Road side unit deployment method for V2X auxiliary positioning

Abstract

The invention relates to a road side unit deployment method for V2X auxiliary positioning, and belongs to the technical field of intelligent traffic. The method comprises the following steps: firstly, acquiring an urban road network of a to-be-deployed area of road side units and vehicle driving tracks in different time periods, and acquiring a sensor measurement error, a GNSS error, the to-be-deployed number of the road side units and the like; constructing a V2X assisted positioning performance expression; and mapping the candidate positions of the road side units into individuals in a natural population, and solving the optimal population of the road side units based on a genetic algorithm and a cuckoo local optimization algorithm by taking a maximum V2X assisted positioning performance expression as a fitness function, thereby obtaining the optimal candidate positions of the road side units, and performing deployment of the road side units. Aiming at the problems of weak local search capability and poor stability of the genetic algorithm, the discrete cuckoo algorithm is applied to the search of the optimal layout scheme, and the method has good support for improving the vehicle-road cooperative positioning precision and improving the road side unit application efficiency.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent transportation technology, and specifically relates to a method for deploying roadside units for V2X assisted positioning. Background Art

[0002] V2V (Vehicle To Vehicle, V2V) collaborative positioning mainly uses shared location and distance information to improve positioning accuracy. However, due to the high-speed mobility of vehicles, this method is very challenging. On the one hand, the time provided for data exchange between vehicle nodes is short; on the other hand, the addition of new nodes and the exit of old nodes cause the network structure to change rapidly, which is not conducive to maintaining the stability of the collaborative positioning system. Therefore, people consider deploying road side units (Road Side Unit, RSU) along the road, hoping to improve communication performance through V2I (Vehicle To Infrastructure, V2I) to achieve more reliable position estimation. However, RSU has expensive installation and maintenance costs, and it is impractical to implement V2I collaborative positioning in the entire urban road network. It is of great significance to use a limited number of RSUs to provide auxiliary positioning services to more vehicles.

[0003] Most existing RSU deployment methods either only optimize network indicators such as coverage, number of service vehicles, and average reporting time, or only focus on economic costs such as equipment installation and energy loss, while few studies involve vehicle positioning issues. In addition, when evaluating the impact of RSU deployment schemes on positioning performance, the literature usually uses customized single-lane or two-lane straight roads without considering traffic conditions in real scenarios, such as vehicle mobility and urban road layout.

[0004] Therefore, how to overcome the shortcomings of existing technologies is an urgent problem to be solved in the field of intelligent transportation technology. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the prior art and to provide a method for roadside unit deployment for V2X assisted positioning. This method addresses the problems of weak local search capability and poor stability of genetic algorithms, and applies a discretized cuckoo algorithm to the search for the optimal deployment plan, which provides good support for improving the accuracy of vehicle-road collaborative positioning and enhancing the application efficiency of roadside units.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0007] The method for deploying a roadside unit for V2X assisted positioning includes the following steps:

[0008] Step 1: Obtain the urban road network and vehicle movement data at different time periods in the area where the roadside unit is to be deployed, and determine the number of roadside units to be deployed and the V2X assisted positioning coverage of the roadside units. The vehicle movement data includes speed, acceleration, heading angle and GNSS data.

[0009] Step 2: Construct V2X assisted positioning performance expression;

[0010] Step 3: Map the candidate positions of the roadside unit to individuals in the natural population, use the maximum V2X assisted positioning performance expression as the fitness function, solve the optimal population of the roadside unit based on the genetic algorithm and the cuckoo local optimization algorithm, and thus obtain the optimal candidate position of the roadside unit;

[0011] Step 4: Deploy the roadside unit according to the best candidate position of the roadside unit obtained in step 3.

[0012] Further, preferably, in step one, the vehicle motion data includes speed, acceleration, heading angle and GNSS data acquired through INS.

[0013] Further, preferably, in step 2, the V2X assisted positioning performance expression is as follows:

[0014]

[0015] Where ξ represents the V2X assisted positioning performance, N represents the number of roadside units to be deployed; C i is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit i V2X assisted positioning; C j It is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit jV2X auxiliary positioning;

[0016] Vehicle i The calculation formula for positioning accuracy improvement is as follows:

[0017]

[0018] Among them, c i Indicates vehicle v i Positioning accuracy improvement value; p i gx With p i gy Represents vehicle v i GNSS components on the x and y axes; p i tx With p i ty Represents vehicle v i The actual position on the x and y axes; p i ax With pi ay They represent the vehicle v obtained by V2X assisted positioning. i Estimated coordinates on the x and y axes.

[0019] Further, preferably, in step 3, the specific method for solving the optimal population of roadside units based on the genetic algorithm and the cuckoo local optimization algorithm is:

[0020] Step (1), input the urban road network of the area where the roadside unit is to be deployed, the number of roadside units to be deployed, and the V2X assisted positioning coverage of the roadside unit; then initialize the genetic algorithm parameters and cuckoo algorithm parameters according to the number of roadside units to be deployed; the population size of the genetic algorithm needs to correspond to; and the maximum V2X assisted positioning performance expression is used as the fitness function;

[0021] Step (2), creating an initial population including a number of individuals, each individual having the same characteristics as the number of candidate locations where the roadside unit can be deployed; creating two arrays to store the best fitness value and the average fitness value in each iteration process respectively;

[0022] Step (3), calculate the fitness value of each individual in the population and arrange them in order from high to low;

[0023] Step (4), select individuals with higher fitness values ​​from the current population according to the replacement ratio to serve as the survivors of the old generation after the survival of the fittest, and then randomly select these survivors to become parents, and obtain another part of the new population individuals through single-point crossover;

[0024] Step (5), randomly generate a random number between 0 and 1 and compare the value with the mutation probability to determine whether the chromosome of the new population individual has mutated, and randomly switch the bits of the chromosome gene digital code within the valid number range, and the maximum digital number does not exceed the total number of candidate locations that can be deployed by the roadside unit;

[0025] Step (6), using the discretized cuckoo algorithm to perform local search on the generated population;

[0026] Step (7), repeating steps (3) to (6), continuously performing fitness evaluation, selection, crossover, mutation, and local search until a preset number of iterations is reached;

[0027] Step (8), finally, the new chromosome returned by the iteratively updated local search is merged with the existing chromosome to obtain a new generation population of the genetic algorithm, that is, the corresponding optimized roadside unit deployment position coordinates are obtained as the best candidate position of the roadside unit.

[0028] Further, preferably, in step (1), the genetic algorithm parameters include population size, number of genetic algorithm iterations, permutation ratio, crossover probability and mutation probability;

[0029] The cuckoo algorithm parameters include the number of initialized nests, the number of cuckoo algorithm iterations, the discovery probability and the step size factor.

[0030] Further, preferably, in step (6), the specific method of using the discretized cuckoo algorithm to perform local search is:

[0031] First, the elements in the offspring array are shuffled to initialize a cuckoo population with a size equal to the number of initialized nests; then, the Levy flight strategy is used to update the location of the cuckoo's nest that replaces the host bird's eggs; if the host bird unfortunately finds the disguised bird eggs, the cuckoo needs to find a replacement nest; finally, the cuckoo individual with the highest fitness value in the above process is returned.

[0032] The V2I assisted positioning method refers to the roadside unit receiving the motion data of vehicles within the communication range and completing the positioning estimation through a filtering algorithm.

[0033] The V2X assisted positioning method refers to establishing a GNSS error information correction model based on V2I positioning estimation data and vehicle motion data, using V2V technology to share the model parameters with more vehicles, and expanding the influence range of roadside unit positioning.

[0034] In the present invention, the V2X assisted positioning coverage refers to the influence range of the roadside unit through the V2X assisted positioning method, which is expanded relative to the influence range of the roadside unit assisted positioning of the original V2I assisted positioning method. The V2X assisted positioning coverage can be measured experimentally.

[0035] In step (1) of the present invention, other genetic algorithm and cuckoo algorithm parameters except the population size are initialized according to the convergence of the algorithm and adaptively adjusted.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention provides a roadside unit deployment method for V2X assisted positioning. By designing a fitness function for improving V2X positioning performance, a deployment solution is provided for V2X assisted positioning. By applying the discretized cuckoo algorithm to the offspring individuals created by the genetic algorithm, the roadside unit deployment algorithm based on the genetic algorithm is improved, and the probability of premature convergence of the algorithm is reduced. In the experiment set by the present invention, after 15 iterations of the algorithm, the optimal fitness value obtained by CLOGA in the Xi'an Electronic City scenario is improved by 79.31% and 41.81% respectively compared with the genetic algorithm and the RSU optimization deployment algorithm under the meme framework. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of a cuckoo local optimization genetic algorithm proposed by the present invention for urban roadside unit deployment with the goal of improving V2X assisted positioning performance;

[0039] Figure 2 is a schematic diagram of RSU deployment in genetic algorithm; (a) is a diagram showing individual RSU layout positions in genetic algorithm, where dots represent candidate positions and (b) represents an example of chromosome gene sequence coding;

[0040] Figure 3 It is a map of the Xi'an Electronics City area;

[0041] Figure 4 In the simulation experiment, the fitness values ​​of the algorithms deployed in the electronics city area are compared;

[0042] Figure 5 It is a comparison of the fitness values ​​of the deployment algorithms under different numbers of RSUs in the electronics city area in the simulation experiment. DETAILED DESCRIPTION

[0043] The present invention is further described in detail below in conjunction with embodiments.

[0044] Those skilled in the art will appreciate that the following examples are only used to illustrate the present invention and should not be considered to limit the scope of the present invention. If no specific techniques or conditions are specified in the examples, the techniques or conditions described in the literature in the art or the product specifications are used. If the manufacturer of the materials or equipment used is not specified, they are all conventional products that can be purchased.

[0045] Example 1

[0046] The method for deploying a roadside unit for V2X assisted positioning includes the following steps:

[0047] Step 1: Obtain the urban road network and vehicle movement data at different time periods in the area where the roadside unit is to be deployed, and determine the number of roadside units to be deployed and the V2X assisted positioning coverage of the roadside units. The vehicle movement data includes speed, acceleration, heading angle and GNSS data.

[0048] Step 2: Construct V2X assisted positioning performance expression;

[0049] Step 3: Map the candidate positions of the roadside unit to individuals in the natural population, use the maximum V2X assisted positioning performance expression as the fitness function, solve the optimal population of the roadside unit based on the genetic algorithm and the cuckoo local optimization algorithm, and thus obtain the optimal candidate position of the roadside unit;

[0050] Step 4: Deploy the roadside unit according to the best candidate position of the roadside unit obtained in step 3.

[0051] Example 2

[0052] The method for deploying a roadside unit for V2X assisted positioning includes the following steps:

[0053] Step 1: Obtain the urban road network and vehicle movement data at different time periods in the area where the roadside unit is to be deployed, and determine the number of roadside units to be deployed and the V2X assisted positioning coverage of the roadside units. The vehicle movement data includes speed, acceleration, heading angle and GNSS data.

[0054] Step 2: Construct V2X assisted positioning performance expression;

[0055] Step 3: Map the candidate positions of the roadside unit to individuals in the natural population, use the maximum V2X assisted positioning performance expression as the fitness function, solve the optimal population of the roadside unit based on the genetic algorithm and the cuckoo local optimization algorithm, and thus obtain the optimal candidate position of the roadside unit;

[0056] Step 4: Deploy the roadside unit according to the best candidate position of the roadside unit obtained in step 3.

[0057] In step 1, the vehicle motion data includes speed, acceleration, heading angle and GNSS data obtained through INS.

[0058] In step 2, the V2X assisted positioning performance expression is as follows:

[0059]

[0060] Where ξ represents the V2X assisted positioning performance, N represents the number of roadside units to be deployed; C i is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit iV2X assisted positioning; C j It is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit jV2X auxiliary positioning;

[0061] Vehicle i The calculation formula for positioning accuracy improvement is as follows:

[0062]

[0063] Among them, c i Indicates vehicle v i Positioning accuracy improvement value; p i gx With p i gy Represents vehicle v i GNSS components on the x and y axes; pi tx With p i ty Represents vehicle v i The actual position on the x and y axes; p i ax With p i ay They represent the vehicle v obtained by V2X assisted positioning. i Estimated coordinates on the x and y axes.

[0064] In step 3, the specific method for solving the optimal population of roadside units based on genetic algorithm and cuckoo local optimization algorithm is:

[0065] Step (1), input the urban road network of the area where the roadside unit is to be deployed, the number of roadside units to be deployed, and the V2X assisted positioning coverage of the roadside unit; then initialize the genetic algorithm parameters and cuckoo algorithm parameters according to the number of roadside units to be deployed; the population size of the genetic algorithm needs to correspond to; and the maximum V2X assisted positioning performance expression is used as the fitness function;

[0066] Step (2), creating an initial population including a number of individuals, each individual having the same characteristics as the number of candidate locations where the roadside unit can be deployed; creating two arrays to store the best fitness value and the average fitness value in each iteration process respectively;

[0067] Step (3), calculate the fitness value of each individual in the population and arrange them in order from high to low;

[0068] Step (4), select individuals with higher fitness values ​​from the current population according to the replacement ratio to serve as the survivors of the old generation after the survival of the fittest, and then randomly select these survivors to become parents, and obtain another part of the new population individuals through single-point crossover;

[0069] Step (5), randomly generate a random number between 0 and 1 and compare the value with the mutation probability to determine whether the chromosome of the new population individual has mutated, and randomly switch the bits of the chromosome gene digital code within the valid number range, and the maximum digital number does not exceed the total number of candidate locations that can be deployed by the roadside unit;

[0070] Step (6), using the discretized cuckoo algorithm to perform local search on the generated population;

[0071] Step (7), repeating steps (3) to (6), continuously performing fitness evaluation, selection, crossover, mutation, and local search until a preset number of iterations is reached;

[0072] Step (8), finally, the new chromosome returned by the iteratively updated local search is merged with the existing chromosome to obtain a new generation population of the genetic algorithm, that is, the corresponding optimized roadside unit deployment position coordinates are obtained as the best candidate position of the roadside unit.

[0073] Example 3

[0074] The method for deploying a roadside unit for V2X assisted positioning includes the following steps:

[0075] Step 1: Obtain the urban road network and vehicle movement data at different time periods in the area where the roadside unit is to be deployed, and determine the number of roadside units to be deployed and the V2X assisted positioning coverage of the roadside units. The vehicle movement data includes speed, acceleration, heading angle and GNSS data.

[0076] Step 2: Construct V2X assisted positioning performance expression;

[0077] Step 3: Map the candidate positions of the roadside unit to individuals in the natural population, use the maximum V2X assisted positioning performance expression as the fitness function, solve the optimal population of the roadside unit based on the genetic algorithm and the cuckoo local optimization algorithm, and thus obtain the optimal candidate position of the roadside unit;

[0078] Step 4: Deploy the roadside unit according to the best candidate position of the roadside unit obtained in step 3.

[0079] In step 1, the vehicle motion data includes speed, acceleration, heading angle and GNSS data obtained through INS.

[0080] In step 2, the V2X assisted positioning performance expression is as follows:

[0081]

[0082] Where ξ represents the V2X assisted positioning performance, N represents the number of roadside units to be deployed; C i is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit iV2X assisted positioning; C j It is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit jV2X auxiliary positioning;

[0083] Vehicle i The calculation formula for positioning accuracy improvement is as follows:

[0084]

[0085] Among them, c i Indicates vehicle v i Positioning accuracy improvement value; p i gx With p igy Represents vehicle v i GNSS components on the x and y axes; p i tx With p i ty Represents vehicle v i The actual position on the x and y axes; p i ax With p i ay They represent the vehicle v obtained by V2X assisted positioning. i Estimated coordinates on the x and y axes.

[0086] In step 3, the specific method for solving the optimal population of roadside units based on genetic algorithm and cuckoo local optimization algorithm is:

[0087] Step (1), input the urban road network of the area where the roadside unit is to be deployed, the number of roadside units to be deployed, and the V2X assisted positioning coverage of the roadside unit; then initialize the genetic algorithm parameters and cuckoo algorithm parameters according to the number of roadside units to be deployed; the population size of the genetic algorithm needs to correspond to; and the maximum V2X assisted positioning performance expression is used as the fitness function;

[0088] Step (2), creating an initial population including a number of individuals, each individual having the same characteristics as the number of candidate locations where the roadside unit can be deployed; creating two arrays to store the best fitness value and the average fitness value in each iteration process respectively;

[0089] Step (3), calculate the fitness value of each individual in the population and arrange them in order from high to low;

[0090] Step (4), select individuals with higher fitness values ​​from the current population according to the replacement ratio to serve as the survivors of the old generation after the survival of the fittest, and then randomly select these survivors to become parents, and obtain another part of the new population individuals through single-point crossover;

[0091] Step (5), randomly generate a random number between 0 and 1 and compare the value with the mutation probability to determine whether the chromosome of the new population individual has mutated, and randomly switch the bits of the chromosome gene digital code within the valid number range, and the maximum digital number does not exceed the total number of candidate locations that can be deployed by the roadside unit;

[0092] Step (6), using the discretized cuckoo algorithm to perform local search on the generated population;

[0093] Step (7), repeating steps (3) to (6), continuously performing fitness evaluation, selection, crossover, mutation, and local search until a preset number of iterations is reached;

[0094] Step (8), finally, the new chromosome returned by the iteratively updated local search is merged with the existing chromosome to obtain a new generation population of the genetic algorithm, that is, the corresponding optimized roadside unit deployment position coordinates are obtained as the best candidate position of the roadside unit.

[0095] In step (1), the genetic algorithm parameters include population size, number of genetic algorithm iterations, permutation ratio, crossover probability and mutation probability;

[0096] The cuckoo algorithm parameters include the number of initialized nests, the number of cuckoo algorithm iterations, the discovery probability and the step size factor.

[0097] In step (6), the specific method of using the discretized cuckoo algorithm to perform local search is:

[0098] First, the elements in the offspring array are shuffled to initialize a cuckoo population with a size equal to the number of initialized nests; then, the Levy flight strategy is used to update the location of the cuckoo's nest that replaces the host bird's eggs; if the host bird unfortunately finds the disguised bird eggs, the cuckoo needs to find a replacement nest; finally, the cuckoo individual with the highest fitness value in the above process is returned.

[0099] Example 4

[0100] The method for deploying a roadside unit for V2X assisted positioning includes the following steps:

[0101] Step 1: Obtain the urban road network and vehicle motion data at different time periods in the area where the roadside unit is to be deployed, and determine the number of roadside units to be deployed and the V2X assisted positioning coverage of the roadside units. The vehicle motion data includes speed, acceleration, heading angle and GNSS data obtained through INS.

[0102] Step 2: Construct V2X assisted positioning performance expression;

[0103] In order to strike a balance between positioning performance and economic cost, the present invention sets the number of deployable roadside units as the input value and focuses on optimizing the positioning performance represented by ξ. For ξ, the aim is to find the sum of the positioning accuracy improvement values ​​of all vehicles within the coverage of each roadside unit V2X assisted positioning. i The positioning accuracy improvement value is obtained by adding the calculation amounts in the x and y axes, as shown in formula (1).

[0104]

[0105] Among them, c i Indicates vehicle v i Positioning accuracy improvement value; p i gx With p igy Represents vehicle v i GNSS components on the x and y axes; p i tx With p i ty Represents vehicle v i The actual position on the x and y axes; p i ax With p i ay They represent the vehicle v obtained by V2X assisted positioning. i Estimated coordinates on the x and y axes.

[0106] Let C i is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit i V2X assisted positioning, C j is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage of roadside unit jV2X assisted positioning; then, the optimal intersection number covered by roadside unit i and roadside unit j is C i +C j –(C i ∩C j )(The vehicles with common coverage are not counted repeatedly). Therefore, the positioning performance ξ in the roadside unit layout is expressed as:

[0107]

[0108] Where ξ represents the V2X assisted positioning performance, and N represents the deployable number of roadside units.

[0109] Step 3: Map the candidate positions of the roadside unit to individuals in the natural population, use the maximum V2X assisted positioning performance expression as the fitness function, solve the optimal population of the roadside unit based on the genetic algorithm and the cuckoo local optimization algorithm, and thus obtain the optimal candidate position of the roadside unit;

[0110] The present invention aims to maximize ξ in a given urban road network and proposes a Cuckoo Local Optimization Genetic Algorithm (CLOGA genetic algorithm). Specifically, the genetic algorithm first uses the original population to perform a global search in the environment, and then after the mutation phase, the discretized cuckoo algorithm is used to perform a local search around the chromosome, and then the chromosome with a higher fitness value is selected in the next phase. These phases continue until the iteration of the genetic algorithm is completed.

[0111] The goal of the genetic algorithm is to find the optimal location in a given urban road network, that is, the phenotype is the candidate location where the roadside unit can be deployed. There needs to be a clear and unique correspondence between the chromosome and the phenotype. Considering that there are many candidate locations for roadside units and binary numbers or alphabets cannot assign unique numbers to them, the present invention uses a digital sequence to represent the chromosome. Its dimension depends on the number of deployable roadside units (taking 6 as an example), such as Figure 2 As shown. In this way, repeated deployment of roadside units at a candidate location is avoided, which improves the robustness of the genetic algorithm. It should be noted that the intersection is selected as the candidate location where the roadside unit can be deployed by default because of its large traffic volume and wide space.

[0112] Next, the present invention will give a detailed overview of the CLOGA genetic algorithm:

[0113] 1. Set parameters: coverage, deployable number, genetic algorithm parameters (population size, GA iteration number ga_generation, replacement ratio selection_ratio, crossover probability Pc and mutation probability Pm), cuckoo algorithm parameters (initialized nest number N, CS iteration number cs_generation, discovery probability Pa and step size factor α), and given urban road network.

[0114] 2. Initialization: Generate an initial population P of size ga_population. Each individual represents an RSU deployment plan, including count RSU positions. To avoid repeated deployment, individuals are represented as digital sequences. Create two arrays to store the best fitness value and average fitness value in each iteration.

[0115] 3. Fitness evaluation: Calculate the fitness value of each individual based on the improvement of vehicle positioning accuracy within the RSU coverage area, and use the aforementioned expression of maximizing V2X assisted positioning performance as the optimization target of the fitness function. Sort the individuals in population P from high to low according to their fitness values.

[0116] 4. Select excellent individuals: According to the replacement ratio selection_ratio, select individuals with higher fitness from the current population as the survivors to form part of the new generation population.

[0117] 5. Single-point crossover: Randomly select two individuals from the selected excellent individuals as parents. Randomly find a gene "cut" point on the parent chromosome, generate new individuals through single-point crossover, and form another part of the new population individuals.

[0118] 6. Mutation operation: A mutation probability is preset to indicate the probability of chromosome mutation, which is usually set to a smaller value to avoid excessive damage to the population structure. A random number between 0 and 1 is randomly generated and compared with the mutation probability to determine whether the chromosome of the new population individual mutates. The bits of the chromosome gene digital code are randomly switched within the valid number range, and the maximum digital number does not exceed the total number of RSU candidate positions.

[0119] 7. Local search: Use local search techniques on the offspring generated by the genetic algorithm to more quickly find local optimal values ​​and converge the search process. Here, the discretized cuckoo algorithm is used for local search: first, shuffle the elements in the offspring array offspring to initialize a cuckoo population of size N; then, use the Levy flight strategy to update the location of the cuckoo's nest that replaces the host bird's eggs. If the host bird unfortunately finds the disguised bird eggs, the cuckoo needs to find a replacement nest. Finally, return the cuckoo individual with the highest fitness value in the above process;

[0120] 8. Repeat steps 3 to 7, continuously performing fitness evaluation, selection, crossover, mutation, and local search until the preset number of iterations is reached.

[0121] 9. Finally, the new chromosomes returned by the local search after multiple iterations are merged with the existing chromosomes to obtain the new generation population of GA, that is, the corresponding optimized RSU deployment position coordinates are obtained.

[0122] Step 4: Deploy the roadside unit according to the best candidate position of the roadside unit obtained in step 3.

[0123] Simulation experiment

[0124] The inventors of the present application verified the performance of the proposed CLOGA in simulation experiments. As for the urban scene to be simulated, the Xi'an Electronic City area was selected, and the map was selected in the OpenStreetMap format with street and intersection layout, such as Figure 3 SUMO generates the driving trajectory of each vehicle in different time periods.

[0125] In order to better demonstrate the fitness of optimizing the position of the roadside unit at each iteration, the present invention compares the proposed method with the genetic algorithm and the Memetic Framework-based Optimal RSUDeployment (MFRD). Moreover, each calculation result is the average value of multiple (set to 3) runs, which ensures the representativeness of the experiment. In the proposed algorithm, the genetic algorithm first randomly generates an initial population for global search, and after mutation to produce offspring, a discretized cuckoo is used to perform a local search around it, and then the individual with the highest fitness value is selected for replacement until the end of the genetic algorithm iteration. For the setting of relevant parameters, see Table 1.

[0126] Table 1 Experimental parameter settings

[0127]

[0128]

[0129] A. Performance comparison between CLOGA and existing methods under a given number of deployable RSUs

[0130] Figure 4 It means that when the number of deployable RSUs is limited to 4, the fitness values ​​of the best and average individuals in each generation of population are calculated using different algorithms in the electronic city scenario. The fitness function is related to the auxiliary positioning performance of RSU under certain economic cost constraints. CLOGA aims to maximize the fitness function to reduce the number of vehicles that cross-cover between different RSUs. As shown in the figure, CLOGA applies the improved CS to the local search process in GA, which improves the accuracy of traditional GA. The results show that CLOGA has the highest fitness value, proving its performance is better than other algorithms. Table 2 shows that after 15 iterations, the best fitness value obtained by CLOGA in the electronic city scenario is 79.31% and 41.81% higher than that of GA and MFRD respectively. In addition, after a generation of population evolution, the fitness values ​​of the best and average individuals have been significantly improved. This is attributed to the mutation operator and the local search process that allows searching for new maximum values ​​in the solution space. Compared with CLOGA, although GA converges faster, it has a larger gap in fitness value. This is mainly because LF finds the optimal value by random walk in the local search process. The same is true for MFRD.

[0131] Table 2 Fitness values ​​of different algorithms after 15 iterations

[0132]

[0133] B. Performance comparison between CLOGA and existing methods under different numbers of RSUs that can be deployed

[0134] In order to evaluate the impact of all test algorithms under different deployable numbers of roadside units, the present invention increases the input value from 3 to 8. Figure 5 It can be seen that the fitness value increases with the increase in the number of roadside units. Before the area is fully covered, the communication range of each newly deployed roadside unit will not completely overlap with other roadside units. In addition, Figure 5 The fitness values ​​in also show that the CLOGA genetic algorithm has better performance in all cases compared with the genetic algorithm and MFRD.

[0135] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for deploying roadside units for V2X assisted positioning, characterized in that: The steps include: Step 1: Obtain the urban road network and vehicle movement data at different time periods in the area where the roadside unit is to be deployed, and determine the number of roadside units to be deployed and the V2X assisted positioning coverage of the roadside units. The vehicle movement data includes speed, acceleration, heading angle and GNSS data. Step 2: Construct V2X assisted positioning performance expression; Step 3: Map the candidate positions of the roadside unit to individuals in the natural population, use the maximum V2X assisted positioning performance expression as the fitness function, solve the optimal population of the roadside unit based on the genetic algorithm and the cuckoo local optimization algorithm, and thus obtain the optimal candidate position of the roadside unit; Step 4: Deploy the roadside unit according to the best candidate position of the roadside unit obtained in step 3.

2. The method for roadside unit deployment for V2X assisted positioning according to claim 1, characterized in that: In step 1, the vehicle motion data includes speed, acceleration, heading angle and GNSS data obtained through INS.

3. The method for roadside unit deployment for V2X assisted positioning according to claim 1, characterized in that: In step 2, the V2X assisted positioning performance expression is as follows: Where ξ represents the V2X assisted positioning performance, N represents the number of roadside units to be deployed; C i is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit i V2X assisted positioning; C j It is the sum of the positioning accuracy improvement values ​​of vehicles within the coverage area of ​​the roadside unit jV2X auxiliary positioning; Vehicle i The calculation formula for positioning accuracy improvement is as follows: Among them, c i Indicates vehicle v i Positioning accuracy improvement value; p i gx With p i gy Represents vehicle v i GNSS components on the x and y axes; p i tx With p i ty Represents vehicle v i The actual position on the x and y axes; p i ax With p i ay They represent the vehicle v obtained by V2X assisted positioning. i Estimated coordinates on the x and y axes.

4. The method for roadside unit deployment for V2X assisted positioning according to claim 1, characterized in that: In step 3, the specific method for solving the optimal population of roadside units based on genetic algorithm and cuckoo local optimization algorithm is: Step (1), input the urban road network of the area where the roadside unit is to be deployed, the number of roadside units to be deployed, and the V2X assisted positioning coverage of the roadside unit; then initialize the genetic algorithm parameters and cuckoo algorithm parameters according to the number of roadside units to be deployed; the population size of the genetic algorithm needs to correspond to; and the maximum V2X assisted positioning performance expression is used as the fitness function; Step (2), creating an initial population including a number of individuals, each individual having the same characteristics as the number of candidate locations where the roadside unit can be deployed; Create two arrays to store the best fitness value and average fitness value in each iteration; Step (3), calculate the fitness value of each individual in the population and arrange them in order from high to low; Step (4), select individuals with higher fitness values ​​from the current population according to the replacement ratio to serve as the survivors of the old generation after the survival of the fittest, and then randomly select these survivors to become parents, and obtain another part of the new population individuals through single-point crossover; Step (5), randomly generate a random number between 0 and 1 and compare the value with the mutation probability to determine whether the chromosome of the new population individual has mutated, and randomly switch the bits of the chromosome gene digital code within the valid number range, and the maximum digital number does not exceed the total number of candidate locations that can be deployed by the roadside unit; Step (6), using the discretized cuckoo algorithm to perform local search on the generated population; Step (7), repeating steps (3) to (6), continuously performing fitness evaluation, selection, crossover, mutation, and local search until a preset number of iterations is reached; Step (8), finally, the new chromosome returned by the iteratively updated local search is merged with the existing chromosome to obtain a new generation population of the genetic algorithm, that is, the corresponding optimized roadside unit deployment position coordinates are obtained as the best candidate position of the roadside unit.

5. The method for deploying roadside units for V2X assisted positioning according to claim 4, characterized in that: In step (1), the genetic algorithm parameters include population size, number of genetic algorithm iterations, permutation ratio, crossover probability and mutation probability; The cuckoo algorithm parameters include the number of initialized nests, the number of cuckoo algorithm iterations, the discovery probability and the step size factor.

6. The method for deploying roadside units for V2X assisted positioning according to claim 4, characterized in that: In step (6), the specific method of using the discretized cuckoo algorithm to perform local search is: First, the elements in the offspring array are shuffled to initialize a cuckoo population with a size equal to the number of initialized nests; then, the Levy flight strategy is used to update the location of the cuckoo's nest that replaces the host bird's eggs; if the host bird unfortunately finds the disguised bird eggs, the cuckoo needs to find a replacement nest; finally, the cuckoo individual with the highest fitness value in the above process is returned.

Images