Joint simulation software parameter correction method for automatic driving

Abstract

The invention discloses a joint simulation software parameter correction method for automatic driving. The method comprises the following steps: (1) building an SUMO simulation scene; (2) selecting evaluation indexes; (3) according to initial evaluation of parameter default values, judging whether requirements are met, and if yes, entering step (6); if not, performing the step (4); (4) determining parameters to be corrected and parameter values; (5) evaluating the model after the parameter values are adjusted, judging whether requirements are met or not, if yes, entering the step (6), and if not, returning to the step (4); (6) completing the construction of a joint simulation platform; (7) selecting an evaluation index of joint simulation interaction data accuracy; and (8) preliminarily evaluating the data accuracy of the joint simulation model according to a default time interval of data updating, judging whether requirements are met or not, if not, adjusting the time interval of data updating, and if yes, ending. The precision of a traffic simulation model is effectively improved, and the credibility of a simulation result is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of parameter correction during joint simulation of multiple simulation software in an intelligent transportation Internet of Things, and in particular to a joint simulation software parameter correction method for autonomous driving. Background Art

[0002] With the rapid development of new information technologies such as big data and artificial intelligence, intelligence and networking have gradually become important directions for the development of the new generation of automobiles. At present, the research on intelligent transportation Internet of Things is mainly focused on the technical system related to autonomous driving vehicles, information interaction, and cross-field research. Due to the requirements of the test environment for autonomous driving vehicles, related research is mostly in the simulation experiment stage. A small number of studies obtain research results by building a real traffic environment. This method is costly and difficult to reflect various scene problems. Therefore, building simulation experiment simulation software and simulation platforms is an important link that restricts the research and development and application of technologies in this field.

[0003] The accuracy of simulation can be effectively improved by using a variety of simulation software, building simulation platforms, and constructing joint simulation models. Among them, the accuracy of the parameters in the joint simulation model directly affects whether the entire simulation system can accurately restore the actual scene. Therefore, effectively correcting the parameters in the joint simulation model can ensure the effectiveness of the simulation system. Since the simulation software that is currently widely used mostly sets default parameters based on foreign conditions, there may be certain differences between its background traffic flow parameters and the traffic operation conditions in my country, so it is necessary to reasonably correct the parameters of the traffic simulation model. Although there are a lot of studies on the parameter correction of traffic simulation models or network simulation models, there is still a lack of research on the interaction between joint simulation software. Therefore, how to ensure the real simulation of traffic scenes while maximizing the accuracy of the transmission of traffic parameter information between software has become the key to joint simulation parameter correction. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the technical problem to be solved by this patent application is how to provide a joint simulation software parameter correction method for autonomous driving, by correcting the parameters of the traffic simulation model so that the error between the traffic simulation data and the actual observation data is within an acceptable range, thereby effectively improving the accuracy of the traffic simulation model; at the same time, the interval for data exchange between the joint simulation software is adjusted to ensure that the traffic flow parameters can be accurately reflected in the network simulation software.

[0005] Using the joint simulation platform for testing and evaluation is a key step in the development and application of intelligent connected vehicle technology. The accuracy of the joint simulation model determines whether the simulation evaluation results can provide scientific and effective theoretical support for subsequent research. Therefore, reasonable correction of the joint simulation parameters is an important prerequisite for ensuring the credibility of the simulation results output by the joint simulation model.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] A method for correcting parameters of a joint simulation software for autonomous driving comprises the following steps:

[0008] (1) Construction of SUMO simulation scene;

[0009] (2) Selecting evaluation indicators for traffic simulation parameter correction;

[0010] (3) According to the initial evaluation of the parameter default values, determine whether the initial SUMO simulation model meets the requirements. If so, proceed to step (6); if not, proceed to step (4);

[0011] (4) Determine the parameters to be calibrated and their values;

[0012] (5) Evaluate the traffic simulation model after adjusting the parameter values ​​to determine whether the model after parameter correction meets the requirements. If so, proceed to step (6); if not, return to step (4);

[0013] (6) Complete the construction of the joint simulation platform based on the Veins framework;

[0014] (7) Select evaluation indicators for the accuracy of joint simulation interaction data;

[0015] (8) Perform a preliminary evaluation on the data accuracy of the joint simulation model according to the default time interval for data update to determine whether the initial joint simulation model meets the requirements. If not, adjust the time interval for data update. If yes, end the process.

[0016] Specifically, in step (1), when the SUMO simulation scene is constructed, the construction of the simulated road network is completed in SUMO by investigating the current status of the scene road network structure, road width, and intersection channelization. At the same time, the traffic volume, road section flow direction, average vehicle speed, and traffic composition data obtained from the investigation are used as traffic parameters to input into the simulation.

[0017] Specifically, in step (2), road flow, traffic capacity, and travel time are selected as evaluation indicators for traffic simulation parameter correction.

[0018] Specifically, the process of step (3) is as follows:

[0019] Based on the initial simulation results, the sample standard deviation is used to estimate the minimum number of simulations for each evaluation index. The reasonable number of simulations is determined through analysis. The formula is as follows:

[0020]

[0021] Where: n is the minimum number of simulations, S is the sample standard deviation, K is the confidence level, and e is the allowable error value;

[0022] Run the simulation n times according to the default values ​​of the parameters in the simulation model, and evaluate the initial simulation results using the absolute percentage error to determine whether the simulation software model under the default values ​​needs parameter correction. The formula is as follows:

[0023]

[0024] Where: T is the target threshold, y r is the measured value of the evaluation index, y s Output the average value of the evaluation index for the simulation system;

[0025] If the simulation evaluation results of the simulation model under the condition of default parameter values ​​and the error between the measured values ​​have met the requirements, there is no need to calibrate the parameters of the simulation model;

[0026] If the simulation evaluation results of the simulation model under the default parameter value conditions and the error between the measured values ​​cannot meet the requirements, the simulation model needs to be calibrated.

[0027] Specifically, in step (4), before determining the parameters to be calibrated and the parameter values, it is necessary to perform a preliminary selection of the parameters to be calibrated and a parameter sensitivity analysis.

[0028] Specifically, in step (4), the process of determining the parameters to be calibrated and the parameter values ​​is as follows:

[0029] ① Parameter preliminary selection:

[0030] The selection range of the parameters to be corrected is defined according to the simulation modeling scenario, and the parameters to be corrected are preliminarily selected according to the influencing factors of the simulation object. The specific parameters are shown in Table 1:

[0031] Table 1 Preliminary selection of parameters to be corrected

[0032]

[0033] ②Parameter sensitivity analysis:

[0034] First, the default value is taken for each candidate parameter, 10 different random number seeds are selected for simulation, and the range R of the evaluation index caused by the change of random number is calculated based on the 10 simulation results.m ;

[0035] Secondly, each parameter to be corrected is simulated by selecting its various level values, during which the values ​​of other parameters remain unchanged, and the range R of the evaluation index caused by the change of each parameter is calculated. p i ;

[0036] Finally, the following formula is used to determine whether the change in simulation results caused by the parameter change is significant, and the appropriate parameters to be corrected are selected according to the sensitivity ranking of the significant parameters;

[0037] R p i >R m

[0038] Where: R p i is the extreme difference of the evaluation index caused by the change of the value of the i-th parameter p, R m The evaluation index generated for the simulation seed change is extremely poor;

[0039] If the index change range R caused by a certain parameter change p i Greater than the index change range R caused by the simulation seed change m , then this parameter needs to be corrected;

[0040] If the index change range R caused by a certain parameter change p i Not greater than the index change range R caused by the simulation seed change m , then this parameter does not need to be included in the set to be corrected.

[0041] ③ Determination of parameter values:

[0042] According to the actual traffic operation conditions of the simulation scenario and the recommended value range of the parameters in the simulation software user manual, the values ​​of the parameters to be corrected are preliminarily determined.

[0043] Specifically, in step (5), the process of evaluating the traffic simulation model after adjusting the parameter values ​​is as follows:

[0044] After resetting the values ​​of the parameters to be calibrated, run the simulation and evaluate the simulation results through the evaluation index calculation formula to determine whether the simulation software model under the parameter values ​​is satisfied; if the evaluation index calculation formula is satisfied, the traffic simulation parameter calibration is terminated; if the evaluation result cannot meet the error requirements, continue to adjust the key parameter values ​​until the traffic simulation model meets the requirements. The evaluation index calculation formula is as follows:

[0045]

[0046] Where: T' is the target threshold, y r is the measured value of the evaluation index, y s ' is the average value of the simulation output evaluation index after parameter calibration, and E is the allowable error value.

[0047] Specifically, in step (6), the process of building a joint simulation platform based on the Veins framework is as follows:

[0048] Import the Veins framework into the OMNet++ network simulation software, complete the network parameter settings in OMNet++, and use the Veins framework to connect OMNet++ with the SUMO traffic simulation model that has completed traffic parameter correction to complete the construction of the joint simulation platform.

[0049] Specifically, in step (7), the evaluation index for the accuracy of the joint simulation interaction data is selected as follows:

[0050] Vehicle speed, x-coordinate, y-coordinate, and acceleration traffic parameters are selected as evaluation indicators for the accuracy of interactive data in the joint simulation platform.

[0051] Specifically, the process of step (8) is as follows:

[0052] First, run the joint simulation according to the default data update time interval, and retrieve the vehicle speed and position traffic parameter information of each simulation step in the SUMO and OMNet++ simulation process, and make a preliminary evaluation of the accuracy of the interactive data in the joint simulation model to determine whether the initial joint simulation model meets the requirements. The formula is as follows:

[0053]

[0054] Where: T L is the target threshold of traffic parameter interaction accuracy under joint simulation, P s is the value of the traffic parameter in SUMO, P o is the value of the traffic parameter in OMNet++, E L is the allowable error value of traffic parameter interaction under joint simulation;

[0055] If the accuracy of the joint simulation model has met the error requirement, then the process ends;

[0056] If the error requirement is not met, the time interval for data update in TraCIScenarioManager is adjusted until the accuracy of data interaction of the joint simulation model meets the requirement.

[0057] In summary, this method of joint simulation software parameter correction for autonomous driving can effectively correct the parameters of the joint simulation model so that its simulation can more accurately reflect the characteristics of my country's road traffic and simulate traffic conditions that are more in line with actual scenarios. At the same time, the data exchange interval between the joint simulation software is adjusted to ensure that the traffic flow parameters can be accurately reflected in the network simulation software. The joint simulation parameter correction method is used to improve the accuracy of the joint simulation model and the credibility of the simulation results, providing scientific support for promoting the widespread application of Internet of Vehicles technology in the field of smart cars. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a flow chart of a method for correcting parameters of a joint simulation software for autonomous driving according to the present invention. DETAILED DESCRIPTION

[0059] The present invention is further described in detail below in conjunction with the accompanying drawings. In the description of the present invention, it should be understood that the directions or positional relationships indicated by directional words such as "upper, lower" and "top, bottom" are usually based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description. Unless otherwise stated, these directional words do not indicate or imply that the devices or components referred to must have a specific direction or be constructed and operated in a specific direction, and therefore cannot be understood as limiting the scope of protection of the present invention; the directional words "inside and outside" refer to the inside and outside relative to the contours of each component itself.

[0060] like Figure 1 As shown, a method for correcting parameters of a joint simulation software for autonomous driving comprises the following steps:

[0061] (1) Construction of SUMO simulation scene;

[0062] (2) Selecting evaluation indicators for traffic simulation parameter correction;

[0063] (3) According to the initial evaluation of the parameter default values, determine whether the initial SUMO simulation model meets the requirements. If so, proceed to step (6); if not, proceed to step (4);

[0064] (4) Determine the parameters to be calibrated and their values;

[0065] (5) Evaluate the traffic simulation model after adjusting the parameter values ​​to determine whether the model after parameter correction meets the requirements. If so, proceed to step (6); if not, return to step (4);

[0066] (6) Complete the construction of the joint simulation platform based on the Veins framework;

[0067] (7) Select evaluation indicators for the accuracy of joint simulation interaction data;

[0068] (8) Perform a preliminary evaluation on the data accuracy of the joint simulation model according to the default time interval for data update to determine whether the initial joint simulation model meets the requirements. If not, adjust the time interval for data update. If yes, end the process.

[0069] Specifically, in step (1), when the SUMO simulation scene is constructed, the construction of the simulated road network is completed in SUMO by investigating the current status of the scene road network structure, road width, and intersection channelization. At the same time, the traffic volume, road section flow direction, average vehicle speed, and traffic composition data obtained from the investigation are used as traffic parameters to input into the simulation.

[0070] Specifically, in step (2), road flow, traffic capacity, and travel time are selected as evaluation indicators for traffic simulation parameter correction.

[0071] Specifically, the process of step (3) is as follows:

[0072] Based on the initial simulation results, the sample standard deviation is used to estimate the minimum number of simulations for each evaluation index. The reasonable number of simulations is determined through analysis. The formula is as follows:

[0073]

[0074] Where: n is the minimum number of simulations, S is the sample standard deviation, K is the confidence level, and e is the allowable error value;

[0075] Run the simulation n times according to the default values ​​of the parameters in the simulation model, and evaluate the initial simulation results using the absolute percentage error to determine whether the simulation software model under the default values ​​needs parameter correction. The formula is as follows:

[0076]

[0077] Where: T is the target threshold, y r is the measured value of the evaluation index, y s Output the average value of the evaluation index for the simulation system;

[0078] If the simulation evaluation results of the simulation model under the condition of default parameter values ​​and the error between the measured values ​​have met the requirements, there is no need to calibrate the parameters of the simulation model;

[0079] If the simulation evaluation results of the simulation model under the default parameter value conditions and the error between the measured values ​​cannot meet the requirements, the simulation model needs to be calibrated.

[0080] Specifically, in step (4), before determining the parameters to be calibrated and the parameter values, it is necessary to perform a preliminary selection of the parameters to be calibrated and a parameter sensitivity analysis.

[0081] Specifically, in step (4), the process of determining the parameters to be calibrated and the parameter values ​​is as follows:

[0082] ① Parameter preliminary selection:

[0083] The selection range of the parameters to be corrected is defined according to the simulation modeling scenario, and the parameters to be corrected are preliminarily selected according to the influencing factors of the simulation object. The specific parameters are shown in Table 1:

[0084] Table 1 Preliminary selection of parameters to be corrected

[0085]

[0086] ②Parameter sensitivity analysis:

[0087] First, the default value is taken for each candidate parameter, 10 different random number seeds are selected for simulation, and the range R of the evaluation index caused by the change of random number is calculated based on the 10 simulation results. m ;

[0088] Secondly, each parameter to be corrected is simulated by selecting its various level values, during which the values ​​of other parameters remain unchanged, and the range R of the evaluation index caused by the change of each parameter is calculated. p i ;

[0089] Finally, the following formula is used to determine whether the change in simulation results caused by the parameter change is significant, and the appropriate parameters to be corrected are selected according to the sensitivity ranking of the significant parameters;

[0090] R p i >R m

[0091] Where: R p i is the extreme difference of the evaluation index caused by the change of the value of the i-th parameter p, R m The evaluation index generated for the simulation seed change is extremely poor;

[0092] If the index change range R caused by a certain parameter change p i Greater than the index change range R caused by the simulation seed change m , then this parameter needs to be corrected;

[0093] If the index change range R caused by a certain parameter change p i Not greater than the index change range R caused by the simulation seed change m , then this parameter does not need to be included in the set to be corrected.

[0094] ③ Determination of parameter values:

[0095] According to the actual traffic operation conditions of the simulation scenario and the recommended value range of the parameters in the simulation software user manual, the values ​​of the parameters to be corrected are preliminarily determined.

[0096] Specifically, in step (5), the process of evaluating the traffic simulation model after adjusting the parameter values ​​is as follows:

[0097] After resetting the values ​​of the parameters to be calibrated, run the simulation and evaluate the simulation results through the evaluation index calculation formula to determine whether the simulation software model under the parameter values ​​is satisfied; if the evaluation index calculation formula is satisfied, the traffic simulation parameter calibration is terminated; if the evaluation result cannot meet the error requirements, continue to adjust the key parameter values ​​until the traffic simulation model meets the requirements. The evaluation index calculation formula is as follows:

[0098]

[0099] Where: T' is the target threshold, y r is the measured value of the evaluation index, y s ' is the average value of the simulation output evaluation index after parameter calibration, and E is the allowable error value.

[0100] Specifically, in step (6), the process of building a joint simulation platform based on the Veins framework is as follows:

[0101] Import the Veins framework into the OMNet++ network simulation software, complete the network parameter settings in OMNet++, and use the Veins framework to connect OMNet++ with the SUMO traffic simulation model that has completed traffic parameter correction to complete the construction of the joint simulation platform.

[0102] Specifically, in step (7), the evaluation index for the accuracy of the joint simulation interaction data is selected as follows:

[0103] Vehicle speed, x-coordinate, y-coordinate, and acceleration traffic parameters are selected as evaluation indicators for the accuracy of interactive data in the joint simulation platform.

[0104] Specifically, the process of step (8) is as follows:

[0105] First, run the joint simulation according to the default data update time interval, and retrieve the vehicle speed and position traffic parameter information of each simulation step in the SUMO and OMNet++ simulation process, and make a preliminary evaluation of the accuracy of the interactive data in the joint simulation model to determine whether the initial joint simulation model meets the requirements. The formula is as follows:

[0106]

[0107] Where: T L is the target threshold of traffic parameter interaction accuracy under joint simulation, P s is the value of the traffic parameter in SUMO, P o is the value of the traffic parameter in OMNet++, E L is the allowable error value of traffic parameter interaction under joint simulation;

[0108] If the accuracy of the joint simulation model has met the error requirement, then the process ends;

[0109] If the error requirement is not met, the time interval for data update in TraCIScenarioManager is adjusted until the accuracy of data interaction of the joint simulation model meets the requirement.

[0110] This joint simulation software parameter correction method for autonomous driving can effectively correct the parameters of the joint simulation model so that its simulation can more accurately reflect the characteristics of my country's road traffic and simulate traffic conditions that are more in line with actual scenarios. At the same time, the data exchange interval between the joint simulation software is adjusted to ensure that the traffic flow parameters can be accurately reflected in the network simulation software. The joint simulation parameter correction method can improve the accuracy of the joint simulation model and the credibility of the simulation results, providing scientific support for promoting the widespread application of Internet of Vehicles technology in the field of smart cars.

[0111] Finally, it should be noted that those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for calibrating parameters of joint simulation software for autonomous driving, characterized in that: The following steps are involved: (1) Construction of SUMO simulation scene; (2) Selecting evaluation indicators for traffic simulation parameter correction; (3) According to the initial evaluation of the parameter default values, determine whether the initial SUMO simulation model meets the requirements. If so, proceed to step (6); if not, proceed to step (4); (4) Determine the parameters to be calibrated and their values; (5) Evaluate the traffic simulation model after adjusting the parameter values ​​to determine whether the model after parameter correction meets the requirements. If so, proceed to step (6); if not, return to step (4); (6) Complete the construction of the joint simulation platform based on the Veins framework; (7) Select evaluation indicators for the accuracy of joint simulation interaction data; (8) Perform a preliminary evaluation on the data accuracy of the joint simulation model according to the default time interval for data update to determine whether the initial joint simulation model meets the requirements. If not, adjust the time interval for data update. If yes, end the process.

2. A method for calibrating parameters of joint simulation software for autonomous driving according to claim 1, characterized in that: In step (1), when building the SUMO simulation scene, the simulation road network is built in SUMO by investigating the current status of the scene road network structure, road width, and intersection channelization. At the same time, the traffic volume, road section flow direction, average vehicle speed, and traffic composition data obtained from the investigation are used as traffic parameters to input into the simulation.

3. A method for calibrating parameters of joint simulation software for autonomous driving according to claim 2, characterized in that: In step (2), road flow, traffic capacity and travel time are selected as evaluation indicators for traffic simulation parameter correction.

4. The method for calibrating parameters of a joint simulation software for autonomous driving according to claim 1, characterized in that: The process of step (3) is as follows: Based on the initial simulation results, the sample standard deviation is used to estimate the minimum number of simulations for each evaluation index. The reasonable number of simulations is determined through analysis. The formula is as follows: Where: n is the minimum number of simulations, S is the sample standard deviation, K is the confidence level, and e is the allowable error value; Run the simulation n times according to the default values ​​of the parameters in the simulation model, and evaluate the initial simulation results using the absolute percentage error to determine whether the simulation software model under the default values ​​needs parameter correction. The formula is as follows: Where: T is the target threshold, y r is the measured value of the evaluation index, y s Output the average value of the evaluation index for the simulation system; If the simulation evaluation results of the simulation model under the condition of default parameter values ​​and the error between the measured values ​​have met the requirements, there is no need to calibrate the parameters of the simulation model; If the simulation evaluation results of the simulation model under the default parameter value conditions and the error between the measured values ​​cannot meet the requirements, the simulation model needs to be calibrated.

5. The method for calibrating parameters of a joint simulation software for autonomous driving according to claim 1, characterized in that: In step (4), before determining the parameters to be corrected and their values, it is necessary to conduct a preliminary selection of the parameters to be corrected and a parameter sensitivity analysis.

6. A method for calibrating parameters of joint simulation software for autonomous driving according to claim 2, characterized in that: In step (4), the process of determining the parameters to be calibrated and the parameter values ​​is as follows: ① Parameter preliminary selection: The selection range of the parameters to be corrected is defined according to the simulation modeling scenario, and the parameters to be corrected are preliminarily selected according to the influencing factors of the simulation object. The specific parameters are shown in Table 1: Table 1 Preliminary selection of parameters to be corrected ②Parameter sensitivity analysis: First, the default value is taken for each candidate parameter, 10 different random number seeds are selected for simulation, and the range R of the evaluation index caused by the change of random number is calculated based on the 10 simulation results. m ; Secondly, each parameter to be corrected is simulated by selecting its various level values, during which the values ​​of other parameters remain unchanged, and the range R of the evaluation index caused by the change of each parameter is calculated. p i ; Finally, the following formula is used to determine whether the change in simulation results caused by the parameter change is significant, and the appropriate parameters to be corrected are selected according to the sensitivity ranking of the significant parameters; R p i >R m Where: R p i is the extreme difference of the evaluation index caused by the change of the value of the i-th parameter p, R m The evaluation index generated for the simulation seed change is extremely poor; If the index change range R caused by a certain parameter change p i Greater than the index change range R caused by the simulation seed change m , then this parameter needs to be corrected; If the index change range R caused by a certain parameter change p i Not greater than the index change range R caused by the simulation seed change m , then this parameter does not need to be included in the set to be corrected. ③ Determination of parameter values: According to the actual traffic operation conditions of the simulation scenario and the recommended value range of the parameters in the simulation software user manual, the values ​​of the parameters to be corrected are preliminarily determined.

7. The method for correcting parameters of a joint simulation software for autonomous driving according to claim 1, characterized in that: In step (5), the process of evaluating the traffic simulation model after adjusting the parameter values ​​is as follows: After resetting the values ​​of the parameters to be calibrated, run the simulation and evaluate the simulation results through the evaluation index calculation formula to determine whether the simulation software model under the parameter values ​​is satisfied; if the evaluation index calculation formula is satisfied, the traffic simulation parameter calibration is terminated; if the evaluation result cannot meet the error requirements, continue to adjust the key parameter values ​​until the traffic simulation model meets the requirements. The evaluation index calculation formula is as follows: Where: T' is the target threshold, y r is the measured value of the evaluation index, y s ' is the average value of the simulation output evaluation index after parameter calibration, and E is the allowable error value.

8. The method for calibrating parameters of a joint simulation software for autonomous driving according to claim 1, characterized in that: In step (6), the process of building a joint simulation platform based on the Veins framework is as follows: Import the Veins framework into the OMNet++ network simulation software, complete the network parameter settings in OMNet++, and use the Veins framework to connect OMNet++ with the SUMO traffic simulation model that has completed traffic parameter correction to complete the construction of the joint simulation platform.

9. The method for correcting parameters of joint simulation software for autonomous driving according to claim 1, characterized in that: In step (7), the evaluation index for the accuracy of the joint simulation interaction data is selected as follows: Vehicle speed, x-coordinate, y-coordinate, and acceleration traffic parameters are selected as evaluation indicators for the accuracy of interactive data in the joint simulation platform.

10. The method for calibrating parameters of a joint simulation software for autonomous driving according to claim 1, characterized in that: The process of step (8) is as follows: First, run the joint simulation according to the default data update time interval, and retrieve the vehicle speed and position traffic parameter information of each simulation step in the SUMO and OMNet++ simulation process, and make a preliminary evaluation of the accuracy of the interactive data in the joint simulation model to determine whether the initial joint simulation model meets the requirements. The formula is as follows: Where: T L is the target threshold of traffic parameter interaction accuracy under joint simulation, P s is the value of the traffic parameter in SUMO, P o is the value of the traffic parameter in OMNet++, E L is the allowable error value of traffic parameter interaction under joint simulation; If the accuracy of the joint simulation model has met the error requirement, then the process ends; If the error requirement is not met, the time interval for data update in TraCIScenarioManager is adjusted until the accuracy of data interaction of the joint simulation model meets the requirement.

Images