach phase, the lane-changing sequence planning and emergency lane utilization strategy for the rescue vehicles are calculated. In the exit phase, the rescue vehicles are guided away from the accident scene according to a pre-set exit plan. The specific steps are as follows. Step E21: Calculate the shortest feasible path from the starting point to the accident point for the rescue vehicle based on the starting position of the rescue vehicle, the location of the accident, the predetermined destination, the preset entry path constraints, and the preset exit path constraints. Then, form the rescue vehicle's entry route based on the shortest feasible path and evaluate the estimated arrival time of the rescue vehicle, as shown in formula (11). (11) in, The estimated arrival time of the rescue vehicle. The shortest feasible route distance for rescue vehicles from their departure point to the accident site. For rescue vehicles on the route The theoretical speed at that point, For path Traffic flow impact coefficient at the location, For the average theoretical speed, This is the average impact coefficient of the path; Step E22 involves planning the timing of lane changes for rescue vehicles, as shown in formula (12). (12) in, For the first The trigger time for the next lane change. For the rescue departure time, For the first Distance threshold for secondary lane changes The average speed of the rescue vehicles; Step E23: Set the arrival criteria for rescue vehicles. This includes determining the distance along the road from the rescue vehicle to the accident site. Less than the preset threshold When the time is right, it is determined that the rescue vehicle has arrived at the accident site and its status is switched to "operation in progress", as shown in formula (13). (13) in, For rescue vehicles, The location where the accident occurred. The speed limit for the section of highway where the accident occurred; Step E24: Upon arrival at the accident site, the rescue vehicle will perform on-site handling operations and continue the scheduled operation for the prescribed time. Once the operation is completed, the process of the rescue vehicle leaving the site is triggered.

[0010] The aforementioned highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow includes step F, which involves performing vehicle-road cooperative management on the road infrastructure side during the rescue process based on processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan, and rescue vehicle travel trajectory, and obtaining the vehicle-road cooperative management results. The specific steps are as follows. Step F1 involves creating a road condition database, specifically by grouping the lanes subject to temporary traffic restrictions. Record the original set of permitted vehicle categories for each lane before the traffic restriction begins. And form a filing matrix; Step F2, dynamic traffic restriction update, specifically, when the distance between the rescue vehicle and the accident site enters the predetermined threshold window, the allowed traffic category of the next target lane is temporarily updated to include only emergency vehicles, as shown in formulas (14) and (15). (14); (15) in, As the trigger threshold, To allow for advance planning, To control the set of temporary traffic permissions that are in effect for this lane during the implementation period. For the identification of rescue vehicles; Step F3: Calculate the priority coefficient for rescue vehicles, as shown in formula (16). (16) in, Let be the right-of-way coefficient, and let be the right-of-way coefficient. , This is the vehicle-road cooperative gain coefficient. For vehicle-road cooperative switching; Step F4, Road Status Restoration, wherein the road status restoration is used to trigger the restoration of permissions for the previously controlled lanes immediately after the rescue vehicle leaves the scene, specifically based on the backup. The matrix restores the permitted traffic categories for each lane to their original configuration.

[0011] The aforementioned highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow includes step G, which involves modeling the temporary work area at the accident site based on the vehicle-road cooperation control results and generating a spatial model of the temporary work area. Then, the impact of the temporary work area on traffic flow is assessed based on the spatial model, and the traffic flow impact assessment results are obtained. The specific steps are as follows. Step G1 involves modeling the temporary work area at the accident site based on the vehicle-road cooperative management results and generating a spatial model of the temporary work area at the accident site, specifically according to a specified length. and target lane assembly Automatic cone isolation zones are set up, and the work area is then abstracted into an occupied strip aligned with the accident point, thereby obtaining the spatial set covered by the work area, as shown in formula (17). (17) in, The set of spaces covered by the work area. This is the longitudinal distance measured from the point of the accident. Lane numbering, For a specified length, For the target lane set; Step G2 involves assessing the traffic flow impact of the temporary work area at the accident site based on the spatial model of the temporary work area and obtaining the traffic flow impact assessment results. The assessment process specifically involves evaluating both the macroscopic capacity model and the microscopic behavioral constraint model. The specific steps are as follows. Step G21 involves assessing the impact on traffic flow using a macroscopic capacity model. The specific steps are as follows. Step G211 introduces a driver caution coefficient and uses a basic traffic flow graph model to describe the relationship between speed, flow rate, and density, as shown in formulas (18) and (19). (18); (19) in, Traffic density is At that time, the balanced driving speed of vehicles on the road; For free flow velocity, For driver caution factor, For congestion density, This refers to the capacity of the road cross-section under normal conditions. Step G212: In cases where there are closed lanes or obstacles in the accident work area, a capacity reduction factor is introduced. The basic capacity is reduced to represent the degree of impairment of traffic capacity caused by the accident, as shown in formula (20). (20) in, The effective capacity of the accident area; Step G213 involves using a shock wave model to characterize the propagation speed of the vehicle queue and congestion wave caused by the accident, as shown in formula (21). (twenty one) in, For the propagation speed of congestion waves, This represents the normal traffic flow on the upstream section of the road where the accident occurred. The traffic density corresponding to the upstream section of the accident. This refers to the traffic flow at the cross-section of the accident work area. Traffic density in the accident congestion area; Step G22, the micro-behavioral constraint model, specifically reduces micro-traffic capacity at the vehicle level by restricting the behavior of vehicles entering the work area, and further describes the more cautious and conservative driving behavior of drivers in the accident section by adjusting following distance and safe distance.

[0012] The aforementioned highway accident rescue and control method integrating vehicle-road cooperation and dual-mode traffic flow includes step H, which involves completing the accident scene rescue task based on traffic flow impact assessment results and removing the rescue vehicles and accident vehicles from the accident scene. After the rescue vehicles and accident vehicles have left the accident scene, road control is lifted and traffic operation index data is output, thus completing the highway accident rescue and control operation. The specific steps are as follows. Step H1 involves completing the accident scene rescue mission based on the traffic flow impact assessment results and removing the rescue vehicles and accident vehicles from the accident scene. The specific steps are as follows: Step H11, departure trigger: When the duration of the rescue operation reaches the preset time, the departure stage of the rescue vehicle is triggered. At this time, the rescue vehicle has completed its on-site task and is ready to leave the accident scene. Step H12, the exit process, directs rescue vehicles to leave the accident section according to the pre-configured exit route strategy. The exit route strategy includes a lead-out mode and a tow truck towing mode. The lead-out mode uses rescue vehicles to slowly guide vehicles behind the accident to restore traffic flow. The tow truck towing mode uses rescue vehicles to tow the accident vehicles away from the scene and then drive away together. Step H13, Status Restoration: Immediately after the rescue vehicles have moved away from the accident area, the previously implemented lane restrictions and control measures are lifted, and the permitted vehicle categories for each controlled lane are restored to the configuration before the accident. Step H2 involves lifting road control and outputting traffic operation index data after the rescue traffic flow and accident vehicles have left the accident scene, thereby completing the highway accident rescue control operation. The traffic operation index data includes congestion intensity index, spatiotemporal average speed, queue length dynamic index, and traffic safety risk proxy index.

[0013] A highway accident rescue control system integrating vehicle-road cooperation and dual-mode traffic flow includes a parameter verification module, a distance measurement module, a traffic flow modeling module, a vehicle type generation module, a trajectory generation module, a collaborative control module, an impact assessment module, and a control release module. The parameter verification module is used to uniformly configure the accident scene parameters and obtain the configured accident scene parameters, and then verify the configured accident scene parameters and generate a scene initialization instruction set. The distance measurement module is used to abstract the road network into a directed graph based on the scene initialization instruction set and initialize the lane set and road segment set. Then, based on the initialized lane set and road segment set, it constructs a straight-line distance measurement method between vehicles and a distance measurement method along the road driving path. The traffic flow modeling module is used to model the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance between vehicles and generate the focused traffic flow within the accident-affected area. Then, it models the background traffic in the area based on the distance along the road travel path and generates random traffic flow across the entire road network. The vehicle model generation module is used to process the case where the expected number of vehicles generated in each step under the discrete step length is a non-integer by using a traffic flow intensity margin accumulation compensation mechanism based on focused traffic flow and random traffic flow to obtain the processed focused traffic flow and processed random traffic flow. Then, the probability of the vehicle type of the processed focused traffic flow and processed random traffic flow is set and the vehicle model is generated based on multivariate distribution sampling. The trajectory generation module is used to construct the accident event triggering conditions and the accident state evolution process, and then, based on the accident event triggering conditions and the accident state evolution process, to generate a rescue vehicle dispatch plan and rescue vehicle driving trajectory in the accident response stage after the accident occurs. The collaborative management module is used to perform vehicle-road collaborative management on the road infrastructure side during the rescue process based on the processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan and rescue vehicle driving trajectory, and to obtain the vehicle-road collaborative management results. The impact assessment module is used to model the temporary work area at the accident site based on the vehicle-road cooperative management and control results and generate a spatial model of the temporary work area at the accident site. Then, it assesses the traffic flow impact caused by the temporary work area at the accident site based on the spatial model of the temporary work area at the accident site and obtains the traffic flow impact assessment results. The control release module is used to complete the accident scene rescue task based on the traffic flow impact assessment results and remove the rescue traffic flow and accident vehicles from the accident scene. After the rescue traffic flow and accident vehicles leave the accident scene, the road control is released and traffic operation index data is output to complete the highway accident rescue control operation.

[0014] The beneficial effects of this invention are as follows: This invention provides a highway accident rescue control method and system integrating vehicle-road cooperation and dual-mode traffic flow. First, accident scenario parameters are uniformly configured and the configured accident scenario parameters are obtained. Then, the configured accident scenario parameters are verified and a scenario initialization instruction set is generated. Next, based on the scenario initialization instruction set, the road network is abstracted into a directed graph, and lane sets and road segment sets are initialized. Then, based on the initialized lane sets and road segment sets, a method for measuring straight-line distance between vehicles and a method for measuring distance along the road travel path are constructed. Subsequently, based on the method for measuring straight-line distance between vehicles, the concentrated traffic flow affected by the accident near the accident site is modeled, and a focused traffic flow within the accident-affected area is generated. Then, based on the method for measuring distance along the road travel path, the regional background traffic is modeled, and a random traffic flow across the entire road network is generated. Then, based on the focused traffic flow and random traffic flow, a traffic flow intensity margin accumulation compensation mechanism is used to process cases where the expected number of vehicles generated at each step under the dispersion step length is not an integer, resulting in processed focused traffic flow and processed random traffic flow. Finally, the processed focused traffic flow and processed random traffic flow are further processed... The system sets probabilities for vehicle types in the flow and generates vehicle models based on multivariate sampling. Next, it constructs accident event triggering conditions and the accident state evolution process. Then, based on these conditions, it enters the accident response phase after the accident, generating a rescue vehicle dispatch plan and rescue vehicle trajectories. Subsequently, based on the processed focused traffic flow, processed random traffic flow, vehicle models, rescue vehicle dispatch plan, and rescue vehicle trajectories, it performs vehicle-road cooperative management on the road infrastructure side during the rescue process and obtains the vehicle-road cooperative management results. Then, based on the vehicle-road cooperative management results, it models the temporary work area at the accident site and generates a spatial model of the temporary work area. Next, it assesses the traffic flow impact caused by the temporary work area based on the spatial model and obtains the traffic flow impact assessment results. Finally, based on the traffic flow impact assessment results, it completes the accident site rescue task and removes the rescue traffic flow and accident vehicles from the accident site. After the rescue traffic flow and accident vehicles leave the accident site, it lifts road control and outputs traffic operation index data, thus completing the highway accident rescue control operation.This highway accident rescue control method and system effectively achieves precise control of the highway accident rescue process by integrating vehicle-road cooperation and dual-mode traffic flow. It not only improves the scientific level of accident rescue and traffic management but also generates multi-dimensional benefits across different groups and application areas. Furthermore, by conducting multi-scenario simulations and quantitative assessments before accident handling, it optimizes rescue routes and lane control strategies, shortens response time, reduces the risk of congestion spread, and significantly improves emergency management efficiency and decision-making scientificity. Through capacity reduction analysis and queuing prediction models, it can rationally allocate emergency resources and work area layout schemes, reducing the impact of accidents on traffic capacity and minimizing operational losses. Simultaneously, by providing dynamic ETA assessment and collaborative priority passage mechanisms, it helps improve rescue arrival efficiency and on-site safety assurance levels, reducing delays caused by accidents and the risk of secondary accidents, thereby improving travel reliability and transportation timeliness. Attached Figure Description

[0015] Figure 1 This is an overall flowchart of a highway accident rescue and control method that integrates vehicle-road cooperation and dual-mode traffic flow according to the present invention. Figure 2 This is an overview diagram of a major accident scenario in the fourth lane, as described in an embodiment of the present invention. Figure 3 This is a schematic diagram of scheme G in the embodiment of the present invention where lane-level remote speed limiting is not enabled; Figure 4 This is a schematic diagram illustrating the configuration of lane-level remote speed limiting strategy in embodiment H of the present invention; Figure 5 This is a simulation trend diagram of the average speed and the number of vehicles in scheme G of the present invention; Figure 6 This is a simulation trend diagram of the average speed and number of vehicles in scheme H of the present invention. Detailed Implementation

[0016] The present invention will now be further described with reference to the accompanying drawings.

[0017] like Figure 1 As shown, the present invention provides a highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow, comprising the following steps: Step A involves uniformly configuring the accident scenario parameters and obtaining the configured accident scenario parameters. Then, the configured accident scenario parameters are validated, and a scenario initialization instruction set is generated. The specific steps are as follows. Step A1: Configure the accident scenario parameters uniformly and obtain the configured accident scenario parameters. The accident scenario parameters include the accident road segment identifier, the longitudinal position of the accident point, the accident lane, the accident vehicle triggering conditions, the rescue vehicle departure position, the rescue vehicle delay time, the rescue vehicle operation time, the rescue path constraints, the cone deployment parameters, the vehicle-road cooperative switch, and the traffic flow parameters. Step A2 involves verifying the configured accident scenario parameters and generating a scenario initialization instruction set, as shown in formula (1). (1) in, The result of the parameter set consistency check, and only if After the accident scenario parameters are configured, they pass the verification. It is an indicator function, and the indicator function If the condition is met, the value is 1; otherwise, the indicator function is used. Set to 0; For the first Each scene parameter This represents the total number of scene parameters. and Parameters The minimum and maximum valid values.

[0018] Step B involves abstracting the road network into a directed graph based on the scene initialization instruction set and initializing the lane set and road segment set. Then, based on the initialized lane set and road segment set, a method for measuring the straight-line distance between vehicles and a method for measuring the distance along the road travel path are constructed. The specific steps are as follows. Step B1: Based on the scenario initialization instruction set, the road network is abstracted into a directed graph and the lane set and road segment set are initialized, as shown in formula (2). (2) in, It is a directed graph. For the set of all nodes in the road network, It is a set of directed edges; Step B2 involves constructing a method for measuring straight-line distance between vehicles and a method for measuring distance along the road travel path based on the initialized lane set and road segment set. The specific steps are as follows: Step B21: Construct a method for measuring the straight-line distance between vehicles, wherein the method for measuring the straight-line distance between vehicles is used to determine close-range interaction and collision detection, as shown in formula (3). (3) in, For vehicles and vehicles The straight-line distance between vehicles. and vehicles x and y coordinates and vehicles The x and y coordinates; Step B22: Construct a method for measuring the distance along the road travel path. This method is used for calculating the estimated arrival time of rescue vehicles, controlling the length of the cone queue, and propagating congestion waves, as shown in formula (4). (4) in, For vehicles and vehicles The distance of the driving path along the road between them and vehicles and vehicles The section of road where it is located For from vehicles The section of road To the vehicle The section of road The shortest path length, and vehicles and vehicles The longitudinal position within the lane, For vehicles The section of road Its total physical length For path The Middle The path correction length for each turning node.

[0019] Step C involves modeling the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance metric between vehicles, generating focused traffic flow within the accident-affected area, and then modeling the background traffic in the area based on the distance metric along the road travel path, generating random traffic flow across the entire road network. The specific steps are as follows. Step C1 involves modeling the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance measurement method between vehicles, and generating the focused traffic flow within the accident-affected area, as shown in formula (5). (5) in, Let t be the intensity of the traffic flow. To focus on peak traffic intensity, It is a natural exponential function. The time of the accident. For peak traffic delay time, The standard deviation of the time distribution; Step C2 involves modeling the regional background traffic based on the distance measurement method along the road travel path and generating random traffic flow across the entire road network, as shown in formula (6). (6) in, Let be the intensity of the random traffic flow at time t. The average intensity of random traffic flow. For fluctuation coefficient, The start time, This refers to the traffic flow fluctuation cycle.

[0020] The dual-mode generation logic focusing on traffic flow plus random traffic flow, combined with traffic flow intensity margin accumulation compensation and multi-vehicle probability model, can restore the real traffic flow characteristics of local disturbance and global randomness in accident scenarios, and solve the problem of deviation between traditional single traffic flow mode and real scene. Step D involves using a traffic flow intensity margin accumulation compensation mechanism to process the case where the expected number of vehicles generated per step is a non-integer under the distance step length, based on the focused traffic flow and random traffic flow, to obtain the processed focused traffic flow and processed random traffic flow. Then, probabilities are set for the vehicle types of the processed focused traffic flow and processed random traffic flow, and vehicle models are generated based on multinomial distribution sampling. The specific steps are as follows: Step D1: Based on the focused traffic flow and random traffic flow, for the case where the expected number of vehicles generated per step under the distance step length is not an integer, a traffic flow intensity margin accumulation compensation mechanism is used to process and obtain the processed focused traffic flow and processed random traffic flow. The specific traffic flow intensity margin accumulation compensation mechanism is shown in formula (7). (7) in, The actual number of vehicles generated in each step. The desired number of vehicles to generate per simulation step. To round down, To accumulate the margin, The expected number of vehicles to be generated per simulation step. The decimal part, To keep only the fractional part and discard the integer part for any real number, This represents the total number of vehicles generated during the entire cycle. Total duration; The traffic flow intensity margin accumulation compensation mechanism can stably achieve accurate injection of any real number of traffic flow intensity within a discrete time step, reducing traffic flow fluctuation deviations caused by step size discretization.

[0021] Step D2 involves setting probabilities for vehicle types in the processed focused traffic flow and the processed random traffic flow, and generating vehicle models based on multinomial distribution sampling, as shown in formulas (8) and (9). (8); (9) in, To generate vehicle models The probability, For vehicle types The generation weights, This represents the total number of vehicle models. Let m be the weight of the m-th type of vehicle. This is the vehicle category number.

[0022] Step E involves constructing the accident event triggering conditions and the accident state evolution process. Based on these conditions and process, a rescue vehicle dispatch plan and rescue vehicle travel trajectories are generated during the accident response phase after the accident occurs. The specific steps are as follows: Step E1: Construct the accident event triggering conditions and the accident state evolution process. The specific steps are as follows: Step E11: Construct the accident event triggering conditions, as shown in formula (10). (10) in, For the accident trigger indication function, and when At that time, an accident was triggered. For the trigger position range, The longitudinal position of the vehicle involved in the accident. This is the starting value at the accident trigger location. The length of the trigger position interval, The speed of the vehicle involved in the accident. To trigger the speed threshold, This is a function of the geometric distance between the accident vehicle and the vehicle involved in the conflict. For vehicles involved in the conflict, Distance is used to determine collisions; Step E12: Construct the accident state evolution process. Specifically, write the accident event information into the system's event timeline, and then trigger the subsequent rescue dispatch, traffic control measures, and indicator statistics window switching. The accident triggering adopts a collision event subscription or callback mechanism for real-time detection. Step E2 involves generating a rescue vehicle dispatch plan and rescue vehicle trajectories based on the accident event triggering conditions and the accident state evolution process after the accident occurs, entering the accident response phase. During the approach phase, the lane-changing sequence planning and emergency lane utilization strategy for the rescue vehicles are calculated. In the exit phase, the rescue vehicles are guided away from the accident scene according to a pre-set exit plan. The specific steps are as follows. Step E21: Calculate the shortest feasible path from the starting point to the accident point for the rescue vehicle based on the starting position of the rescue vehicle, the location of the accident, the predetermined destination, the preset entry path constraints, and the preset exit path constraints. Then, form the rescue vehicle's entry route based on the shortest feasible path and evaluate the estimated arrival time of the rescue vehicle, as shown in formula (11). (11) in, The estimated arrival time of the rescue vehicle. The shortest feasible route distance for rescue vehicles from their departure point to the accident site. For rescue vehicles on the route The theoretical speed at that point, For path Traffic flow impact coefficient at the location, For the average theoretical speed, This is the average impact coefficient of the path; Step E22 involves planning the timing of lane changes for rescue vehicles, as shown in formula (12). (12) in, For the first The trigger time for the next lane change. For the rescue departure time, For the first Distance threshold for secondary lane changes The average speed of the rescue vehicles; Step E23: Set the arrival criteria for rescue vehicles. This includes determining the distance along the road from the rescue vehicle to the accident site. Less than the preset threshold When the time is right, it is determined that the rescue vehicle has arrived at the accident site and its status is switched to "operation in progress", as shown in formula (13). (13) in, For rescue vehicles, The location where the accident occurred. The speed limit for the section of highway where the accident occurred; Step E24: Upon arrival at the accident site, the rescue vehicle will perform on-site handling operations and continue the scheduled operation for the prescribed time. Once the operation is completed, the process of the rescue vehicle leaving the site is triggered.

[0023] Step F involves performing vehicle-road cooperative management on the road infrastructure side during the rescue process based on the processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan, and rescue vehicle travel trajectory, and obtaining the vehicle-road cooperative management results. The specific steps are as follows. Step F1 involves creating a road condition database, specifically by grouping the lanes subject to temporary traffic restrictions. Record the original set of permitted vehicle categories for each lane before the traffic restriction begins. And form a filing matrix; Step F2, dynamic traffic restriction update, specifically, when the distance between the rescue vehicle and the accident site enters the predetermined threshold window, the allowed traffic category of the next target lane is temporarily updated to include only emergency vehicles, as shown in formulas (14) and (15). (14); (15) in, As the trigger threshold, To allow for advance planning, To control the set of temporary traffic permissions that are in effect for this lane during the implementation period. For the identification of rescue vehicles; Dynamic traffic restriction updates are synchronously recorded in the backup matrix to ensure recovery in subsequent stages. By sequentially setting up emergency lanes in each lane ahead of the direction of travel of rescue vehicles, it ensures that rescue vehicles can quickly approach the accident scene; this clears other vehicles from the target lane in front of the rescue vehicles in advance, avoiding obstruction when the rescue vehicles change lanes; at the same time, when the traffic restriction measures are implemented, the original lane's passage permission is recorded in the backup matrix to ensure subsequent restoration; the traffic restriction can be triggered repeatedly for each lane that the rescue vehicle needs to change lanes, until the rescue vehicle successfully reaches the accident scene.

[0024] Step F3: Calculate the priority coefficient for rescue vehicles, as shown in formula (16). (16) in, Let be the right-of-way coefficient, and let be the right-of-way coefficient. , This is the vehicle-road cooperative gain coefficient. For vehicle-road cooperative switching; Step F4, Road Status Restoration, wherein the road status restoration is used to trigger the restoration of permissions for the previously controlled lanes immediately after the rescue vehicle leaves the scene, specifically based on the backup. The matrix restores the permitted traffic categories for each lane to their original configuration.

[0025] The restoration of road conditions ensures that road traffic organization returns to normal; through the closed-loop control mechanism of state backup, traffic restriction update and state restoration, the reversibility of control measures in the simulation is guaranteed, and the traffic order after the accident is handled is not affected.

[0026] Step G involves modeling the temporary work area at the accident site based on the vehicle-road cooperative management results and generating a spatial model of the temporary work area. Then, the traffic flow impact caused by the temporary work area is assessed based on this spatial model, and the traffic flow impact assessment results are obtained. The specific steps are as follows. Step G1 involves modeling the temporary work area at the accident site based on the vehicle-road cooperative management results and generating a spatial model of the temporary work area at the accident site, specifically according to a specified length. and target lane assembly Automatic cone isolation zones are set up, and the work area is then abstracted into an occupied strip aligned with the accident point, thereby obtaining the spatial set covered by the work area, as shown in formula (17). (17) in, The set of spaces covered by the work area. This is the longitudinal distance measured from the point of the accident. Lane numbering, For a specified length, For the target lane set; Step G2 involves assessing the traffic flow impact of the temporary work area at the accident site based on the spatial model of the temporary work area and obtaining the traffic flow impact assessment results. The assessment process specifically involves evaluating both the macroscopic capacity model and the microscopic behavioral constraint model. The specific steps are as follows. Step G21 involves assessing the impact on traffic flow using a macroscopic capacity model. The specific steps are as follows. Step G211 introduces a driver caution coefficient and uses a basic traffic flow graph model to describe the relationship between speed, flow rate, and density, as shown in formulas (18) and (19). (18); (19) in, Traffic density is At that time, the balanced driving speed of vehicles on the road; For free flow velocity, For driver caution factor, For congestion density, This refers to the capacity of the road cross-section under normal conditions. Step G212: In cases where there are closed lanes or obstacles in the accident work area, a capacity reduction factor is introduced. The basic capacity is reduced to represent the degree of impairment of traffic capacity caused by the accident, as shown in formula (20). (20) in, The effective capacity of the accident area; when This indicates no discount. The smaller the value, the greater the capacity loss caused by the accident; Step G213 involves using a shock wave model to characterize the propagation speed of the vehicle queue and congestion wave caused by the accident, as shown in formula (21). (twenty one) in, For the propagation speed of congestion waves, This represents the normal traffic flow on the upstream section of the road where the accident occurred. The traffic density corresponding to the upstream section of the accident. This refers to the traffic flow at the cross-section of the accident work area. Traffic density in the accident congestion area; If the calculation yields This indicates that the congestion wave travels at a speed of Propagation upstream helps assess the extent to which the accident site affects overall traffic flow.

[0027] Step G22, the micro-behavioral constraint model, specifically reduces micro-traffic capacity at the vehicle level by restricting the behavior of vehicles entering the work area, and further describes the more cautious and conservative driving behavior of drivers in the accident section by adjusting following distance and safe distance.

[0028] Step H involves completing the accident scene rescue mission based on the traffic flow impact assessment results, removing the rescue vehicles and accident vehicles from the accident scene, and then lifting road control measures and outputting traffic operation indicator data after the rescue vehicles and accident vehicles have left the accident scene, thereby completing the highway accident rescue control operation. The specific steps are as follows. Step H1 involves completing the accident scene rescue mission based on the traffic flow impact assessment results and removing the rescue vehicles and accident vehicles from the accident scene. The specific steps are as follows: Step H11, departure trigger: When the duration of the rescue operation reaches the preset time, the departure stage of the rescue vehicle is triggered. At this time, the rescue vehicle has completed its on-site task and is ready to leave the accident scene. Step H12, the exit process, directs rescue vehicles to leave the accident section according to the pre-configured exit route strategy. The exit route strategy includes a lead-out mode and a tow truck towing mode. The lead-out mode uses rescue vehicles to slowly guide vehicles behind the accident to restore traffic flow. The tow truck towing mode uses rescue vehicles to tow the accident vehicles away from the scene and then drive away together. Step H13, Status Restoration: Immediately after the rescue vehicles have moved away from the accident area, the previously implemented lane restrictions and control measures are lifted, and the permitted vehicle categories for each controlled lane are restored to the configuration before the accident. Meanwhile, if temporary facilities such as traffic cones were deployed, they were removed in sequence or removed by simulated vehicle loading. This ensured that road traffic conditions were restored to their pre-accident state, achieving reversibility of control measures and a smooth transition at the scene.

[0029] Step H2 involves lifting road control and outputting traffic operation index data after the rescue traffic flow and accident vehicles have left the accident scene, thereby completing the highway accident rescue control operation. The traffic operation index data includes congestion intensity index, spatiotemporal average speed, queue length dynamic index, and traffic safety risk proxy index.

[0030] A highway accident rescue control system integrating vehicle-road cooperation and dual-mode traffic flow includes a parameter verification module, a distance measurement module, a traffic flow modeling module, a vehicle type generation module, a trajectory generation module, a collaborative control module, an impact assessment module, and a control release module. The parameter verification module is used to uniformly configure the accident scene parameters and obtain the configured accident scene parameters, and then verify the configured accident scene parameters and generate a scene initialization instruction set. The distance measurement module is used to abstract the road network into a directed graph based on the scene initialization instruction set and initialize the lane set and road segment set. Then, based on the initialized lane set and road segment set, it constructs a straight-line distance measurement method between vehicles and a distance measurement method along the road driving path. The traffic flow modeling module is used to model the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance between vehicles and generate the focused traffic flow within the accident-affected area. Then, it models the background traffic in the area based on the distance along the road travel path and generates random traffic flow across the entire road network. The vehicle model generation module is used to process the case where the expected number of vehicles generated in each step under the discrete step length is a non-integer by using a traffic flow intensity margin accumulation compensation mechanism based on focused traffic flow and random traffic flow to obtain the processed focused traffic flow and processed random traffic flow. Then, the probability of the vehicle type of the processed focused traffic flow and processed random traffic flow is set and the vehicle model is generated based on multivariate distribution sampling. The trajectory generation module is used to construct the accident event triggering conditions and the accident state evolution process, and then, based on the accident event triggering conditions and the accident state evolution process, to generate a rescue vehicle dispatch plan and rescue vehicle driving trajectory in the accident response stage after the accident occurs. The collaborative management module is used to perform vehicle-road collaborative management on the road infrastructure side during the rescue process based on the processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan and rescue vehicle driving trajectory, and to obtain the vehicle-road collaborative management results. The impact assessment module is used to model the temporary work area at the accident site based on the vehicle-road cooperative management and control results and generate a spatial model of the temporary work area at the accident site. Then, it assesses the traffic flow impact caused by the temporary work area at the accident site based on the spatial model of the temporary work area at the accident site and obtains the traffic flow impact assessment results. The control release module is used to complete the accident scene rescue task based on the traffic flow impact assessment results and remove the rescue traffic flow and accident vehicles from the accident scene. After the rescue traffic flow and accident vehicles leave the accident scene, the road control is released and traffic operation index data is output to complete the highway accident rescue control operation.

[0031] To better illustrate the effects of the present invention, a specific embodiment of the control method and system of the present invention is described below.

[0032] like Figure 2 As shown, this embodiment focuses on a major accident in the fourth lane of a two-way ten-lane highway. By constructing a standardized experimental environment and setting up a core simulation task of graded speed limits for rescue sections, the effectiveness of intelligent rescue strategies that need to be focused on in this scenario is analyzed. In this way, the impact of different dynamic graded speed limit strategies on rescue efficiency and overall traffic efficiency of the road section is quantitatively evaluated.

[0033] (1) Input basic parameters; The basic parameter inputs for the major and serious accident in the fourth lane of this embodiment are shown in Table 1. The configuration parameter types include six types: road parameters, traffic flow parameters, accident parameters, control parameters, environmental parameters, and other parameters, as shown in Table 1. Table 1, Input Information Table for Scene Simulation in this Embodiment;

[0034] (2) Scenario rescue simulation configuration in this embodiment In the rescue simulation experiment of this embodiment, the core simulation task of graded speed limits for rescue road sections was set, and a control experiment design was used to carry out simulation configuration and analysis.

[0035] In the graded speed limit module for rescue road sections, the impact of graded speed limit strategies on rescue efficiency and the overall road network operation status is evaluated by comparing Scheme G (speed limit not enabled) and Scheme H (lane-level remote speed limit enabled). The rescue simulation configuration table for this embodiment is shown in Table 2.

[0036] Table 2, Scenario 3 Rescue Simulation Configuration Table;

[0037] (3) Scene simulation process and output in this embodiment Accident occurrence stage Following the accident, simulation tests were conducted on the graded speed limit module for the rescue route.

[0038] like Figure 3 As shown, Scheme G is to not activate the lane-level remote speed limit strategy: ① At the far upstream end of the accident, the information board issues an initial warning of "accident on the right 10 kilometers ahead"; ② At the near upstream end of the accident, the information board issues a secondary warning of "accident on the right 5 kilometers ahead"; ③ At the location of the accident, the information board issues a road closure notice of "the right lane ahead is closed", and at the same time, the 4th and 5th lanes where the accident is located are closed.

[0039] like Figure 4 As shown, Scheme H implements a lane-level remote speed limit strategy within 10 kilometers upstream, controlling traffic flow through lane-specific and gradient-based methods: ① In the 6-10 kilometer range upstream of the accident, lane 1 remains open, lanes 2-3 are limited to 100 km / h, and lanes 4-5 are limited to 80-100 km / h; ② In the 3-5 kilometer range upstream of the accident, the speed limit gradient is further tightened, with lanes 1-2 maintaining 100 km / h, lane 3 reduced to 80 km / h, and lanes 4-5 simultaneously reduced to 60 km / h; ③ In the area adjacent to the accident upstream, the control intensity reaches its peak, with lane 1 still maintaining 100 km / h, lanes 2 and 3 reduced to 80 km / h and 60 km / h respectively, while lanes 4-5, where the accident occurred, are closed, forming a closed loop from remote warning to near-end control.

[0040] Rescue vehicle entry phase Option G follows the rescue and control strategy of Option H during the accident phase. Rescue vehicles enter the highway from the right emergency lane and gradually reach the accident site in lane 4 to carry out rescue operations. Simulation results show that congestion occurs at 167 seconds, with the longest congestion distance being 270 meters.

[0041] Option H follows the rescue and control strategy of Option G during the accident phase. Rescue vehicles enter the highway from the right emergency lane and gradually reach the accident site in lane 4 to carry out rescue operations. Simulation results show that congestion occurs at 130 seconds, with the longest congestion distance being 110 meters.

[0042] Clearance and rescue phase During this phase, the established rescue and control strategy will continue to be implemented without adding new simulation variables. The simulation will only capture the continuous change pattern of traffic flow parameters with the duration of the 30-minute rescue operation after the closure of lanes 4 and 5.

[0043] Rescue vehicle departure phase Rescue control measures were gradually lifted, and rescue vehicles changed lanes downstream of the accident site, eventually reaching the right emergency lane to exit the road.

[0044] (4) Simulation conclusions of this embodiment like Figure 5 and Figure 6 As shown, lane-level remote speed limits are beneficial for high-volume accident rescue. ① In the short period after an accident, Option G, without speed limits, will not reduce average speed or congestion in a short time; Option H, with speed limits, reduces average speed more quickly, visually causing localized congestion. ② In the longer period after an accident, the average speed of Option G will drop rapidly at some point, and large-scale congestion is more likely to induce secondary traffic accidents. With the same traffic flow, the average speed of Option G is lower; the average speed of Option H drops more slowly, resulting in localized congestion but not large-scale congestion. ③ After the accident rescue is completed, the average speed of Option G recovers more slowly than Option H, meaning the congestion dissipates more slowly.

[0045] Remote speed limiting strategies have significant advantages in high-volume accident rescue. ① In the initial stage of an accident, Scheme G, which did not implement lane-level remote speed limiting, did not show a significant decrease in vehicle speed or congestion in the short term; while Scheme H, through lane-specific and tiered speed limiting 10 kilometers upstream, caused the traffic flow to gradually slow down in advance. Although this visually created a localized decrease in vehicle speed and congestion, it laid the foundation for the orderly management of subsequent traffic flow. ② In the longer period after the accident, Scheme G, due to sudden changes in near-end traffic flow, experienced a rapid drop in average vehicle speed in a short period, which could easily induce large-scale congestion and increase the risk of secondary accidents. Under the same traffic input conditions, the overall average vehicle speed was significantly lower; Scheme H, through continuous guidance via remote speed limiting, made the average vehicle speed decrease more gradually, resulting in only localized and controllable congestion and effectively avoiding the formation of large-scale congestion. ③ In the traffic recovery phase after the rescue, Scheme G, affected by the accumulation of previous congestion, had a slower average vehicle speed recovery rate than Scheme H, and its congestion dissipation efficiency was lower. Scheme H, thanks to the orderly management in the early stages, achieved a smoother traffic flow recovery.

[0046] In summary, the present invention provides a highway accident rescue control method and system integrating vehicle-road cooperation and dual-mode traffic flow. First, accident scenario parameters are uniformly configured to obtain the configured accident scenario parameters. Then, the configured accident scenario parameters are verified, and a scenario initialization instruction set is generated. Next, based on the scenario initialization instruction set, the road network is abstracted into a directed graph, and lane sets and road segment sets are initialized. Then, based on the initialized lane sets and road segment sets, a method for measuring straight-line distance between vehicles and a method for measuring distance along the road travel path are constructed. Subsequently, based on the method for measuring straight-line distance between vehicles, the concentrated traffic flow affected by the accident near the accident site is modeled, and a focused traffic flow within the accident-affected area is generated. Then, based on the method for measuring distance along the road travel path, the regional background traffic is modeled, and a random traffic flow across the entire road network is generated. Then, based on the focused traffic flow and random traffic flow, a traffic flow intensity margin accumulation compensation mechanism is used to process cases where the expected number of vehicles generated at each step under the dispersion step length is not an integer, resulting in processed focused traffic flow and processed random traffic flow. Finally, the processed focused traffic flow and processed random traffic flow are further processed... The system sets probabilities for vehicle types and generates vehicle models based on multivariate sampling. Next, it constructs accident event triggering conditions and accident state evolution processes. Then, based on these conditions and evolution processes, it enters the accident response phase after the accident, generating a rescue vehicle dispatch plan and rescue vehicle trajectories. Subsequently, based on processed focused traffic flow, processed random traffic flow, vehicle models, the rescue vehicle dispatch plan, and the rescue vehicle trajectories, it performs vehicle-road cooperative management on the road infrastructure side during the rescue process and obtains the results. Then, based on the vehicle-road cooperative management results, it models the temporary work area at the accident site and generates a spatial model of the temporary work area. Next, it assesses the traffic flow impact caused by the temporary work area based on the spatial model and obtains the traffic flow impact assessment results. Finally, based on the traffic flow impact assessment results, it completes the accident site rescue task and removes the rescue vehicles and accident vehicles from the accident site. After the rescue vehicles and accident vehicles leave the accident site, it lifts road control and outputs traffic operation index data, thus completing the highway accident rescue control operation.This highway accident rescue control method and system effectively achieves precise control of the highway accident rescue process by integrating vehicle-road cooperation and dual-mode traffic flow. It not only improves the scientific level of accident rescue and traffic management but also generates multi-dimensional benefits across different groups and application areas. Furthermore, by conducting multi-scenario simulations and quantitative assessments before accident handling, it optimizes rescue routes and lane control strategies, shortens response time, reduces the risk of congestion spread, and significantly improves emergency management efficiency and decision-making scientificity. Through capacity reduction analysis and queuing prediction models, it can rationally allocate emergency resources and work area layout schemes, reducing the impact of accidents on traffic capacity and minimizing operational losses. Simultaneously, by providing dynamic ETA assessment and collaborative priority passage mechanisms, it helps improve rescue arrival efficiency and on-site safety assurance levels, reducing delays caused by accidents and the risk of secondary accidents, thereby improving travel reliability and transportation timeliness.

[0047] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A highway accident rescue and control method integrating vehicle-road cooperation and dual-mode traffic flow, characterized in that: Includes the following steps, Step A: Configure the accident scenario parameters in a unified manner and obtain the configured accident scenario parameters. Then, verify the configured accident scenario parameters and generate a scenario initialization instruction set. Step B: Based on the scene initialization instruction set, the road network is abstracted into a directed graph and the lane set and road segment set are initialized. Then, based on the initialized lane set and road segment set, a method for measuring the straight-line distance between vehicles and a method for measuring the distance along the road travel path are constructed. Step C: Based on the straight-line distance between vehicles, model the concentrated traffic flow affected by the accident near the accident site and generate the focused traffic flow within the accident-affected area. Then, based on the distance measurement along the road travel path, model the regional background traffic and generate random traffic flow across the entire road network. Step D: Based on the focused traffic flow and random traffic flow, the traffic flow intensity margin accumulation compensation mechanism is used to process the case where the expected number of vehicles generated in each step under the distance step length is not an integer, and the processed focused traffic flow and processed random traffic flow are obtained. Then, the probability of the vehicle type of the processed focused traffic flow and processed random traffic flow is set, and the vehicle type is generated based on multivariate distribution sampling. Step E involves constructing the accident event triggering conditions and the accident state evolution process. Based on these conditions and process, a rescue vehicle dispatch plan and rescue vehicle travel trajectories are generated during the accident response phase after the accident occurs. The specific steps are as follows: Step E1: Construct the accident event triggering conditions and the accident state evolution process. The specific steps are as follows: Step E11: Construct the accident event triggering conditions, as shown in formula (10). （10） in, For the accident trigger indication function, and when At that time, an accident was triggered. For the trigger position range, The longitudinal position of the vehicle involved in the accident. This is the starting value at the accident trigger location. The length of the trigger position interval, The speed of the vehicle involved in the accident. To trigger the speed threshold, This is a function of the geometric distance between the accident vehicle and the vehicle involved in the conflict. For vehicles involved in the conflict, Distance is used to determine collisions; Step E12: Construct the accident state evolution process. Specifically, write the accident event information into the system's event timeline, and then trigger the subsequent rescue dispatch, traffic control measures, and indicator statistics window switching. The accident triggering adopts a collision event subscription or callback mechanism for real-time detection. Step E2 involves generating a rescue vehicle dispatch plan and rescue vehicle trajectories based on the accident event triggering conditions and the accident state evolution process after the accident occurs, entering the accident response phase. During the approach phase, the lane-changing sequence planning and emergency lane utilization strategy for the rescue vehicles are calculated. In the exit phase, the rescue vehicles are guided away from the accident scene according to a pre-set exit plan. The specific steps are as follows. Step E21: Calculate the shortest feasible path from the starting point to the accident point for the rescue vehicle based on the starting position of the rescue vehicle, the location of the accident, the predetermined destination, the preset entry path constraints, and the preset exit path constraints. Then, form the rescue vehicle's entry route based on the shortest feasible path and evaluate the estimated arrival time of the rescue vehicle, as shown in formula (11). （11） in, The estimated arrival time of the rescue vehicle. The shortest feasible route distance for rescue vehicles from their departure point to the accident site. For rescue vehicles on the route The theoretical speed at that point, For path Traffic flow impact coefficient at the location, For the average theoretical speed, This represents the average impact coefficient of the path. Step E22 involves planning the timing of lane changes for rescue vehicles, as shown in formula (12). （12） in, For the first The trigger time for the next lane change. For the rescue departure time, For the first Distance threshold for secondary lane changes The average speed of the rescue vehicles; Step E23: Set the arrival criteria for rescue vehicles. This includes determining the distance along the road from the rescue vehicle to the accident site. Less than the preset threshold When the time is right, it is determined that the rescue vehicle has arrived at the accident site and its status is switched to "operation in progress", as shown in formula (13). （13） in, For rescue vehicles, The location where the accident occurred. The speed limit for the section of highway where the accident occurred; Step E24: Upon arrival at the accident site, the rescue vehicle will perform on-site handling operations and continue the scheduled operation for the prescribed time. Once the operation is completed, the departure procedure for the rescue vehicle is triggered; Step F involves performing vehicle-road cooperative management on the road infrastructure side during the rescue process based on the processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan, and rescue vehicle travel trajectory, and obtaining the vehicle-road cooperative management results. Step G: Based on the vehicle-road cooperative management and control results, model the temporary work area at the accident site and generate a spatial model of the temporary work area at the accident site. Then, based on the spatial model of the temporary work area at the accident site, assess the traffic flow impact caused by the temporary work area at the accident site and obtain the traffic flow impact assessment results. Step H involves completing the accident scene rescue task based on the traffic flow impact assessment results and removing the rescue vehicles and accident vehicles from the accident scene. After the rescue vehicles and accident vehicles have left the accident scene, the road control is lifted and traffic operation index data is output, thereby completing the highway accident rescue control operation.

2. The highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow as described in claim 1, characterized in that: Step A involves uniformly configuring the accident scenario parameters and obtaining the configured accident scenario parameters. Then, the configured accident scenario parameters are validated, and a scenario initialization instruction set is generated. The specific steps are as follows. Step A1: Configure the accident scenario parameters uniformly and obtain the configured accident scenario parameters. The accident scenario parameters include the accident road segment identifier, the longitudinal position of the accident point, the accident lane, the accident vehicle triggering conditions, the rescue vehicle departure position, the rescue vehicle delay time, the rescue vehicle operation time, the rescue path constraints, the cone deployment parameters, the vehicle-road cooperative switch, and the traffic flow parameters. Step A2 involves verifying the configured accident scenario parameters and generating a scenario initialization instruction set, as shown in formula (1). （1） in, The result of the parameter set consistency check, and only if After the accident scenario parameters are configured, they pass the verification. It is an indicator function, and the indicator function If the condition is met, the value is 1; otherwise, the indicator function is used. Set to 0; For the first Each scene parameter This represents the total number of scene parameters. and Parameters The minimum and maximum valid values.

3. The highway accident rescue and control method integrating vehicle-road cooperation and dual-mode traffic flow according to claim 2, characterized in that: Step B involves abstracting the road network into a directed graph based on the scene initialization instruction set and initializing the lane set and road segment set. Then, based on the initialized lane set and road segment set, a method for measuring straight-line distance between vehicles and a method for measuring distance along the road travel path are constructed. The specific steps are as follows. Step B1: Based on the scenario initialization instruction set, the road network is abstracted into a directed graph and the lane set and road segment set are initialized, as shown in formula (2). （2） in, It is a directed graph. For the set of all nodes in the road network, It is a set of directed edges; Step B2 involves constructing a method for measuring straight-line distance between vehicles and a method for measuring distance along the road travel path based on the initialized lane set and road segment set. The specific steps are as follows: Step B21: Construct a method for measuring the straight-line distance between vehicles, wherein the method for measuring the straight-line distance between vehicles is used to determine close-range interaction and collision detection, as shown in formula (3). （3） in, For vehicles and vehicles The straight-line distance between vehicles. and vehicles x and y coordinates and vehicles The x and y coordinates; Step B22: Construct a method for measuring the distance along the road travel path. This method is used for calculating the estimated arrival time of rescue vehicles, controlling the length of the cone queue, and propagating congestion waves, as shown in formula (4). （4） in, For vehicles and vehicles The distance of the driving path along the road between them and vehicles and vehicles The section of road where it is located For from vehicles The section of road To the vehicle The section of road The shortest path length, and vehicles and vehicles The longitudinal position within the lane, For vehicles The section of road Its total physical length For path The Middle The path correction length for each turning node.

4. The highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow according to claim 3, characterized in that: Step C involves modeling the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance metric between vehicles, generating focused traffic flow within the accident-affected area, and then modeling the background traffic in the area based on the distance metric along the road travel path, generating random traffic flow across the entire road network. The specific steps are as follows. Step C1 involves modeling the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance measurement method between vehicles, and generating the focused traffic flow within the accident-affected area, as shown in formula (5). （5） in, Let t be the intensity of the traffic flow. To focus on peak traffic intensity, It is a natural exponential function. The time of the accident. For peak traffic delay time, The standard deviation of the time distribution; Step C2 involves modeling the regional background traffic based on the distance measurement method along the road travel path and generating random traffic flow across the entire road network, as shown in formula (6). （6） in, Let be the intensity of the random traffic flow at time t. The average intensity of random traffic flow. For fluctuation coefficient, The start time, This refers to the traffic flow fluctuation cycle.

5. The highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow according to claim 4, characterized in that: Step D involves using a traffic flow intensity margin accumulation compensation mechanism to process the case where the expected number of vehicles generated per step is a non-integer under the distance step length, based on the focused traffic flow and random traffic flow, to obtain the processed focused traffic flow and processed random traffic flow. Then, probabilities are set for the vehicle types of the processed focused traffic flow and processed random traffic flow, and vehicle models are generated based on multinomial distribution sampling. The specific steps are as follows: Step D1: Based on the focused traffic flow and random traffic flow, for the case where the expected number of vehicles generated per step under the distance step length is not an integer, a traffic flow intensity margin accumulation compensation mechanism is used to process and obtain the processed focused traffic flow and processed random traffic flow. The specific traffic flow intensity margin accumulation compensation mechanism is shown in formula (7). （7） in, The actual number of vehicles generated in each step. The desired number of vehicles to generate per simulation step. To round down, To accumulate the margin, The expected number of vehicles to be generated per simulation step. The decimal part, To keep only the fractional part and discard the integer part for any real number, This represents the total number of vehicles generated during the entire cycle. Total duration; Step D2 involves setting probabilities for vehicle types in the processed focused traffic flow and the processed random traffic flow, and generating vehicle models based on multinomial distribution sampling, as shown in formulas (8) and (9). （8）； （9） in, To generate vehicle models The probability, For vehicle types The generation weights, This represents the total number of vehicle models. Let m be the weight of the m-th type of vehicle. This is the vehicle category number.

6. The highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow according to claim 5, characterized in that: Step F involves performing vehicle-road cooperative management on the road infrastructure side during the rescue process based on the processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan, and rescue vehicle travel trajectory, and obtaining the vehicle-road cooperative management results. The specific steps are as follows. Step F1 involves creating a road condition database, specifically by grouping the lanes subject to temporary traffic restrictions. Record the original set of permitted vehicle categories for each lane before the traffic restriction begins. And form a filing matrix; Step F2, dynamic traffic restriction update, specifically, when the distance between the rescue vehicle and the accident site enters the predetermined threshold window, the allowed traffic category of the next target lane is temporarily updated to include only emergency vehicles, as shown in formulas (14) and (15). （14）； （15） in, As the trigger threshold, To allow for advance planning, To control the set of temporary traffic permissions that are in effect for this lane during the implementation period. For the identification of rescue vehicles; Step F3: Calculate the priority coefficient for rescue vehicles, as shown in formula (16). （16） in, Let be the right-of-way coefficient, and let be the right-of-way coefficient. , This is the vehicle-road cooperative gain coefficient. For vehicle-road cooperative switching; Step F4, Road Status Restoration, wherein the road status restoration is used to trigger the restoration of permissions for the previously controlled lanes immediately after the rescue vehicle leaves the scene, specifically based on the backup. The matrix restores the permitted traffic categories for each lane to their original configuration.

7. The highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow according to claim 6, characterized in that: Step G involves modeling the temporary work area at the accident site based on the vehicle-road cooperative management results and generating a spatial model of the temporary work area. Then, the traffic flow impact caused by the temporary work area is assessed based on this spatial model, and the traffic flow impact assessment results are obtained. The specific steps are as follows. Step G1 involves modeling the temporary work area at the accident site based on the vehicle-road cooperative management results and generating a spatial model of the temporary work area at the accident site, specifically according to a specified length. and target lane assembly Automatic cone isolation zones are set up, and the work area is then abstracted into an occupied strip aligned with the accident point, thereby obtaining the spatial set covered by the work area, as shown in formula (17). （17） in, The set of spaces covered by the work area. This is the longitudinal distance measured from the point of the accident. Lane numbering, For a specified length, For the target lane set; Step G2 involves assessing the traffic flow impact of the temporary work area at the accident site based on the spatial model of the temporary work area and obtaining the traffic flow impact assessment results. The assessment process specifically involves evaluating both the macroscopic capacity model and the microscopic behavioral constraint model. The specific steps are as follows. Step G21 involves assessing the impact on traffic flow using a macroscopic capacity model. The specific steps are as follows. Step G211 introduces a driver caution coefficient and uses a basic traffic flow graph model to describe the relationship between speed, flow rate, and density, as shown in formulas (18) and (19). （18）； （19） in, Traffic density is At that time, the balanced driving speed of vehicles on the road; For free flow velocity, For driver caution factor, For congestion density, This refers to the capacity of the road cross-section under normal conditions. Step G212: In cases where there are closed lanes or obstacles in the accident work area, a capacity reduction factor is introduced. The basic capacity is reduced to represent the degree of impairment of traffic capacity caused by the accident, as shown in formula (20). （20） in, The effective capacity of the accident area; Step G213 involves using a shock wave model to characterize the propagation speed of the vehicle queue and congestion wave caused by the accident, as shown in formula (21). （21） in, For the propagation speed of congestion waves, This represents the normal traffic flow on the upstream section of the road where the accident occurred. The traffic density corresponding to the upstream section of the accident. This refers to the traffic flow at the cross-section of the accident work area. Traffic density in the accident congestion area; Step G22, the micro-behavioral constraint model, specifically reduces micro-traffic capacity at the vehicle level by restricting the behavior of vehicles entering the work area, and further describes the more cautious and conservative driving behavior of drivers in the accident section by adjusting following distance and safe distance.

8. The highway accident rescue control method integrating vehicle-road cooperation and dual-mode traffic flow according to claim 7, characterized in that: Step H involves completing the accident scene rescue mission based on the traffic flow impact assessment results, removing the rescue vehicles and accident vehicles from the accident scene, and then lifting road control measures and outputting traffic operation indicator data after the rescue vehicles and accident vehicles have left the accident scene, thereby completing the highway accident rescue control operation. The specific steps are as follows. Step H1 involves completing the accident scene rescue mission based on the traffic flow impact assessment results and removing the rescue vehicles and accident vehicles from the accident scene. The specific steps are as follows: Step H11, departure trigger: When the duration of the rescue operation reaches the preset time, the departure stage of the rescue vehicle is triggered. At this time, the rescue vehicle has completed its on-site task and is ready to leave the accident scene. Step H12, the exit process, directs rescue vehicles to leave the accident section according to the pre-configured exit route strategy. The exit route strategy includes a lead-out mode and a tow truck towing mode. The lead-out mode uses rescue vehicles to slowly guide vehicles behind the accident to restore traffic flow. The tow truck towing mode uses rescue vehicles to tow the accident vehicles away from the scene and then drive away together. Step H13, Status Restoration: Immediately after the rescue vehicles have moved away from the accident area, the previously implemented lane restrictions and control measures are lifted, and the permitted vehicle categories for each controlled lane are restored to the configuration before the accident. Step H2 involves lifting road control and outputting traffic operation index data after the rescue traffic flow and accident vehicles have left the accident scene, thereby completing the highway accident rescue control operation. The traffic operation index data includes congestion intensity index, spatiotemporal average speed, queue length dynamic index, and traffic safety risk proxy index.

9. A highway accident rescue control system integrating vehicle-road cooperation and dual-mode traffic flow, wherein the specific rescue control process of the highway accident rescue control system is based on the highway accident rescue control method according to any one of claims 1-8, characterized in that: It includes a parameter verification module, a distance measurement module, a traffic flow modeling module, a vehicle type generation module, a trajectory generation module, a collaborative management and control module, an impact assessment module, and a management and control deactivation module. The parameter verification module is used to uniformly configure the accident scene parameters and obtain the configured accident scene parameters, and then verify the configured accident scene parameters and generate a scene initialization instruction set. The distance measurement module is used to abstract the road network into a directed graph based on the scene initialization instruction set and initialize the lane set and road segment set. Then, based on the initialized lane set and road segment set, it constructs a straight-line distance measurement method between vehicles and a distance measurement method along the road driving path. The traffic flow modeling module is used to model the concentrated traffic flow affected by the accident near the accident site based on the straight-line distance between vehicles and generate the focused traffic flow within the accident-affected area. Then, it models the background traffic in the area based on the distance along the road travel path and generates random traffic flow across the entire road network. The vehicle model generation module is used to process the case where the expected number of vehicles generated in each step under the discrete step length is a non-integer by using a traffic flow intensity margin accumulation compensation mechanism based on focused traffic flow and random traffic flow to obtain the processed focused traffic flow and processed random traffic flow. Then, the probability of the vehicle type of the processed focused traffic flow and processed random traffic flow is set and the vehicle model is generated based on multivariate distribution sampling. The trajectory generation module is used to construct the accident event triggering conditions and the accident state evolution process, and then, based on the accident event triggering conditions and the accident state evolution process, to generate a rescue vehicle dispatch plan and rescue vehicle driving trajectory in the accident response stage after the accident occurs. The collaborative management module is used to perform vehicle-road collaborative management on the road infrastructure side during the rescue process based on the processed focused traffic flow, processed random traffic flow, vehicle type, rescue vehicle dispatch plan and rescue vehicle driving trajectory, and to obtain the vehicle-road collaborative management results. The impact assessment module is used to model the temporary work area at the accident site based on the vehicle-road cooperative management results and generate a spatial model of the temporary work area at the accident site. Then, it assesses the traffic flow impact caused by the temporary work area at the accident site based on the spatial model of the temporary work area at the accident site and obtains the traffic flow impact assessment results. The control release module is used to complete the accident scene rescue task based on the traffic flow impact assessment results and remove the rescue vehicles and accident vehicles from the accident scene. After the rescue vehicles and accident vehicles leave the accident scene, the road control is released and traffic operation index data is output to complete the highway accident rescue control operation.

Images