Automatic driving simulation test scene data generation method and simulation test method

Abstract

The invention provides an automatic driving simulation test scene data generation method and a simulation test method, and relates to the technical field of automatic driving, and the method comprises the steps: obtaining traffic environment data collected by a sensor, and extracting a plurality of key variables in the traffic environment data; obtaining a specified traffic participant type in the simulation test scene and a first key variable type corresponding to the specified traffic participant type, and determining a generalized traffic participant type and a second key variable type corresponding to the generalized traffic participant type in the traffic participant types corresponding to the key variables; and recombining the key variables to obtain a first combination result corresponding to the first key variable type and a second combination result corresponding to the second key variable type, and taking the first combination result and the second combination result as simulation test scene data. According to the method and the device, a simulation scene with higher representativeness and complexity can be generated, and the coverage degree and the effect of a simulation test are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and in particular to an autonomous driving simulation test scenario data generation method and a simulation test method. Background Art

[0002] Autonomous driving simulation testing mainly uses mathematical modeling to digitally restore the application scenarios of autonomous driving, establish a system model that is as close to the real world as possible, and achieve the purpose of testing and verifying the autonomous driving system and algorithm by directly performing simulation testing through software without the need for a real car. With the development of autonomous driving technology, extensive and in-depth testing and verification of autonomous driving technology has become the key to ensuring its safety and reliability. Traditional simulation test scenarios often rely on preset rules or script generation, which limits their diversity and complexity and makes it difficult to cover complex interactive behaviors in real traffic. Summary of the invention

[0003] The embodiments of the present application provide a method for generating scene data for an autonomous driving simulation test and a simulation test method to solve one or more of the above-mentioned technical problems.

[0004] In a first aspect, an embodiment of the present application provides a method for generating scene data for an autonomous driving simulation test, comprising: obtaining traffic environment data collected by a sensor, and extracting a plurality of key variables from the traffic environment data; wherein the key variables are used to describe the characteristics of traffic participants contained in the simulation scene; obtaining a specified traffic participant type in the simulation test scene and a first key variable type corresponding to the specified traffic participant type, determining a generalized traffic participant type and a second key variable type corresponding to the generalized traffic participant type among the traffic participant types corresponding to the key variables; recombining the key variables to obtain a first combination result corresponding to the first key variable type and a second combination result corresponding to the second key variable type, and using the first combination result and the second combination result as simulation test scene data.

[0005] In a second aspect, an embodiment of the present application provides a simulation testing method, comprising: obtaining simulation test scenario data; wherein the simulation test scenario data is generated according to the above-mentioned method; generating a simulation scenario based on the simulation test scenario data, and using the simulation scenario to perform a simulation test of the autonomous driving system.

[0006] In a third aspect, an embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory, wherein the processor implements any of the above methods when executing the computer program.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of the above is implemented.

[0008] Compared with the related art, this application has the following advantages:

[0009] The present application obtains the traffic environment data collected by the sensor, so as to obtain the environmental data when the vehicle is running in the real road environment, extracts a variety of key variables in the traffic environment data, and then realizes the disassembly of the real data, wherein the key variables are used to describe the characteristics of the traffic participants contained in the simulation scene; then, obtains the specified traffic participant type in the simulation test scene and the first key variable type corresponding to the specified traffic participant type, so that the obtained simulation test scene data can meet the actual scene requirements, determines the generalized traffic participant type and the second key variable type corresponding to the generalized traffic participant type in the traffic participant type corresponding to the key variable, so as to improve the diversity of the simulation test scene data, and finally, recombines the key variables to obtain the first combination result corresponding to the first key variable type and the second combination result corresponding to the second key variable type, and uses the first combination result and the second combination result as the simulation test scene data, which improves the complexity of the simulation test scene data. According to the embodiment of the present application, a more representative and complex simulation scene can be generated, which significantly improves the coverage and effect of the simulation test.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar parts or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments according to the present application and should not be regarded as limiting the scope of the present application.

[0012] Figure 1 A flowchart of a method for generating scene data for an autonomous driving simulation test provided in an embodiment of the present application is shown;

[0013] Figure 2 A flowchart of a simulation test method provided in an embodiment of the present application is shown;

[0014] Figure 3A structural block diagram of an automatic driving simulation test scenario data generating device provided in an embodiment of the present application is shown;

[0015] Figure 4 A structural block diagram of a simulation test device provided in an embodiment of the present application is shown;

[0016] Figure 5 A block diagram of an electronic device used to implement an embodiment of the present application is shown. DETAILED DESCRIPTION

[0017] In the following, only some exemplary embodiments are briefly described. As those skilled in the art will appreciate, the described embodiments may be modified in various ways without departing from the concept or scope of the present application. Therefore, the drawings and descriptions are considered to be exemplary in nature and not restrictive.

[0018] To facilitate understanding of the technical solutions of the embodiments of the present application, the following describes the related technologies of the embodiments of the present application. The following related technologies can be combined with the technical solutions of the embodiments of the present application as optional solutions, and they all belong to the protection scope of the embodiments of the present application.

[0019] The present application provides a method for generating data for an autonomous driving simulation test scenario and a simulation test method. The method is based on real vehicle sensor data, disassembles the key variables of the interaction scenario, and generates a simulation test scenario by recombining these variables and setting corresponding rules.

[0020] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The several specific embodiments listed can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0021] The present application embodiment provides a method for generating scene data for an autonomous driving simulation test. Figure 1 The flowchart of a method for generating scene data for an autonomous driving simulation test according to an embodiment of the present application is shown. The method may include:

[0022] Step S101, acquiring traffic environment data collected by sensors, and extracting a variety of key variables from the traffic environment data; wherein the key variables are used to describe the characteristics of traffic participants included in the simulation scenario.

[0023] In an embodiment of the present application, the sensor is a sensor carried by a real vehicle, which may include but is not limited to: a camera, a millimeter wave radar, a laser radar, and a GPS (Global Positioning System). Through the sensor, a large amount of traffic environment data is collected, and these traffic environment data include various state information of the vehicle when it is running in a real road environment and interactive behaviors with other traffic participants. Among them, the state information included in the traffic environment data can be used to characterize the position and speed conditions of the vehicle, and may specifically include information such as being in a stationary state, a moving state, and / or the positional relationship between the vehicle and the lane line. The interactive data included in the traffic environment data can be used to describe the relationship between the vehicle and other traffic participants. Among them, traffic participants are all individuals and entities that use roads and related facilities in the traffic system. Traffic participants can be in a stationary state, such as a tree on the side of the road, or in a moving state, such as a moving vehicle.

[0024] It should be noted that the scope of traffic participants is very broad, covering various types of road users. In the autonomous driving scenario, traffic participants cover various elements in the road traffic system, including: vehicles, other autonomous vehicles: other autonomous vehicles that interact with autonomous vehicles. Human-driven vehicles, traditional vehicles driven by humans, including cars, trucks, etc. Pedestrians: pedestrians, including stationary pedestrians and pedestrians in various postures such as squatting, lying, half-squatting, running pedestrians, and also accompanied by the generalization of various forms of pedestrians such as pedestrians holding umbrellas, people on scooters, people in wheelchairs, pulling suitcases, pushing baby strollers, etc. and crowds. Irregular behavior in traffic scenes and sudden entry into lanes are major challenges to the perception and response of autonomous driving systems. Cyclists: Bicycles, electric bicycles / motorcycles, etc., as a common traffic participant, have speeds and behavior patterns between pedestrians and vehicles. Non-motorized vehicles, often referring to tricycles or carts, electric scooters / transportation tools, are becoming more and more common in cities, usually mixed with pedestrians and sharing the road with vehicles. Special vehicles: such as public transport vehicles, buses and trams, whose running trajectories and stop behaviors affect traffic flow; emergency vehicles, such as ambulances, fire trucks, police cars, etc., require the autonomous driving system to make timely decisions such as giving way; freight vehicles, such as trucks and logistics vehicles, especially for special scenarios such as autonomous truck platoons. Road infrastructure: traffic lights, changes in lights affect the movement and stopping of vehicles. Road signs and markings: including speed limit signs, turn signs, lane markings, etc. Obstacles, such as roadblocks, construction areas, etc., may be temporary or permanent traffic participants. Dynamic environmental factors, such as animals, in some areas, the autonomous driving system needs to consider the situation where wild animals or pets suddenly cross the road. Floating or rolling objects: such as leaves blown by the wind, trash cans, etc., which have a dynamic impact on the vehicle path. Each traffic participant has different behavioral characteristics and challenges to the autonomous driving system. Therefore, the generated simulation scenarios need to take into account the diverse traffic participants and their interactions with each other.

[0025] In the embodiment of the present application, the traffic environment data includes multiple key variables, and each key variable in the traffic environment data is extracted separately. The key variables are used to describe the characteristics of the traffic participants included in the simulation scene, wherein the characteristics of the traffic participants can be the status of the traffic participants and / or the relationship between the traffic participants and other traffic participants.

[0026] In this step, real traffic environment data can be collected through sensors, and multiple key variables in the traffic environment data can be extracted so that they can be recombined based on these key variables to obtain more realistic combination results. Using real combination results can restore more realistic scenes than using combination results generated by algorithms. For example, for a scene where pedestrians cross the road, the speed of pedestrians in a simulation scene is configured using the results generated by an algorithm. It is possible to set a speed value for each time period in a continuous time period, but in reality pedestrians rarely walk at the set speed value. Using real combination results essentially configures the actual walking speed of pedestrians crossing the road in historical data into the simulation scene, so it is more in line with the real scene.

[0027] Step S102, obtaining a specified traffic participant type in a simulation test scenario and a first key variable type corresponding to the specified traffic participant type, and determining a generalized traffic participant type and a second key variable type corresponding to the generalized traffic participant type among the traffic participant types corresponding to the key variables.

[0028] In an embodiment of the present application, a simulation test scenario is a test scenario that a user wants to construct. The specified traffic participant type is a traffic participant type given by a user, for example, a pedestrian, a vehicle, or a traffic light, etc. The key variable type corresponding to the specified traffic participant type is used as the first key variable type. For example, for a pedestrian, the first key variable type may include: speed, acceleration, posture, and orientation, etc. The generalized traffic participant type is a traffic participant type obtained by calculation. The generalized traffic participant type may be any one or more of the traffic participant types corresponding to the key variables. The key variable type corresponding to the generalized traffic participant type is used as the second key variable type.

[0029] Step S103, recombining the key variables to obtain a first combination result corresponding to the first key variable type and a second combination result corresponding to the second key variable type, and using the first combination result and the second combination result as simulation test scenario data.

[0030] In an embodiment of the present application, the key variables are recombined, and the combination result obtained from the key variables corresponding to the first key variable type is used as the first combination result, and the combination result obtained from the key variables corresponding to the second key variable type is used as the second combination result.

[0031] By recombining the key variables extracted above, a variety of simulation scenarios can be generated. These recombinations can systematically cover more possible driving scenarios and improve the comprehensiveness of the test. The formula is as follows: S = {V1, V2, ..., Vn}, where S represents the simulation scenario combination, and Vi represents the recombinant result obtained after recombining the key variables extracted from the real data, which can be the first recombinant result or the second recombinant result. The combination of these key variables can be selected according to actual needs to generate a variety of scenarios.

[0032] The present application obtains the traffic environment data collected by the sensor, so as to obtain the environmental data when the vehicle is running in the real road environment, extracts a variety of key variables in the traffic environment data, and then realizes the disassembly of the real data, wherein the key variables are used to describe the characteristics of the traffic participants contained in the simulation scene; then, obtains the specified traffic participant type in the simulation test scene and the first key variable type corresponding to the specified traffic participant type, so that the obtained simulation test scene data can meet the actual scene requirements, determines the generalized traffic participant type and the second key variable type corresponding to the generalized traffic participant type in the traffic participant type corresponding to the key variable, so as to improve the diversity of the simulation test scene data, and finally, recombines the key variables to obtain the first combination result corresponding to the first key variable type and the second combination result corresponding to the second key variable type, and uses the first combination result and the second combination result as the simulation test scene data, which improves the complexity of the simulation test scene data. According to the embodiment of the present application, a more representative and complex simulation scene can be generated, which significantly improves the coverage and effect of the simulation test.

[0033] In a possible implementation, extracting multiple key variables from the traffic environment data includes at least: extracting one or more of the following types of variables from the traffic environment data: variables for describing posture and motion, variables for describing interaction and collision, variables for describing path and position, and variables for describing environment. Specifically, the variables for describing posture and motion include at least one or more of the following variables: heading angle, heading angle change rate, speed, acceleration, and jerk; the variables for describing interaction and collision include at least one or more of the following variables: collision time, collision angle, relative distance, relative speed, and following distance; the variables for describing path and position include at least one or more of the following variables: path deviation, position, and road curvature; the variables for describing environment include at least one or more of the following variables: relative position, sensor visibility, and environmental complexity.

[0034] In this possible implementation, during the data analysis process, key variables including but not limited to the following are identified and extracted for describing and generating interactive scenarios: variables for describing posture and motion, including but not limited to: heading angle, heading angle change rate, speed, acceleration and jerk. Among them, the heading angle (Yaw) of the vehicle describes the angle of the vehicle relative to the north direction. The heading angle change rate (Yaw Rate) describes the turning speed of the vehicle. Velocity (Velocity): The instantaneous driving speed of the vehicle. Acceleration (Acceleration): The rate of change of vehicle speed, which describes the acceleration or deceleration behavior of the vehicle. Jerk (Jerk): The rate of change of acceleration, which reflects the acceleration comfort of the vehicle.

[0035] Variables used to describe interactions and collisions include, but are not limited to, collision time, collision angle, relative distance, relative speed, and following distance. Among them, the collision time (TTC, Time to Collision) indicates the estimated time when the vehicle collides with surrounding obstacles while maintaining its current speed and direction. Collision Angle: The angle between the two vehicles when the collision occurs. Relative Distance: The relative distance to other traffic participants. Relative Velocity: The speed difference with other traffic participants. Following distance, also known as safety distance (Safety Distance): The minimum safe following distance calculated based on speed and vehicle type.

[0036] Variables used to describe the path and position, including but not limited to: path deviation, position, and road curvature. Path Deviation: The deviation between the vehicle's current path and the planned path. Position: The vehicle's GPS coordinates or location on the map. Road Curvature: Describes the curvature of the current road, which affects the vehicle's steering requirements.

[0037] Variables used to describe the environment, including but not limited to: relative position, sensor visibility, and environmental complexity. Relative Position: The relative position to other traffic participants, usually expressed in a local coordinate system. Visibility: Reflects the impact of factors such as weather and light on vehicle sensors. Environmental Complexity: The density and complexity of environmental elements such as surrounding traffic participants, road signs, and buildings.

[0038] In an embodiment of the present application, a scenario library is established by collecting massive amounts of sensor data, wherein the scenario library includes a variety of key variables so as to be recombined based on the elements in the scenario library.

[0039] It should be noted that the autonomous driving scenario library contains a series of key elements from static road elements to dynamic traffic participants, environmental conditions, sensor simulation, etc. These elements together constitute a comprehensive virtual traffic test environment, which can test and verify the behavior and performance of the autonomous driving system under various complex conditions in a highly controllable manner. This is crucial for the development, optimization and certification of autonomous driving systems. The key variables in the scenario library can be used to describe, but are not limited to, the following: 1. Scenario templates: including road layouts: such as urban roads, highways, rural roads, roundabouts, intersections, etc. The road layout defines the basic structure of the scene, such as the number of lanes, lane width, forks, intersection design, etc. It also includes infrastructure: including traffic signs, traffic lights, roadside signs, roadblocks, guardrails and other facilities that affect driving decisions. It also includes static elements: such as parked vehicles on the roadside, buildings, green belts, road signs, etc. Static elements have an impact on the perception system and can also be used to test the performance of the perception system. 2. Traffic participant models, vehicles: including autonomous driving vehicles, ordinary human-driven vehicles, trucks, buses, etc. Each vehicle may have different physical characteristics such as acceleration, deceleration, maximum speed, turning radius, etc. Pedestrians: simulate pedestrians crossing the road, waiting on the side of the road, and suddenly entering the lane. Pedestrian behavior patterns may change randomly and are difficult to predict. Non-motorized vehicles: such as bicycles, electric vehicles, motorcycles, etc., have different speeds and behavior patterns, and often have complex interactions with vehicles and pedestrians. Special vehicles: emergency vehicles such as ambulances, fire trucks, and police cars, whose special driving rules require the autonomous driving system to make specific yielding or avoidance operations. 3. Environmental conditions, including: Weather conditions: such as sunny, cloudy, rainy, snowy, foggy, etc. Different weather conditions will affect the perception ability of sensors (such as rain affecting cameras, fog and haze affecting the ranging of lidar) and the driving conditions of the road (such as slippery roads). Lighting conditions: The scene library can simulate different lighting environments, such as daytime, dusk, night, etc., which affect the working performance of visual sensors. Road conditions: simulate different road conditions, such as dry, slippery, icy, snowy, potholes, ramps, etc. Changes in road conditions will directly affect the vehicle's driving performance and the grip of the tires. 4. Dynamic events, including: emergencies: such as pedestrians suddenly crossing the road, vehicles changing lanes urgently, animals suddenly entering the road, etc. These dynamic events place high demands on the decision-making and response capabilities of the autonomous driving system. Traffic accidents: simulate traffic accidents or emergency stops ahead to examine how autonomous driving vehicles respond to such incidents. Vehicle failures: including sudden vehicle breakdowns, emergency brake failures, tire bursts, and other emergencies that may affect driving safety. 5. Traffic flow model. Traffic flow: The traffic flow in the scenario library can be set to different conditions of low, medium, and high density to test the performance of the autonomous driving system under different traffic congestion conditions.Behavior model: The behavior model of different vehicles and traffic participants can be rule-driven (such as keeping a safe distance and avoiding pedestrians) or data-driven (behavior model trained by real data). Interaction complexity: includes interaction with other vehicles, pedestrians, and cyclists. For example, vehicles giving way and changing lanes at complex intersections. 6. Sensor simulation, including: Perception sensor: simulates various sensors on the vehicle, including cameras, lidar, radar, ultrasonic sensors, GPS, etc. Sensor simulation provides input data for the autonomous driving system in different scenarios. Sensor accuracy and error: By simulating the sensor's perception accuracy, noise, occlusion, signal delay and other issues, test how the autonomous driving system handles the uncertainty and errors of perception data. 7. Behavior and decision algorithms, including: Autonomous driving decision model: used to generate decision-making behaviors of autonomous driving vehicles in different traffic scenarios, including path planning, vehicle speed control, obstacle avoidance strategies, etc. Risk assessment: The scenario library can provide different driving risk assessment mechanisms, such as too close distance, too fast speed, risk level (calculated by TTC, Jerk and other indicators), etc., to test how autonomous driving vehicles identify and avoid risks. 8. Interaction events, including: Lane change: simulates lane change behavior between autonomous vehicles and other vehicles, such as autonomous vehicles changing lanes, avoiding, and inserting. Interaction at intersections: simulates the interaction of vehicles at complex intersections, including right turns, left turns, U-turns, and avoidance of pedestrians and non-motorized vehicles. Emergency avoidance: tests the emergency avoidance response of the autonomous driving system to emergencies (such as emergency braking of the vehicle ahead or sudden entry of a side vehicle into the lane).

[0040] In one possible implementation, obtaining a specified traffic participant type in a simulation test scenario and a first key variable type corresponding to the specified traffic participant type includes: obtaining a first selection instruction for the specified traffic participant type through an interactive interface, and determining the specified traffic participant type from a plurality of preset traffic participant types according to the first selection instruction; wherein the interactive interface is used to display the plurality of preset traffic participant types and the plurality of preset key variable types; obtaining a second selection instruction for the first key variable type corresponding to the specified traffic participant type through the interactive interface, and determining the first key variable type from the plurality of preset key variable types according to the second selection instruction.

[0041] In this possible implementation, the interactive interface is an interface for a user to interact with a computer system or device. The interactive interface is used to display the preset multiple types of traffic participants and the preset multiple types of key variables. It should be noted that each preset type of traffic participant corresponds to one or more types of key variables. The one or more types of key variables corresponding to the same type of traffic participant meet the preset rules. The preset rules include but are not limited to: speed distance rules: used to simulate scenes such as following a vehicle at different distances, sudden acceleration, and sudden deceleration. Turning path rules: generate scenes when the vehicle turns or changes lanes. TTC collision rules: simulate scenes when a collision is about to occur, and detect the emergency avoidance ability of the system. Dynamic behavior rules: simulate scenes such as the sudden appearance of pedestrians and non-motor vehicles or the sudden braking of vehicles. Violation of traffic regulations: simulate scenes where vehicles, pedestrians, and non-motor vehicles do not drive according to traffic regulations. The simulation scenarios generated by these rules can cover a variety of possible traffic scenarios, especially those that are relatively rare on actual roads but may have high risks. For example, if the traffic participant type is a tree, the corresponding key variable type may not include key variables such as speed and acceleration. In the case of including key variables such as speed and acceleration, the values ​​of these key variables should be 0.

[0042] In an embodiment of the present application, based on the key variables in the scenario library, the elements that the user is concerned about can be retrieved to obtain the first key variable type.

[0043] In a possible implementation, a generalized traffic participant type and a second key variable type corresponding to the generalized traffic participant type are determined among the traffic participant types corresponding to the key variables, including: randomly selecting a generalized traffic participant type among the traffic participant types corresponding to the key variables; and randomly selecting a second key variable type among the key variable types corresponding to the randomly selected generalized traffic participant type.

[0044] In this possible implementation, a generalized traffic participant type is randomly selected from the traffic participant types corresponding to the key variables; and a second key variable type is randomly selected from the key variable types corresponding to the randomly selected generalized traffic participant type, which can improve the diversity and complexity of the simulation scenarios.

[0045] In the embodiment of the present application, based on the various key variables in the scenario library, elements may also be randomly determined to obtain a second key variable type.

[0046] In a possible implementation, the key variables are recombined, including: recombining the key variables according to one or more of the following methods: linear combination, nonlinear combination, sorting according to time series, statistical analysis of the key variables, probability sampling based on statistical analysis results, recombining the key variables based on probability sampling results, recombining key variables of different dimensions, and recombining the key variables based on a plurality of pre-set threshold ranges and the correlation between the plurality of threshold ranges.

[0047] In this possible implementation, linear combination, i.e., weighted summation of multiple key variables can be performed to obtain a recombined result. Non-linear combination, including but not limited to the following processing of key variables: multiplication calculation, exponential calculation, logarithmic calculation, trigonometric function calculation, etc. Sorting by time series can be performed in ascending or descending order.

[0048] In the specific implementation, the key variables are recombined, including: generating simulation scenarios by linearly combining different variables. For example, for the two variables of speed and acceleration, multiple combination scenarios can be constructed. For example: s=a1v1+a2v2... Nonlinear combination allows more complex relationships between variables and can capture more complex interactive behaviors. It can be any nonlinear function, such as a polynomial function, a logarithmic function, an exponential function, etc.

[0049] The key variables are recombined, including: through nonlinear combination, complex scenarios can be generated to simulate more unpredictable dynamic driving situations. The dynamic behavior in traffic scenes is often closely related to time, so time series combination is a very important recombining method. By combining key variables at different time points, scenarios where variables change over time can be simulated. For example, by combining variables such as the speed, acceleration, Yaw, etc. of a vehicle at different times, dynamic scenarios of the vehicle during acceleration and steering can be generated. In this way, complex scenarios such as dynamic changes before and after a traffic accident, emergency avoidance and braking of vehicles, etc. can be simulated.

[0050] The key variables are recombined, including: by statistically analyzing the key variables in the traffic data, the distribution characteristics of the variables (such as normal distribution, Poisson distribution, etc.) can be obtained. Based on these distributions, a scene combination that conforms to statistical laws can be generated through a probability sampling method. For example, possible collision scenes are generated based on the distribution characteristics of TTC (Time to Collision), and a simulation scene that reflects the laws of actual traffic accidents can be generated through probability-based recombining.

[0051] The key variables are recombined, including the key variables of different dimensions, that is, by combining the key variables of multiple dimensions, a highly complex scene can be generated. For example, by combining variables such as speed, acceleration, Yaw, YawRate, relative distance, TTC, etc., a multi-dimensional interactive scene is generated: the multi-dimensional combination scene can simulate various complex traffic conditions such as multi-vehicle intersections, emergency avoidance, complex intersections, etc. In the embodiment of the present application, dimensions are used to describe different types of key variables, such as speed, acceleration, Yaw, Yaw Rate, relative distance, TTC, which belong to key variables of different dimensions.

[0052] Recombining the key variables may also include: recombining the key variables based on a plurality of pre-set threshold ranges and the correlation between the plurality of threshold ranges. In an embodiment of the present application, a plurality of pre-set threshold ranges and the correlation between the plurality of threshold ranges can be used to determine different combination rules to obtain different scenarios. That is, in addition to simple variable combinations, specific scenarios can also be generated through these rules. For example: speed-distance rule: when the speed reaches a certain value, a certain safety distance is maintained. TTC-collision rule: according to different TTC values, potential collision scenarios are generated. Steering-path rule: according to the vehicle steering angle and path planning, a scenario of vehicle changing lanes or turning is generated. In an embodiment of the present application, for example, for the speed-distance rule, the threshold range that can be set includes a speed threshold range and a distance threshold range, and the correlation between the speed threshold range and the distance threshold range must also be set. Based on the set speed threshold range, distance threshold range and the correlation between the two, the key variables are recombined.

[0053] In practical applications, the above-mentioned multiple recombinations can be combined but not limited to generate more complex and diverse scenarios. For example, nonlinear combinations can be combined with time series combinations to generate scenarios where variables change over time and have complex interactions, or multi-dimensional combinations and rule-based combinations can be used to generate highly complex scenarios for specific test targets.

[0054] The implementation process of the method is described below with a specific embodiment, which specifically includes the following steps:

[0055] 1. Data collection: Through real vehicles equipped with sensors, vehicle status and interaction data are collected in a variety of different traffic environments.

[0056] 2. Data processing and variable extraction: Analyze the collected data and extract key variables that affect the interaction scenario.

[0057] 3. Variable recombination: Based on the extracted variables, different recombination strategies are designed to generate diverse simulation scenarios.

[0058] 4. Rule generation: Generate specific scenario combinations based on preset traffic rules or test requirements to form simulation test cases.

[0059] 5. Simulation testing and evaluation: Run the generated scenarios in the simulation platform to comprehensively test and evaluate the performance of the autonomous driving system.

[0060] The present application provides a method for generating scene data for autonomous driving simulation test, which can be used to generate simulation scenes by combining the recombination of key variables and rules. The present application can significantly improve the complexity and coverage of scenes. Among them, the complexity of scenes is achieved through hierarchical processing of variables and refinement of rule combinations. Coverage is achieved by generating a large number of test scenes by covering more driving scenarios, thereby improving the comprehensiveness and effectiveness of simulation tests.

[0061] The present application also provides a simulation test method, such as Figure 2 The flowchart of a simulation test method according to an embodiment of the present application is shown, and the method may include:

[0062] Step S201, obtaining simulation test scenario data; wherein the simulation test scenario data is generated according to the above-mentioned method for generating automatic driving simulation test scenario data.

[0063] In this possible implementation, any of the above-mentioned autonomous driving simulation test scene data generation methods can be used to obtain simulation test scene data.

[0064] Step S202: Generate a simulation scenario based on the simulation test scenario data, and use the simulation scenario to perform a simulation test on the autonomous driving system.

[0065] In this possible implementation, in the simulation platform, the simulation test scenario data can be used as an input to a computer to generate a simulation scenario through the computer, that is, a simulation scenario is generated according to the simulation test scenario data, and the generated scenario is run, thereby comprehensively testing and evaluating the performance of the autonomous driving system.

[0066] The present application provides a simulation test method, which obtains simulation test scene data, wherein the simulation test scene data is obtained according to the following scheme: traffic environment data collected by sensors are obtained, so that environmental data of the vehicle when running in a real road environment can be obtained, and multiple key variables in the traffic environment data are extracted, thereby realizing the disassembly of the real data, wherein the key variables are used to describe the characteristics of the traffic participants contained in the simulation scene; then, the specified traffic participant type in the simulation test scene and the first key variable type corresponding to the specified traffic participant type are obtained, so that the obtained simulation test scene data can meet the actual scene requirements, and the generalized traffic participant type and the second key variable type corresponding to the generalized traffic participant type are determined in the traffic participant type corresponding to the key variable, so as to improve the diversity of the simulation test scene data, and finally, the key variables are recombined to obtain the first combination result corresponding to the first key variable type and the second combination result corresponding to the second key variable type, and the first combination result and the second combination result are used as the simulation test scene data, thereby improving the complexity of the simulation test scene data. According to the embodiment of the present application, a more representative and complex simulation scene can be generated, which significantly improves the coverage and effect of the simulation test.

[0067] Corresponding to the application scenario and method of the method for generating scene data for an autonomous driving simulation test provided in the embodiment of the present application, the embodiment of the present application also provides a device for generating scene data for an autonomous driving simulation test. Figure 3 The figure is a structural block diagram of a device for generating scene data for an autonomous driving simulation test according to an embodiment of the present application, and the device may include:

[0068] The extraction module 301 is used to obtain the traffic environment data collected by the sensor and extract multiple key variables in the traffic environment data; wherein the key variables are used to describe the characteristics of the traffic participants included in the simulation scene; the type module 302 is used to obtain the specified traffic participant type in the simulation test scene and the first key variable type corresponding to the specified traffic participant type, determine the generalized traffic participant type and the second key variable type corresponding to the generalized traffic participant type in the traffic participant type corresponding to the key variable; the reorganization module 303 is used to recombine the key variables to obtain a first combination result corresponding to the first key variable type and a second combination result corresponding to the second key variable type, and use the first combination result and the second combination result as the simulation test scene data.

[0069] The present application obtains the traffic environment data collected by the sensor, so as to obtain the environmental data when the vehicle is running in the real road environment, extracts a variety of key variables in the traffic environment data, and then realizes the disassembly of the real data, wherein the key variables are used to describe the characteristics of the traffic participants contained in the simulation scene; then, obtains the specified traffic participant type in the simulation test scene and the first key variable type corresponding to the specified traffic participant type, so that the obtained simulation test scene data can meet the actual scene requirements, determines the generalized traffic participant type and the second key variable type corresponding to the generalized traffic participant type in the traffic participant type corresponding to the key variable, so as to improve the diversity of the simulation test scene data, and finally, recombines the key variables to obtain the first combination result corresponding to the first key variable type and the second combination result corresponding to the second key variable type, and uses the first combination result and the second combination result as the simulation test scene data, which improves the complexity of the simulation test scene data. According to the embodiment of the present application, a more representative and complex simulation scene can be generated, which significantly improves the coverage and effect of the simulation test.

[0070] In one possible implementation, extracting multiple key variables from the traffic environment data includes at least: extracting one or more of the following types of variables from the traffic environment data: variables for describing posture and movement, variables for describing interaction and collision, variables for describing path and position, and variables for describing the environment.

[0071] In one possible implementation, obtaining a specified traffic participant type in a simulation test scenario and a first key variable type corresponding to the specified traffic participant type includes: obtaining a first selection instruction for the specified traffic participant type through an interactive interface, and determining the specified traffic participant type from a plurality of preset traffic participant types according to the first selection instruction; wherein the interactive interface is used to display the plurality of preset traffic participant types and the plurality of preset key variable types; obtaining a second selection instruction for the first key variable type corresponding to the specified traffic participant type through the interactive interface, and determining the first key variable type from the plurality of preset key variable types according to the second selection instruction.

[0072] In a possible implementation, a generalized traffic participant type and a second key variable type corresponding to the generalized traffic participant type are determined among the traffic participant types corresponding to the key variables, including: randomly selecting a generalized traffic participant type among the traffic participant types corresponding to the key variables; and randomly selecting a second key variable type among the key variable types corresponding to the randomly selected generalized traffic participant type.

[0073] In a possible implementation, the key variables are recombined, including: recombining the key variables according to one or more of the following methods: linear combination, nonlinear combination, sorting according to time series, statistical analysis of the key variables, probability sampling based on statistical analysis results, recombining the key variables based on probability sampling results, recombining key variables of different dimensions, and recombining the key variables based on a plurality of pre-set threshold ranges and the correlation between the plurality of threshold ranges.

[0074] In a possible implementation, the variables used to describe posture and motion include at least one or more of the following variables: heading angle, heading angle change rate, speed, acceleration and jerk; the variables used to describe interaction and collision include at least one or more of the following variables: collision time, collision angle, relative distance, relative speed and following distance; the variables used to describe path and position include at least one or more of the following variables: path deviation, position and road curvature; the variables used to describe the environment include at least one or more of the following variables: relative position, sensor visibility and environmental complexity.

[0075] Corresponding to the application scenario and method of the method provided in the embodiment of the present application, the embodiment of the present application also provides a simulation test device. Figure 4 The structure block diagram of a simulation test device according to an embodiment of the present application is shown, and the device may include:

[0076] The acquisition module 401 is used to acquire simulation test scenario data; wherein the simulation test scenario data is generated by the above method; the test module 402 is used to generate a simulation scenario based on the simulation test scenario data, and use the simulation scenario to perform simulation testing of the autonomous driving system.

[0077] The present application provides a simulation test method, which obtains simulation test scene data, wherein the simulation test scene data is obtained according to the following scheme: traffic environment data collected by sensors are obtained, so that environmental data of the vehicle when running in a real road environment can be obtained, and multiple key variables in the traffic environment data are extracted, thereby realizing the disassembly of the real data, wherein the key variables are used to describe the characteristics of the traffic participants contained in the simulation scene; then, the specified traffic participant type in the simulation test scene and the first key variable type corresponding to the specified traffic participant type are obtained, so that the obtained simulation test scene data can meet the actual scene requirements, and the generalized traffic participant type and the second key variable type corresponding to the generalized traffic participant type are determined in the traffic participant type corresponding to the key variable, so as to improve the diversity of the simulation test scene data, and finally, the key variables are recombined to obtain the first combination result corresponding to the first key variable type and the second combination result corresponding to the second key variable type, and the first combination result and the second combination result are used as the simulation test scene data, thereby improving the complexity of the simulation test scene data. According to the embodiment of the present application, a more representative and complex simulation scene can be generated, which significantly improves the coverage and effect of the simulation test.

[0078] The functions of each module in each device in the embodiments of the present application can be found in the corresponding description in the above method, and have corresponding beneficial effects, which will not be repeated here.

[0079] Figure 5 FIG. 1 is a block diagram of an electronic device used to implement an embodiment of the present application. Figure 5 As shown, the electronic device includes: a memory 501 and a processor 502. The memory 501 stores a computer program that can be run on the processor 502. When the processor 502 executes the computer program, the method in the above embodiment is implemented. The number of the memory 501 and the processor 502 can be one or more.

[0080] The electronic device also includes:

[0081] The communication interface 503 is used to communicate with external devices and perform data exchange transmission.

[0082] If the memory 501, the processor 502 and the communication interface 503 are implemented independently, the memory 501, the processor 502 and the communication interface 503 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0083] Optionally, in a specific implementation, if the memory 501, the processor 502 and the communication interface 503 are integrated on a chip, the memory 501, the processor 502 and the communication interface 503 can communicate with each other through an internal interface.

[0084] An embodiment of the present application provides a computer-readable storage medium storing a computer program, which implements the method provided in the embodiment of the present application when the program is executed by a processor.

[0085] An embodiment of the present application provides a computer program product, wherein the computer program product includes a computer program, and when the computer program is executed by a processor, the method provided in the embodiment of the present application is implemented.

[0086] An embodiment of the present application also provides a chip, which includes a processor for calling and executing instructions stored in the memory from the memory, so that a communication device equipped with the chip executes the method provided by the embodiment of the present application.

[0087] An embodiment of the present application also provides a chip, including: an input interface, an output interface, a processor and a memory, wherein the input interface, the output interface, the processor and the memory are connected via an internal connection path, and the processor is used to execute the code in the memory. When the code is executed, the processor is used to execute the method provided in the embodiment of the application.

[0088] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. It is worth noting that the processor may be a processor supporting the Advanced RISC Machines (ARM) architecture.

[0089] Further, optionally, the above-mentioned memory may include a read-only memory and a random access memory. The memory may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory may include a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may include a random access memory (RAM), which is used as an external cache. By way of exemplary but not limiting description, many forms of RAM are available. For example, static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct memory bus random access memory (DR RAM).

[0090] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function according to the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium.

[0091] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0092] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0093] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0094] Any process or method described in the flow chart or otherwise described herein can be understood as a module, fragment or portion of a code representing one or more executable instructions for implementing the steps of a specific logical function or process. And the scope of the preferred embodiment of the present application includes other implementations, in which the functions may not be performed in the order shown or discussed, including in a substantially simultaneous manner or in a reverse order according to the functions involved.

[0095] The logic and / or steps described in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, which can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or used in combination with these instruction execution systems, devices or apparatuses.

[0096] It should be understood that the various parts of the present application can be implemented with hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented with software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above embodiment method can be completed by instructing the relevant hardware through a program, which can be stored in a computer-readable storage medium, and when the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0097] In addition, each functional unit in each embodiment of the present application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the above-mentioned integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The storage medium can be a read-only memory, a disk or an optical disk, etc.

[0098] The above is only an exemplary embodiment of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various changes or substitutions within the technical scope recorded in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for generating scene data for an autonomous driving simulation test, comprising: Acquire traffic environment data collected by sensors, and extract multiple key variables from the traffic environment data; wherein the key variables are used to describe the characteristics of traffic participants included in the simulation scene; Obtaining a specified traffic participant type in a simulation test scenario and a first key variable type corresponding to the specified traffic participant type, and determining a generalized traffic participant type and a second key variable type corresponding to the generalized traffic participant type from the traffic participant types corresponding to the key variables; The key variables are recombined to obtain a first combination result corresponding to the first key variable type and a second combination result corresponding to the second key variable type, and the first combination result and the second combination result are used as simulation test scenario data.

2. The method according to claim 1, wherein: Extracting multiple key variables from the traffic environment data, including at least: One or more of the following types of variables are extracted from the traffic environment data: variables for describing posture and movement, variables for describing interaction and collision, variables for describing path and position, and variables for describing environment.

3. The method according to claim 1, wherein: Obtaining a specified traffic participant type in a simulation test scenario and a first key variable type corresponding to the specified traffic participant type, including: Obtaining a first selection instruction for specifying a traffic participant type through an interactive interface, and determining the specified traffic participant type from a plurality of preset traffic participant types according to the first selection instruction; wherein the interactive interface is used to display the plurality of preset traffic participant types and the plurality of preset key variable types; A second selection instruction of the first key variable type corresponding to the designated traffic participant type is obtained through the interactive interface, and the first key variable type is determined from among the preset multiple key variable types according to the second selection instruction.

4. The method according to claim 1, wherein: Determining a generalized traffic participant type and a second key variable type corresponding to the generalized traffic participant type from the traffic participant types corresponding to the key variables includes: Randomly selecting a generalized traffic participant type from the traffic participant types corresponding to the key variables; A second key variable type is randomly selected from the key variable types corresponding to the randomly selected generalized traffic participant type.

5. The method according to claim 1, wherein: The key variables are reorganized to include: The key variables are recombined according to one or more of the following methods: linear combination, nonlinear combination, sorting according to time series, statistical analysis of the key variables, probability sampling based on statistical analysis results, recombining the key variables based on probability sampling results, recombining key variables of different dimensions, and recombining the key variables based on multiple pre-set threshold ranges and the correlation between the multiple threshold ranges.

6. The method according to claim 2, wherein: The variables used to describe posture and motion include at least one or more of the following variables: heading angle, heading angle change rate, speed, acceleration and jerk; the variables used to describe interaction and collision include at least one or more of the following variables: collision time, collision angle, relative distance, relative speed and following distance; the variables used to describe path and position include at least one or more of the following variables: path deviation, position and road curvature; the variables used to describe the environment include at least one or more of the following variables: relative position, sensor visibility and environmental complexity.

7. A simulation test method, comprising: Acquire simulation test scenario data; wherein the simulation test scenario data is generated according to the method according to any one of claims 1-6; A simulation scenario is generated based on the simulation test scenario data, and the simulation scenario is used to perform a simulation test on the autonomous driving system.

8. An electronic device comprising a memory, a processor and a computer program stored in the memory, wherein the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer program product, wherein: The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images