CN113868778B

CN113868778B - Simulation scene management method and device

Info

Publication number: CN113868778B
Application number: CN202111456058.3A
Authority: CN
Inventors: 魏桢; 王劲
Original assignee: Ciic Technology Co ltd
Current assignee: Ciic Technology Co ltd; Tianyi Transportation Technology Co ltd
Priority date: 2021-12-02
Filing date: 2021-12-02
Publication date: 2022-03-11
Anticipated expiration: 2041-12-02
Also published as: CN113868778A

Abstract

The invention provides a simulation scene management method and a device, the method decouples automatic driving vehicles, each traffic participant and each monitoring task by setting various time baselines based on time synchronization, each time baseline is only responsible for respective management tasks, when a new simulation scene needs to be constructed, only needs to determine each item time base line required to participate in the simulation scene according to management requirement parameters and respectively update the original management information in each target time base line to obtain target management information, because of the low coupling between the management objects of each time base line, the information update of each target time base line is independently carried out without influencing other target time base lines, therefore, the invention can rapidly and flexibly construct a plurality of different simulation scenes according to the management requirement parameters and the original scene management files and support generalization, and the management cost and difficulty are lower, the efficiency is higher, and the requirement of simulation test can be met.

Description

Simulation scene management method and device

Technical Field

The invention relates to the technical field of simulation, in particular to a simulation scene management method and device.

Background

In an autopilot system, a simulation environment is typically constructed to debug or test the autopilot algorithm. At present, in a scene file describing a simulation scene, the simulation scene is mostly defined and managed in a behavior tree manner, however, in this manner, behaviors of autonomous vehicles and all traffic participants in the scene, various local monitoring tasks and the like are mixed in the same behavior tree, high coupling among elements causes high editing cost and high difficulty of the scene, and when editing operations such as adding or reducing traffic participants, changing state manners of the traffic participants, changing the monitoring tasks and the like need to be performed in the scene, the editing cost is no better than rewriting a new behavior tree, so that the method does not support a scene generalization function. In addition, when simulation testing is carried out, each test must run the whole scene from beginning to end according to the logic of the behavior tree, simulation in a specific time period is not supported, and if a certain node which consumes a long time is encountered in the running process, blocking can occur, so that the running of subsequent nodes is influenced, and the simulation efficiency is low. Therefore, the current simulation scene management method is difficult to meet the requirements of simulation tests.

Disclosure of Invention

The invention provides a simulation scene management method, a simulation scene management device, electronic equipment and a computer-readable storage medium, which are used for solving the technical problem that the requirements of simulation tests are difficult to meet in the existing simulation scene management method.

In order to solve the technical problems, the invention provides the following technical scheme:

the invention provides a simulation scene management method, which comprises the following steps:

acquiring an original simulation scene management file, wherein the original simulation scene management file comprises a scene time base line based on time synchronization, a behavior time base line group and a monitoring time base line group; the scene time base line manages the automatic driving vehicle based on first original management information, the behavior time base line group comprises at least one behavior time base line, each behavior time base line manages different traffic participants based on second original management information, the monitoring time base line group comprises at least one monitoring time base line, and each monitoring time base line manages different monitoring tasks based on third original management information;

acquiring management demand parameters, wherein the management demand parameters comprise scene demand parameters of a target simulation scene, traffic participant demand parameters and monitoring demand parameters;

updating the first original management information according to the scene demand parameters to obtain first target management information of the automatic driving vehicle, and obtaining a target scene time baseline according to the first target management information;

determining and updating second original management information of each target traffic participant according to the traffic participant demand parameters to obtain second target management information of each target traffic participant, and obtaining a target behavior time base line according to the second target management information;

determining and updating third original management information of each target monitoring task according to the monitoring demand parameters to obtain third target management information of each target monitoring task, and obtaining a target monitoring time base line according to the third target management information;

and managing the target simulation scene according to the target scene time base line, the target behavior time base line and the target monitoring time base line.

The invention also provides a simulation scene management device, which comprises:

the system comprises a first acquisition module, a second acquisition module and a monitoring module, wherein the first acquisition module is used for acquiring an original simulation scene management file, and the original simulation scene management file comprises a scene time base line, a behavior time base line group and a monitoring time base line group based on time synchronization; the scene time base line manages the automatic driving vehicle based on first original management information, the behavior time base line group comprises at least one behavior time base line, each behavior time base line manages different traffic participants based on second original management information, the monitoring time base line group comprises at least one monitoring time base line, and each monitoring time base line manages different monitoring tasks based on third original management information;

the second acquisition module is used for acquiring management demand parameters, wherein the management demand parameters comprise scene demand parameters of a target simulation scene, traffic participant demand parameters and monitoring demand parameters;

the first updating module is used for updating the first original management information according to the scene demand parameters to obtain first target management information of the automatic driving vehicle, and obtaining a target scene time base line according to the first target management information;

the second updating module is used for determining and updating second original management information of each target traffic participant according to the traffic participant demand parameters to obtain second target management information of each target traffic participant and obtain a target behavior time base line according to the second target management information;

the third updating module is used for determining and updating third original management information of each target monitoring task according to the monitoring demand parameters to obtain third target management information of each target monitoring task, and obtaining a target monitoring time base line according to the third target management information;

and the management module is used for managing the target simulation scene according to the target scene time base line, the target behavior time base line and the target monitoring time base line.

The invention also provides an electronic device comprising a memory and a processor; the memory stores an application program, and the processor is configured to run the application program in the memory to perform any of the steps in the simulation scene management method.

The present invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps in the simulation scenario management method according to any one of the above.

Has the advantages that: the invention provides a simulation scene management method and a device, the method decouples automatic driving vehicles, each traffic participant and each monitoring task by setting various time baselines based on time synchronization, each time baseline is only responsible for respective management tasks, when a new simulation scene needs to be constructed, only needs to determine each item time base line required to participate in the simulation scene according to management requirement parameters and respectively update the original management information in each target time base line to obtain target management information, because of the low coupling between the management objects of each time base line, the information update of each target time base line is independently carried out without influencing other target time base lines, therefore, the invention can rapidly and flexibly construct a plurality of different simulation scenes according to the management requirement parameters and the original scene management files and support generalization, and the management cost and difficulty are lower, the efficiency is higher, and the requirement of simulation test can be met.

Drawings

The technical solution and other advantages of the present invention will become apparent from the following detailed description of specific embodiments of the present invention, which is to be read in connection with the accompanying drawings.

Fig. 1 is a scene schematic diagram applicable to the simulation scene management method of the present invention.

Fig. 2 is a schematic flow chart of a simulation scene management method according to the present invention.

Fig. 3 is a schematic diagram of scene management based on a time baseline in the present invention.

FIG. 4 is a diagram illustrating scene management of a scene time baseline according to the present invention.

FIG. 5 is a diagram illustrating a scenario management of behavior time baselines according to the present invention.

Fig. 6 is a schematic view of scene management of monitoring time base line according to the present invention.

Fig. 7 is a schematic view of scene management of recording time base line in the present invention.

Fig. 8 is a scene diagram of a car following case.

Fig. 9 is a schematic view of following case scene management based on a behavior tree in the prior art.

Fig. 10 is a schematic view of an operation flow of each node in a following case scene based on a behavior tree in the prior art.

Fig. 11 is a schematic view of the following case scene management based on the time base line in the present invention.

Fig. 12 is a schematic diagram of a simulation scenario management apparatus according to the present invention.

Fig. 13 is a schematic structural diagram of an electronic device according to the present invention.

Detailed Description

The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The invention provides a simulation scene management method and device, which are used for relieving the technical problem that the requirements of simulation tests are difficult to meet in the conventional simulation scene management method.

Referring to fig. 1, fig. 1 is a schematic view of a scenario to which the simulation scenario management method of the present invention is applicable, where the scenario may include terminals and servers, and the terminals, the servers, and the terminals and the servers are connected and communicated through the internet formed by various gateways, and the like, where the application scenario includes a simulation scenario 11 and a simulation platform 12; wherein:

the simulation scene 11 is a virtual scene constructed for testing an automatic driving algorithm, and each object in the simulation scene 11 is a virtual object and comprises an automatic driving vehicle 111 and a plurality of traffic participants 112;

the simulation platform 12 is used for constructing the simulation scene 11, generating an environment and objects in the simulation scene 11, managing the simulation scene 11, controlling the automatically driven vehicles in the simulation scene 11 to run based on a preset automatic algorithm, simultaneously controlling the traffic participants 112 to run based on a preset behavior, monitoring all or part of links in the running process, positioning potential problems of the preset automatic algorithm, and testing the robustness of relevant codes.

The simulation platform 12 first obtains an original simulation scenario management file, which includes a scenario time baseline, a behavior time baseline group, and a monitoring time baseline group based on time synchronization. The scene time base lines manage the automatic driving vehicles 111 based on the first original management information, the behavior time base line group comprises at least one behavior time base line, the behavior time base lines are independent from each other, each behavior time base line manages different traffic participants 112 based on the second original management information, the monitoring time base line group comprises at least one monitoring time base line, and each monitoring time base line manages different monitoring tasks based on the third original management information. In addition, according to needs, the original simulation scene management file may further include a recording time base set synchronized with other time bases based on time, where the recording time base set includes at least one recording time base, and each recording time base manages different recording tasks based on fourth original management information.

When a new target simulation scene needs to be constructed, the simulation platform 12 obtains management requirement parameters, where the management requirement parameters include scene requirement parameters, traffic participant requirement parameters, and monitoring requirement parameters of the target simulation scene. When the original simulation scene management file comprises the recording time base line group, the management requirement parameters also comprise recording requirement parameters.

The simulation platform 12 updates the first original management information according to the scene demand parameter to obtain first target management information of the autonomous vehicle. With this step, the scene time baseline of the autonomous vehicle is managed in the original simulation scene management file based on the first original management information as the target scene time baseline of the autonomous vehicle is managed in the target simulation scene management file based on the first target management information.

The simulation platform 12 determines all target traffic participants who need to participate in the target simulation scene according to the traffic participant demand parameters, further determines each behavior time base line for managing each target traffic participant, and updates the second original management information of each behavior time base line according to the traffic participant demand parameters to obtain the second target management information of each target traffic participant. Through this step, the part of the behavior time base of the target traffic participant is managed based on the second original management information in the original simulation scene management file, and is used as the target behavior time base of each target traffic participant managed based on the second target management information in the target simulation scene management file.

The simulation platform 12 determines all target monitoring tasks required by the target simulation scene according to the monitoring demand parameters, further determines each monitoring time baseline for managing each target monitoring task, and updates the third original management information of each monitoring time baseline according to the monitoring demand parameters to obtain the third target management information of each target monitoring task. Through the step, the part of the monitoring time base line of the target monitoring task based on the third original management information in the original simulation scene management file is used as the target monitoring time base line of each target monitoring task based on the third target management information in the target simulation scene management file.

The simulation platform 12 determines all target recording tasks required by the target simulation scene according to the recording requirement parameters, further determines each recording time baseline for managing each target recording task, and updates the fourth original management information of each recording time baseline according to the recording requirement parameters to obtain the fourth target management information of each target recording task. Through the step, the part of the recording time base line of the target recording task based on the fourth original management information in the original simulation scene management file is used as the target recording time base line of each target recording task based on the fourth target management information in the target simulation scene management file.

Finally, the simulation platform 12 manages the target simulation scene based on the target scene time baseline, the target behavior time baseline, the target monitoring time baseline and the target recording time baseline obtained in the above steps, and tests the performance of the automatic driving algorithm in the target simulation scene.

It should be noted that the system scenario diagram shown in fig. 1 is only an example, the server and the scenario described in the present invention are for more clearly illustrating the technical solution of the present invention, and do not constitute a limitation to the technical solution provided by the present invention, and it is known to those skilled in the art that as the system evolves and a new service scenario appears, the technical solution provided by the present invention is also applicable to similar technical problems. The following are detailed below. It should be noted that the following description of the embodiments is not intended to limit the preferred order of the embodiments.

Referring to fig. 2, fig. 2 is a schematic flow chart of a simulation scenario management method according to the present invention, the method includes:

s201: acquiring an original simulation scene management file, wherein the original simulation scene management file comprises a scene time base line, a behavior time base line group and a monitoring time base line group based on time synchronization; the scene time baseline manages the automatic driving vehicle based on the first original management information, the behavior time baseline group comprises at least one behavior time baseline, each behavior time baseline manages different traffic participants based on the second original management information, the monitoring time baseline group comprises at least one monitoring time baseline, and each monitoring time baseline manages different monitoring tasks based on the third original management information.

In the debugging process of the automatic driving algorithm, scene generalization is an important link. Scene generalization refers to deriving a plurality of new simulation scenes by increasing or decreasing the number and types of traffic participants, changing the initial states or running routes of the traffic participants and the like based on one original simulation scene, and then repeatedly debugging an automatic driving algorithm in the plurality of simulation scenes to locate potential problems of the algorithm and test the robustness of related codes. Whether the simulation scene is an original simulation scene or a derived simulation scene, a corresponding simulation scene management file is needed to manage.

The original simulation scene management file is used for managing an original simulation scene, and as shown in fig. 3, the original simulation scene management file includes a scene time baseline, a behavior time baseline group and a monitoring time baseline group, where the scene time baseline has only one, the behavior time baseline group includes at least one behavior time baseline, for example, three behavior time baselines including a behavior time baseline 1, a behavior time baseline 2 and a behavior time baseline 3 shown in fig. 3, and the monitoring time baseline group includes at least one monitoring time baseline, for example, two monitoring time baselines including a monitoring time baseline 1 and a monitoring time baseline 2 shown in fig. 3. Each time base line in the original simulation scene management file has a plurality of continuous moments, and each time base line is based on time synchronization.

The scenario time baseline manages the autonomous vehicle based on first original management information, which refers to information for planning, guiding and controlling various states and behaviors of the autonomous vehicle in the simulation scenario, and may include, for example, a first original life cycle of the autonomous vehicle, which refers to a whole period of time from the beginning to the end of the operation of the original simulation scenario, and initial state information, etc., and the autonomous vehicle exists throughout the process, and thus also refers to a period of time that the autonomous vehicle exists throughout the original simulation scenario. The initial state information refers to state information obtained by the autonomous vehicle after a series of operations such as generating an autonomous vehicle model, setting an initial position, setting an initial speed, generating a vehicle-mounted laser radar, and the like are performed on the autonomous vehicle at the start time of the first original life cycle, and includes spatial information, time information, attribute information, and the like of the autonomous vehicle. At each time on the scene time baseline after the start time, the autonomous vehicle travels according to a preset autonomous algorithm until the end time of the first original life cycle is reached.

Each behavior time base line manages different traffic participants based on second original management information, the traffic participants refer to simulation objects except automatic driving vehicles in the original simulation scene and comprise various types of vehicles, pedestrians, obstacles and the like, the second original management information refers to information for planning, guiding and controlling various states and behaviors of the traffic participants in the simulation scene, for example, the second original life cycle and the continuous state information of the traffic participant can be included, the second original life cycle refers to the period of time that the traffic participant exists in the first original life cycle of the original simulation scene, and according to different test requirements, the traffic participant may be present in only one time period, or may be present in two discrete time periods, the corresponding behavioral time baseline for each traffic participant may have at least one second original life cycle. The continuous state information refers to the sum of the state information of the traffic participants at each moment in each second original life cycle, and the state information at each moment comprises the spatial information, the time information, the attribute information and the like of the traffic participants. When generating the continuous state information, the user can describe the state at any time point in the second original life cycle, and then the transition between the two time points with the state description is completed by means of interpolation or reading the pre-recorded information.

Each monitoring time base line manages different monitoring tasks based on third original management information, the monitoring tasks refer to monitoring whether a certain area or a certain event and the like in an original simulation scene need to be monitored in a certain time period, if the automatic driving vehicle needs to be monitored whether the automatic driving vehicle reaches a preset terminal, whether the automatic driving vehicle collides with other traffic participants, how many times the automatic driving vehicle collides, whether the vehicle speed is in a specific interval after the automatic driving vehicle runs from a common road to a speed-limiting road and the like, and the required monitoring tasks are different according to different testing requirements. The third original management information refers to information for planning, guiding and controlling the monitoring tasks in the simulation scenario, and may include, for example, original monitoring trigger conditions, original monitoring execution conditions, original monitoring termination conditions, and the like of the corresponding monitoring tasks. On a monitoring time base line, a monitoring task sequentially passes through three stages of triggering, executing and terminating, wherein an original monitoring triggering condition is a condition for judging whether the monitoring task starts, such as whether an automatic driving vehicle enters a starting point of a certain monitoring interval or not, an original monitoring executing condition is a specific executing condition of the monitoring task, such as whether the speed of the monitoring automatic driving vehicle in the monitoring interval is 60-80 or not, and an original monitoring terminating condition is a condition for judging whether the monitoring task ends, such as whether the automatic driving vehicle runs out of a terminal point of the monitoring interval or not. Each condition can be presented in the form of a trigger function, a monitoring function and a termination function, three stages required by the monitoring task are sequentially completed in the modes of a registration function and an operation function, and the time period from the registration of the trigger function to the condition that the termination function meets is taken as the third original life cycle of the monitoring task.

S202: and acquiring management demand parameters, wherein the management demand parameters comprise scene demand parameters of a target simulation scene, traffic participant demand parameters and monitoring demand parameters.

When the original simulation scene needs to be managed, namely a new target simulation scene is obtained on the basis of the original simulation scene, management requirement parameters are generated and acquired by the simulation platform. The management demand parameters comprise scene demand parameters of the target simulation scene, traffic participant demand parameters and monitoring demand parameters.

The scene requirement parameters comprise scene life cycle requirement parameters and scene state requirement parameters, the scene life cycle requirement parameters refer to life cycles of the automatic driving vehicle in the target simulation scene, namely the actual life cycles of the target simulation scene, and the scene state requirement parameters refer to initial states of the automatic driving vehicle in the target simulation scene life cycle, including a vehicle model, an initial position, an initial orientation, an initial speed and the like.

The traffic participant demand parameters comprise traffic participant attribute demand parameters, traffic participant life cycle demand parameters and traffic participant state demand parameters, the traffic participant attribute demand parameters comprise types, numbers and the like of traffic participants needing to participate in the target simulation scene, the traffic participant life cycle demand parameters comprise life cycles needing to exist in the target simulation scene of the traffic participants, and the traffic participant state demand parameters comprise continuous states needing to be shown by the traffic participants in each life cycle, including positions, speeds, accelerations, orientations and the like at all times.

The monitoring requirement parameters comprise monitoring task requirement parameters and monitoring condition requirement parameters, the monitoring task requirement parameters are monitoring tasks required by a target simulation scene, such as overspeed task monitoring, collision frequency task monitoring, terminal task monitoring and the like, and the monitoring condition requirement parameters refer to triggering conditions, execution conditions and termination conditions required to be met by each monitoring task of the target simulation scene during specific implementation.

S203: and updating the first original management information according to the scene demand parameters to obtain first target management information of the automatic driving vehicle, and obtaining a target scene time baseline according to the first target management information.

After the scene demand parameters are acquired, the first original management information corresponding to the scene time baseline is updated, the updated management information is used as first target management information of the automatic driving vehicle, and then the scene time baseline is used as a target scene time baseline of the target simulation scene, and the automatic driving vehicle can be managed based on the first target management information.

In one embodiment, S203 specifically includes: updating the starting time and the ending time of the first original life cycle according to the scene life cycle demand parameters to obtain the target starting time and the target ending time of the target simulation scene, and obtaining a first target life cycle according to the target starting time and the target ending time; updating the state information of the automatic driving vehicle at the target starting moment according to the scene state demand parameters to obtain target initial state information; and obtaining first target management information of the automatic driving vehicle according to the first target life cycle and the target initial state information.

The scene life cycle requirement parameters are used for representing life cycles required to exist in the target simulation scene of the automatic driving vehicle, the required life cycles are different for different simulation scenes, the life cycles can be prolonged, the life cycles can also be shortened, the starting time and the ending time of the first original life cycle are updated to obtain the target starting time and the target ending time, the time period between the target starting time and the target ending time is the first target life cycle of the target simulation scene, and the first target life cycle can be of any length so as to meet the requirements of different scene tests. The scene state demand parameter is used for representing an initial state required by the automatic driving vehicle in a first target life cycle, after the life cycle of the scene is updated, the state of the automatic driving vehicle at a target starting moment is also required to be updated according to the scene state demand parameter to obtain target initial state information, the state of the corresponding moment in the original simulation scene can be used as the target initial state during updating, the state of the moment can also be reset, and the new state is used as the target initial state. After the above operation is performed, first target management information of the autonomous vehicle is obtained.

Specifically, as shown in fig. 4, in the original simulation scene, the first original lifecycle is 90 seconds, the start time when the scene starts is 0 seconds, and the end time when the scene ends is 90 seconds, and it is found that a problem of a preset autopilot algorithm occurs between 35 th second and 70 th second in the debugging process, the target simulation scene may be obtained by simulating only a part of the length of the first original lifecycle, specifically, the start time is updated to 35 th second, the end time is updated to 70 th second, the finally obtained first target lifecycle is 35 seconds, and in the time periods before and after the first target lifecycle, the target scene time baselines are both in a silent state and are represented by silent. Further, the state information of the autonomous vehicle at 35 seconds in the original simulation scenario may be acquired and taken as the target initial state information of the autonomous vehicle at the start time of the target simulation scenario.

In the traditional scene management method based on the behavior tree, the whole scene can only be run from beginning to end when the behavior tree scene is operated, the simulation in a specific time period is not supported, and the requirement of efficient and rapid iteration is not met. In the invention, the scene management method based on the time base line can randomly adjust the positions of the starting point and the end point of the scene, modify the life cycle length of the scene, simulate only a local scene and greatly improve the simulation efficiency.

S204: and determining and updating second original management information of each target traffic participant according to the traffic participant demand parameters to obtain second target management information of each target traffic participant, and obtaining a target behavior time base line according to the second target management information.

After the traffic participant demand parameters are obtained, firstly determining which traffic participants are needed in a target simulation scene, taking the traffic participants as target traffic participants, then determining which behavior time baselines are used for managing the target traffic participants from a behavior time base line group, updating second original management information corresponding to each determined behavior time base line, taking the updated management information as second target management information of the corresponding target traffic participants, taking the behavior time baselines as target behavior time baselines of the target simulation scene, and respectively managing the respective target traffic participants based on the second target management information.

In one embodiment, S204 specifically includes: determining target traffic participants required by a target simulation scene according to the attribute demand parameters of the traffic participants; updating a second original life cycle corresponding to each target traffic participant according to the life cycle demand parameters of the traffic participants to obtain at least one second target life cycle of each target traffic participant in the target simulation scene; updating the continuous state information of each target traffic participant in each second target life cycle according to the traffic participant state demand parameters to obtain the target continuous state information of each target traffic participant in each second target life cycle; and obtaining second target management information of each target traffic participant according to the target traffic participant, the second target life cycle and the target continuous state information.

The traffic participant attribute demand parameters are used for representing how many traffic participants are needed in the target simulation scene, and what types of the traffic participants are (cars, trucks, pedestrians, etc.), the traffic participants are used as the target traffic participants, then the behavior time baselines of the management target traffic participants are determined from the behavior time baselines group, for example, if the behavior time baselines of 30 management cars, 30 management trucks, and 40 management pedestrians exist in the behavior time baselines group, and the target traffic participants include 10 cars and 5 pedestrians, only the time baselines of 10 management cars and the behavior time baselines of 5 management pedestrians need to be taken out from the behavior time baselines group to participate in the management of the target simulation scene. Certainly, the present invention is not limited to this, if the required traffic participant does not have a corresponding behavior time baseline in the original behavior time baseline group to be managed, the required behavior time baseline can be newly added at this time to participate in the management of the target simulation scene, and the required behavior time baseline is added to the original behavior time baseline group at the same time. Through the attribute demand parameters of the traffic participants, the traffic participants can be increased or decreased arbitrarily in the target simulation scene.

The traffic participant life cycle demand parameters are used for expressing the life cycle of each traffic participant in a target simulation scene, and specifically include the occurrence frequency and the number of the start time and the end time of each occurrence, for the 15 obtained behavior time baselines, updating a second original life cycle of each behavior time baseline according to the traffic participant life cycle demand parameters, and each updated behavior time baseline has at least one second target life cycle. And the traffic participant state demand parameter is used for expressing the state of each traffic participant in each second target life cycle, and the continuous state information of the target traffic participants in each second target life cycle is updated according to the traffic participant state demand parameter to obtain the target continuous state information in each second target life cycle. After the 15 behavior time baselines are updated, second target management information of the 15 target traffic participants is obtained.

In one embodiment, the step of updating the continuous state information of each target traffic participant in each second target life cycle according to the traffic participant state demand parameter to obtain the target continuous state information of each target traffic participant in each second target life cycle includes: determining a state information source of each target traffic participant according to the state demand parameters of the traffic participants, wherein the state information source comprises at least one of a manual simulation state information source and a road end real state information source; and acquiring target continuous state information of each target traffic participant in each second target life cycle from the state information source corresponding to each target traffic participant.

In each second target life cycle of the target behavior baseline, the target continuous state information corresponding to each target traffic participant can be obtained from two sources, one is obtained by manual editing based on experience, and the other is obtained by acquiring actual traffic flow information at the road end. Different simulation scenes have different state requirements on target traffic participants, and may only need to manually edit data for testing, may only need to test road end real data, and may also need to test two types of data at the same time, so that for each target traffic participant, a required state information source can be selected from a manual simulation state information source and a road end real state information source according to corresponding traffic participant state requirement parameters, and respective target continuous state information can be obtained from the state information sources.

In the traditional scene management method based on the behavior tree, the artificial simulation state information source and the road end real state information source are distinct, and the importing of the state information in one information source is supported each time, so that the information source is single, the application scene is narrow, and the final test effect is not good. In the invention, the scene management method based on the time base line can simultaneously support the information import of two different information sources, namely the state information of the two sources can coexist, so that the states of traffic participants in a simulation scene can be more diversified, the application range of a test scene can be widened, and a better test effect can be obtained on the basis.

Specifically, as shown in fig. 5, car-1 and car-2 are set as target traffic participants, and action time baseline 1 and action time baseline 2 are respectively the target action time baseline of car-1 and the target action time baseline of car-2 determined according to the traffic participant attribute demand parameters, after updating, action time baseline 1 has two second target lifecycles, and the rest of the periods on action time baseline 1 except for the two second target lifecycles are in a silent state. Behavioral time baseline 2 has a second target life cycle, and other times on behavioral time baseline 2 except for the second target life cycle are in a silent state. For the behavior time baseline 1, according to the traffic participant state demand parameters, it is determined that the state information source is a manual simulation state information source, then manual editing is performed in both the second target life cycles, during editing, the state information of the start time and the end time of each second target life cycle can be obtained by editing first, the state information is represented by circles corresponding to the two times in fig. 5, and then the state completion is performed between the two circles by using a difference method. And for the behavior time baseline 2, determining that the state information source is a road end real state information source according to the state demand parameters of the traffic participants, loading a data packet in the second target life cycle, receiving continuous state information in the data packet and updating in real time.

In the traditional behavior tree-based scene management method, when a batch test loop of an automatic driving algorithm is performed, the algorithm debugging link is not friendly, because traffic participants need to be continuously added and removed in the planning algorithm development process or the accident scene analysis process, the reason of scene failure is determined, because the behaviors of a plurality of traffic participants are located in the same behavior tree and are mutually coupled, the behavior tree-based scene does not support the addition and deletion of the traffic participants, the cost of the addition and deletion is not inferior to or even higher than that of rewriting a new scene behavior tree, the overall cost is higher, and the efficiency is lower. In the invention, the scene management method based on the time base line decouples the traffic participants by means of a time base line isolation mode, each traffic participant occupies one time base line, state editing can be freely carried out on the base line, a plurality of traffic participants form a base line group through tags, and the mode of managing the base line and the base line group can realize a rapid generalization function, thereby reducing the cost and improving the efficiency. On each behavior time base line, a user is allowed to carry out state description at any time, transition is completed between two state description points in an interpolation mode or a prerecorded information reading mode, state synchronization is conveniently completed in the initialization process of traffic participants, and the problem that the current initialized traffic participants have no initial state is solved.

S205: and determining and updating third original management information of each target monitoring task according to the monitoring demand parameters to obtain third target management information of each target monitoring task, and obtaining a target monitoring time base line according to the third target management information.

After the monitoring demand parameters are obtained, firstly determining which monitoring tasks are needed by the target simulation scene, taking the monitoring tasks as target monitoring tasks, then determining which monitoring time baselines are used for managing the target monitoring tasks from the monitoring time baselines, updating third original management information corresponding to each determined monitoring time baseline, taking the updated management information as third target management information of the corresponding target monitoring tasks, taking the monitoring time baselines as target monitoring time baselines of the target simulation scene, and respectively managing the respective target monitoring tasks based on second target management information.

In one embodiment, S205 specifically includes: determining a target monitoring task required by a target simulation scene according to the monitoring task demand parameter; updating the original monitoring triggering condition, the original monitoring executing condition and the original monitoring terminating condition corresponding to each target monitoring task according to the monitoring condition demand parameters to obtain the target monitoring triggering condition, the target monitoring executing condition and the target monitoring terminating condition corresponding to each target monitoring task; and obtaining third target management information of each target monitoring task according to the target monitoring triggering condition, the target monitoring executing condition and the target monitoring terminating condition.

The monitoring task demand parameters are used for expressing which monitoring tasks are needed by a target simulation scene, the monitoring tasks are used as target monitoring tasks, then monitoring time baselines for managing the target monitoring tasks are determined from a monitoring time baseline group, the monitoring time baselines comprise various monitoring time baselines for managing and monitoring overspeed tasks, collision number monitoring tasks, arrival monitoring tasks, own vehicle deviation error monitoring and the like, the target monitoring tasks comprise the overspeed monitoring tasks and the arrival monitoring task, and only 2 monitoring time baselines for managing the 2 target monitoring tasks need to be taken out from the monitoring time baseline group to participate in the management of the target simulation scene. Similarly, the present invention is not limited to this, if the required target monitoring task has no corresponding monitoring time baseline in the original monitoring time baseline set to be managed, the required monitoring time baseline can be newly added at this time to participate in the management of the target simulation scenario, and the monitoring time baseline is added to the original monitoring time baseline set at the same time. By monitoring the task demand parameters, the monitoring tasks can be increased or decreased arbitrarily in the target simulation scene.

The monitoring condition demand parameter is used for expressing what conditions each monitoring task of the target simulation scene needs to meet when being specifically implemented, and comprises a triggering condition, an execution condition and a termination condition. And for the obtained 2 monitoring behavior baselines, correspondingly updating the original monitoring triggering condition, the original monitoring executing condition and the original monitoring terminating condition according to the specific requirements of respective monitoring tasks. For example, in the original simulation scenario, for monitoring the overspeed task, the triggering condition is that the autonomous vehicle enters the lane B from the lane a, the executing condition is that whether the speed of the autonomous vehicle is within the interval from S1 to S2 is monitored, and the terminating condition is that the autonomous vehicle leaves the lane B; in the target simulation scenario, also for monitoring the overspeed task, the triggering condition is that the autonomous vehicle enters the C lane from the A lane, the executing condition is that whether the speed of the autonomous vehicle is within the interval of S3-S4 is monitored, and the terminating condition is that the autonomous vehicle leaves the C lane. Although both are overspeed monitoring tasks, different monitoring condition requirements can be met in different scenes, so that the conditions on the monitoring time base line can be updated to obtain third target management information of each target monitoring task.

Specifically, as shown in fig. 6, the monitoring time baseline 1 is a target monitoring time baseline corresponding to a certain target monitoring task, after updating, the monitoring time baseline 1 is just in a silent state until a circle registers three functions into the monitoring time baseline 1, namely a trigger function for judging whether monitoring is started, a monitoring function for executing the monitoring task, and a termination function for judging whether monitoring is stopped, the trigger function is operated after registration is completed, and the state of monitoring the autonomous vehicle is judged whether to start, as shown in an interval corresponding to a "detection condition" in fig. 6; after the condition of the trigger function is met, the trigger function terminates operation, the monitoring function and the termination function are started, as shown in a section corresponding to a monitoring section in fig. 6, the monitoring function monitors the vehicle operation state and executes a monitoring task, after the condition of the termination function is met, the monitoring task is completed, the monitoring time base 1 is discarded, and the time period from the registration of the trigger function to the condition that the termination function is met on the monitoring time base 1 is taken as a third target life cycle of the target monitoring task.

In the traditional behavior tree-based scene management method, nodes in a behavior tree are all global, the scene starts to be monitored when the scene starts, and the monitoring is terminated when the scene runs, so that the behavior tree-based scene only supports global monitoring, if local monitoring requires writing monitoring conditions and monitoring processes into the behavior tree, external monitoring behaviors are coupled with the scene, and the readability of the scene is greatly reduced. In the invention, the monitoring tasks are decoupled from other elements, each monitoring task occupies a time base line, and global monitoring or local monitoring can be selected by freely editing on the base line, so that the readability of the scene is high.

In one embodiment, the original simulation scene management file further includes a recording time baseline group synchronized with other time baselines based on time, the recording time baseline group includes at least one recording time baseline, each recording time baseline manages different recording tasks based on fourth original management information, the management requirement parameters further include recording requirement parameters, and after the step of obtaining the management requirement parameters, the method further includes: determining and updating fourth original management information of each target recording task according to the recording demand parameters to obtain fourth target management information of each target recording task, and obtaining a target recording time base line according to the fourth target management information; and managing the target simulation scene according to the target recording time base line.

The original simulation scene management file further includes a recording time base set including at least one recording time base, such as recording time base 1 and recording time base 2 shown in fig. 3, and similarly, the recording time base has a plurality of continuous time instants and is synchronized with other time bases based on time.

Each recording time baseline manages different recording tasks based on fourth original management information, the recording tasks refer to the fact that all or part of the running processes of the objects in the original simulation scene need to be recorded and the results are stored in a specific time period, and the fourth original management information refers to the information for planning, guiding and controlling the recording tasks, for example, the fourth original life cycle of the corresponding recording tasks can be included, namely, the time period from the beginning of recording to the end of recording.

When the target simulation scene has a recording requirement, the corresponding management requirement parameter comprises a recording requirement parameter, the recording requirement parameter comprises a recording task requirement parameter and a recording life cycle parameter, the recording task requirement parameter indicates that the target simulation scene needs to be recorded for several times, and the recording life cycle parameter indicates that the recording starts from which moment to which moment ends each time. After the recording requirement parameters are obtained, firstly, several recording tasks needed by the target simulation scene are determined, the recording tasks are determined to be the target recording tasks, then, the corresponding number of recording time baselines are determined from the recording time baselines, the fourth original management information corresponding to each determined recording time baseline is updated, the updated management information is used as the fourth target management information corresponding to the recording tasks, and then the recording time baselines are used as the target recording time baselines of the target simulation scene, and the respective target recording tasks can be managed respectively based on the fourth target management information.

In an embodiment, the step of determining and updating the fourth original management information of each target recording task according to the recording requirement parameter to obtain the fourth target management information of each target recording task includes: determining a target recording task required by a target simulation scene according to the recording task demand parameter; updating a fourth original life cycle corresponding to each target recording task according to the recording life cycle demand parameters to obtain a fourth target life cycle corresponding to each target recording task; and obtaining fourth target management information of each target recording task according to the fourth target life cycle.

The recording requirement parameters are used for indicating that several recording time baselines are required to record different objects of the target simulation scene, according to the recording requirement parameters, a corresponding number of recording time baselines are taken out from the target recording time baselines, for example, 2 recording time baselines are taken out, the first recording time baseline is used for recording the whole target simulation scene, the second recording time baseline is used for recording one traffic participant, the recording life cycle requirement parameters are used for indicating the time period that each recording task needs to be recorded, for example, the first recording time baseline is recorded from 0 th second to 90 th second of the target simulation scene, and the second recording time baseline can be recorded from 35 th second to 60 th second of the target simulation scene. And updating each recording time base line according to the recording life cycle demand parameters to obtain a fourth target life cycle corresponding to each target monitoring task. And after the operation, obtaining fourth target management information of each target monitoring task.

In one embodiment, the step of managing the target simulation scenario according to the target recording time baseline includes: updating the mode information of each target recording time baseline into a recording mode in a first target life cycle of a target simulation scene so as to enable each target recording time baseline to execute a corresponding target recording task and obtain recording information; and after the first target life cycle of the target simulation scene, updating the mode information of the at least one target recording time base line into a playback mode so as to enable the at least one target recording time base line to play back the corresponding recording information. Each recording time base line has two modes, namely a recording mode and a playback mode, the recording mode can record the state information of the automatic driving vehicle and the traffic participants in the scene operation process, and the playback mode can play the recorded information. Specifically, as shown in fig. 7, a circle indicates registration of a recording task, recording is started after the registration is completed, and any selection of recording time can be realized by adjusting a life cycle of a recording time baseline.

In the traditional scene management method based on the behavior tree, result playback is not supported, an automatic driving algorithm needs to be accessed to each running scene, namely, simulation is carried out again, and the persistence is lacked, so that the result reproduction is very troublesome, and the comparison of simulation results of multiple algorithms or multi-version algorithms is not supported. In the scene management method based on the time base line, the recording time base line can randomly adjust the recording life cycle to select the recording time interval, the current scene state information can be recorded, the prerecorded scene state information can also be loaded, and in the algorithm iteration process, the recorded information can be played based on the playback mode, the multi-speed viewing is supported, and the synchronous playing of a plurality of recorded information is supported, so that the new and old scenes can be played simultaneously to visually compare the algorithm improvement condition, the algorithm problem is quickly positioned, and the algorithm comparison cost is reduced.

S206: and managing the target simulation scene according to the target scene time base line, the target behavior time base line and the target monitoring time base line.

After updating, the target scene time base line, each target behavior time base line, each target monitoring time base line and each target monitoring time base line are still based on time synchronization, each time base line forms a target simulation scene management file, the target simulation scene can be managed by the file, the driving condition of the automatic driving vehicle in the target simulation scene based on the preset automatic driving algorithm is tested, and the algorithm is debugged according to the test result. By repeatedly executing the processes, more target simulation scenes can be obtained to participate in the test, and the generalized debugging is realized.

The target scene time base line of the target simulation scene can continuously adjust the self simulation time according to the information processing speed of other automatic driving modules in the simulation process, and because the adjusted time can come in and go out with the real time to a certain extent, other target time base lines are kept synchronous with the time of the target scene time base line instead of the real time. Specifically, when the target simulation scenario is managed, the time baselines are associated in a time stamp manner. The target scene time base line is responsible for pushing the simulation time to advance, and continuously issues the simulation time stamps outwards, other types of target time base lines trigger events on the corresponding time base lines according to the received simulation time stamps, if each target behavior time base line can trigger the updating of the state of the corresponding target traffic participant according to the simulation time stamps, each target monitoring time base line can register and monitor the corresponding target monitoring task according to the simulation time stamps, each recording time base line can trigger the corresponding target recording task according to the simulation time stamps, and when the simulation scene or the scene recorded by the road end is played back, the functions of double-speed playing, progress bar jumping playing and the like are realized on the basis of the simulation time stamps.

It should be noted that, in the above embodiments, each time baseline of the target simulation scene is updated, but the present invention is not limited thereto, and if the management information of a certain time baseline required by the target simulation scene and the original simulation scene does not change, if the performance of a certain traffic participant in two different scenes is the same, the target simulation scene may be managed directly by using the time baseline without updating.

In the above embodiment, the lifecycle management is divided into three levels, which are the lifecycle of the scene, the lifecycle of the single baseline, and the lifecycle of the baseline group in turn. The priority of the scene life cycle is highest, and only one continuous life cycle exists; the baseline group priority is ranked second, allowing multiple discrete life cycles; the single baseline is the lowest priority and the other characteristics are the same as the baseline group.

In one embodiment, S206 is preceded by: acquiring state information of all time baselines in an original simulation scene management file; and updating the state information of all the target time baselines into a first state so that all the target time baselines participate in the target simulation scene, and updating the state information of other time baselines into a second state so that other time baselines do not participate in the target simulation scene.

In the present invention, all the target scene time baselines, the target behavior time baselines, the target monitoring time baselines and the target recording time baselines determined in the foregoing embodiments are the target time baselines, each time baseline in the original simulation scene management file has corresponding state information, and the state information includes a first state "activate" and a second state "un _ activate", that is, an activated state and an inactivated state. When in the activated state, the time base is indicated to be capable of participating in the management of the target simulation scene, and when in the inactivated state, the time base is indicated not to be capable of participating in the management of the target simulation scene. In the above embodiments, after all the target time baselines required by the target simulation scene are determined, the states of the target time baselines and the non-target time baselines can be correspondingly updated, so that the time baselines participating in scene interaction and objects managed by the time baselines can be conveniently and quickly selected. In addition, all the time baselines in the same base line group are provided with a tag label, the time baselines and corresponding management objects, such as traffic participants, can be removed or added in batches by changing the state of the tag label, so that the efficiency is improved. Of all time baselines, the scene time baseline has the highest priority, affecting the other time baselines, and only the other time baselines are in the active state during the scene life cycle. The scene lifecycle must be continuous, uninterrupted, and each scene can only have a certain lifecycle.

It can be known from the above embodiments that the simulation scenario management method of the present invention decouples the autonomous vehicle, each traffic participant and each monitoring task by setting various time baselines based on time synchronization, each time baseline is only responsible for its own management task, when a new simulation scenario needs to be constructed, it is only necessary to determine each item time baseline needed to participate in the simulation scenario according to the management requirement parameters, and update the original management information in each target time baseline to obtain the target management information, because of the low coupling between the management objects of each time baseline, the information update of each target time baseline is performed independently, and no influence is generated on other target time baselines, so that the present invention can construct a plurality of different simulation scenarios rapidly and flexibly according to the management requirement parameters and the original scenario management files, and support scenario generalization, and the management cost and difficulty are lower, the efficiency is higher, and the requirement of simulation test can be met.

The following describes a behavior tree-based scene management method and a time-based scene management method by using a typical car following case, with reference to fig. 8 to 11.

As shown in fig. 8, an automatic driving vehicle is represented by EV, a traffic participant is represented by RV, and the vehicle enters a second stage after the first stage of high-speed driving of RV, high-speed following of EV and 3 seconds of following of EV; in the second stage, the RV decelerates to drive, the EV changes speed to follow the vehicle, and the vehicle smoothly reaches the target point.

The scene files based on the behavior tree are recorded by using xml, and each xml file corresponds to one test case. The scene file includes four main parts, namely, a parameter list (params _ list), an initialization (initialization), a behavior tree (behavior _ tree), and a scene evaluation list (criterion _ list). Behavior tree nodes can be divided into three categories: the system comprises a behavior node, a trigger node and a composite node. The behavior and trigger are the minimum logic units of the scene tree, namely leaf nodes. behavior represents an action or action (action) such as "Waypoint follow" to give the target vehicle an action to follow waypoint; trigger represents a block check or monitor, such as "StandStill" to check whether the target vehicle is parked, and if so, return to success and execute the next node, otherwise continue to monitor vehicle speed and subsequent nodes will not be operated. The composite compound node is used for managing the child nodes, does not do any influence on the test scene, and is a logical concept. The typical composite nodes "sequence", "parallel _ success _ on _ one" and "parallel _ success _ on _ all" abstract the complex scene logic into three basic elements, and realize all scene behaviors through free combination thereof.

As shown in fig. 9, in the behavior tree based management method, a root is a root node of the behavior tree and is also an entry node. The root node is a sequence composite node and comprises three sub-nodes of 'place', 'period _ 1' and 'period _ 2', the three nodes are sequentially executed, and when the three nodes are all completed, the operation of the simulation scene is finished.

As shown in fig. 10, a "place" node is first run, the node is a parallel _ success _ on _ all composite node, two child nodes "spawn _ lidar" and "actor transform router" are executed in parallel, and when both nodes return success, the "place" node returns success.

Then, the "period _ 1" node is operated again, the node is a parallel _ success _ on _ one composite node, two child nodes of the node, namely "other _ operator _0_ moving" and "ego _ period _ following _ condition _ 1", are executed in parallel, wherein the "other _ operator _0_ moving" node blocks and does not return success, only the "ego _ period _ following _ condition _ 1" node completes, and the "period _ 1" node can complete. The 'ego _ cause _ following _ condition _ 1' node is also a parallel _ success _ on _ one type composite node, and its two child nodes are 'trigger _ following' and 'trigger _ delete _ distance', respectively, where the 'trigger _ following' node is executed first and returns success, then the 'ego _ cause _ following _ condition _ 1' node is executed completely and returns success, and meanwhile, the period _1 node returns success.

And finally, operating the 'period _ 2' node, wherein the 'period _ 2' node and the 'period _ 1' node are the same in principle.

As shown in fig. 11, a time baseline-based management method sets a scene time baseline, a behavior time baseline 1, a monitoring time baseline 2, and a recording time baseline 1, the scene time baseline manages EV state initialization and a scene life cycle, the initialization includes generating a vehicle model, setting an initial position, setting an initial speed, generating a vehicle-mounted laser radar, and the like, the maximum duration of a scene is 90 seconds, and the scene can be terminated when a monitoring EV in the monitoring time baseline 1 reaches a target point. The behavior time baseline 1 records the state of the RV, driving at high speed for 3 seconds, then decelerating for 1 second, and then keeping driving at low speed until the scene ends. Monitoring whether the EV reaches a target point or not by a monitoring time baseline 1, keeping the first half in a silent state, registering a trigger function, a monitoring function and a termination function at a circle, wherein the trigger condition is that the distance from the vehicle to the target point is less than 50m, the monitoring function monitors whether the EV is stopped and the distance from the EV to the target point is less than 1m, and the termination function stops a scene. Monitoring time base line 2 monitors whether the EV collides, a function is registered when a scene starts, the triggering function and the terminating function are empty, and the monitoring function counts the collision times. Recording time baseline 1 records the whole process of a scene, and a recording object EV is registered into a recording module when recording starts.

As can be seen from the comparison, in the management method based on the behavior tree, behaviors of a plurality of traffic participants and local monitoring behaviors are mixed in one behavior tree and are associated with each other, so that the successful operation of a certain node needs to depend on the successful operation of one or more nodes, and if the certain node runs to a node which consumes a long time, a block occurs, thereby affecting the operation of subsequent nodes. Although the blocking problem can be solved by using a large number of compound nodes, the writing difficulty of the behavior tree is increased, and the readability is reduced. In addition, the behavior tree can theoretically realize a complex scene with multiple traffic participants, but the number of the traffic participants and the editing difficulty also increase exponentially. Therefore, the behavior tree-based scene management method is not suitable for the debugging stage, and the completed scene does not support the dynamic generalization function. In the same scene, the time line-based mode is relatively low in logic and scene editing difficulty, space is reserved for scene expansion, and the application range is wider, so that the requirements of a debugging stage and a large-scale integration testing stage can be met.

Accordingly, fig. 12 is a schematic structural diagram of the artificial scene management device of the present invention, please refer to fig. 12, the artificial scene management device includes:

a first obtaining module 110, configured to obtain an original simulation scene management file, where the original simulation scene management file includes a scene time baseline, a behavior time baseline group, and a monitoring time baseline group based on time synchronization; the scene time base line manages the automatic driving vehicle based on the first original management information, the behavior time base line group comprises at least one behavior time base line, each behavior time base line manages different traffic participants based on the second original management information, the monitoring time base line group comprises at least one monitoring time base line, and each monitoring time base line manages different monitoring tasks based on the third original management information;

a second obtaining module 120, configured to obtain management requirement parameters, where the management requirement parameters include a scene requirement parameter of a target simulation scene, a traffic participant requirement parameter, and a monitoring requirement parameter;

the first updating module 130 is configured to update the first original management information according to the scene demand parameter to obtain first target management information of the autonomous vehicle, and obtain a target scene time baseline according to the first target management information;

the second updating module 140 is configured to determine and update second original management information of each target traffic participant according to the traffic participant demand parameter, obtain second target management information of each target traffic participant, and obtain a target behavior time baseline according to the second target management information;

the third updating module 150 is configured to determine and update third original management information of each target monitoring task according to the monitoring demand parameter, obtain third target management information of each target monitoring task, and obtain a target monitoring time baseline according to the third target management information;

and the management module 160 is configured to manage the target simulation scenario according to the target scenario time baseline, the target behavior time baseline, and the target monitoring time baseline.

In one embodiment, the first original management information includes a first original lifecycle and initial status information of the autonomous vehicle, and the first updating module 130 is configured to update a starting time and an ending time of the first original lifecycle according to a scene lifecycle requirement parameter, to obtain a target starting time and a target ending time of the target simulation scene, and to obtain a first target lifecycle according to the target starting time and the target ending time; updating the state information of the automatic driving vehicle at the target starting moment according to the scene state demand parameters to obtain target initial state information; and obtaining first target management information of the automatic driving vehicle according to the first target life cycle and the target initial state information.

In one embodiment, the second original management information includes second original life cycle and continuous status information of the corresponding traffic participant, and the second updating module 140 is configured to determine a target traffic participant required by the target simulation scene according to the traffic participant attribute demand parameter; updating a second original life cycle corresponding to each target traffic participant according to the life cycle demand parameters of the traffic participants to obtain at least one second target life cycle of each target traffic participant in the target simulation scene; updating the continuous state information of each target traffic participant in each second target life cycle according to the traffic participant state demand parameters to obtain the target continuous state information of each target traffic participant in each second target life cycle; and obtaining second target management information of each target traffic participant according to the target traffic participant, the second target life cycle and the target continuous state information.

In one embodiment, the second updating module 140 is configured to determine a status information source of each target traffic participant according to the traffic participant status demand parameter, where the status information source includes at least one of a manual simulation status information source and a road-end real status information source; and acquiring target continuous state information of each target traffic participant in each second target life cycle from the state information source corresponding to each target traffic participant.

In an embodiment, the third original management information includes an original monitoring trigger condition, an original monitoring execution condition, and an original monitoring termination condition corresponding to the monitoring task, and the third updating module 150 is configured to determine, according to the monitoring task demand parameter, a target monitoring task required by the target simulation scenario; updating the original monitoring triggering condition, the original monitoring executing condition and the original monitoring terminating condition corresponding to each target monitoring task according to the monitoring condition demand parameters to obtain the target monitoring triggering condition, the target monitoring executing condition and the target monitoring terminating condition corresponding to each target monitoring task; and obtaining third target management information of each target monitoring task according to the target monitoring triggering condition, the target monitoring executing condition and the target monitoring terminating condition.

In one embodiment, the original simulation scene management file further includes a recording time base line group based on time synchronization with other time bases, the recording time base line group includes at least one recording time base line, each recording time base line manages different recording tasks based on fourth original management information, the management requirement parameters further include recording requirement parameters, the simulation scene management device further includes a fourth updating module, the fourth updating module is used for determining and updating fourth original management information of each target recording task according to the recording requirement parameters to obtain fourth target management information of each target recording task, and the target recording time base line is obtained according to the fourth target management information; and managing the target simulation scene according to the target recording time base line.

In one embodiment, the fourth original management information includes a fourth original life cycle corresponding to the recording task, and the fourth updating module is configured to determine a target recording task required by the target simulation scene according to the recording task demand parameter; updating a fourth original life cycle corresponding to each target recording task according to the recording life cycle demand parameters to obtain a fourth target life cycle corresponding to each target recording task; and obtaining fourth target management information of each target recording task according to the fourth target life cycle.

In an embodiment, the fourth updating module is configured to update the mode information of each target recording time baseline into a recording mode in a first target life cycle of the target simulation scene, so that each target recording time baseline executes a corresponding target recording task to obtain recording information; and after the first target life cycle of the target simulation scene, updating the mode information of the at least one target recording time base line into a playback mode so as to enable the at least one target recording time base line to play back the corresponding recording information.

In an embodiment, the simulation scenario management apparatus further includes a fifth updating module, where the fifth updating module is configured to obtain state information of all time baselines in the original simulation scenario management file; and updating the state information of all the target time baselines into a first state so that all the target time baselines participate in the target simulation scene, and updating the state information of other time baselines into a second state so that other time baselines do not participate in the target simulation scene.

Different from the prior art, the simulation scene management device provided by the invention decouples the automatic driving vehicle, each traffic participant and each monitoring task by setting various time baselines based on time synchronization, each time baseline is only responsible for respective management tasks, when a new simulation scene needs to be constructed, only the time baselines of each item required to participate in the simulation scene need to be determined according to the management requirement parameters, and the original management information in each target time baseline is respectively updated to obtain the target management information, because of the low coupling between the management objects of each time baseline, the information updating of each target time baseline is independently carried out without influencing other target time baselines, therefore, the invention can rapidly and flexibly construct a plurality of different simulation scenes according to the management requirement parameters and the original scene management files, and supports generalization scene, and the management cost and difficulty are lower, the efficiency is higher, and the requirement of simulation test can be met.

Accordingly, the present invention also provides an electronic device, as shown in fig. 13, which may include radio frequency circuit 1301, memory 1302 including one or more computer-readable storage media, input unit 1303, display unit 1304, sensor 1305, audio circuit 1306, WiFi module 1307, processor 1308 including one or more processing cores, and power supply 1309. Those skilled in the art will appreciate that the electronic device configuration shown in fig. 13 does not constitute a limitation of the electronic device and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components. Wherein:

the rf circuit 1301 may be used for receiving and transmitting signals during information transmission and reception or during a call, and in particular, receives downlink information of a base station and then sends the received downlink information to one or more processors 1308 for processing; in addition, data relating to uplink is transmitted to the base station. The memory 1302 may be used to store software programs and modules, and the processor 1308 may execute various functional applications and data processing by operating the software programs and modules stored in the memory 1302. The input unit 1303 may be used to receive input numeric or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

The display unit 1304 may be used to display information input by or provided to a user as well as various graphical user interfaces of the electronic device, which may be made up of graphics, text, icons, video, and any combination thereof.

The electronic device may also include at least one sensor 1305, such as a light sensor, motion sensor, and other sensors. The audio circuitry 1306 includes speakers that may provide an audio interface between a user and the electronic device.

WiFi belongs to a short-distance wireless transmission technology, and the electronic device can help a user send and receive e-mails, browse webpages, access streaming media and the like through the WiFi module 1307, and provides wireless broadband internet access for the user. Although fig. 13 shows the WiFi module 1307, it is understood that it does not belong to the essential constitution of the electronic device, and may be omitted entirely as needed within the scope not changing the essence of the invention.

The processor 1308 is a control center of the electronic device, connects various parts of the entire cellular phone by various interfaces and lines, and performs various functions of the electronic device and processes data by operating or executing software programs and/or modules stored in the memory 1302 and calling data stored in the memory 1302, thereby integrally monitoring the cellular phone.

The electronic device also includes a power supply 1309 (such as a battery) for powering the various components, which may preferably be logically connected to the processor 1308 via a power management system that may be used to manage charging, discharging, and power consumption.

Although not shown, the electronic device may further include a camera, a bluetooth module, and the like, which are not described in detail herein. Specifically, in this embodiment, the processor 1308 in the electronic device loads an executable file corresponding to a process of one or more application programs into the memory 1302 according to the following instructions, and the processor 1308 runs the application programs stored in the memory 1302, so as to implement the following functions:

acquiring an original simulation scene management file, wherein the original simulation scene management file comprises a scene time base line, a behavior time base line group and a monitoring time base line group based on time synchronization; the scene time base line manages the automatic driving vehicle based on the first original management information, the behavior time base line group comprises at least one behavior time base line, each behavior time base line manages different traffic participants based on the second original management information, the monitoring time base line group comprises at least one monitoring time base line, and each monitoring time base line manages different monitoring tasks based on the third original management information; acquiring management demand parameters, wherein the management demand parameters comprise scene demand parameters of a target simulation scene, traffic participant demand parameters and monitoring demand parameters; updating the first original management information according to the scene demand parameters to obtain first target management information of the automatic driving vehicle, and obtaining a target scene time baseline according to the first target management information; determining and updating second original management information of each target traffic participant according to the traffic participant demand parameters to obtain second target management information of each target traffic participant, and obtaining a target behavior time base line according to the second target management information; determining and updating third original management information of each target monitoring task according to the monitoring demand parameters to obtain third target management information of each target monitoring task, and obtaining a target monitoring time base line according to the third target management information; and managing the target simulation scene according to the target scene time base line, the target behavior time base line and the target monitoring time base line.

In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor.

To this end, the present invention provides a computer readable storage medium having stored therein a plurality of instructions that are loadable by a processor to cause the following functions:

The above operations can be implemented in the foregoing embodiments, and are not described in detail herein.

Wherein the computer-readable storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.

Since the instructions stored in the computer-readable storage medium can execute the steps of any method provided by the present invention, the beneficial effects that any method provided by the present invention can achieve can be achieved, for details, see the foregoing embodiments, and are not described herein again.

The simulation scene management method, the simulation scene management device, the electronic device and the storage medium provided by the invention are described in detail, a specific example is applied in the description to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the technical scheme and the core idea of the invention; those of ordinary skill in the art will understand that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims

1. A simulation scene management method is characterized by comprising the following steps:

acquiring an original simulation scene management file, wherein the original simulation scene management file comprises a scene time base line based on time synchronization, a behavior time base line group and a monitoring time base line group; the scene time base line manages the automatic driving vehicle based on first original management information, the behavior time base line group comprises at least one behavior time base line, each behavior time base line manages different traffic participants based on second original management information, the monitoring time base line group comprises at least one monitoring time base line, and each monitoring time base line manages different monitoring tasks based on third original management information; the first original management information comprises a first original life cycle and initial state information of the automatic driving vehicle, and the second original management information comprises a second original life cycle and continuous state information of the corresponding traffic participant;

updating the first original management information according to the scene demand parameters to obtain first target management information of the automatic driving vehicle, and obtaining a target scene time baseline according to the first target management information; the method comprises the following steps: updating the starting time and the ending time of the first original life cycle according to the scene life cycle demand parameters to obtain the target starting time and the target ending time of the target simulation scene, and obtaining a first target life cycle according to the target starting time and the target ending time; updating the state information of the automatic driving vehicle at the target starting moment according to the scene state demand parameters to obtain target initial state information; obtaining first target management information of the automatic driving vehicle according to the first target life cycle and the target initial state information;

determining and updating second original management information of each target traffic participant according to the traffic participant demand parameters to obtain second target management information of each target traffic participant, and obtaining a target behavior time base line according to the second target management information; the method comprises the following steps: determining target traffic participants required by a target simulation scene according to the attribute demand parameters of the traffic participants; updating a second original life cycle corresponding to each target traffic participant according to the life cycle demand parameters of the traffic participants to obtain at least one second target life cycle of each target traffic participant in the target simulation scene; updating the continuous state information of each target traffic participant in each second target life cycle according to the traffic participant state demand parameters to obtain the target continuous state information of each target traffic participant in each second target life cycle; obtaining second target management information of each target traffic participant according to the target traffic participant, the second target life cycle and the target continuous state information;

2. The simulation scenario management method of claim 1, wherein the step of updating the continuous status information of each target traffic participant in each second target life cycle according to the traffic participant status demand parameter to obtain the target continuous status information of each target traffic participant in each second target life cycle comprises:

determining a state information source of each target traffic participant according to the state demand parameters of the traffic participants, wherein the state information source comprises at least one of a manual simulation state information source and a road end real state information source;

and acquiring target continuous state information of each target traffic participant in each second target life cycle from the state information source corresponding to each target traffic participant.

3. The simulation scene management method according to claim 1, wherein the third original management information includes an original monitoring trigger condition, an original monitoring execution condition, and an original monitoring termination condition corresponding to the monitoring task, and the step of determining and updating the third original management information of each target monitoring task according to the monitoring requirement parameter to obtain the third target management information of each target monitoring task includes:

determining a target monitoring task required by a target simulation scene according to the monitoring task demand parameter;

updating the original monitoring triggering condition, the original monitoring executing condition and the original monitoring terminating condition corresponding to each target monitoring task according to the monitoring condition demand parameters to obtain the target monitoring triggering condition, the target monitoring executing condition and the target monitoring terminating condition corresponding to each target monitoring task;

and obtaining third target management information of each target monitoring task according to the target monitoring triggering condition, the target monitoring executing condition and the target monitoring terminating condition.

4. The simulation scene management method according to claim 1, wherein the original simulation scene management file further includes a recording time base set synchronized with other time bases based on time, the recording time base set includes at least one recording time base, each recording time base manages a different recording task based on fourth original management information, the management requirement parameter further includes a recording requirement parameter, and after the step of obtaining the management requirement parameter, the method further includes:

determining and updating fourth original management information of each target recording task according to the recording demand parameters to obtain fourth target management information of each target recording task, and obtaining a target recording time base line according to the fourth target management information;

and managing the target simulation scene according to the target recording time base line.

5. The method for managing the simulation scene according to claim 4, wherein the fourth original management information includes a fourth original life cycle of the corresponding recording task, and the step of determining and updating the fourth original management information of each target recording task according to the recording requirement parameter to obtain the fourth target management information of each target recording task includes:

determining a target recording task required by a target simulation scene according to the recording task demand parameter;

updating a fourth original life cycle corresponding to each target recording task according to the recording life cycle demand parameters to obtain a fourth target life cycle corresponding to each target recording task;

and obtaining fourth target management information of each target recording task according to the fourth target life cycle.

6. The simulation scenario management method of claim 4, wherein the step of managing the target simulation scenario according to the target recording time baseline comprises:

updating the mode information of each target recording time baseline into a recording mode in a first target life cycle of a target simulation scene so as to enable each target recording time baseline to execute a corresponding target recording task and obtain recording information;

after the first target life cycle of the target simulation scene, updating the mode information of at least one target recording time base line into a playback mode so as to enable the at least one target recording time base line to play back the corresponding recording information.

7. The simulation scenario management method of any of claims 1 to 6, further comprising, before the step of managing the target simulation scenario according to the target scenario time baseline, the target behavior time baseline and the target monitoring time baseline:

acquiring state information of all time baselines in the original simulation scene management file;

and updating the state information of all the target time baselines into a first state so that all the target time baselines participate in the target simulation scene, and updating the state information of other time baselines into a second state so that the other time baselines do not participate in the target simulation scene.

8. A simulation scene management apparatus, comprising:

the system comprises a first acquisition module, a second acquisition module and a monitoring module, wherein the first acquisition module is used for acquiring an original simulation scene management file, and the original simulation scene management file comprises a scene time base line, a behavior time base line group and a monitoring time base line group based on time synchronization; the scene time base line manages the automatic driving vehicle based on first original management information, the behavior time base line group comprises at least one behavior time base line, each behavior time base line manages different traffic participants based on second original management information, the monitoring time base line group comprises at least one monitoring time base line, and each monitoring time base line manages different monitoring tasks based on third original management information; the first original management information comprises a first original life cycle and initial state information of the automatic driving vehicle, and the second original management information comprises a second original life cycle and continuous state information of the corresponding traffic participant;

the first updating module is used for updating the first original management information according to the scene demand parameters to obtain first target management information of the automatic driving vehicle, and obtaining a target scene time base line according to the first target management information; the method comprises the following steps: updating the starting time and the ending time of the first original life cycle according to the scene life cycle demand parameters to obtain the target starting time and the target ending time of the target simulation scene, and obtaining a first target life cycle according to the target starting time and the target ending time; updating the state information of the automatic driving vehicle at the target starting moment according to the scene state demand parameters to obtain target initial state information; obtaining first target management information of the automatic driving vehicle according to the first target life cycle and the target initial state information;

the second updating module is used for determining and updating second original management information of each target traffic participant according to the traffic participant demand parameters to obtain second target management information of each target traffic participant and obtain a target behavior time base line according to the second target management information; the method comprises the following steps: determining target traffic participants required by a target simulation scene according to the attribute demand parameters of the traffic participants; updating a second original life cycle corresponding to each target traffic participant according to the life cycle demand parameters of the traffic participants to obtain at least one second target life cycle of each target traffic participant in the target simulation scene; updating the continuous state information of each target traffic participant in each second target life cycle according to the traffic participant state demand parameters to obtain the target continuous state information of each target traffic participant in each second target life cycle; obtaining second target management information of each target traffic participant according to the target traffic participant, the second target life cycle and the target continuous state information;