A dynamic service system for ramp merging of intelligent connected vehicles

Abstract

The present invention relates to the technical field of intelligent networked vehicles, and discloses a dynamic service system for merging intelligent networked vehicles on ramps. A game model is established for the main road vehicles with time-space conflicts when the ramp intelligent networked vehicles merge, and the type of the main road vehicle is determined: when the main road vehicle is an intelligent networked vehicle, the game benefit coefficient is calibrated to determine the game benefit coefficient under the current driving intention; when the main road conflicting vehicle is a manually driven vehicle, the driving intention of the manually driven vehicle is classified and the game benefit coefficient is calibrated, and the current driving intention is dynamically identified to determine the game benefit coefficient under the current driving intention; by calculating the benefits and equilibrium points, the optimal control strategy of the ramp intelligent networked vehicle during merging is generated, the time-space conflict is eliminated, the problem of low efficiency of merging of ramp intelligent networked vehicles and the risk of collision is solved, and the operation efficiency of the merging process is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent networked vehicles, and in particular to a dynamic service system for ramp merging of intelligent networked vehicles. Background Art

[0002] After years of development, intelligent connected technology has made significant progress. However, due to issues such as technology maturity, legal supervision and social acceptance, it is still difficult to achieve complete unmanned driving. It will take a long period of time for intelligent connected vehicles and manually driven cars to coexist, that is, intelligent connected vehicles need to interact with surrounding vehicles of different operating categories.

[0003] Ramp is an indispensable part of interchange. When ramp vehicles enter the main road traffic, the merging ramp vehicles will seriously interfere with the main road traffic flow and have various negative impacts on traffic efficiency and safe driving. Under the heterogeneous traffic flow composed of manually driven vehicles and intelligent connected vehicles, this mixed driving environment will become more complicated.

[0004] When merging on ramps, there are a large number of time-space conflict scenarios. When a ramp intelligent connected vehicle conflicts with a main road vehicle, the traditional method is to use the roadside perception system to determine whether the ramp entrance is in a congested state, and then design an ideal merging control distance through safe acceleration to control each area separately (CN116386385B), or to form a variety of strategy sets for mixed traffic, and select the optimal strategy with the goal of minimizing travel time loss (CN114999152B). However, the above methods do not take into account the important factors that the conflicting vehicles are the control differences between intelligent connected vehicles and manually driven vehicles, as well as the changes in driving intentions of manually driven vehicles. As a result, when the conflicting objects on the main road are different, it is impossible to adapt to the changes in driving intentions in a timely manner to adopt an effective merging strategy for intelligent connected vehicles, resulting in reduced merging efficiency. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a dynamic service system for merging ramps of intelligent networked vehicles, which establishes a game model by screening out the main road vehicles that conflict with the ramp intelligent networked vehicles. When the main road vehicle is an intelligent networked vehicle, the game benefit coefficient is calibrated to determine the game benefit coefficient under the current driving intention; when the main road conflicting vehicle is a manually driven vehicle, the driving intention of the manually driven vehicle is classified and the game benefit coefficient is calibrated, and the current driving intention is dynamically identified to determine the game benefit coefficient under the current driving intention; by calculating the benefits and equilibrium points, the optimal control strategy of the ramp intelligent networked vehicle during merging is generated, and the time and space conflicts between the main road manually driven vehicles, intelligent networked vehicles and ramp intelligent networked vehicles are eliminated during the merging process.

[0006] The present invention provides a dynamic service system for ramp merging of intelligent networked vehicles, comprising: a main road vehicle, a ramp vehicle, a roadside communication unit, a roadside sensing unit and a roadside computing unit; during the merging process, the main road vehicle is an intelligent networked vehicle or a manually driven vehicle, and the ramp vehicle is an intelligent networked vehicle;

[0007] The roadside communication unit is wirelessly connected to the on-board communication unit of the intelligent networked vehicle, and is used to send the driving intention and motion state data of the intelligent networked vehicle to the roadside computing unit;

[0008] The roadside sensing unit is used to sense the motion state data of the manually driven vehicle on the main road, and send the motion state data of the manually driven vehicle to the roadside computing unit;

[0009] The roadside computing unit is used to determine whether there is a time-space conflict between the main road vehicles and the ramp intelligent networked vehicles during the merging process, and to establish a game model for the ramp intelligent networked vehicles and the main road vehicles that have a time-space conflict;

[0010] The roadside computing unit is also used to determine that the main road vehicle with time-space conflict is an intelligent connected vehicle, receive different driving intentions and motion state data of the ramp intelligent connected vehicle and the main road intelligent connected vehicle sent by the roadside communication unit, and calibrate the game benefit coefficient to determine the game benefit coefficient under the current driving intention, and calculate the benefits and equilibrium points to generate the optimal control strategy of the ramp intelligent connected vehicle during merging, so as to eliminate the time-space conflict;

[0011] The roadside computing unit is also used to determine that when the main road vehicle with time-space conflict is a manually driven vehicle, it receives the motion state data of the main road manually driven vehicle sent by the roadside perception unit, and classifies the driving intention of the manually driven vehicle on the main road and calibrates the game profit coefficient. By identifying the driving intention of the manually driven vehicle on the main road at the current moment, the game profit coefficient under the current driving intention is determined, and the profit and equilibrium point are calculated to generate the optimal control strategy for the intelligent connected vehicle on the ramp during merging, thereby eliminating time-space conflicts.

[0012] Furthermore, the game model includes game objects, game strategies and game benefits;

[0013] The game objects are conflicting intelligent connected vehicles on ramps, as well as intelligent connected vehicles on the main road or manually driven vehicles on the main road;

[0014] Game strategies include merging or not merging on the ramp, and slowing down or not giving way to vehicles on the main road;

[0015] Gaming benefits include safety benefits, speed benefits, efficiency benefits, and comfort benefits.

[0016] Furthermore, when the roadside computing unit determines that the conflicting main road vehicle is an intelligent connected vehicle, it further includes:

[0017] The driving intentions of the intelligent connected vehicles on the ramp include accelerating merging AM, merging in the original state NM, and decelerating merging DM. The driving intentions of the intelligent connected vehicles on the main road include accelerating passing AP, decelerating passing DP, and passing in the original state NP.

[0018] Movement status including following distance l ,speed v , acceleration a and impact J .

[0019] Furthermore, the roadside computing unit collects the driving intention and motion state data of the ramp intelligent networked vehicles and the main road intelligent networked vehicles with time and space conflicts, and calibrates the game benefit coefficient, specifically including:

[0020] The root mean square values ​​of the motion state data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles under three types of driving intentions are calculated respectively;

[0021] The sum of the root mean square values ​​of each type of motion state data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles under different driving intentions is calculated respectively, and the proportion of the root mean square value of each type of motion state data of each type of driving intention in the sum of the corresponding root mean square values ​​is further calculated to obtain the dimensionless data corresponding to each type of motion state for each type of driving intention; the sum of the dimensionless data corresponding to all motion state data of each type of driving intention of the ramp intelligent connected vehicles and the main road intelligent connected vehicles is further calculated, and the proportion of the dimensionless data corresponding to each type of motion state data of each type of driving intention in the sum of the corresponding dimensionless data is calculated to obtain the game benefit coefficient corresponding to different benefits under each type of driving intention.

[0022] Furthermore, the roadside communication unit collects the driving intention and motion status data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles at the current moment, and determines the game profit coefficient under the current driving intention.

[0023] Furthermore, when the roadside computing unit determines that the conflicting main road vehicle is a manually driven vehicle, the method further includes:

[0024] The motion status data of manually driven vehicles on the main road includes the following distance l ,speed v , acceleration a and impact J ;

[0025] The driving intentions of manually driven vehicles on the main road include accelerating merging AM, merging in place NM, and decelerating merging DM.

[0026] Furthermore, the roadside computing unit classifies the driving intentions of the manually driven vehicles on the main road with time-space conflicts and calibrates the game benefit coefficients, specifically including:

[0027] A number of time-varying motion state data of manually driven vehicles are collected in advance by the roadside perception unit, and a self-organizing feature mapping neural network is used for unsupervised learning to output the categories of motion states of manually driven vehicles. Then, cluster analysis is performed on the categories of motion states of manually driven vehicles to generate cluster centers of different motion state data of three types of driving intentions: accelerated merging AM, original merging NM, and decelerated merging DM.

[0028] The sum of different driving intentions of the cluster centers of each type of motion state data of manually driven vehicles on the main road is calculated, and the proportion of the sum of the cluster centers of each type of motion state data and different driving intentions is further calculated to obtain the dimensionless data corresponding to each type of motion state for each type of driving intention; the sum of the dimensionless data corresponding to all motion state data of each type of driving intention of manually driven vehicles on the main road is further calculated, and the proportion of the dimensionless data corresponding to each type of motion state data of each type of driving intention in the sum of the corresponding dimensionless data is calculated to obtain the game benefit coefficient corresponding to different benefits under each type of driving intention.

[0029] Furthermore, the roadside calculation unit classifies the driving intention of the manual driving vehicle on the main road with time-space conflict and calibrates the game benefit coefficient, further including:

[0030] The roadside computing unit labels the motion state data of the driving intention classification of manually driven vehicles on the main road with driving intention labels to obtain a driving intention recognition training set.

[0031] Furthermore, the roadside perception unit collects the motion state data of the manually driven vehicles on the main road at the current moment. The roadside calculation unit uses a support vector machine to analyze the driving intention of the current motion state of the manually driven vehicles on the main road based on the driving intention recognition training set, and determines the game profit coefficient under the current driving intention.

[0032] Furthermore, if the roadside computing unit determines that the main road vehicle and the ramp intelligent connected vehicle have not completed the merging process, it will cyclically collect the motion status data of the main road vehicle and the ramp intelligent connected vehicle and dynamically update the merging strategy to eliminate the merging conflict between vehicles.

[0033] The embodiments of the present invention have the following technical effects:

[0034] The present invention provides a dynamic service system for ramp merging of intelligent connected vehicles, establishes a game model for main road and ramp vehicles with time-space conflicts, realizes the calibration of game benefit coefficients by collecting driving intention and motion state data of intelligent connected vehicles, realizes the classification of driving intentions, calibration of game benefit coefficients and recognition of driving intentions by collecting motion state data of manually driven vehicles, calculates the vehicle game benefit at the current moment based on the calibrated game benefit coefficients and the driving intention at the current moment, obtains the optimal strategy for merging of intelligent connected vehicles by calculating the equilibrium point, and dynamically updates the optimal strategy through the vehicle motion state to eliminate the time-space conflicts between main road vehicles and intelligent connected vehicles on the ramp during the merging process, solves the problems of low ramp merging efficiency and collision risks, and improves the operation efficiency of the transportation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 is a schematic diagram of ramp merging provided by an embodiment of the present invention;

[0037] Figure 2 It is a flow chart of a dynamic service method for ramp merging of intelligent connected vehicles provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0039] like Figure 1 The schematic diagram of the intelligent connected vehicle ramp merging dynamic service system provided by an embodiment of the present invention is shown. The intelligent connected vehicle ramp merging dynamic service system specifically includes: a main road vehicle, a ramp vehicle, a roadside communication unit, a roadside perception unit and a roadside computing unit.

[0040] During the merging process, the main road vehicle is an intelligent connected vehicle or a manually driven vehicle, and the ramp vehicle is an intelligent connected vehicle.

[0041] The roadside communication unit is wirelessly connected to the on-board communication unit of the intelligent networked vehicle to send the driving intention and motion state data of the intelligent networked vehicle to the roadside computing unit;

[0042] The roadside sensing unit is used to sense the motion state data of the manually driven vehicle on the main road, and send the motion state data of the manually driven vehicle to the roadside computing unit;

[0043] The roadside computing unit is used to determine whether there is a time-space conflict between the vehicles on the main road and the intelligent connected vehicles on the ramp during the merging process: If the roadside computing unit determines that there is a time-space conflict between the vehicles on the main road and the intelligent connected vehicles on the ramp during the merging process, the roadside computing unit establishes a game model and action set for the intelligent connected vehicles on the ramp and the main road vehicles, where the game model includes game objects, game strategies and game benefits.

[0044] The game objects are conflicting intelligent networked vehicles on ramps, and intelligent networked vehicles on the main road or manually driven vehicles on the main road;

[0045] The game strategy includes ramp vehicles merging, not merging, and main road vehicles slowing down to avoid, or not avoiding;

[0046] The gaming benefits include safety benefits, speed benefits, efficiency benefits, and comfort benefits.

[0047] The roadside computing unit determines the type of vehicle on the main road that has a merging conflict with the ramp intelligent connected vehicle. If the conflicting vehicle on the main road is an intelligent connected vehicle, the roadside communication unit collects the driving intention and motion parameters of the ramp intelligent connected vehicle and the main road intelligent connected vehicle and sends them to the roadside computing unit.

[0048] The driving intentions of the ramp intelligent networked vehicles include accelerated merging AM, original merging NM, and decelerated merging DM, and the driving intentions of the main road intelligent networked vehicles include accelerated passing AP, original passing NP, and decelerated passing DP;

[0049] The motion status data includes the following distance l ,speed v , acceleration a and impact J .

[0050] At the same time, the roadside calculation unit calibrates and calculates the game benefit coefficients of the ramp and main road intelligent connected vehicles, including respectively calculating the sum of the root mean square values ​​of each type of motion state data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles under different driving intentions, and further calculating the proportion of the root mean square value of each type of motion state data of each type of driving intention to the corresponding sum of the root mean square values, and obtaining the dimensionless data corresponding to each type of motion state for each type of driving intention; further calculating the sum of the dimensionless data corresponding to all motion state data of each type of driving intention of the ramp intelligent connected vehicles and the main road intelligent connected vehicles, calculating the proportion of the dimensionless data corresponding to each type of motion state data of each type of driving intention to the corresponding sum of the dimensionless data, and obtaining the game benefit coefficients corresponding to different benefits under each type of driving intention;

[0051] The roadside computing unit determines that the vehicle type on the main road that has a merging conflict with the ramp intelligent connected vehicle is a manually driven vehicle, and the roadside sensing unit collects the motion state data of the manually driven vehicle on the main road and sends it to the roadside computing unit;

[0052] The motion status data of manually driven vehicles on the main road includes the following distance l ,speed v , acceleration a and impact J .

[0053] The roadside sensing unit collects several pieces of motion state data of manually driven vehicles that change over time in advance, uses a self-organizing feature mapping neural network for unsupervised learning, outputs the categories of motion states of manually driven vehicles, and then performs cluster analysis on the categories of motion states of manually driven vehicles to generate three types of driving intention cluster centers: accelerated passing AP, decelerated passing DP, and original passing NP.

[0054] The sum of different driving intentions of each type of motion state data cluster center of manually driven vehicles on the main road is calculated, and the proportion of each type of motion state data cluster center to the sum of the different driving intentions is further calculated to obtain dimensionless data corresponding to each type of motion state for each type of driving intention; the sum of dimensionless data corresponding to all motion state data of each type of driving intention of manually driven vehicles on the main road is further calculated, and the proportion of dimensionless data corresponding to each type of motion state data of each type of driving intention to the corresponding sum of dimensionless data is calculated to obtain the game benefit coefficient corresponding to different benefits under each type of driving intention.

[0055] The roadside perception unit collects the motion state data of manual driving on the main road at the current moment. The roadside calculation unit labels several pre-collected motion state data with driving intention labels according to the driving intention classification results, and uses a support vector machine to analyze the driving intention of the current motion state to determine the game profit coefficient under the current driving intention.

[0056] The roadside calculation unit calculates the benefits of the conflicting vehicles based on the game profit coefficient, and calculates the equilibrium point to generate the optimal merging strategy for the ramp intelligent connected vehicles. If the merging is not completed, the motion state data of the conflicting vehicles is cyclically collected and the merging strategy is dynamically updated to eliminate the spatiotemporal conflict between the ramp intelligent connected vehicles and the main road vehicles during the merging process.

[0057] The embodiment of the present application may also be a dynamic service method for intelligent networked vehicle ramp merging, such as Figure 2 The flowchart of the intelligent networked vehicle ramp merging dynamic service method shown in the figure includes the following steps:

[0058] The merging dynamic service system collects the motion status data of the ramp intelligent connected vehicles and the motion status data of the main road intelligent connected vehicles and manually driven vehicles, and determines whether there is a conflict between the main road vehicles and the ramp intelligent connected vehicles. If there is a conflict, a game model is established.

[0059] If the main road intelligent connected vehicle conflicts with the ramp intelligent connected vehicle, the root mean square values ​​of the motion state data of the ramp intelligent connected vehicle under the three driving intentions of acceleration merging AM, original merging NM, and deceleration merging DM and the main road intelligent connected vehicle under the three driving intentions of acceleration passing AP, original passing NP, and deceleration passing DP are calculated respectively, and the sum of the root mean square values ​​of each type of motion state data of the ramp intelligent connected vehicle and the main road intelligent connected vehicle under different driving intentions is calculated respectively, and the root mean square value of each type of motion state data of each type of driving intention is further calculated to account for the corresponding average The dimensionless data corresponding to each type of motion state of each type of driving intention is obtained by taking the proportion of the sum of the square root values; the sum of the dimensionless data corresponding to all motion state data of each type of driving intention of the ramp intelligent connected vehicles and the main road intelligent connected vehicles is further calculated, and the proportion of the dimensionless data corresponding to each type of motion state data of each type of driving intention in the corresponding sum of the dimensionless data is calculated, so as to obtain the game profit coefficient corresponding to different benefits under each type of driving intention; the game profit coefficient at the current moment is determined by collecting the driving intentions of the ramp and main road intelligent connected vehicles at the current moment.

[0060] If a conflict occurs between a manually driven vehicle on the main road and an intelligent connected vehicle on the ramp, the motion state data of the manually driven vehicle on the main road is collected, and the self-organizing feature mapping neural network is used for unsupervised learning to output the category of the motion state of the manually driven vehicle. Then, cluster analysis is performed on the category of the motion state of the manually driven vehicle to generate three types of driving intention cluster centers: accelerated passage AP, original passage NP, and decelerated passage DP.

[0061] Calculate the sum of different driving intentions of each type of motion state data cluster center of manually driven vehicles on the main road, further calculate the proportion of each type of motion state data cluster center and the sum of the different driving intentions, and obtain dimensionless data corresponding to each type of motion state for each type of driving intention; further calculate the sum of dimensionless data corresponding to all motion state data of each type of driving intention of manually driven vehicles on the main road, calculate the proportion of dimensionless data corresponding to each type of motion state data of each type of driving intention in the sum of the corresponding dimensionless data, and obtain the game benefit coefficient corresponding to different benefits under each type of driving intention;

[0062] The motion state data of manual driving on the main road at the current moment is collected. The roadside computing unit labels several pieces of motion state data collected in advance with driving intention labels according to the driving intention classification results, and uses support vector machine to analyze the driving intention of the current motion state to determine the game benefit coefficient under the current driving intention.

[0063] Based on the obtained game profit coefficient, the benefits of the conflicting vehicles are calculated, and the equilibrium point is calculated to generate the optimal merging strategy for the ramp intelligent connected vehicles, eliminating the spatiotemporal conflict between the ramp intelligent connected vehicles and the main road vehicles during the merging process.

[0064] In an embodiment of the present application, the process of establishing the vehicle game model includes:

[0065] Strategy set for main road HDV or human-driven vehicle (HDV) {1: slow down and avoid, 2: do not avoid}, strategy set of ramp intelligent connected vehicle (CAV) {1: confluence, 2: non-confluence}, as shown in the following table:

[0066]

[0067] Gaming benefits include safety benefits, speed benefits, efficiency benefits, and comfort benefits.

[0068] In the embodiment of the present application, the method for calibrating the game benefit coefficient of the intelligent networked vehicle is: collecting the motion state data of the intelligent networked vehicles with the intention of accelerating and merging on N ramps, , calculate the RMS value of each motion parameter,

[0069] , , , .

[0070] Similarly, the RMS values ​​of the motion state data of the ramp intelligent connected vehicle under the original merging NM and deceleration merging DM intentions can be obtained, which are , , , and , , , .

[0071] The sum of the root mean square values ​​of each type of motion state data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles under different driving intentions is calculated respectively, and the proportion of the root mean square value of each type of motion state data of each type of driving intention to the corresponding sum of the root mean square values ​​is further calculated to obtain the dimensionless data corresponding to each type of motion state for each type of driving intention:

[0072] , , ;

[0073] , , ;

[0074] , , ;

[0075] , , .

[0076] The sum of dimensionless data corresponding to all motion state data of the ramp intelligent networked vehicle under the intention of accelerating merging is further calculated, and the proportion of dimensionless data corresponding to each type of motion state data in the sum of the corresponding dimensionless data is calculated, and the game benefit coefficients of safety benefit, speed benefit, efficiency benefit, and comfort benefit of the ramp intelligent networked vehicle under the intention of accelerating merging are obtained as follows:

[0077] , ,

[0078] , .

[0079] Similarly, the game benefit coefficients of the safety benefit, speed benefit, efficiency benefit, and comfort benefit of the ramp intelligent connected vehicle under the original merging intention and the deceleration merging intention can be obtained as follows: , , , and , , , .

[0080] Similarly, the game benefit coefficients of safety benefit, speed benefit, efficiency benefit, and comfort benefit of the main road intelligent connected vehicle under accelerated passage AP, original passage NP, and decelerated passage DP can be obtained as follows: , , , , , , , and , , , .

[0081] In an embodiment of the present application, the method for classifying driving intentions of manually driven vehicles on the main road is: collecting K pieces of motion state data of manually driven vehicles on the main road, , self-organizing feature mapping neural network is used for unsupervised learning, the calculation layer size is set to 4*4, and quantization error is introduced Evaluate the clustering effect, is the weight vector, and 16 output categories are obtained. Then, the K-means method is used to cluster the 16 categories output by the self-organizing feature mapping neural network into three types of driving intentions, namely, accelerating, passing as is, and decelerating, and the cluster centers of different motion state data under the three types of driving intentions are obtained:

[0082] Manually driven vehicles on the main road accelerate , , , ; Manually driven vehicles on the main road pass as they are: , , , ; Manually driven vehicles on the main road slow down: , , , .

[0083] In the embodiment of the present application, the method for calibrating the gaming benefit coefficient of manually driven vehicles on the main road is:

[0084] The sum of the cluster center of each type of motion state data of manually driven vehicles on the main road in accelerated passage, original state passage and decelerated passage is calculated, and the proportion of the cluster center of each type of motion state data to the sum is further calculated to obtain the dimensionless data corresponding to each type of driving intention and each type of motion state:

[0085] , , ;

[0086] , , ;

[0087] , , ;

[0088] , , .

[0089] The sum of dimensionless data corresponding to each type of motion state data under the acceleration intention of the manual driving vehicle on the main road is further calculated, and the proportion of dimensionless data corresponding to each type of motion state data under the acceleration intention to the sum of the corresponding dimensionless data is calculated, and the game benefit coefficients of safety benefit, speed benefit, efficiency benefit, and comfort benefit under the acceleration intention of the manual driving vehicle on the main road when merging are obtained:

[0090] , , , .

[0091] Similarly, the game benefit coefficients of safety benefit, speed benefit, efficiency benefit, and comfort benefit of manual driving vehicles on the main road under the intention of original passage and deceleration when merging can be calculated as follows: , , , and , , , .

[0092] In the embodiment of the present application, the method for identifying the driving intention of a main road manually driven vehicle is as follows: the motion state data of the main road manually driven vehicle driving intention classification is marked with a driving intention label to obtain a driving intention identification training set ,in is the motion status data, =1, 2, 3, representing accelerated passage, original passage, and decelerated passage, respectively. . Solve for the separating hyperplane Get the classification decision function ,in and are the coefficients and intercepts of the hyperplane equation. By inputting the motion state data of the manual driving vehicle on the main road at the current moment, the driving intention at the current moment can be obtained, and the game benefit coefficient at the current moment can be determined based on the obtained driving intention.

[0093] In the embodiment of the present application, the method for calculating the conflicting vehicle benefit function and solving the equilibrium point to generate the optimal merging strategy is: represents the revenue of the ramp intelligent connected vehicle CAV, and represents the benefits of the main road intelligent connected vehicle CAV or the manually driven vehicle HDV. The problem of solving the equilibrium point is transformed into the problem of finding the optimal solution for the benefits of both parties in the game.

[0094] The game payoff matrix of ramp intelligent networked vehicles and main road manually driven vehicles is shown in the following table. If there is a strategy between ramp intelligent networked vehicles and main road intelligent networked vehicles or manually driven vehicles So that equation (1) holds, then is the Nash equilibrium solution of the current game, The maximum benefit of the ramp intelligent connected vehicles and manually driven vehicles under the current game, and It is the optimal pure strategy for both the ramp intelligent connected vehicles and the main road intelligent connected vehicles / manually driven vehicles.

[0095]

[0096] (1)

[0097] A game profit function is established, where the profit function of the ramp intelligent connected vehicle is:

[0098]

[0099] The revenue function of the main road intelligent connected vehicle is:

[0100]

[0101] The revenue function of a manually driven vehicle on the main road is:

[0102]

[0103] in, The driving intention of the main road intelligent connected vehicle is one of the following: accelerated merging AM, original merging NM, and decelerated merging DM. The driving intention of the main road intelligent connected vehicle or manually driven vehicle is one of acceleration AP, original state NP, and deceleration DP.

[0104] in , is the safety benefit of the ramp intelligent connected vehicles and the main road intelligent connected vehicles / manually driven vehicles, and is calculated as follows:

[0105]

[0106] In the formula, is the minimum safe distance, is the distance between the ramp intelligent connected vehicle and the main road intelligent connected vehicle / manually driven vehicle at time t, is the expected vehicle spacing.

[0107] in , is the speed gain of the ramp intelligent connected vehicles and the main road intelligent connected vehicles / manually driven vehicles, calculated as:

[0108]

[0109]

[0110] In the formula, , is the current driving speed of the intelligent connected vehicle on the ramp and the maximum allowed driving speed of the ramp, , The current driving speed of the intelligent connected vehicle / manually driven vehicle on the main road and the maximum allowed driving speed on the main road.

[0111] in , is the efficiency gain of the ramp intelligent connected vehicles and the main road intelligent connected vehicles / manually driven vehicles, calculated as follows:

[0112]

[0113]

[0114] In the formula, , , , , , is the initial speed of ramp vehicles and main road vehicles, , The expected speed of ramp vehicles and main road vehicles to execute a certain merging strategy, , The acceleration of the ramp vehicle and the main road vehicle when executing a merging strategy, , is the initial distance between the ramp vehicle and the main road vehicle when they reach the merging point, , is the time it takes for ramp vehicles and main road vehicles to reach the merging point at their initial speeds, , The time it takes for ramp vehicles and main road vehicles to reach the merging point after executing a certain merging strategy.

[0115] in , is the comfort benefit of the ramp intelligent connected vehicles and the main road intelligent connected vehicles / manually driven vehicles, and is calculated as follows:

[0116]

[0117]

[0118] In the formula, , is the impact degree between the current ramp vehicles and the main road vehicles, , It is the maximum impact degree allowed for vehicles on ramps and main roads.

[0119] An embodiment of the present application may also be an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the steps of a dynamic service method for ramp merging of intelligent connected vehicles provided in any embodiment of the present application.

[0120] The computer program instructions may be written in any combination of one or more programming languages ​​to form program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user computing device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0121] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0122] It should be noted that the terms used in this application are only for describing specific embodiments, rather than limiting the scope of this application. As shown in the specification and claims of this application, unless the context clearly indicates an exception, the words "one", "a", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method or device including the elements.

[0123] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and its core ideas of this application. The above are only preferred implementation methods of this application. It should be pointed out that due to the limitations of textual expression and the objective existence of infinite specific structures, ordinary technicians in this technical field can make several improvements, modifications or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes or combinations, or the direct application of the concept and technical solution of the invention to other occasions without improvement, should be regarded as the scope of protection of this application.

Claims

1. A dynamic service system for intelligent networked vehicle ramp merging, characterized in that: include: A main road vehicle, a ramp vehicle, a roadside communication unit, a roadside sensing unit, and a roadside computing unit; during the merging process, the main road vehicle is an intelligent networked vehicle or a manually driven vehicle, and the ramp vehicle is an intelligent networked vehicle; The roadside communication unit is wirelessly connected to the on-board communication unit of the intelligent networked vehicle, and is used to send the driving intention and motion state data of the intelligent networked vehicle to the roadside computing unit; The roadside sensing unit is used to sense the motion state data of the manually driven vehicle on the main road, and send the motion state data of the manually driven vehicle to the roadside computing unit; The roadside computing unit is used to determine whether there is a time-space conflict between the main road vehicle and the ramp intelligent networked vehicle during the merging process, and to establish a game model for the ramp intelligent networked vehicle and the main road vehicle that have a time-space conflict; The roadside calculation unit is also used to determine that when the main road vehicle with time-space conflict is an intelligent connected vehicle, it receives different driving intentions and motion state data of the ramp intelligent connected vehicle and the main road intelligent connected vehicle sent by the roadside communication unit, and calibrates the game profit coefficient to determine the game profit coefficient under the current driving intention, and generates the optimal control strategy of the ramp intelligent connected vehicle when merging by calculating the profit and equilibrium point to eliminate the time-space conflict, wherein the driving intentions of the ramp intelligent connected vehicle include accelerated merging AM, original merging NM and decelerated merging DM, the driving intentions of the main road intelligent connected vehicle include accelerated passage AP, decelerated passage DP and original passage NP, and the motion state data includes the following vehicle distance l ,speed v , acceleration a and impact J ; The roadside computing unit is also used to determine that when the main road vehicle with time-space conflict is a manually driven vehicle, receive the motion state data of the main road manually driven vehicle sent by the roadside sensing unit, and classify the driving intention of the manually driven vehicle on the main road and calibrate the game benefit coefficient, determine the game benefit coefficient under the current driving intention by identifying the driving intention of the manually driven vehicle on the main road at the current moment, and calculate the benefit and equilibrium point to generate the optimal control strategy of the ramp intelligent connected vehicle during merging, eliminate the time-space conflict, wherein the motion state data of the main road manually driven vehicle includes the following vehicle spacing l ,speed v , acceleration a and impact J， The driving intention of the manual vehicle on the main road includes acceleration merging AM, original merging NM and deceleration merging DM; The roadside computing unit collects the driving intention and motion state data of the ramp intelligent networked vehicles and the main road intelligent networked vehicles with time and space conflicts, and calibrates the game benefit coefficient, specifically including: The root mean square values ​​of the motion state data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles under three types of driving intentions are calculated respectively; The sum of the root mean square values ​​of each type of motion state data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles under different driving intentions is calculated respectively, and the proportion of the root mean square value of each type of motion state data of each type of driving intention in the corresponding sum of the root mean square values ​​is further calculated to obtain the dimensionless data corresponding to each type of motion state of each type of driving intention; the sum of the dimensionless data corresponding to all motion state data of each type of driving intention of the ramp intelligent connected vehicles and the main road intelligent connected vehicles is further calculated, and the proportion of the dimensionless data corresponding to each type of motion state data of each type of driving intention in the corresponding sum of the dimensionless data is calculated to obtain the game benefit coefficient corresponding to different benefits under each type of driving intention; The roadside calculation unit classifies the driving intention of the main road manually driven vehicle with time and space conflict and calibrates the game benefit coefficient, specifically including: A number of time-varying motion state data of manually driven vehicles are collected in advance by the roadside perception unit, and a self-organizing feature mapping neural network is used for unsupervised learning to output the categories of motion states of manually driven vehicles. Then, cluster analysis is performed on the categories of motion states of manually driven vehicles to generate cluster centers of different motion state data of three types of driving intentions: accelerated merging AM, original merging NM, and decelerated merging DM. The sum of different driving intentions of each type of motion state data cluster center of manually driven vehicles on the main road is calculated, and the proportion of each type of motion state data cluster center to the sum of the different driving intentions is further calculated to obtain dimensionless data corresponding to each type of motion state for each type of driving intention; the sum of dimensionless data corresponding to all motion state data of each type of driving intention of manually driven vehicles on the main road is further calculated, and the proportion of dimensionless data corresponding to each type of motion state data of each type of driving intention to the corresponding sum of dimensionless data is calculated to obtain the game benefit coefficient corresponding to different benefits under each type of driving intention.

2. The intelligent networked vehicle ramp merging dynamic service system according to claim 1, characterized in that: The game model includes game objects, game strategies and game benefits; The game objects are conflicting intelligent networked vehicles on ramps, and intelligent networked vehicles on the main road or manually driven vehicles on the main road; The game strategy includes whether ramp vehicles merge or not merge, and whether main road vehicles slow down to avoid or not avoid; The gaming benefits include safety benefits, speed benefits, efficiency benefits and comfort benefits.

3. The intelligent networked vehicle ramp merging dynamic service system according to claim 1, characterized in that: The roadside communication unit collects the driving intention and motion status data of the ramp intelligent connected vehicles and the main road intelligent connected vehicles at the current moment, and determines the game profit coefficient under the current driving intention.

4. The intelligent networked vehicle ramp merging dynamic service system according to claim 1, characterized in that: The roadside calculation unit classifies the driving intention of the main road manually driven vehicle with time and space conflict and calibrates the game benefit coefficient, further comprising: The roadside computing unit labels the motion state data of the driving intention classification of the manually driven vehicles on the main road with a driving intention label to obtain a driving intention recognition training set.

5. The intelligent networked vehicle ramp merging dynamic service system according to claim 4, characterized in that: The roadside perception unit collects the motion state data of manually driven vehicles on the main road at the current moment. The roadside calculation unit uses a support vector machine to analyze the driving intention of the current motion state of manually driven vehicles on the main road based on the driving intention recognition training set, and determines the game profit coefficient under the current driving intention.

6. The intelligent networked vehicle ramp merging dynamic service system according to claim 1 or 5, characterized in that: If the roadside computing unit determines that the main road vehicle and the ramp intelligent connected vehicle have not completed the merging process, it will cyclically collect the motion status data of the main road vehicle and the ramp intelligent connected vehicle and dynamically update the merging strategy to eliminate the merging conflicts between vehicles.

Images