Dynamic routing method and device for vehicle-mounted self-organizing network in C-V2X environment

Abstract

The present invention discloses a method and device for dynamic routing of a vehicle self-organizing network in a C‑V2X environment, the method comprising the following steps: S01. constructing a routing model of a vehicle self-organizing network based on a genetic algorithm model according to the parameters of the current vehicle self-organizing network, and using connectivity probability and connection stability index as evaluation parameters of the fitness function; S02. receiving the driving status information of each vehicle node uploaded by each vehicle-mounted C‑V2XOBU device to obtain the current network real-time dynamic topology; S03. inputting the current network real-time dynamic topology into the routing model to perform optimal node sequence analysis and generate a dynamic routing table; S04. when a vehicle node exits the network, re-triggering the network for routing selection. The present invention has the advantages of simple implementation method, high stability and reliability, strong robustness, low resource overhead, and the ability to achieve rapid network convergence and global search.

Description

Technical Field

[0001] The present invention relates to the field of vehicle networking communication technology, and in particular to a method and device for dynamic routing of a vehicle-mounted self-organizing network in a C-V2X (Cellular-Vehicle to Everything) environment. Background Art

[0002] IoV communications include V2V between vehicles, V2I between vehicles and roads, V2P between vehicles and people, and V2N between vehicles and networks, which have stringent requirements such as low latency and high reliability. Based on the environmental perception, information interaction and collaborative control capabilities provided by V2X communications, IoV can support applications in various scenarios such as traffic safety, traffic efficiency, autonomous driving, and infotainment. Compared with traditional network communications, IoV communications have increasingly stringent requirements on communication performance, from telematics to vehicle-road collaboration, from single-vehicle intelligence to networked intelligence, which are manifested in the need for lower latency, higher reliability, and higher rates. For example, the message sending frequency changes from 10Hz for traditional basic safety services to 100Hz for autonomous driving applications, and the message size changes from the traditional 300-400Byte to the required 6000-12000Byte, etc.

[0003] Traditional mobile ad hoc networks (MANETs) all adopt a communication mode of multi-hop transmission between nodes, but in vehicle ad hoc networks (VANETs), mobile vehicles are used as network nodes, which have the following characteristics: ① The node moves fast, the multi-hop connection formed by self-organization is unstable, and the service quality is difficult to guarantee; ② The network topology changes dynamically. The network topology of the Internet of Vehicles has been changing all the time, which has caused great difficulties in the implementation of information transmission between vehicles; ③ Vehicle nodes can only travel along the road, so the data packet transmission can only move forward in the direction of the road. Therefore, traditional routing algorithms for mobile ad hoc networks MANETs (such as AODV, GPSR, etc.) cannot effectively adapt to the environment of vehicle ad hoc networks (VANETs), especially in the scenario of high node density of vehicle ad hoc networks. The routing reliability is low and the resource overhead is large, so it is difficult to take into account both routing performance and resource overhead. Therefore, it is urgent to provide a dynamic routing method for vehicle ad hoc networks that can adapt to C-V2X scenarios, so that both routing performance and resource overhead can be taken into account in the scenario of high node density or low node density of vehicle ad hoc networks. Summary of the invention

[0004] The technical problem to be solved by the present invention is: in response to the technical problems existing in the prior art, the present invention provides a method and device for dynamic routing of a vehicle-mounted self-organizing network in a C-V2X environment, which has a simple implementation method, high stability and reliability, strong robustness and low resource overhead, and can well match the vehicle-mounted self-organizing network in the C-V2X environment to achieve high-performance dynamic routing, and can also achieve rapid network convergence and global search.

[0005] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0006] A dynamic routing method for a vehicle-mounted self-organizing network in a C-V2X environment, comprising the following steps:

[0007] S01. Constructing a routing model of a vehicle self-organizing network based on a genetic algorithm model according to the parameters of the current vehicle self-organizing network, wherein the routing model uses a genetic algorithm to iteratively select the optimal node, and uses a connectivity probability and a connection stability index as evaluation parameters of a fitness function, wherein the connection stability index is calculated based on the statistical value of the node's historical connection interruption probability;

[0008] S02. Receive the driving status information of each vehicle node uploaded by each on-board C-V2X OBU device to obtain the real-time dynamic topology of the current network;

[0009] S03. The current real-time dynamic network topology is input into the routing model for optimal node sequence analysis to generate a dynamic routing table;

[0010] S04. When a vehicle node exits the network, return to step S02 to re-trigger the network for routing selection.

[0011] Furthermore, the driving status information of the vehicle node includes any one or more of position coordinates, speed, heading, and acceleration.

[0012] Furthermore, in the step S02, the real-time dynamic topology of the current network is formed by mapping the acquired driving status information of each vehicle node and the node information of each vehicle node onto the LDM map.

[0013] Furthermore, the connectivity probability is calculated based on the position p of each vehicle within the specified range of the road at the current time tc and the distance d between two adjacent vehicles. ij The probability F is less than the communication distance R ij Calculated.

[0014] Furthermore, according to the position p of each vehicle and the distance d between any two adjacent vehicles, ij The probability distribution state is obtained by ij The calculation expression is:

[0015]

[0016] Among them, μ pi , μ pj are the means corresponding to the normal distribution of the positions of the i-th and j-th vehicle nodes, are the variances corresponding to the normal distribution of the locations of the i-th and j-th vehicle nodes respectively.

[0017] Furthermore, the vehicle position p follows the mean μ corresponding to the normal distribution p With variance Calculate according to the following formula:

[0018]

[0019]

[0020] Among them, t c ,t r With t a are the current time, record update time, vehicle travel time, μ fv is the recorded mean, a is the vehicle node acceleration, is the variance at the previous moment, and the initial value of the variance is 1, which conforms to the standard normal distribution.

[0021] Furthermore, in the routing model in step S03, the step of inputting the current network real-time dynamic topology into the optimal node sequence analysis includes:

[0022] S301. Randomly generate an initial population based on the current real-time dynamic topology of the network;

[0023] S302. Determine whether the fitness of the current individual meets the preset requirements or reaches the preset number of loop iterations. If so, proceed to step S305, otherwise, execute step S303;

[0024] S303. Calculate the connection probability and connection stability index of the current network, and evaluate the fitness according to the calculated connection probability and connection stability index, obtain the selection probability according to the ratio of the fitness of each node sequence to the sum of the fitness of all sequences, and select individual nodes according to the selection probability;

[0025] S304. After performing crossover and mutation on the node sequence according to the current real-time dynamic topology of the network, the next generation population is generated, and the process returns to step S302;

[0026] S305. Output the currently selected individual as the best individual, and obtain the corresponding optimal solution output.

[0027] Furthermore, the selection of individual nodes in step S303 specifically includes: obtaining the network load and routing hop count of each node within a specified time period, and selecting individual nodes based on the size of the comprehensive fitness, the network load of each node, and the routing hop count between each node.

[0028] A vehicle-mounted self-organizing network dynamic routing device in a C-V2X environment, comprising:

[0029] A routing model module is used to construct a routing model of a vehicle self-organizing network based on a genetic algorithm model according to the parameters of the current vehicle self-organizing network. The routing model uses a genetic algorithm to iteratively select the optimal node, and uses a connection probability and a connection stability index as evaluation parameters of the fitness function, wherein the connection stability index is calculated based on the statistical value of the node's historical connection interruption probability;

[0030] The real-time topology acquisition module is used to receive the driving status information of each vehicle node uploaded by each on-board C-V2X OBU device to obtain the real-time dynamic topology of the current network;

[0031] A dynamic routing module, used for inputting the current network real-time dynamic topology into the routing model to perform optimal node sequence analysis and generate a dynamic routing table;

[0032] The route reselection control module is used to return to the real-time topology acquisition module to re-trigger the network for route selection when a vehicle node exits the network.

[0033] A computer device comprises a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to perform the above method.

[0034] Compared with the prior art, the advantages of the present invention are as follows: the present invention obtains the vehicle-borne driving status information as prior information by utilizing the vehicle-borne C-V2X OBU device, and realizes the selection of the optimal node sequence in the dynamic routing of the vehicle-borne self-organizing network under the C-V2X environment in combination with the routing model constructed by the genetic algorithm, and can quickly and accurately select the next hop node to avoid the influence of the vehicle-borne self-organizing network interference under the C-V2X environment on data transmission, thereby ensuring the accuracy and reliability of the routing selection process, and can still guarantee a stable arrival rate when the network topology environment changes rapidly. The dynamic routing generation process does not require excessive intervention and influence on the system, and can also effectively improve the applicability, and realize the direct balance between the performance and overhead of the routing protocol. At the same time, the connectivity probability and the connection stability index are used as the evaluation parameters of the fitness function, and the routing process can also realize fast global search and fast convergence. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1It is a schematic diagram of the implementation process of the dynamic routing method of the vehicle-mounted self-organizing network in the C-V2X environment of this embodiment.

[0036] Figure 2 1 is a schematic diagram of the implementation process of iterative selection using a genetic algorithm in this embodiment. DETAILED DESCRIPTION

[0037] The present invention is further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.

[0038] like Figure 1 As shown, the steps of the vehicle-mounted self-organizing network dynamic routing method under the C-V2X environment of this embodiment include:

[0039] S01. According to the parameters of the current vehicle self-organizing network (including vehicle driving state parameters, dynamic routing table, etc.), a routing model of the vehicle self-organizing network is constructed based on a genetic algorithm model. The routing model uses a genetic algorithm to iteratively select the optimal node, and uses the connection probability and connection stability index as the evaluation parameters of the fitness function, wherein the connection stability index is calculated based on the statistical value of the node's historical connection interruption probability;

[0040] S02. Receive the driving status information of each vehicle node uploaded by each on-board C-V2X OBU device to obtain the real-time dynamic topology of the current network;

[0041] S03. Input the current real-time dynamic network topology into the routing model to perform optimal node sequence analysis and generate a dynamic routing table;

[0042] S04. When a vehicle node exits the network, return to step S02 to re-trigger the network for routing selection.

[0043] The genetic algorithm calculates the optimal solution by imitating the selection and inheritance of nature, has good fast convergence and global search capabilities, and the chromosome sequence description is very similar to the multi-hop node sequence in structure. This embodiment takes into account the above characteristics of the genetic algorithm, and uses the on-board C-V2X OBU device to obtain the on-board driving status information as a priori information, and combines the routing model constructed by the genetic algorithm to realize the selection of the optimal node sequence in the dynamic routing of the on-board self-organizing network in the C-V2X environment. It can quickly and accurately select the next hop node and avoid the impact of the on-board self-organizing network interference in the C-V2X environment on data transmission, ensuring the accuracy and reliability of the routing selection process, so that a stable arrival rate can still be guaranteed when the network topology environment changes rapidly. At the same time, by making full use of the on-board C-V2X OBU device to obtain the on-board driving status information as a priori information, the dynamic routing generation process does not need to intervene and affect the system too much, effectively improving the applicability, so that it can adapt to C-V2X. OBU equipment resource constraints are used to achieve a direct balance between the performance and overhead of the routing protocol. Since the vehicle-mounted self-organizing network has the characteristics of rapid changes in network topology, the connectivity of the vehicle-mounted nodes will greatly affect the impact of network routing selection. In the iterative selection process of the optimal node, the connectivity probability and connection stability index are used as the evaluation parameters of the fitness function, which can ensure the accuracy of node selection and enable the routing process to achieve fast global search and fast convergence. Even in the scenario of high node density in the vehicle self-organizing network, high-performance and low-overhead dynamic routing can still be achieved.

[0044] In this embodiment, the driving status information of the vehicle node specifically includes position coordinates, speed, heading, acceleration, etc., to determine the real-time dynamic topology of the current network. Subsequently, the real-time dynamic topology of the current network containing the position, speed and other information of each vehicle node is used as a priori information and input into the routing model for routing selection, so as to make full use of the vehicle's prior information and combine with the genetic algorithm to select better nodes in the iterative process to achieve dynamic routing. Specifically, during the driving process of the vehicle, the above-mentioned driving status information broadcasted by other on-board C-V2X OBU devices is obtained through the on-board C-V2X OBU device, and the driving status information of the vehicle itself is obtained at the same time, so as to determine the real-time dynamic topology of the current network. It can be understood that the above-mentioned driving status information can of course also select other state information parameter combinations according to actual needs.

[0045] In step S02 of this embodiment, the driving state information of each vehicle node and the node information of each vehicle node are mapped to the LDM (Local Dynamic Map) map, combined with the driving state information broadcast by other on-board C-V2XOBU devices and the basic information and driving information of the vehicle itself, to form the current network real-time dynamic topology, and then input the current network real-time dynamic topology into the routing model for optimal node sequence analysis. The node information of the above vehicle node includes basic information such as the node ID number and vehicle type.

[0046] In order to ensure the single-hop network connectivity between two vehicle nodes, the distance d between two adjacent vehicles ji must be smaller than the communication transmission range R. In this embodiment, the connection probability is specifically based on the position p of each vehicle within the specified range of the road at the current time tc, and the distance d between two adjacent vehicles. ij The probability P is less than the communication distance R ij In scenarios where C-V2X nodes are densely deployed, such as expressways or trunk roads, where vehicle speeds vary little and network topology changes at a low rate, when the traffic volume is small, vehicles are in a free-flow state. In the free-flow state, the position p of each vehicle on the road at the current time tc and the probability Pi that the distance between two adjacent vehicles is less than the communication distance R can be directly calculated based on the recorded data in the vehicle information table of the intersection service node. j , and then determine the network connectivity probability.

[0047] Considering that the free flow speed of a single vehicle follows a normal distribution, and the travel distance is the speed multiplied by the time, the current position p also follows a normal distribution, then its mean μ p With variance The calculation expression is:

[0048]

[0049]

[0050] Among them, t c ,t r With t a are the current time, record update time, vehicle travel time, μ fv is the recorded mean, a is the vehicle node acceleration, is the variance at the previous moment, and the initial value of the variance is 1, which conforms to the standard normal distribution. As shown in formula (1), when (t c -t r )<t a When p use It is calculated that when (t c-t r )≥t a When using Calculate the mean.

[0051] In the free flow state, the free flow speeds of different vehicles are independent and identically distributed. According to the properties of normal distribution, it can be further concluded that the distance between any two vehicle nodes i and j also obeys the normal distribution, that is:

[0052]

[0053] Then, according to the position p of each vehicle and the distance d between any two adjacent vehicles ij The probability distribution state is obtained, and the probability P ij The calculation expression is:

[0054]

[0055] Among them, μ pi , μ pj are the means corresponding to the normal distribution of the positions of the i-th and j-th vehicle nodes, are the variances corresponding to the normal distribution of the locations of the i-th and j-th vehicle nodes respectively.

[0056] In step S03 of this embodiment, the step of inputting the current network real-time dynamic topology into the routing model for optimal node sequence analysis includes:

[0057] S301. Randomly generate an initial population based on the current real-time dynamic topology of the network;

[0058] S302. Determine whether the fitness of the current individual meets the preset requirements or reaches the preset number of loop iterations. If so, proceed to step S305, otherwise, execute step S303;

[0059] S303. Calculate the connection probability and connection stability index of the current network, evaluate the fitness based on the calculated connection probability and connection stability index, and select individuals according to the fitness;

[0060] S304. After performing crossover and mutation on the node sequence according to the current real-time dynamic topology of the network, the next generation population is generated, and the process returns to step S302;

[0061] S305. Output the currently selected individual as the best individual, and obtain the corresponding optimal solution output.

[0062] In the above step S301, a number of nodes are first selected to form different sequences to construct an initial population. Each sequence must meet the value range requirement of the number of nodes. At the same time, the fitness of each node is calculated, and the selection probability is obtained according to the ratio of the fitness of each node sequence to the sum of the fitness of all sequences.

[0063] In the above step S303, the fitness is calculated specifically according to the connectivity probability of each sequence and the connection stability index, and then the selection probability is calculated according to the fitness, and the selection and pairing of the optimal nodes are performed according to the selection probability.

[0064] In the above step S304, for the paired node sequence, a position is randomly selected as a crossover point, and a crossover operation is performed on the crossover point to create a descendant sequence. Each position in the descendant sequence obtained after the crossover is then generated according to a preset probability P. mut Random mutation is performed to generate the next generation population. The above selection, crossover, and mutation operations are performed repeatedly until the fitness of the optimal sequence reaches a given threshold or the number of cycles reaches a preset value, and the dynamic routing table can be generated by the optimal solutions obtained by the genetic algorithm.

[0065] This embodiment uses a genetic algorithm to select the optimal node sequence in the dynamic routing of the vehicle-mounted self-organizing network under the C-V2X environment. In the process of selecting the optimal node and generating the population, the driving status information, connectivity probability, and connection stability index of each vehicle node are fully utilized, and the next hop node can be selected quickly and accurately, avoiding the impact of the vehicle-mounted self-organizing network interference in the C-V2X environment on data transmission, ensuring the accuracy and reliability of the routing selection process, and ensuring a stable arrival rate when the network topology environment changes rapidly. At the same time, the dynamic route generation process does not need to intervene and affect the system too much, and effectively achieves a balance between the performance and overhead of the routing protocol.

[0066] When data packets are transmitted on the road, the data packet queues in certain vehicle nodes at key locations often become longer, causing the buffer area to overflow and resulting in packet loss at the network layer. When selecting individual nodes in step S303 of this embodiment, the network load and routing hop count of each node in the specified time period are obtained, and the individual nodes are selected based on the size of the comprehensive fitness, the network load of each node, and the routing hop count between each node, so as to achieve balanced processing of the network load of each node in a certain time period under the premise of ensuring that the individual fitness is as high as possible. For example, when selecting a new routing node sequence, try to avoid those nodes with large loads to ensure balanced network load. At the same time, since the number of nodes in the multi-hop transmission sequence is more (i.e., the number of hops is more), the number of communications required is also more, and the probability of transmission failure between end-to-end will increase. Therefore, by also taking the routing hop count as a key consideration when selecting nodes, the routing hop count can be effectively reduced and the routing efficiency can be improved.

[0067] The vehicle-mounted self-organizing network dynamic routing device under the C-V2X environment of this embodiment includes:

[0068] The routing model module is used to construct a routing model of the vehicle self-organizing network based on the genetic algorithm model according to the parameters of the current vehicle self-organizing network. The routing model uses the genetic algorithm to iteratively select the optimal node, and uses the connection probability and connection stability index as the evaluation parameters of the fitness function;

[0069] The real-time topology acquisition module is used to receive the driving status information of each vehicle node uploaded by each on-board C-V2X OBU device to obtain the real-time dynamic topology of the current network;

[0070] Dynamic routing module, used to input the current network real-time dynamic topology into the routing model to perform optimal node sequence analysis and generate a dynamic routing table;

[0071] The route reselection control module is used to return to the real-time topology acquisition module to re-trigger the network for route selection when a vehicle node exits the network.

[0072] The above-mentioned routing model module can be specifically constructed in an offline manner, such as offline constructing a routing model of the vehicle self-organizing network based on the genetic algorithm model according to the parameters of the current vehicle self-organizing network in the above manner, and then storing the program module corresponding to the routing model. When dynamic routing needs to be executed, the real-time topology acquisition module obtains the real-time dynamic topology of the current network, and inputs the current real-time dynamic topology of the current network as input data into the stored routing model, so as to generate a dynamic routing table. The routing reselection control module judges the changing status of the nodes in the network in real time. If a vehicle node exits, the network is re-triggered for routing selection.

[0073] The above-mentioned real-time topology acquisition module, dynamic routing module and route reselection control module can be implemented by discrete devices or integrated modules. For example, one device can simultaneously implement the above-mentioned real-time topology acquisition, dynamic routing and route reselection control functions. The specific structural form can be configured according to actual needs.

[0074] The vehicle-mounted self-organizing network dynamic routing device under the C-V2X environment of this embodiment corresponds one to one with the vehicle-mounted self-organizing network dynamic routing method under the C-V2X environment mentioned above, and will not be described one by one here.

[0075] This embodiment also provides a computer device, including a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to perform the above method.

[0076] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment, it is not intended to limit the present invention. Therefore, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the scope of protection of the technical solution of the present invention.

Claims

1. A dynamic routing method for a vehicle-mounted self-organizing network in a C-V2X environment, characterized in that the steps include: S01. A routing model of a vehicle self-organizing network is constructed based on a genetic algorithm model according to the parameters of the current vehicle self-organizing network. The routing model uses a genetic algorithm to iteratively select the optimal node, and uses a connection probability and a connection stability index as evaluation parameters of the fitness function, wherein the connection stability index is calculated based on the statistical value of the probability of historical connection interruption of the node, and the connection probability is calculated based on the position p of each vehicle within the specified range of the road at the current time tc, and the distance between two adjacent vehicles. The probability of being less than the communication distance R Calculated, According to the position p of each vehicle, the distance between any two adjacent vehicles The probability distribution state is obtained by The calculation expression is: in, are the means corresponding to the normal distribution of the positions of the i-th and j-th vehicle nodes, are the variances corresponding to the normal distribution of the positions of the i-th and j-th vehicle nodes respectively; S02. Receive the driving status information of each vehicle node uploaded by each on-board C-V2X OBU device to obtain the real-time dynamic topology of the current network; S03. The current real-time dynamic network topology is input into the routing model for optimal node sequence analysis to generate a dynamic routing table; S04. When a vehicle node exits the network, return to step S02 to re-trigger the network for routing selection.

2. The method for dynamic routing of a vehicle-mounted self-organizing network in a C-V2X environment according to claim 1, characterized in that: The driving status information of the vehicle node includes any one or more of position coordinates, speed, heading, and acceleration.

3. The method for dynamic routing of a vehicle-mounted self-organizing network in a C-V2X environment according to claim 1, characterized in that: In the step S02, the real-time dynamic topology of the current network is formed by mapping the acquired driving status information of each vehicle node and the node information of each vehicle node onto the LDM map.

4. The method for dynamic routing of a vehicle-mounted self-organizing network in a C-V2X environment according to claim 1, characterized in that: The mean value corresponding to the normal distribution of the vehicle position p With variance Calculate according to the following formula: in, , and They are the current time, record update time, and vehicle travel time. is the recorded mean, is the vehicle node acceleration, is the variance of the previous moment.

5. The method for dynamic routing of a vehicle-mounted self-organizing network in a C-V2X environment according to claim 4, characterized in that: The step of inputting the current network real-time dynamic topology into the routing model for optimal node sequence analysis in step S03 includes: S301. Randomly generate an initial population based on the current real-time dynamic topology of the network; S302. Determine whether the fitness of the current individual node meets the preset requirements or reaches the preset number of loop iterations. If so, proceed to step S305, otherwise, execute step S303; S303. Calculate the connection probability and connection stability index of the current network, and evaluate the fitness according to the calculated connection probability and connection stability index, obtain the selection probability according to the ratio of the fitness of each node sequence to the sum of the fitness of all sequences, and select individual nodes according to the selection probability; S304. After performing crossover and mutation on the node sequence according to the current real-time dynamic topology of the network, the next generation population is generated, and the process returns to step S302; S305. Output the currently selected individual node as the best individual node, and obtain the corresponding optimal solution output.

6. The method for dynamic routing of a vehicle-mounted self-organizing network in a C-V2X environment according to claim 5, characterized in that: The step S303 of selecting individual nodes specifically includes: obtaining the network load and routing hop count of each node within a specified time period, and selecting individual nodes based on the size of the comprehensive fitness, the network load of each node, and the routing hop count between each node.

7. A dynamic routing device for a vehicle-mounted self-organizing network in a C-V2X environment, characterized in that: include: The routing model module is used to construct a routing model of the vehicle self-organizing network based on the genetic algorithm model according to the parameters of the current vehicle self-organizing network. The routing model uses a genetic algorithm to iteratively select the optimal node, and uses the connection probability and connection stability index as the evaluation parameters of the fitness function, wherein the connection stability index is calculated based on the statistical value of the node's historical connection interruption probability, and the connection probability is calculated based on the position p of each vehicle within the specified range of the road at the current time tc, and the distance between two adjacent vehicles. The probability of being less than the communication distance R Calculated, According to the position p of each vehicle, the distance between any two adjacent vehicles The probability distribution state is obtained by The calculation expression is: in, are the means corresponding to the normal distribution of the positions of the i-th and j-th vehicle nodes, are the variances corresponding to the normal distribution of the positions of the i-th and j-th vehicle nodes respectively; The real-time topology acquisition module is used to receive the driving status information of each vehicle node uploaded by each on-board C-V2X OBU device to obtain the real-time dynamic topology of the current network; A dynamic routing module, used for inputting the current network real-time dynamic topology into the routing model to perform optimal node sequence analysis and generate a dynamic routing table; The route reselection control module is used to return to the real-time topology acquisition module to re-trigger the network for route selection when a vehicle node exits the network.

8. A computer device comprising a processor and a memory, wherein the memory is used to store a computer program, wherein: The processor is configured to execute the computer program to perform the method according to any one of claims 1 to 6.

Images