Closed-loop method for simulation test and site test involving dynamic scenarios oriented towards autonomous driving test

Abstract

Provided in the present invention is a closed-loop method for a simulation test and a site test involving dynamic scenarios oriented towards an autonomous driving test. The method comprises the following steps: step 1, establishing a dynamic scenario oriented towards a simulation test, and carrying out a simulation test; step 2, establishing a dynamic scenario oriented towards a site test, and carrying out a site test; step 3, converting the simulation-test dynamic scenario into the site-test dynamic scenario, and then carrying out a site test; and step 4, converting the site-test dynamic scenario into the simulation-test dynamic scenario, and then carrying out a simulation test. The present invention combines a virtual simulation test with a closed-site test to generate a dynamic scenario including dynamic traffic participants, and achieves mutual conversion between a simulation-test dynamic scenario and a site-test dynamic scenario.

Description

A closed-loop method of dynamic scene simulation and site test for automatic driving test TECHNICAL FIELD

[0001] The present application belongs to the field of automatic driving test, and particularly relates to a closed-loop method of dynamic scene simulation and site test for automatic driving test. BACKGROUND

[0002] The current automatic driving test scene generation method includes the following steps: firstly, a scene modeling tool is used to model the static road environment such as road network, traffic signs, traffic lights, intersection structure, etc., usually using map data or virtual modeling software for modeling, usually using OpenDrive standard; then a dynamic scene editing tool is used to generate various dynamic traffic participants, including vehicles, pedestrians, bicycles, etc., and the motion trajectory of the traffic participants is generated through trajectory editing, including path planning, speed control, etc., to simulate their motion in the scene, usually using OpenScenario standard; finally, the generated dynamic and static scenes are loaded into the simulation platform for testing and evaluation, which usually includes physical engine, sensor simulation, etc.

[0003] From the above, the existing technology has the defect that simulation test and site test are disconnected, and it is difficult to meet the authenticity and reproducibility at the same time. Specifically, after the dynamic and static scene design meets the scene generation and simulation visualization, it cannot be deployed in the field test, so it cannot meet the authenticity requirement; and the traditional standard site test cannot meet the consistency and reproducibility of the test scene because it needs to be controlled by people. SUMMARY

[0004] The purpose of the present application is to provide a closed-loop method of dynamic scene simulation and site test system for automatic driving test, which combines virtual simulation and closed site test to generate dynamic scenes containing dynamic traffic participants, and realizes the mutual conversion between simulation and site test dynamic scenes. The technical scheme adopted is:

[0005] A closed-loop method of dynamic scene simulation and site test for automatic driving test, including the following steps:

[0006] Step 1, based on the dynamic scene protocol FX for intelligent driving simulation test, a simulation scene information file F1 for simulation test is established, and simulation test is carried out.

[0007] Among them, the first simulation scene information file F1 is created by a dynamic scene editing tool, and the first simulation scene information file F1 contains road network information and traffic participant information;

[0008] The road network information includes an OpenDrive map and an OpenSceneGraph scene model.

[0009] The traffic participant information includes a name, a category, a traffic participant rendering model, an initial position, an initial speed, a trajectory information list, and a speed change.

[0010] Step 2: Based on the dynamic scene protocol CX for site testing, a first site scene information file C1 is established for site testing, and site testing is performed.

[0011] The first site scene information file C1 is established by a dynamic scene editing tool.

[0012] The first site scene information file C1 includes the following information: a scene name, a scene number, a scene description, a number of traffic participants, and traffic participant information.

[0013] The traffic participant information includes an OBU number, a number of trajectory points, and a trajectory information list.

[0014] Step 3: The dynamic scene for simulation testing is converted into a dynamic scene for site testing, and then site testing is performed, including the following steps:

[0015] Step 3A: The first simulation scene information file F1 is converted into a second site scene information file C2.

[0016] Step 3A1: Based on the dynamic scene protocol CX for site testing, a second site scene information file C2 is created for site testing.

[0017] The second site scene information file C2 has the same fields as the first site scene information file C1.

[0018] All fields in the second site scene information file C2 have information in a temporary absence state.

[0019] Step 3A2: According to the dynamic scene protocol FX for simulation testing and the dynamic scene protocol CX for site testing, data conversion is performed, and the converted data is filled into the corresponding fields in the second site scene information file C2.

[0020] Step 3A3: Supplementary data is filled into the unfilled fields in the second site scene information file C2.

[0021] The fields that need to be supplemented include a scene name, a scene number, a scene description, a number of cloud-controlled simulated traffic participants, and a number of trajectory points for each cloud-controlled simulated traffic participant.

[0022] Step 3B: The second site scene information file C2 output in step 3A3 is stored in a cloud database as a JSON file.

[0023] Step 3C, adding the OBU number corresponding to the respective communication unit to the cloud-controlled simulated traffic participants to realize correct delivery of trajectory information;

[0024] Step 3D, conducting a site test:

[0025] The cloud server reads the information of each traffic participant in the No. 2 site scenario information file C2 in the cloud database, and according to the OBU number, correspondingly transmits the trajectory information list to the communication unit of the cloud-controlled simulated traffic participant through 5G;

[0026] The cloud-controlled simulated traffic participants include mannequins and car models placed in the actual test site; the cloud-controlled simulated traffic participants can move in the actual test site and are each installed with a communication unit OBU.

[0027] Preferably, after step 3, the following steps are further included:

[0028] Step 4, converting the site test dynamic scenario into a simulation test dynamic scenario, and then conducting a simulation test. Specifically, the following steps are included:

[0029] Step 4A, converting the site test dynamic scenario into a simulation test dynamic scenario:

[0030] Step 4A1, the communication unit of each cloud-controlled simulated traffic participant uploads the test data in the running of the site test dynamic scenario in step 3D, including GPS positioning coordinate information, heading angle information, category and OBU number, to the cloud for receiving the test data of each cloud-controlled simulated traffic participant;

[0031] The GPS positioning coordinate information is provided by a GPS receiver in real time, the heading angle information is provided by a self-contained inertial navigation unit in real time, and the category and OBU number are obtained by reading the self-state information file pre-stored in the storage unit;

[0032] The No. 3 site scenario information file C3 is generated using the site test dynamic scenario protocol CX; in the No. 3 site scenario information file C3, there is a field, and the data in the field is to be filled in;

[0033] The collected test data of each cloud-controlled simulated traffic participant is filled into the corresponding field of the No. 3 site scenario information file C3;

[0034] Step 4A2, based on the dynamic scenario protocol FX for intelligent driving simulation test, a No. 2 simulation scenario information file F2 for simulation test is established;

[0035] The No. 2 simulation scenario information file F2 has the same fields as the No. 1 simulation scenario information file F1;

[0036] All fields in the second simulation scene information file F2, the information in which is in a temporary absence state;

[0037] Step 4A3, according to the dynamic scene protocol FX of the simulation test and the dynamic scene protocol CX of the site test, data in the third site scene information file C3 is converted into corresponding fields in the second simulation scene information file F2;

[0038] Step 4A4, supplementary data is filled into the fields in the second simulation scene information file F2 which are not filled;

[0039] The fields which need to be supplemented include: rendering model of each traffic participant, OpenDrive map, and OpenSceneGraph scene model.

[0040] Step 4B, the second simulation scene information file F2 storing data in step 4A4 is loaded into a simulation software supporting OpenScenario protocol to perform virtual simulation test.

[0041] Preferably, step 3A2 specifically includes the following conversion steps:

[0042] The name is converted into OBU number, the coordinate in the initialization position is converted into the longitude and latitude at 0 time, the attitude angle h in the initialization position is converted into the heading angle, the initialization speed is converted into the speed at 0 time, the time t is converted, and the current speed is converted.

[0043] Preferably, step 4A2 specifically includes the following conversion steps:

[0044] The OBU number is converted into the name, the longitude and latitude at 0 time is converted into the coordinate in the initialization position, the heading angle at 0 time is converted into the attitude angle h in the initialization position, the speed at 0 time is converted into the initialization speed, the time t is converted, and the current speed is converted.

[0045] Compared with the prior art, the application has the following advantages:

[0046] 1. The automatic driving dynamic scene is converted and regenerated between simulation test and site test.

[0047] Through scene conversion and on-site deployment in site test, the conversion from virtual environment test to actual road test is realized, so that the simulation test scene can be directly applied to actual site test, the authenticity of automatic driving test is ensured, and the consistency and reproducibility of automatic driving site test are improved.

[0048] 2. The full-process closed loop from virtual scene to site test is realized, which provides a comprehensive and efficient solution for key edge test of automatic driving system.

[0049] 3. After the dynamic and static scene designs are generated, or after the scenes are regenerated, simulation visualization can be performed. By performing simulation visualization verification in a virtual environment, problems can be discovered and optimized in advance, reducing unnecessary repetitive testing during field testing, and significantly reducing the cost and time of field testing.

[0050] 4. Through dynamic scene design, transformation, and regeneration, the performance and safety of autonomous driving systems in various complex scenarios can be evaluated more realistically, cost-effectively, and quickly, promoting the development and application of autonomous driving evaluation technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flow chart of a closed-loop method for dynamic scenario simulation and field testing for autonomous driving testing;

[0052] Figure 2 shows a dynamic scenario simulation test protocol for multiple traffic participants based on the OpenScenario standard;

[0053] Figure 3 shows the action elements of the multi-traffic participant simulation motion control method;

[0054] Figure 4 shows the dynamic scenario protocol for field testing of cloud-controlled simulated traffic participants;

[0055] Figure 5 shows the cloud-controlled simulation method for converting and generating trajectories of traffic participants;

[0056] Figure 6 shows the cloud-controlled simulated traffic participant field test scenario regeneration method. DETAILED DESCRIPTION

[0057] The following, with reference to schematic diagrams, describes in more detail the closed-loop method for dynamic scene simulation and field testing for autonomous driving testing of the present invention. These schematic diagrams illustrate preferred embodiments of the present invention. It should be understood that those skilled in the art may modify the invention described herein while still achieving the beneficial effects of the present invention. Therefore, the following description should be understood as generally known to those skilled in the art and not as a limitation of the present invention.

[0058] As shown in Figures 1 to 6, the closed-loop simulation and field testing method proposed in this invention focuses on achieving the transformation and regeneration of multi-target dynamic traffic scene simulation and real target trajectories in field testing, so as to further improve the authenticity, consistency and reproducibility of autonomous driving testing.

[0059] Step 1: Based on the dynamic scenario protocol FX for intelligent driving simulation testing, create a simulation scenario information file F1 for simulation testing and perform simulation testing.

[0060] Among them, the dynamic scenario protocol FX for intelligent driving simulation testing is developed based on the OpenScenario standard.

[0061] OpenScenario is an open protocol for describing autonomous driving scenarios, mainly used to describe the synchronous operation of multiple entities involved in autonomous driving scenarios, such as lane changing or gear shifting actions of vehicles, and is commonly used for the definition of dynamic scenarios in autonomous driving simulation tests. The multi-traffic participant dynamic scenario protocol for simulation tests in this embodiment is developed based on the OpenScenario 1.0 protocol, providing a simple and standardized way to define complex traffic scenarios, including vehicles, pedestrians and other traffic participants.

[0062] The dynamic scenario protocol for simulation tests in this embodiment FX carries all the necessary dynamic scenario information through the OpenScenario 1.0 standard, the motion trajectory of the traffic participant is realized by the trajectory following action, the speed change of the traffic participant is realized by the gear shifting action, and the time trigger is used as the execution condition of the traffic participant action, as shown in FIG. 2. Among them, 1-3 are trajectory points.

[0063] Among them, the first simulation scenario information file F1 is created through a dynamic scenario editing tool. In this embodiment, the fields of the first simulation scenario information file F1 and the second simulation scenario information file F2 are the same.

[0064] The first simulation scenario information file F1 includes road network information and traffic participant information, and the detailed composition is shown in FIG. 3.

[0065] Among them, the road network information includes OpenDrive map and OpenSceneGraph scene model.

[0066] The traffic participant information includes name, category, traffic participant rendering model, initial position, initial speed, trajectory information list, and speed change (v within t).

[0067] The detailed steps of creating the first simulation scenario information file F1 based on the protocol FX are as follows:

[0068] Import high-precision maps in OpenDrive format, in order to maintain the consistency of simulation and field tests, it is necessary to obtain the closed test site OpenDrive map and OpenSceneGraph scene model made by high-precision map suppliers in advance.

[0069] Use a driving scenario editor that supports the OpenScenario standard to create a dynamic scenario, add the traffic participants required by the dynamic scenario, set the name for each participant, and establish a continuous trajectory. Not only do you need to select the trajectory point coordinates on the map, but you also need to adjust the vehicle speed and heading angle corresponding to different trajectory points, as well as set the waiting time.

[0070] The protocol FX-based scene information file F made by the above method also meets the OpenScenario standard, is an xosc file, and can be loaded into automatic driving simulation software such as Carla, Carmaker, VTD, 51SimOne, and the like for virtual simulation testing.

[0071] The first simulation scene information file F1 and the second simulation scene information file F2 both belong to the scene information file F in FIG. 1.

[0072] Specifically, the simulation test step in step 1: the first simulation scene information file F1 is loaded into simulation software supporting the OpenScenario protocol for virtual simulation testing.

[0073] Step 2, based on the site test dynamic scene protocol CX (i.e., the delivery protocol in FIG. 4), a first site scene information file C1 for site testing is established, and site testing is performed.

[0074] In this embodiment, the first site scene information file C1 and the second site scene information file C2 both belong to the scene information file C in FIG. 1.

[0075] The first site scene information file C1 is established by a dynamic scene editing tool.

[0076] The site test dynamic scene protocol CX describes the motion information of cloud-controlled simulated traffic participants at different times in a dynamic scene.

[0077] The cloud-controlled simulated traffic participants include human models and vehicle models placed in an actual test site.

[0078] The cloud-controlled simulated traffic participants can all move in the actual test site and are all installed with a communication unit OBU.

[0079] The site test dynamic scene protocol CX is developed based on JSON syntax. JSON syntax is a subset of JavaScript Object Notation syntax, and the two structures of JSON are object and array. JSON syntax is a prior art.

[0080] An object is saved by a pair of braces {} and is an unordered collection of name / value pairs. An object starts with a left brace { and ends with a right brace}. Each "key" is followed by a colon :, and the name / value pairs are separated by commas.

[0081] An array is saved by a pair of square brackets [] and is an ordered collection of values. An array starts with a left square bracket [ and ends with a right square bracket ], and the values are separated by commas.

[0082] The first site scene information file C1 based on the field test dynamic scene protocol CX contains: scene name, scene number, scene description, number of traffic participants, and traffic participant information, and the detailed composition is shown in FIG. 4.

[0083] The traffic participant information includes: OBU number (number of communication unit), number of trajectory points, and trajectory information list. The trajectory point is a discrete point, and each point includes time, latitude and longitude information, heading angle, and speed information.

[0084] Specifically, the field test step in step 2:

[0085] The first site scene information file C1 is stored in the cloud database in the form of a JSON file;

[0086] The cloud-controlled simulated traffic participants are added with OBU numbers corresponding to the respective communication units to realize correct delivery of trajectory information;

[0087] The cloud server reads the traffic participant information of the second site scene information file C2 in the cloud database, and according to the OBU number, correspondingly transmits the trajectory information list to the communication unit of the cloud-controlled simulated traffic participant through 5G;

[0088] The cloud-controlled simulated traffic participants include man models and car models placed in the actual test site; the cloud-controlled simulated traffic participants can move in the actual test site and are all installed with communication units OBU.

[0089] Step 3, convert the simulation test dynamic scene into a field test dynamic scene, and then perform field test. Specifically, the following steps are included:

[0090] Step 3A, convert the first simulation scene information file F1 into the second site scene information file C2.

[0091] As shown in FIG. 5, the first simulation scene information file F1 in step 1 is converted into the second site scene information file C2.

[0092] Step 3A1, based on the field test dynamic scene protocol CX, create the second site scene information file C2 for field test;

[0093] The second site scene information file C2 has the same fields as the first site scene information file C1, and the data in the fields are to be filled;

[0094] That is, all field information in the second site scene information file C2 is in a temporary absence state, and needs to be read from the first simulation scene information file F1 and filled by conversion.

[0095] Step 3A2, according to the dynamic scene protocol FX for intelligent driving simulation test and the dynamic scene protocol CX for site test, data conversion is performed into the corresponding fields of the second site scene information file C2.

[0096] The conversion program is deployed on a cloud server, and the trajectory conversion process of the cloud-controlled simulation traffic participants is based on cloud control. The specific process is shown in FIG. 5:

[0097] 1. The name is converted to the OBU number.

[0098] For the names of the simulation scene traffic participants, in order to enable the cloud-controlled simulation traffic participants to be deployed and controlled in the site test, one-to-one mapping operations need to be performed in the dynamic scene of the site test.

[0099] For example, "fcar1" is mapped to the first cloud-controlled simulation traffic participant, and the OBU number is "OBU-0001".

[0100] For example, "fhuman1" is mapped to the tenth cloud-controlled simulation traffic participant, and the OBU number is "OBU-0010".

[0101] 2. The position (x, y) in the initialization position under the dynamic scene protocol FX for simulation test is converted to the longitude and latitude at time 0 under the dynamic scene protocol CX for site test.

[0102] The position in the dynamic scene protocol FX for simulation test uses UTM coordinates, and needs to be switched to WGS84 longitude and latitude coordinates when converted to the dynamic scene protocol CX for site test.

[0103] The principle of coordinate transformation is prior art. Specifically, both coordinates are international coordinate system standards, and the to_latlon function of the Python utm library can be used to achieve this.

[0104] If the UTM coordinates are [x, y, z], the UTM zone symbol is utm_zone, and the north and south hemisphere symbols are "N" and "S" respectively, then the method for calculating the longitude and latitude in the northern hemisphere is lat, lon = utm.to_latlon(x, y, utm_zone, 'N').

[0105] The z coordinate value of the UTM coordinates is not used, and only the x and y coordinates are converted.

[0106] 3. The attitude angle h in the initialization position under the dynamic scene protocol FX for simulation test corresponds to the heading angle under the dynamic scene protocol CX for site test. That is, the attitude angle corresponds to the heading angle. "Corresponds" means that the numerical connotation is consistent, but the unit or resolution and syntax are not consistent. The definition of "corresponds" below is the same as this place.

[0107] For the three-axis attitude angle [h, p, r], since the pitch angle p and roll angle r are close to 0° under normal circumstances, only the heading angle h in the simulation scene trajectory is retained as the heading angle heading in the field test trajectory.

[0108] 4. The initial speed v under the simulation test dynamic scene protocol FX corresponds to the 0-time speed v under the field test dynamic scene protocol CX.

[0109] 5. In the simulation test dynamic scene protocol FX, the time t and the change time t in the speed change in the trajectory information list correspond to the time t under the field test dynamic scene protocol CX. Among them, the two times t in the simulation test dynamic scene protocol FX are the same value.

[0110] 6. The current speed v under the simulation test dynamic scene protocol FX corresponds to the current speed under the field test dynamic scene protocol CX.

[0111] 7. Unit conversion. The unit conversion is as follows:

[0112] According to the actual transmission efficiency needs, in the conversion process from the simulation test dynamic scene protocol FX to the field test dynamic scene protocol CX, the unit needs to be converted, and some values also need to be compressed from floating-point numbers to unsigned integers with specific resolution. Details are as follows.

[0113] Under the field test dynamic scene protocol CX, the time unit is millisecond, and is represented by Int64 signed integer. For example, the time 1.5s under the simulation test dynamic scene protocol FX is converted to the time 1500 under the field test dynamic scene protocol CX.

[0114] Under the field test dynamic scene protocol CX, the speed resolution is 0.02m / s, which is converted to Uint32 unsigned integer, and the effective range is between [-8191, 8191]. For example, the speed 1m / s under the simulation test dynamic scene protocol FX is converted to the speed 50 under the field test dynamic scene protocol CX.

[0115] Under the field test dynamic scene protocol CX, the heading angle resolution is 0.0125 degrees, which is converted to Uint16 unsigned integer, and the effective range is between [0, 28799]. For example, the attitude angle h is 180° under the simulation test dynamic scene protocol FX, which is converted to the heading angle 14400 under the field test dynamic scene protocol CX.

[0116] The latitude and longitude unit is °, and Float64 floating-point number is used.

[0117] Step 3A3, supplement data to the unfilled fields in the second field scene information file C2.

[0118] The fields that need additional information include: scene name, scene number, scene description, number of cloud-controlled simulated traffic participants, and number of trajectory points for each cloud-controlled simulated traffic participant.

[0119] The process of calculating the number of cloud-controlled simulated traffic participants and the number of trajectory points for each cloud-controlled simulated traffic participant is as follows:

[0120] Number of cloud-controlled simulated traffic participants: Count all simulated traffic participants in the first simulation scene information file F1.

[0121] Number of trajectory points for each cloud-controlled simulated traffic participant: Count the length of the trajectory information list for each simulated traffic participant in the first simulation scene information file F1.

[0122] Scene name, scene number, and scene description are used to distinguish the scene information files C in the cloud server scene library and need to be manually added.

[0123] Step 3B, store the second site scene information file C2 output in step 3A3 as a JSON file in the cloud database.

[0124] Step 3C, add OBU numbers corresponding to the communication units of cloud-controlled simulated traffic participants to realize correct trajectory information delivery.

[0125] Step 3D, conduct site testing.

[0126] The cloud server identifies and reads the traffic participant information of the second site scene information file C2 in the cloud database according to the scene name, scene number, and scene description, and correspondingly transmits the trajectory information list to the communication units OBU of cloud-controlled simulated traffic participants (human models and vehicle models) through 5G based on the OBU number.

[0127] After receiving the information output by the communication unit, the control unit of the human model and the vehicle model controls the drive motor and the steering engine to realize trajectory following through the tracking control algorithm. The drive motor and the steering engine form a drive walking module, and the human model and the vehicle model both include the drive walking module.

[0128] Based on the above automatic tracking driving of human models and vehicle models, automatic driving dynamic scene automation testing is realized.

[0129] Among them, the cloud-controlled simulated traffic participants arriving at the specified starting point are given the corresponding trajectory information list in real time, and the specific process is as follows:

[0130] 1. The cloud-controlled simulated traffic participants upload their own coordinates, and the cloud retrieves the nearest neighbor trajectory point sequence number based on the coordinates, and delivers the subsequent fixed number of trajectory points of the corresponding trajectory of the cloud-controlled simulated traffic participants through the sliding window method.

[0131] Wherein, the cloud control simulation traffic participants include GPS modules (GPS receivers), which can obtain real-time positioning, and automatically upload positioning information through OBU and cloud communication.

[0132] Wherein, the method for determining the arrival at the starting point position is that the cloud control simulation traffic participants upload their own coordinates, the cloud calculates the distance and heading angle error from the initial time of the trajectory point according to the coordinates of the cloud control simulation traffic participants, and if the distance is less than a specified threshold, such as 1 m, and the heading angle error is less than 10 degrees, it is considered that the starting point position is reached.

[0133] 2. After the cloud control simulation traffic participants receive the trajectory point list data through the OBU, the predetermined route is driven through the tracking control algorithm.

[0134] 3. When the cloud control simulation traffic participants reach the end of the trajectory, the cloud sends a stop signal to it.

[0135] In addition, the number of cloud control simulation traffic participants is used to confirm the number of cloud control simulation traffic participants required for scheduling, and the number of trajectory points is used to avoid index overflow when the trajectory point is issued.

[0136] Steps 3A-3D, form a trajectory conversion generation method based on cloud control simulation traffic participants.

[0137] Step 4, convert the field test dynamic scene into a simulation test dynamic scene, and then perform simulation testing. Specifically, the following steps are included:

[0138] Step 4A, convert the field test dynamic scene into a simulation test dynamic scene.

[0139] Step 4A1, the communication unit of each cloud control simulation traffic participant uploads the test data in the running of the field test dynamic scene based on 5G communication in step 3D, including GPS positioning coordinate information, heading angle information, category, OBU number, time and speed, and the cloud receives the test data of each cloud control simulation traffic participant.

[0140] Wherein, the GPS positioning coordinate information is provided by the GPS receiver in real time, the heading angle information is provided by the self-contained inertial navigation unit in real time, and the category and OBU number are obtained by reading the self-state information file stored in the storage unit in advance.

[0141] Wherein, in step 4A1, the category and OBU number are pre-modified and stored in the storage unit of the cloud control simulation traffic participant.

[0142] The time is the cumulative time of the control unit after the dynamic scene starts running, which is obtained by the control unit. That is, the time from the start of the scene, the control unit time of the current time minus the control unit time at the scene start time.

[0143] The angular velocity is provided in real time by a wheel speed sensor mounted on the motor shaft driving the walking module. The speed is calculated by the control unit in combination with the angular velocity.

[0144] The collected data is stored in the cloud database. Reason: each cloud-controlled simulated traffic participant only communicates with the cloud, and all trajectory information is aggregated through the cloud.

[0145] The field test dynamic scene protocol CX is used to generate the third field scene information file C3; in the third field scene information file C3, there are all the necessary fields of the protocol CX, and the data in the fields need to be filled in;

[0146] All field information in the third field scene information file C3 is in a temporary absence state, and needs to be filled in by reading the above collected data;

[0147] The collected data is filled into the corresponding field of the third field scene information file C3;

[0148] Step 4A2, based on the dynamic scene protocol FX for intelligent driving simulation test, the second simulation scene information file F2 for simulation test is established;

[0149] The second simulation scene information file F2 has the same fields as the first simulation scene information file F1;

[0150] All field information in the second simulation scene information file F2 is in a temporary absence state, and needs to be filled in by reading the third field scene information file C3 and through conversion.

[0151] Step 4A3, according to the dynamic scene protocol FX for intelligent driving simulation test and the dynamic scene protocol CX for field test, the data of the third field scene information file C3 is converted to the corresponding field of the second simulation scene information file F2.

[0152] The conversion program is deployed on the cloud server, and the conversion process is shown in FIG. 6:

[0153] 1. The OBU number under the field test dynamic scene protocol CX corresponds to the name under the dynamic scene protocol FX for simulation test.

[0154] 2. The category under the field test dynamic scene protocol CX corresponds to the category under the dynamic scene protocol FX for simulation test.

[0155] 3. The longitude and latitude at time 0 under the field test dynamic scene protocol CX are converted to the position (x, y) in the initial position under the dynamic scene protocol FX for simulation test.

[0156] The WGS84 latitude and longitude coordinates of the field test dynamic scene protocol CX are switched to the UTM coordinates of the location in the simulation test dynamic scene protocol FX. The from_latlon function of the Python utm library is used to achieve this. If the latitude and longitude are [lat, lon], the Python calculation method of UTM coordinate information is [x, y, zone, band] = utm.from_latlon(lat, lon). Among them, zone is the UTM partition symbol, and band is the north-south hemisphere symbol, respectively "N" and "S".

[0157] 4. The heading angle under the field test dynamic scene protocol CX corresponds to the attitude angle h in the initialization position under the simulation test dynamic scene protocol FX, and the pitch angle p and roll angle r are set to 0.

[0158] 5. The 0 time speed v under the field test dynamic scene protocol CX corresponds to the initialization speed v under the simulation test dynamic scene protocol FX.

[0159] 6. The time t under the field test dynamic scene protocol CX corresponds to the time t under the simulation test dynamic scene protocol FX.

[0160] 7. The current latitude and longitude under the field test dynamic scene protocol CX is converted into the current position (x, y) in the trajectory information list under the simulation test dynamic scene protocol FX.

[0161] The from_latlon function of the Python utm library is used to achieve this. If the latitude and longitude are [lat, lon], the Python calculation method of UTM coordinate information is [x, y, zone, band] = utm.from_latlon(lat, lon). Among them, zone is the UTM partition symbol, and band is the north-south hemisphere symbol, respectively "N" and "S".

[0162] 8. The current speed under the field test dynamic scene protocol CX corresponds to the current speed v under the simulation test dynamic scene protocol FX.

[0163] 9. Unit conversion.

[0164] The unit conversion is as follows:

[0165] According to the requirements of the OpenScenario protocol, the units need to be converted in the conversion process from the field test dynamic scene protocol CX to the simulation test dynamic scene protocol FX, and some values also need to be restored from specific resolution unsigned integers to floating-point numbers, as follows.

[0166] The time unit in the field test dynamic scene protocol CX is millisecond, and is represented by Int64 signed integer. For example, the time 1500 in the field test dynamic scene protocol CX is converted into 1.5s in the simulation test dynamic scene protocol FX.

[0167] The speed resolution in the field test dynamic scene protocol CX is 0.02m / s, and is represented by Uint32 unsigned integer, with the effective range of [-8191, 8191]. For example, the speed 50 in the field test dynamic scene protocol CX is converted into 1m / s in the simulation test dynamic scene protocol FX.

[0168] The heading angle resolution in the field test dynamic scene protocol CX is 0.0125 degree, and is converted into Uint16 unsigned integer, with the effective range of [0, 28799]. For example, the heading angle 14400 in the field test dynamic scene protocol CX is converted into the attitude angle h of 180° in the simulation test dynamic scene protocol FX.

[0169] Step 4A4, supplement data to the fields not filled in the second simulation scene information file F2.

[0170] The fields that need to be supplemented include: rendering models of various traffic participants, OpenDrive maps, and OpenSceneGraph scene models.

[0171] Step 4B, load the second simulation scene information file F2 storing the data in step 4A4 into a simulation software supporting the OpenScenario protocol, and perform virtual simulation test.

[0172] The simulation software supporting the OpenScenario protocol includes Carla, Carmaker, VTD, 51SimOne, etc.

[0173] Steps 4A-4B form a cloud control simulation traffic participant field test scene regeneration method.

[0174] In summary, the embodiment can reduce the cost and time of field test by converting the virtually generated test scene into a scene deployed in the field and verifying the simulation visualization in the virtual environment to find and optimize problems in advance and reduce unnecessary repeated tests in the field test.

[0175] The program deployment and verification in the intelligent network connection test field have been completed.

[0176] The OpenScenario standard simulation scene can be converted into the field test scene specified by the intelligent network connection test field through a conversion program.

[0177] Through the cloud running scene, the cloud control simulation traffic participants obtain trajectory information from the cloud and perform tracking actions, which can realize real running of intelligent driving site tests containing multiple traffic participants.

[0178] The cloud control simulation traffic participants upload motion trajectories after accessing the cloud communication, and can reproduce the simulation scene of the OpenScenario standard based on the motion trajectories.

[0179] The above is only the preferred embodiment of the present application, and does not have any limiting effect on the present application. Any person skilled in the art can make any form of equivalent replacement or modification of the technical solutions and technical contents disclosed by the present application without departing from the scope of the technical solutions of the present application, which still belongs to the protection scope of the present application.

Claims

1. A closed-loop method for dynamic scene simulation and field testing for autonomous driving testing, characterized by: The following steps are involved: Step 1: Based on the dynamic scenario protocol FX for intelligent driving simulation testing, a simulation scenario information file F1 for simulation testing is created, and a simulation test is performed; The first simulation scene information file F1 is created by a dynamic scene editing tool, and the first simulation scene information file F1 includes road network information and traffic participant information; Road network information includes: OpenDrive map, OpenSceneGraph scene model; Traffic participant information includes: name, category, traffic participant rendering model, initialization position, initialization speed, trajectory information list, speed change; Step 2: Based on the field test dynamic scenario protocol CX, create a field test-oriented field scenario information file C1 and perform field testing; Among them, the scene information file C1 of site 1 is created through the dynamic scene editing tool; The scene information file C1 of site 1 contains the following information: scene name, scene number, scene description, number of traffic participants, and traffic participant information; Traffic participant information includes: OBU number, number of track points, and track information list; Step 3: Convert the simulation test dynamic scene into a field test dynamic scene, and then conduct field testing, which specifically includes the following steps: Step 3A: Convert the first simulation scene information file F1 into the second site scene information file C2: Step 3A1: Based on the field test dynamic scenario protocol CX, create a second field test scenario information file C2 for the field test; The fields of the second site scene information file C2 are the same as those of the first site scene information file C1; All fields in the scene information file C2 of site 2 are temporarily unavailable; Step 3A2: Perform data conversion based on the dynamic scene protocol FX of the simulation test and the dynamic scene protocol CX of the field test, and convert the data into corresponding fields of the second field scene information file C2; Step 3A3: Supplement data to the unfilled fields in the second site scene information file C2; Among them, the fields that require additional information include: scene name, scene number, scene description, number of cloud-controlled simulated traffic participants, and number of trajectory points for each cloud-controlled simulated traffic participant; Step 3B: Store the second site scene information file C2 output in step 3A3 in the cloud database as a JSON file; Step 3C: Add the OBU numbers corresponding to the respective communication units for the cloud-controlled simulated traffic participants to ensure the correct distribution of trajectory information; Step 3D: Conduct field testing: The cloud server reads the information of each traffic participant in the scene information file C2 of site No. 2 in the cloud database, and transmits the trajectory information list to the communication unit of the cloud-controlled simulated traffic participant via 5G according to the OBU number; The cloud-controlled simulated traffic participants include human models and vehicle models placed in the actual test site; all cloud-controlled simulated traffic participants can move in the actual test site and are equipped with a communication unit OBU.

2. The closed-loop method for dynamic scene simulation and field testing for autonomous driving testing according to claim 1, characterized in that: After step 3, the following steps are also included: Step 4: Convert the field test dynamic scene into a simulation test dynamic scene, and then conduct a simulation test. This specifically includes the following steps: Step 4A: Convert the field test dynamic scene into the simulation test dynamic scene: In step 4A1, the communication unit of each cloud-controlled simulated traffic participant uploads the data of the dynamic scene operation of the field test in step 3D via 5G communication, including GPS positioning coordinate information, heading angle information, category, and OBU number. The cloud receives the test data of each cloud-controlled simulated traffic participant. The site test dynamic scene protocol CX is used to generate the site scene information file C3 of the third site. In the site scene information file C3 of the third site, there are fields, and the data in the fields is to be filled. Fill the collected test data of each cloud-controlled simulated traffic participant into the corresponding fields of the No. 3 site scene information file C3; Step 4A2: Based on the dynamic scenario protocol FX for intelligent driving simulation testing, a second simulation scenario information file F2 for simulation testing is created; The fields of the second simulation scenario information file F2 are the same as those of the first simulation scenario information file F1; All fields in the second simulation scenario information file F2 are temporarily unavailable. Step 4A3: According to the simulation test dynamic scene protocol FX and the field test dynamic scene protocol CX, the data of the third field scene information file C3 is converted into the corresponding fields of the second simulation scene information file F2; Step 4A4: Supplement data to the unfilled fields in the second simulation scenario information file F2; Among them, the fields that require additional information include: rendering models of each traffic participant, OpenDrive map, and OpenSceneGraph scene model; Step 4B: Load the second simulation scenario information file F2 containing the data stored in step 4A4 into simulation software supporting the OpenScenario protocol to perform a virtual simulation test.

3. The closed-loop method for dynamic scene simulation and field testing for autonomous driving testing according to claim 2, characterized in that: Step 3A2 specifically includes the following conversion steps: The name is converted to OBU number, the coordinates in the initialization position are converted to the longitude and latitude at time 0, the attitude angle h in the initialization position is converted to the heading angle, the initialization speed is converted to the speed at time 0, the time t is converted, and the current speed is converted.

4. The closed-loop method for dynamic scene simulation and field testing for autonomous driving testing according to claim 2, characterized in that: Step 4A2 specifically includes the following conversion steps: The OBU number is converted into a name, the longitude and latitude at time 0 are converted into the coordinates of the initialization position, the heading angle at time 0 is converted into the attitude angle h at the initialization position, the speed at time 0 is converted into the initialization speed, the time t is converted, and the current speed is converted.

Images