Vehicle function verification method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a vehicle function verification method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining operation scene data corresponding to each operation scene in at least one operation scene; based on the operation scene data corresponding to each operation scene, determining first sensing result information corresponding to a preset vehicle function of the vehicle in each operation scene; on the basis of a perception failure simulation rule, perception failure simulation processing is carried out on the first perception result information, and second perception result information corresponding to the operation scenes is obtained; and based on the second sensing result information, the preset vehicle function is verified, and a verification result is obtained. According to the embodiment of the invention, various perception failure conditions of the perception algorithm are simulated by perceiving the fault injection, the permissible condition of the preset vehicle function to the perception failure of the perception algorithm can be efficiently and quickly determined, and the optimization of the perception algorithm can be conveniently guided, so that the preset vehicle function meets the corresponding expected function safety requirement.

Description

Technical Field

[0001] The present disclosure relates to assisted driving technology, and in particular to a method, device, electronic device and storage medium for verifying vehicle functions. Background Art

[0002] In the fields of autonomous driving and assisted driving, vehicles usually have functions such as autonomous emergency braking (AEB), adaptive cruise control (ACC), navigation on autopilot (NOA), and highway pilot (HWP), etc., to provide users with corresponding functional services. The realization of these functions usually depends on the perception results of the vehicle (also called the self-vehicle) on the surrounding environment of the self-vehicle by the perception function. The surrounding environment may include dynamic objects such as other vehicles, pedestrians, and cyclists around the self-vehicle, as well as static objects such as lane lines and curbs around the self-vehicle. The acquisition of the perception results needs to be realized based on the corresponding perception algorithm of the self-vehicle perception function. The perception algorithm may include, for example, a perception algorithm based on at least one sensor data of vision, laser radar, ultrasonic radar, millimeter wave radar, etc. The performance of the perception algorithm will be affected by factors such as hardware and environmental changes during the application process, which may easily lead to the vehicle function failing to meet the corresponding expected functional safety requirements or functional performance requirements. Summary of the invention

[0003] In order to solve technical problems such as the above-mentioned perception algorithm easily causing vehicle functions to fail to meet expected functional safety requirements, the embodiments of the present disclosure provide a vehicle function verification method, device, electronic device and storage medium, which simulate various perception failure situations of the perception algorithm and verify the tolerance of the vehicle function to the performance of the perception algorithm. This can be used to guide the development and optimization of the perception algorithm so that the perception algorithm can enable the vehicle function to meet the corresponding expected functional safety requirements or functional performance requirements.

[0004] The first aspect of the present disclosure provides a method for verifying vehicle functions, including: obtaining operation scenario data corresponding to each operation scenario in at least one operation scenario; the operation scenario data includes status information of the vehicle and surrounding environment information of the vehicle; based on the operation scenario data corresponding to each operation scenario, determining first perception result information corresponding to a preset vehicle function of the vehicle in each operation scenario; based on a perception failure simulation rule, performing perception failure simulation processing on each of the first perception result information to obtain second perception result information corresponding to each of the operation scenarios; based on each of the second perception result information, verifying the preset vehicle function to obtain a verification result.

[0005] The second aspect of the present disclosure provides a vehicle function verification device, including: an acquisition module, used to obtain operation scenario data corresponding to each operation scenario in at least one operation scenario; the operation scenario data includes status information of the vehicle and surrounding environment information of the vehicle; a first processing module, used to determine first perception result information corresponding to the preset vehicle function of the vehicle in each operation scenario based on the operation scenario data corresponding to each operation scenario; a second processing module, used to perform perception failure simulation processing on each of the first perception result information based on perception failure simulation rules, and obtain second perception result information corresponding to each of the operation scenarios; a third processing module, used to verify the preset vehicle function based on each of the second perception result information, and obtain a verification result.

[0006] According to a third aspect of the present disclosure, a computer-readable storage medium is provided, wherein the storage medium stores a computer program, and the computer program is used to execute the vehicle function verification method described in any of the above embodiments of the present disclosure.

[0007] The fourth aspect of the present disclosure provides an electronic device, which includes: a processor; a memory for storing executable instructions of the processor; the processor is used to read the executable instructions from the memory and execute the instructions to implement the vehicle function verification method described in any of the above embodiments of the present disclosure.

[0008] A fifth aspect of the present disclosure provides a computer program product. When instructions in the computer program product are executed by a processor, the vehicle function verification method provided by any of the above embodiments of the present disclosure is performed.

[0009] Based on the vehicle function verification method, device, electronic device and storage medium provided by the above-mentioned embodiments of the present disclosure, operation scenario data of at least one operation scenario can be obtained, and the first perception result information corresponding to the preset vehicle function of the vehicle in each operation scenario can be determined based on the operation scenario data corresponding to each operation scenario. Then, based on the perception failure simulation rule, the perception failure simulation processing can be performed on each first perception result information to obtain the second perception result information corresponding to each operation scenario, so that the second perception result information contains at least one perception failure situation that may exist in the perception algorithm in the real scene. Based on each second perception result information, the preset vehicle function is verified to determine whether the functional performance of the preset vehicle function in the case of perception failure meets certain expected functional safety requirements or functional performance requirements, so that by simulating various perception failure situations of the perception algorithm, the allowable situation of the preset vehicle function for the perception failure of the perception algorithm can be determined, which is convenient for guiding the development and optimization of the perception algorithm, thereby helping to enable the perception algorithm to enable the preset vehicle function to meet the corresponding expected functional safety requirements or functional performance requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 This is an exemplary application scenario of the vehicle function verification method provided by the present disclosure;

[0011] Figure 2 is a flowchart of a method for verifying a vehicle function provided by an exemplary embodiment of the present disclosure;

[0012] Figure 3 is a flow chart of a method for verifying a vehicle function provided by another exemplary embodiment of the present disclosure;

[0013] Figure 4 is a flowchart of a method for verifying a vehicle function provided by yet another exemplary embodiment of the present disclosure;

[0014] Figure 5 is a schematic diagram of dividing the area around the vehicle provided by an exemplary embodiment of the present disclosure;

[0015] Figure 6 is a flowchart of a method for verifying a vehicle function provided by another exemplary embodiment of the present disclosure;

[0016] Figure 7 is a flowchart of a method for verifying a vehicle function provided by yet another exemplary embodiment of the present disclosure;

[0017] Figure 8 is a flowchart of a method for verifying a vehicle function provided by another exemplary embodiment of the present disclosure;

[0018] Fig. 9is a schematic diagram of the structure of a vehicle function verification device provided by an exemplary embodiment of the present disclosure;

[0019] Fig.10 is a schematic structural diagram of a vehicle function verification device provided by another exemplary embodiment of the present disclosure;

[0020] Fig.11 is a schematic structural diagram of a vehicle function verification device provided by yet another exemplary embodiment of the present disclosure;

[0021] Fig.12 It is a structural diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0022] To explain the present disclosure, example embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. It should be understood that the present disclosure is not limited to the example embodiments.

[0023] It should be noted that the relative arrangement of components and steps, the numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure unless specifically stated otherwise.

[0024] SUMMARY OF THE DISCLOSURE

[0025] In the process of realizing the present disclosure, the inventors found that in the fields of autonomous driving and assisted driving, vehicles usually have functions such as automatic emergency braking (AEB), adaptive cruise control (ACC), navigation on autopilot (NOA), and high-speed cruise control (HWP), etc., to provide users with corresponding functional services. The realization of these functions usually depends on the perception results of the vehicle (also called the self-vehicle) on the surrounding environment of the self-vehicle by the perception function. The surrounding environment may include dynamic objects such as other vehicles, pedestrians, and cyclists around the self-vehicle, as well as static objects such as lane lines and curbs around the self-vehicle. The acquisition of the perception results needs to be realized based on the corresponding perception algorithm of the self-vehicle perception function. The perception algorithm may include, for example, a perception algorithm based on at least one sensor data of vision, laser radar, ultrasonic radar, millimeter wave radar, etc., such as a target detection algorithm, a semantic segmentation algorithm, etc. The performance of the perception algorithm will be affected by factors such as hardware and environmental changes during the application process, which may easily lead to the vehicle function failing to meet the corresponding expected functional safety requirements or functional performance requirements.

[0026] Exemplary Overview

[0027] Figure 1 This is an exemplary application scenario of the vehicle function verification method provided by the present invention.

[0028] like Figure 1 As shown, the verification method of the vehicle function of the present disclosure can be implemented based on an in-loop simulation system including a simulation module, a perception failure simulation (or perception fault injection), a planning control software (referred to as the planning control software), and a dynamics engine. Among them, the simulation module is used to simulate the operation scene based on the operation scene data corresponding to each operation scene in at least one operation scene, and obtain the accurate simulation perception results (i.e., the first perception result information) corresponding to the preset vehicle function of the self-vehicle in each operation scene, and the perception failure simulation is used to perform a perception failure simulation (or fault injection) process on the accurate first perception result information to obtain the second perception result information with a simulated real scene perception failure situation. The planning control software includes various vehicle functions of the self-vehicle, wherein the vehicle function to be verified is the preset vehicle function. The planning control software determines whether the preset vehicle function needs to be triggered based on the triggering rules of the preset vehicle function, and in the case of triggering the preset vehicle function, the control instructions (or control information) of the self-vehicle are calculated based on the functional algorithm of the preset vehicle function, such as lateral control instructions, longitudinal control instructions, etc., for controlling the driving of the self-vehicle. The dynamics engine is used to calculate the driving state information of the vehicle in the future based on the control instructions, and the simulation module determines whether the preset vehicle function of the vehicle can meet the corresponding expected functional safety requirements or functional performance requirements in the case of perception failure based on the driving state information of the vehicle in the future and the surrounding environment information of the vehicle in the future in the operation scenario data. Or it can also include an analysis module, which is used to determine whether the preset vehicle function of the vehicle can meet the corresponding expected functional safety requirements or functional performance requirements in the case of perception failure based on the driving state information of the vehicle in the future and the surrounding environment information of the vehicle in the future in the operation scenario data. For example, whether the AEB function successfully stops without colliding with other surrounding traffic participants (such as potential risk target objects that trigger the AEB function). Based on this, the vehicle function verification based on the in-loop simulation system is helpful to obtain the failure of the perception algorithm that the preset vehicle function can meet the expected functional safety requirements or functional performance requirements, so as to provide effective perception failure reference data for the development and maintenance of the perception algorithm, so as to optimize the perception algorithm, so that when the perception algorithm may fail in the real scene, the preset vehicle function can meet the corresponding expected functional safety requirements or functional performance requirements, thereby improving the driving safety of the vehicle.

[0029] Exemplary Methods

[0030] Figure 2FIG. 1 is a flow chart of a method for verifying a vehicle function provided by an exemplary embodiment of the present disclosure. This embodiment can be applied to electronic devices, such as terminals, servers, and other electronic devices, such as Figure 2 As shown, the following steps are included:

[0031] Step 201: Obtain operation scenario data corresponding to each operation scenario in at least one operation scenario.

[0032] Among them, the operation scene data includes the state information of the vehicle and the surrounding environment information of the vehicle. The state information of the vehicle may include the operation trajectory information of the vehicle within a preset time. The operation trajectory information may include the state of the vehicle at each time point within the preset time, and the state may include one or more of the information such as position, posture (orientation), speed, acceleration, angular velocity, etc. The state information of the vehicle may be the state information under the first coordinate system. The first coordinate system may be, for example, a world coordinate system or a reference coordinate system rigidly connected to the world coordinate system. The surrounding environment information may include one or more of the behavior information, environmental information, road information, etc. of other road participants (which may be called objects) around the vehicle. The behavior information of other road participants may include one or more of the position, posture, speed, acceleration, angular velocity, etc. of other road participants at each time point within the preset time. The environmental information may include at least one of the information such as time information and weather information. Time information may include daytime, night, etc. Weather information may include sunny days, rainy days, snowy days, etc. Road information may include road type (such as expressways, urban roads, etc.), lane line information, road conditions (such as flat roads, uphill roads, downhill roads, etc.), etc. The operation scenario data is used to simulate the operation of the vehicle in a certain environment and provide simulated perception result information for the verification of the preset vehicle functions of the vehicle.

[0033] In some optional embodiments, the surrounding environment information may also include at least one of positioning information, navigation information, high-precision map information, etc., to assist in simulating real operating scenarios.

[0034] In some optional embodiments, the operation scenario data may include a pre-generated scenario description file and a scenario map file. The scenario description file is used to describe the state information of the vehicle and the behavior information of other dynamic objects around the vehicle. The scenario map file is used to describe the environmental information and road information of the operation scenario. The specific representation method of the operation scenario data is not limited.

[0035] In some optional embodiments, the number and specific scenarios of operation scenarios can be set according to the actual needs of the preset vehicle functions. For example, the operation scenarios may include other vehicles behind the vehicle, oncoming vehicles in the adjacent lane, other vehicles with lower speeds in front of the highway, other vehicles cutting in from the front, pedestrians crossing the road in front of the intersection, and so on. Each operation scenario can also be generalized according to the relationship between other objects around the vehicle and the vehicle to obtain more fine-grained operation scenarios. For example, there are multiple operation scenarios of other vehicles at different distances behind the vehicle, multiple operation scenarios of oncoming vehicles with different speeds in the adjacent lane, multiple operation scenarios of other vehicles cutting in from different cutting directions in front, and so on. Take each fine-grained operation scenario as an operation scenario to obtain at least one operation scenario for vehicle function verification.

[0036] Step 202, based on the operation scenario data corresponding to each operation scenario, determine the first perception result information corresponding to the preset vehicle function of the vehicle in each operation scenario.

[0037] Among them, for any operating scenario, the first perception result information of the preset vehicle function of the vehicle in the operating scenario may include behavior information and road information of other objects around the vehicle.

[0038] In some optional embodiments, the first perception result information may be perception result information in the vehicle coordinate system.

[0039] In some optional embodiments, for any operation scenario, the first perception result information may include the perception result information of the vehicle at at least one trajectory point in the operation scenario. The perception result information of the vehicle at any trajectory point includes the behavior information and road information of other objects around the vehicle when the vehicle is at the trajectory point.

[0040] In some optional embodiments, the operation scenario data includes the state information of the vehicle and the surrounding environment information. Therefore, by analyzing the operation scenario data, the traffic flow information around the vehicle can be obtained, and then the first perception result information corresponding to the preset vehicle functions of the vehicle in each operation scenario can be obtained. For example, in the process of vehicle function verification, for each operation scenario being verified, other objects around the vehicle can be determined based on the state information and surrounding environment information of the vehicle at any trajectory point. Other objects may include dynamic objects such as other vehicles, pedestrians, and cyclists around the vehicle, as well as static objects such as lane lines and curbs, and the object state information of other objects relative to the vehicle can be determined. The object state information of other objects can be used as the first perception result information. That is, the first perception result information may include the state information of other objects around the vehicle relative to the vehicle perceived by simulation, such as position, orientation, speed, acceleration, angular velocity, etc. Compared with the perception results obtained through perception algorithms in real environments, the perception results obtained based on operating scenario data are determined based on the status of the vehicle and surrounding objects in the simulated operating scenario, rather than based on actual perception algorithms. They are not affected by factors such as hardware and environmental changes and are accurate perception results without perception failure (or perception fault).

[0041] In some optional embodiments, the types and quantities of other objects included in the first perception result information are set by specific operating scenario data and perception requirements of preset vehicle functions, and are not specifically limited. For example, for the AEB function, it is necessary to perceive obstacle objects around the vehicle that affect the driving of the vehicle to determine whether there is a risk of collision between the obstacle objects and the vehicle, whether the AEB function needs to be triggered, etc.

[0042] Step 203: Based on the perception failure simulation rule, each first perception result information is subjected to a perception failure simulation process to obtain second perception result information corresponding to each operation scenario.

[0043] Among them, the perception failure simulation rules can be set according to the perception failure situations that are prone to occur in the actual operating scenarios of the preset vehicle functions, and are used to perform perception failure simulation processing (or perception fault injection) on the accurate first perception result information to obtain the second perception result information with perception failure.

[0044] In some optional embodiments, the perception failure simulation rule may include trigger conditions for perception failure simulation processing and rules for failure simulation. The trigger conditions may be set according to the verification requirements of the preset vehicle functions. For example, the trigger conditions include triggering the perception failure simulation processing of the perception results of the other objects when the position relationship between the other objects and the vehicle meets certain conditions. The perception failure simulation processing may include simulating at least one of the simulation processing such as missed detection, false detection, detection delay, detection instability, detection distance error, and detection speed error of the other objects. Corresponding simulation rules (i.e., failure simulation rules) may be set for different objects. The simulation rules may include, for example, simulation ranges and simulation accuracy of parameters (which may be referred to as failure parameters or fault parameters) such as missed detection duration, missed detection frequency, false detection duration, false detection frequency, delay time, distance error, and speed error. For example, for the AEB function, after discovering a potential risk object, the AEB function is triggered, and the first perception result information of the risk object is subjected to perception failure simulation processing according to certain simulation rules of missed detection duration and missed detection frequency, so as to obtain the second perception result information with missed detection perception fault.

[0045] In some optional embodiments, for the first perception result information of any other object that can perform perception failure simulation, at least one perception failure simulation process can be performed to obtain corresponding second perception result information.

[0046] Step 204: Based on each second perception result information, verify the preset vehicle function to obtain a verification result.

[0047] Among them, for any second perception result information, verifying the preset vehicle function based on the second perception result information is to verify whether the preset vehicle function can meet the corresponding expected functional safety requirements or functional performance requirements under the second perception result information. The obtained verification result may include at least one of the functional response state of the preset vehicle function under each second perception result information and the perception performance index threshold of the perception algorithm allowed by the preset vehicle function. The perception performance index threshold can be determined based on the functional response state under each second perception result information.

[0048] The vehicle function verification method provided in the present embodiment can obtain operation scenario data of at least one operation scenario, determine the first perception result information corresponding to the preset vehicle function of the vehicle in each operation scenario based on the operation scenario data corresponding to each operation scenario, and then perform perception failure simulation processing on each first perception result information based on the perception failure simulation rule to obtain the second perception result information corresponding to each operation scenario, so that the second perception result information contains at least one perception failure situation that may exist in the perception algorithm in the real scenario, and verify the preset vehicle function based on each second perception result information, so as to determine whether the functional performance of the preset vehicle function in the case of perception failure can meet certain expected functional safety requirements or functional performance requirements, so as to determine the allowable situation of the preset vehicle function to the perception failure of the perception algorithm by simulating various perception failure situations of the perception algorithm, so as to guide the development and optimization of the perception algorithm, thereby helping to enable the perception algorithm to enable the preset vehicle function to meet the corresponding expected functional safety requirements or functional performance requirements.

[0049] Figure 3 It is a flowchart of a method for verifying a vehicle function provided by another exemplary embodiment of the present disclosure.

[0050] In some optional embodiments, the first perception result information includes a first perception result of at least one object around the vehicle. Step 203 performs a perception failure simulation process on each first perception result information based on a perception failure simulation rule to obtain second perception result information corresponding to each operating scenario, including:

[0051] Step 2031: For any operation scenario in each operation scenario, determine a target object that meets the first condition based on the first perception result information in the operation scenario.

[0052] Among them, the first condition can be set according to the preset vehicle function. The first condition may include the relative relationship between the objects around the vehicle and the vehicle in the operation scenario, such as the position relationship between the surrounding objects and the vehicle, the speed relationship, etc. For example, the first condition may be that the distance between the object and the vehicle is less than the distance threshold. The target object may be a dynamic object such as a vehicle or a pedestrian, or a static object such as a lane line or a curb.

[0053] In some optional embodiments, the first perception results of the objects around the vehicle may be perception results in the vehicle coordinate system. The first perception results of the objects around the vehicle may be matched with the first condition to determine the target object that meets the first condition. The specific first conditions for determining different objects as target objects may be different.

[0054] Step 2032: determine the failure mode corresponding to the target object based on the perceived failure simulation rule.

[0055] Among them, the perception failure simulation rules may include mapping rules between different objects and failure modes and simulation rules for different failure modes. The failure modes of different objects may be different. Different failure modes may correspond to different perception failure simulation processing methods (i.e., simulation rules). For example, for a vehicle object cutting in from the front, the perception failure simulation processing corresponding to its failure mode may include simulated missed detection, false detection, detection distance error, etc. For lane line objects, the perception failure simulation processing corresponding to its failure mode may include lateral distance error, etc.

[0056] In some optional embodiments, functional safety and expected functional safety hazard analysis and risk assessment (HARA) can be performed on vehicle functions according to the operational design domain (ODD) of functions such as autonomous driving and assisted driving, and functional safety goals (i.e., goals to be achieved at the functional safety level) can be determined. Functional safety goals include, for example, avoiding lost braking and avoiding unexpected braking. Functional safety goals are analyzed by means of fault tree analysis (FTA) and design failure mode and effects analysis (DFMEA) to obtain failure modes of perception algorithms under corresponding functional safety goals, such as missed detection, false detection, detection delay, unstable detection, detection error, etc. According to the operation scenarios in the HARA analysis and the analysis of the ODD, the parameter ranges of each operation scenario are determined, such as the parameter ranges of variables such as the longitudinal speed, longitudinal distance, and lateral speed of the target object. The parameter range can be associated with the operation scenario to generate operation scenario data for different operation scenarios for verifying vehicle functions and determine the failure mode of the target object under the operation scenario.

[0057] For example, in the hazard analysis and risk assessment of the high-speed cruise function (HWP), there are hazard events such as the vehicle ahead cutting in without slowing down, and the safety goal that the HWP function needs to prevent brake loss. Through FTA and DFMEA, the underlying failure mode can be obtained, such as missed detection of the cutting-in vehicle (that is, the perception algorithm missed the cutting-in vehicle), and another example is that the longitudinal distance of the cutting-in vehicle is too large (that is, the detection distance error causes the perception algorithm to perceive the longitudinal distance of the cutting-in vehicle as large, but it is not actually large). For this safety goal and ODD definition, the parameter range of the longitudinal speed, longitudinal distance, lateral speed and other parameters of the cutting-in vehicle in the cutting-in process is set. The specific parameter distribution of the parameter range can be determined based on the statistical collection of natural driving data, for example, it can be obtained by analyzing and refining road test data and aerial survey data. Missed detection is regarded as a failure mode in the operation scenario where there is a vehicle cutting in front. Through the generalization of missed detection duration and missed detection frequency, multiple failure sub-modes in this operation scenario can be determined, such as long-term missed detection sub-mode and short-term high-frequency missed detection sub-mode. The specific missed detection duration and missed detection frequency of each failure mode or failure sub-mode can be further generalized in a fine-grained manner to obtain fine-grained fault parameters (i.e., failure parameters) under various failure sub-modes, such as any determined missed detection duration value and missed detection frequency value, as a set of fault parameters under the failure mode or failure sub-mode.

[0058] For example, in the hazard analysis and risk assessment of the high-speed cruise function (HWP), there are dangerous events such as the self-braking of a vehicle in the lane ahead (i.e., unexpected braking) and the safety goal of preventing unexpected braking. Through FTA and DFMEA, the underlying failure mode can be obtained, such as inaccurate lateral distance measurement of the side vehicle, the actual lateral distance is large, and the lateral distance perceived by the perception algorithm is small, resulting in unexpected braking of the self-vehicle. For this safety goal, the parameter ranges of variables such as the longitudinal distance, lateral distance, and longitudinal speed of the side ahead vehicle are defined to generate operating scenario data for various operating scenarios. The lateral detection distance error is used as a failure mode, and a variety of fine-grained fault parameters under this failure mode can be obtained by generalizing different lateral detection distance errors. For example, each lateral detection distance error value obtained by generalization is used as a fault parameter under this failure mode.

[0059] Step 2033: Based on the failure mode, a perception failure simulation process is performed on the first perception result of the target object to obtain a second perception result of the perception failure of the target object.

[0060] Among them, after determining the failure mode corresponding to the target object, the first perception result of the target object can be subjected to a perception failure simulation process corresponding to the failure mode based on the failure mode of the target object to obtain a second perception result of the perception failure of the target object.

[0061] Exemplarily, by performing missed detection simulations of different durations and / or different frequencies on the target object, a second perception result under at least one combination of duration and frequency can be obtained. For example, in the first perception result information, the target object is detected from time point A to time point B and triggers the preset vehicle function. Then, in the generated second perception result, in the time period between time point A and time point B, the target object is missed under the conditions of missed detection duration T1 and missed detection frequency F1, to simulate whether the preset vehicle function can respond successfully in this missed detection situation.

[0062] Step 2034: Based on the second perception result of the target object, determine the second perception result information corresponding to the operating scene.

[0063] Among them, the second perception result information may include at least one second perception result corresponding to the target object in the operating scenario.

[0064] In some optional embodiments, for failure modes such as missed detection, false detection, detection distance error, and detection speed error (or perception failures), in order to ensure the accuracy and reliability of vehicle function verification, different second perception results can be generated for different fault parameters (or failure parameters, such as missed detection duration, missed detection frequency, false detection duration, false detection frequency, etc.) of each perception fault. For example, for missed detection faults, multiple fault parameters can be set, and different fault parameters correspond to different missed detection durations and / or different missed detection frequencies, so that second perception results corresponding to multiple fault parameters can be obtained. Then, for any operating scenario, a first perception result of a target object can correspond to one or more second perception results. That is, the second perception result information corresponding to the operating scenario may include at least one second perception result corresponding to each target object in at least one target object.

[0065] This embodiment sets corresponding failure modes for different objects, so that when it is determined that there is a target object that meets the first condition, a perception fault can be injected into the first perception result of the target object based on the failure mode corresponding to the target object, and a second perception result of the target object with the injected perception fault can be obtained, which helps to improve the accuracy and effectiveness of the perception failure simulation.

[0066] Figure 4 It is a flowchart of a method for verifying a vehicle function provided by yet another exemplary embodiment of the present disclosure.

[0067] In some optional embodiments, the step 202 of determining the first perception result information corresponding to the preset vehicle function of the vehicle in each operating scenario based on the operating scenario data corresponding to each operating scenario respectively includes:

[0068] Step 2021, for any operation scenario, based on the operation scenario data corresponding to the operation scenario, determine the first perception result information of the vehicle at at least one trajectory point in the operation scenario.

[0069] Among them, the vehicle has a certain driving trajectory when driving in this operation scenario, and the points on the driving trajectory are called trajectory points. The operation scenario data includes data describing the state information of the vehicle. By parsing the operation scenario data, each trajectory point of the vehicle in the operation scenario can be obtained. Combined with the vehicle's surrounding environment information in the operation scenario data, the first perception result information of the vehicle at any trajectory point can be obtained. As the vehicle moves, the relative relationship between other objects around the vehicle and the vehicle may be different at different trajectory points. That is, other objects in different areas around the vehicle may be different.

[0070] Step 2031 determines, for any one of the operation scenarios, a target object that meets the first condition based on the first perception result information in the operation scenario, including:

[0071] Step 20311, for any of the operation scenarios, determine the target objects in different areas around the vehicle based on the first perception result information corresponding to each trajectory point.

[0072] Among them, for each operation scenario, the target objects in different areas around the vehicle can be determined based on the first perception result information corresponding to each trajectory point of the vehicle in the operation scenario.

[0073] In some optional embodiments, the area around the vehicle may be divided into different areas according to a preset area division rule, for example, the area around the vehicle may be divided into different areas by a grid coordinate system.

[0074] For example, Figure 5 FIG. 1 is a schematic diagram of dividing the area around the vehicle provided by an exemplary embodiment of the present disclosure. Figure 5 As shown, the area around the vehicle is divided into multiple grids through a grid coordinate system (or grid map), and each grid represents an area around the vehicle. xoy represents the vehicle coordinate system of the vehicle. The grid coordinate system can use any position of the vehicle coordinate system as the coordinate origin, such as the origin of the vehicle coordinate system of the vehicle. That is, the relative relationship between the grid coordinate system and the vehicle coordinate system remains unchanged. As the vehicle moves, the vehicle coordinate system changes, and the grid coordinate system changes with the vehicle coordinate system, always representing different areas around the vehicle. For example, as the vehicle and the target object continue to move, the position of the target object relative to the vehicle may change from grid 1 to grid 2 and then to grid 3 in the figure. The size (length and width) of each grid in the grid coordinate system can be set to any size, without specific limitation.

[0075] This embodiment determines the target objects in different areas around the vehicle through the first perception result information at each trajectory point in the operating scenario. On the one hand, it is convenient to determine the triggering of the perception failure simulation, and on the other hand, it helps to verify in a more fine-grained manner the tolerance of the preset vehicle function for the perception failure of the target objects in different areas.

[0076] In some optional embodiments, the in-the-loop simulation system can be used to simulate the driving of the vehicle in any operating scenario, and during the driving process, the current first perception result information can be obtained in real time. The current first perception result information can include the state information of the target objects in different areas around the vehicle. And based on the preset conditions, it is determined whether to trigger the perception fault injection, such as whether the target objects in different areas meet the perception fault injection triggering conditions (such as the target objects in different areas meet the perception fault injection triggering conditions). Figure 5 When the relative position of the vehicle to the ego vehicle is grid 2, the perception fault injection corresponding to grid 2 is triggered). If it is determined that the perception fault injection is triggered, the current first perception result information is subjected to perception fault injection to obtain the current second perception result information, and then based on the second perception result information, the preset vehicle function is verified to determine the functional response state of the preset vehicle function, so that the functional response state of the perception fault injection at each trajectory point during the driving process of the ego vehicle under an operation scenario can be obtained, that is, the functional response state of the perception fault injection in different areas around the ego vehicle can be obtained. It is also possible to perform perception fault injection of multiple failure modes on the first perception result information of the current trajectory point, respectively, so as to obtain the functional response states of multiple perception fault injections at each trajectory point. Based on this, combined with a large number of operation scenarios, multiple failure modes and specific sub-modes, and different trajectory points, different types of perception performance indicator thresholds for different areas around the ego vehicle can be obtained. For example Figure 5 Missed detection duration threshold at middle grid 1, missed detection frequency threshold, longitudinal distance error threshold, lateral distance error threshold, longitudinal speed error threshold, lateral speed error threshold, detection delay threshold, etc.

[0077] In some optional embodiments, the failure parameters can be traversed in order of the impact on the preset vehicle function from small to large, such as first verifying the smaller detection error, missed detection duration, false detection duration, etc. For example, because the error is small, the preset vehicle function can tolerate the error and achieve the expected result, so that the functional response state is successful. As the failure parameter becomes larger and larger, after the perception performance index reaches the allowable boundary of the preset vehicle function, the functional response state of the preset vehicle function will fail, so that the perception performance index threshold that the preset vehicle function can tolerate can be obtained.

[0078] In some optional embodiments, the perception performance index can also be expressed as a perception delay index, a perception recall rate, a detection distance accuracy, a detection speed accuracy, etc. For example, the perception performance index corresponding to a long-term missed detection can be expressed as a perception delay index. The perception performance index corresponding to a short-term high-frequency missed detection can be expressed as a perception recall rate. The perception performance index corresponding to the horizontal and vertical distance error can be expressed as the detection distance accuracy. The perception performance index corresponding to the horizontal and vertical speed error can be expressed as the detection speed accuracy. Thus, the perception delay index threshold, the perception recall rate threshold, the detection distance accuracy threshold, and the detection speed accuracy threshold are used as verification results to guide the development and optimization of the perception algorithm.

[0079] In some optional embodiments, step 204 verifies the preset vehicle function based on each second perception result information to obtain the verification result, including:

[0080] Step 204A, based on each second perception result information, determine the perception performance index thresholds corresponding to the preset vehicle functions in different areas around the vehicle.

[0081] Among them, the perception performance index threshold may include the index threshold corresponding to each failure mode. For example, missed detection duration threshold, missed detection frequency threshold, false detection duration threshold, false detection frequency threshold, detection delay time threshold, detection distance error threshold, detection speed error threshold, etc. The detection distance error threshold can also be subdivided into a lateral distance error threshold and a longitudinal distance error threshold. The detection speed error threshold can also be subdivided into a lateral speed error threshold and a longitudinal speed error threshold. The specific perception performance index division can be set according to the requirements of the preset vehicle functions, and the present disclosure is not limited thereto.

[0082] In some optional embodiments, for each area, it can be determined whether to trigger a preset vehicle function based on the second perception result of the target object in the area, and when the preset vehicle function is triggered, the response state of the preset vehicle function can avoid the occurrence of a hazardous event. For the second perception result of any failure mode, if the response state of the preset vehicle function is successful, it is determined that the area can meet the expected functional safety requirements or functional performance requirements in the failure mode. Based on the response state of the preset vehicle function under a large number of failure modes in the area, it can be determined that the area can meet the expected functional safety requirements or functional performance requirements. Based on these failure modes, the performance indicator threshold corresponding to the area can be determined.

[0083] Step 204B, taking the perception performance index thresholds corresponding to the preset vehicle functions in different areas around the vehicle as the verification result.

[0084] Among them, after obtaining the perception performance index thresholds corresponding to different areas around the vehicle, the perception performance index thresholds corresponding to different areas can be used as verification results to guide the development and optimization of perception algorithms.

[0085] This embodiment determines the perception performance index thresholds corresponding to different areas around the vehicle as verification results, thereby achieving more fine-grained determination of perception performance index thresholds, which helps to provide more accurate and effective perception performance index thresholds for the development and optimization of perception algorithms, and further improves the reliability of perception algorithms.

[0086] Figure 6 It is a flowchart of a method for verifying a vehicle function provided by yet another exemplary embodiment of the present disclosure.

[0087] In some optional embodiments, the failure mode includes at least one type of sub-mode.

[0088] In some optional embodiments, step 2033 performs a perception failure simulation process on the first perception result of the target object based on the failure mode to obtain a second perception result of the perception failure of the target object, including:

[0089] Step 20331: For any seed pattern, based on the simulation sub-rule corresponding to the sub-pattern, perform perception failure simulation processing on the first perception result of the target object to obtain the second perception result of the target object under the sub-pattern.

[0090] Among them, sub-mode is a more fine-grained division of failure mode. For example, the false detection failure mode can be subdivided into multiple sub-modes such as long-term false detection, short-term high-frequency false detection, etc. according to different false detection durations and different false detection frequencies. The detection error can be subdivided into sub-modes such as detection distance error and detection speed error, or subdivided into sub-modes such as lateral distance error, longitudinal distance error, lateral speed error, longitudinal distance error, or subdivided into sub-modes such as errors that make the detection results larger and errors that make the detection results smaller. The sub-mode can also be divided at a finer granularity. The specific sub-mode division can be set according to the specific operating scenario of the preset vehicle function.

[0091] In some optional embodiments, corresponding simulation sub-rules may be set for each sub-mode so as to perform perception failure simulation processing on the first perception result of the target object to obtain the second perception result of the target object under the sub-mode.

[0092] In some optional embodiments, for each sub-mode, the simulation sub-rule corresponding to the sub-mode may include the failure parameter range and failure simulation accuracy corresponding to the sub-mode, so as to determine multiple failure parameters from the failure parameter range, and each failure parameter is used to perform fault injection on the first perception result of the target object to obtain the second perception result corresponding to the failure parameter. The second perception result of the target object under any sub-mode may include the second perception results corresponding to multiple failure parameters. If each failure parameter is taken as a sub-mode, the second perception result of the target object under the sub-mode includes the second perception result corresponding to the failure parameter. No specific limitation is given.

[0093] Determining the second perception result information corresponding to the operation scene based on the second perception result of the target object in step 2034 includes:

[0094] Step 20341, based on the second perception results corresponding to the target object in each sub-mode, determine the second perception result information corresponding to the running scene in each sub-mode.

[0095] Among them, the second perception result information of the running scene in any sub-mode may include the second perception results of each target object in at least one target object in the sub-mode respectively.

[0096] In some optional embodiments, for each operating scenario, the preset vehicle function can be verified in the loop based on the second perception result information corresponding to the operating scenario in each sub-mode, so as to obtain the tolerance of the preset vehicle function to the perception performance indicator of the perception algorithm and obtain the perception performance indicator threshold of the perception algorithm.

[0097] This embodiment divides the failure modes into finer granularity to obtain the second perception result information of each operating scenario in each sub-mode, which helps to verify the preset vehicle functions more accurately and effectively and improve the accuracy and effectiveness of the perception performance indicator threshold of the perception algorithm allowed by the preset vehicle functions.

[0098] In some optional embodiments, step 204 verifies the preset vehicle function based on each second perception result information to obtain the verification result, including:

[0099] Step 2041a, based on the second perception result information corresponding to the operating scenario in each sub-mode, determine the functional response state of the preset vehicle function in each sub-mode of the operating scenario.

[0100] Among them, the function response status can include two states: success and failure. Success means that the preset vehicle function can achieve the expected result in this sub-mode. For example, the AEB function can successfully brake and avoid collision with the target object. Failure means that the preset vehicle function does not achieve the expected result in this sub-mode. For example, for the high-speed cruise function, the longitudinal distance to the vehicle in front is less than the minimum distance safety threshold. The judgment of the success of the function response status can also include the collision time (Time to Collision, abbreviated as: TTC) greater than the safety time threshold, deviating from the center line of the lane by more than the safety distance threshold, etc., which can be set according to the preset vehicle function.

[0101] In some optional embodiments, for each sub-mode of each operation scenario, it can be determined whether to trigger a preset vehicle function based on the second perception result information of the sub-mode. If the preset vehicle function is triggered, the control information of the vehicle is calculated based on the function algorithm of the preset vehicle function. Based on the control information of the vehicle, the driving trajectory of the vehicle under the control of the control information is determined, and combined with the surrounding environment information in the operation scenario data of the operation scenario, it is determined whether the driving trajectory of the vehicle and other surrounding objects can achieve the expected result. If the expected result can be achieved, the function response state is determined to be successful, otherwise the function response state is determined to be failed.

[0102] Step 2042a, based on the functional response state of the preset vehicle function in each sub-mode of the operating scenario, determine the perception performance index threshold of the preset vehicle function in the operating scenario.

[0103] Among them, based on the functional response state of the preset vehicle function in each sub-mode of the operation scenario, the permissible perception performance index of the preset vehicle function in the operation scenario can be determined, that is, the perception performance index (failure parameter) when the functional response state is successful, such as the permissible missed detection duration, missed detection frequency, false detection duration, false detection frequency, detection distance error, detection speed error, etc. Then, based on the permissible perception performance index, the perception performance index threshold of the preset vehicle function in the operation scenario can be determined.

[0104] Step 2043a, based on the perception performance index threshold of the preset vehicle function in each operating scenario, determine the target perception performance index threshold corresponding to the preset vehicle function, and use the target perception performance index threshold as the verification result.

[0105] In some optional embodiments, the perception performance index thresholds of preset vehicle functions in various operating scenarios may be comprehensively considered to determine the target perception performance index thresholds corresponding to the preset vehicle functions as verification results.

[0106] In some optional embodiments, each operation scenario can be classified, and for operation scenarios belonging to the same category, multiple operation scenarios can be combined to determine the target perception performance index threshold of the preset vehicle function in the operation scenario of this category. For example, the target perception performance index threshold under the operation scenario can be determined by combining multiple operation scenarios with vehicles cutting in front. For example, the minimum perception performance index threshold among multiple operation scenarios of the same category of operation scenarios can be used as the target perception performance index threshold of the operation scenario of this category to ensure the functional safety of each operation scenario under this category of operation scenarios.

[0107] This embodiment can determine the allowable situation of perception failure of the preset vehicle function for each sub-mode by determining the functional response state of the preset vehicle function under each sub-mode of each operating scenario, and further determine the perception performance index threshold of the preset vehicle in each operating scenario. The target perception performance index threshold corresponding to the preset vehicle function can be obtained by comprehensively considering the perception performance index threshold of the preset vehicle function in each operating scenario. Through in-the-loop simulation verification of a large number of sub-modes of a large number of operating scenarios, the target perception performance index corresponding to the preset vehicle function can be accurately and effectively determined, which helps to improve verification efficiency.

[0108] Figure 7 It is a flowchart of a method for verifying a vehicle function provided by yet another exemplary embodiment of the present disclosure.

[0109] In some optional embodiments, step 20331 performs perception failure simulation processing on the first perception result of the target object based on the simulation sub-rule corresponding to the sub-mode to obtain the second perception result of the target object under the sub-mode, including:

[0110] Step 203311, based on the simulation sub-rule corresponding to the sub-mode, determine at least one perception performance index value; based on each perception performance index value, update the first perception result of the target object to obtain the second perception result of the target object corresponding to each perception performance index value.

[0111] Among them, the perception performance index value may include missed detection duration, missed detection frequency, false detection duration, false detection frequency, detection distance error, detection speed error, detection delay time, etc.

[0112] In some optional embodiments, at least one perception performance indicator value can be determined based on the sub-mode and the corresponding simulation sub-rule. For example, if the sub-mode is long-term missed detection, the simulation sub-rule may include a missed detection duration range and a sampling accuracy. For example, if the missed detection duration range is [t1, t2] seconds and the sampling accuracy is 0.01 seconds, then each missed detection duration of t1, t1+0.01, t1+0.02, ..., t2 can be determined from the missed detection duration range, and each missed detection duration is used as a perception performance indicator value. Or a combination of missed detection duration and missed detection frequency is used as a perception performance indicator value. There is no specific limitation. For another example, if the sub-mode is longitudinal distance error, the simulation sub-rule may include a longitudinal distance error range and a sampling accuracy, then multiple longitudinal distance error values ​​can be sampled from the longitudinal distance error range based on the sampling accuracy as the perception performance indicator value.

[0113] In some optional embodiments, for any perception performance index value, a perception fault is injected into the first perception result of the target object based on the perception performance index value to obtain a second perception result of the target object corresponding to the perception performance index value. For example, the first perception result of the target object is injected with a missed detection fault according to the missed detection duration, so that the target object is missed for a corresponding duration in the second perception result. For example, if the first perception result includes the state information of the target object at each time point within a period of time, the state information of the target object at the missed detection duration will be removed from the second perception result to simulate the perception fault of the perception algorithm missing the target object within the missed detection duration. For another example, if the first perception result includes the longitudinal distance between the target object and the vehicle, the longitudinal distance of the target object is injected with an error according to the longitudinal distance error value determined above to obtain a second perception result. For example, if the longitudinal distance of the first perception result is x1 and the longitudinal distance error value is △x1, the longitudinal distance in the second perception result is x1+△x1. Based on this, the second perception result of the target object corresponding to each perception performance index value can be obtained, or the second perception result corresponding to the target object under each perception performance index value can be obtained.

[0114] This embodiment facilitates perceptual fault injection into the first perceptual result of the target object by determining at least one perceptual performance indicator value, and obtains the second perceptual result under each perceptual performance indicator value. When used to verify a preset vehicle function, the permissible situation of the preset vehicle function for the perceptual performance indicator value can be determined.

[0115] In some optional examples, the following failure mode perception fault injections may be performed on the first perception results of the dynamic and static target objects:

[0116] 1. The target object is missed for a long time. That is, the target object perceived by the first perception result is continuously discarded for a missed detection period, and the fault injection is terminated after the missed detection period is met.

[0117] 2. Short-term and high-frequency missed detection of target objects. That is, the target objects perceived by the first perception result are discarded according to a certain missed detection frequency, and the corresponding missed detection duration is discarded each time.

[0118] 3. Long-term false detection of the target object. That is, continuous false detection perception fault injection is performed on the target object perceived by the first perception result, and the fault injection is terminated after the false detection duration is met. For example, the state information of the perceived target object is added during the false detection duration of the period in which the target object was not perceived in the first perception result to simulate the perception fault of the false detection target object.

[0119] 4. Short-term and high-frequency false detection of the target object. That is, short-term and high-frequency false detection perception fault injection is performed on the first perception result.

[0120] 5. Target object horizontal and vertical distance error. That is, inject horizontal and vertical distance error fault into the horizontal and vertical distance (including at least one of the horizontal distance and the vertical distance) of the first perception result. For example, inject error perception fault into the horizontal distance between the surrounding vehicles and the vehicle.

[0121] 6. Errors in the lateral and longitudinal speeds of the target object. That is, an error perception fault is injected into the lateral and longitudinal speeds (including at least one of the lateral speed and the longitudinal speed) of the first perception result. For example, an error perception fault is injected into the longitudinal speed of the vehicle in front of the vehicle.

[0122] 7. Target object orientation angle error: That is, the orientation angle (i.e., attitude or heading angle) of the first perception result is subjected to angle error perception fault injection.

[0123] 8. Lane line lateral distance error: This is to inject error perception faults into the lateral distance between the lane line and the vehicle.

[0124] In practical applications, more failure modes can be determined according to the actual operation scenarios, which can be used to verify the preset vehicle functions through the perception fault injection of the in-loop simulation system, and obtain the permissible conditions of the perception performance indicators of the preset vehicle functions for the perception algorithm. Through the rich and diverse perception fault injection, it is helpful to obtain the permissible conditions of more perceptual performance indicators of the preset vehicle functions for the perception algorithm, which can provide more accurate and effective perception performance indicator thresholds for the development and maintenance of the perception algorithm, and further improve the safety of the preset vehicle functions.

[0125] In some optional embodiments, determining the second perception result information corresponding to each sub-mode of the running scene based on the second perception result corresponding to each sub-mode of the target object in step 20341 includes:

[0126] Step 203411, for any sub-mode, based on the second perception results of the target object corresponding to each perception performance indicator value, determine the second perception result information corresponding to each perception performance indicator value.

[0127] Step 204 verifies the preset vehicle function based on each second perception result information to obtain a verification result, including:

[0128] Step 2041b, based on the second perception result information corresponding to each perception performance index value, determine the functional response state of the preset vehicle function corresponding to each perception performance index value.

[0129] Among them, for each second perception result information corresponding to any perception performance index value, the specific operating principle of determining the functional response state of the preset vehicle function can be referred to the aforementioned embodiment and will not be repeated here.

[0130] Step 2042b, based on the functional response states of the preset vehicle functions corresponding to the respective perception performance index values, determine the perception performance index threshold of the preset vehicle function in this sub-mode.

[0131] Among them, each sub-mode may include one or more perception performance index values. Therefore, the functional response state of the preset vehicle function under each perception performance index value can be comprehensively considered to determine the perception performance index threshold of the preset vehicle function under this sub-mode.

[0132] Step 2043b, based on the perception performance index thresholds corresponding to the preset vehicle function in each sub-mode, determine the perception performance index threshold of the preset vehicle function in the operating scenario, and use the perception performance index threshold of the preset vehicle function in the operating scenario as the verification result.

[0133] The specific operation principle of step 2043b can be found in the above-mentioned embodiment and will not be elaborated here.

[0134] This embodiment uses the second perception result information corresponding to each perception performance index value of each sub-mode for verification of the preset vehicle function, so as to obtain the functional response state of the preset vehicle function to each perception performance index value, and thus can accurately obtain the perception performance index threshold value that the preset vehicle function can allow by traversing and verifying the perception performance index value, thereby further improving the efficiency and accuracy of determining the perception performance index threshold value.

[0135] Figure 8 It is a flowchart of a method for verifying a vehicle function provided by yet another exemplary embodiment of the present disclosure.

[0136] In some optional embodiments, step 204 verifies the preset vehicle function based on each second perception result information to obtain the verification result, including:

[0137] Step 2041: For any operation scenario, based on the operation scenario data of the operation scenario, determine the map data of the operation scenario.

[0138] The map data may include at least one of positioning data, navigation data, and high-precision map data. The map data can assist the regulatory control software in planning and control, provide a better simulation environment for planning and control, and more effectively simulate the working conditions of the regulatory control software in real scenarios.

[0139] Step 2042, for any second perception result information of the operating scenario, based on the second perception result information and the map data of the operating scenario, using the function algorithm corresponding to the preset vehicle function, determine the control information of the vehicle.

[0140] Among them, the function algorithm corresponding to the preset vehicle function is an application software algorithm for realizing the preset vehicle function, and is used to calculate the control information of the vehicle based on the second perception result information, map data, etc. For example, after the AEB function is triggered, the control information of the vehicle, such as deceleration, can be calculated, and a deceleration control instruction can be issued to the vehicle, and the vehicle executes the deceleration control instruction to brake, thereby realizing automatic emergency braking. The control information may include at least one of lateral control information, longitudinal control information, etc. For example, the control information may include lateral acceleration, longitudinal acceleration, angle, etc.

[0141] Step 2043, based on the control information of the vehicle and the vehicle dynamics model, determine the driving state information of the vehicle under the control of the control information.

[0142] The vehicle dynamics model is used to simulate the driving trajectory of the vehicle under the control of the control information. Based on the control information of the vehicle and the vehicle dynamics model, the driving state information of the vehicle under the control of the control information can be determined. The driving state information may include the state information of one or more trajectory points (or time points) within a period of time in the future under the control of the control information, such as position, posture, speed, acceleration, angular velocity and other state information.

[0143] Step 2044, based on the driving status information and the operating scenario data corresponding to the second perception result information, determine the functional response state of the preset vehicle function under the second perception result information.

[0144] Among them, based on the operation scenario data corresponding to the second perception result information, the driving trajectories of other objects around the ego vehicle in the future period of time can be determined. Combined with the driving state information of the ego vehicle and the driving trajectories of other objects, it can be determined whether the change in the relative relationship between the ego vehicle and other objects has achieved the expected result, such as whether the ego vehicle has successfully braked due to the AEB function, whether the HWP function has successfully avoided obstacles when a vehicle cuts in, etc. If the expected result can be achieved, the function response state is determined to be successful, and if the expected result cannot be achieved, the function response state is determined to be failed.

[0145] Step 2045, determining the verification result based on the functional response status of the preset vehicle function under each second perception result information.

[0146] The specific operating principle of step 2045 can be found in the aforementioned embodiment and will not be elaborated here.

[0147] This embodiment further combines map data and a vehicle dynamics model to implement verification of preset vehicle functions, which helps to further improve the accuracy of the verification results.

[0148] In some optional embodiments, step 2044 determines the functional response state of the preset vehicle function under the second perception result information based on the driving state information and the operating scenario data corresponding to the second perception result information, including:

[0149] Based on the driving status information and the operating scenario data corresponding to the second perception result information, determine the state of the expected result of the vehicle under the preset vehicle function; in response to the state of the expected result being achieved, determine that the functional response state of the preset vehicle function under the second perception result information is successful; or, in response to the state of the expected result being not achieved, determine that the functional response state of the preset vehicle function under the second perception result information is failed.

[0150] Among them, the expected results can be set according to the functional safety goals of the preset vehicle functions. For example, the AEB function successfully brakes, the HWP function successfully avoids obstacles, etc. Combined with the driving state information of the self-vehicle under the control of the control information and the driving trajectories of other objects in the operation scene data, it can be determined whether the self-vehicle collides with other objects, whether the self-vehicle successfully avoids obstacles, whether the minimum longitudinal distance between the self-vehicle and the target object is greater than the minimum longitudinal safety distance threshold, whether the minimum lateral distance between the self-vehicle and the target object is greater than the minimum lateral safety distance threshold, etc., so as to determine the state of the expected results of the self-vehicle under the preset vehicle function. If the expected result can be achieved, the functional response state of the preset vehicle function is determined to be successful, and if the expected result cannot be achieved, the functional response state is determined to be failed.

[0151] The above-mentioned embodiments of the present disclosure may be implemented separately or in any combination without conflict. The specific configuration may be based on actual needs and the present disclosure does not limit this.

[0152] Any vehicle function verification method provided in the embodiments of the present disclosure may be executed by any appropriate device with data processing capabilities, including but not limited to: a terminal device and a server, etc. Alternatively, any vehicle function verification method provided in the embodiments of the present disclosure may be executed by a processor, such as the processor executing any vehicle function verification method mentioned in the embodiments of the present disclosure by calling corresponding instructions stored in a memory. This will not be described in detail below.

[0153] Exemplary Devices

[0154] Fig. 9 FIG. 1 is a schematic diagram of a vehicle function verification device provided by an exemplary embodiment of the present disclosure. The device of this embodiment can be used to implement a corresponding vehicle function verification method embodiment of the present disclosure, such as Fig. 9 The device shown includes: an acquisition module 501 , a first processing module 502 , a second processing module 503 and a third processing module 504 .

[0155] The acquisition module 501 is used to acquire operation scenario data corresponding to each operation scenario in at least one operation scenario; the operation scenario data includes the state information of the vehicle and the surrounding environment information of the vehicle.

[0156] The first processing module 502 is used to determine the first perception result information corresponding to the preset vehicle functions of the vehicle in each operating scenario based on the operating scenario data corresponding to each operating scenario.

[0157] The second processing module 503 is used to perform perception failure simulation processing on each first perception result information based on the perception failure simulation rule to obtain the second perception result information corresponding to each operation scenario.

[0158] The third processing module 504 is used to verify the preset vehicle function based on each second perception result information to obtain a verification result.

[0159] Fig.10 It is a schematic diagram of the structure of a vehicle function verification device provided by another exemplary embodiment of the present disclosure.

[0160] In some optional embodiments, the first perception result information includes a first perception result of at least one object around the vehicle. The second processing module 503 includes:

[0161] The first processing unit 5031 is used to determine, for any operation scenario in each operation scenario, a target object that meets a first condition based on the first perception result information in the operation scenario.

[0162] The second processing unit 5032 is used to determine the failure mode corresponding to the target object based on the perception failure simulation rule.

[0163] The third processing unit 5033 is used to perform perception failure simulation processing on the first perception result of the target object based on the failure mode, so as to obtain a second perception result of the perception failure of the target object.

[0164] The fourth processing unit 5034 is used to determine the second perception result information corresponding to the operating scene based on the second perception result of the target object.

[0165] In some optional embodiments, the first processing module 502 includes:

[0166] The first determination unit 5021 is used to determine, for any operation scenario, first perception result information of the vehicle at at least one trajectory point in the operation scenario based on the operation scenario data corresponding to the operation scenario.

[0167] The first processing unit 5031 is specifically used to: for any operation scenario in each operation scenario, determine the target objects in different areas around the vehicle based on the first perception result information corresponding to each trajectory point.

[0168] In some optional embodiments, the third processing module 504 is specifically configured to:

[0169] Based on each second perception result information, the perception performance index thresholds corresponding to the preset vehicle function in different areas around the vehicle are determined, and the perception performance index thresholds corresponding to the preset vehicle function in different areas around the vehicle are used as verification results.

[0170] In some optional embodiments, the failure mode includes at least one type of sub-mode.

[0171] In some optional embodiments, the third processing unit 5033 is specifically configured to:

[0172] For any seed pattern, based on the simulation sub-rule corresponding to the sub-pattern, a perception failure simulation process is performed on the first perception result of the target object to obtain a second perception result of the target object under the sub-pattern.

[0173] The fourth processing unit 5034 is specifically used for:

[0174] Based on the second perception results respectively corresponding to the target object in each sub-mode, the second perception result information respectively corresponding to the running scene in each sub-mode is determined.

[0175] In some optional embodiments, the third processing module 504 includes:

[0176] The second determination unit 5041a is used to determine the functional response state of the preset vehicle function in each sub-mode of the operating scenario based on the second perception result information corresponding to the operating scenario in each sub-mode.

[0177] The third determination unit 5042a is used to determine the perception performance index threshold of the preset vehicle function in the operating scenario based on the functional response state of the preset vehicle function in each sub-mode of the operating scenario.

[0178] The fourth determination unit 5043a is used to determine the target perception performance index threshold corresponding to the preset vehicle function based on the perception performance index threshold of the preset vehicle function in each operating scenario, and use the target perception performance index threshold as the verification result.

[0179] In some optional embodiments, the third processing unit 5033 is specifically configured to:

[0180] Based on the simulation sub-rule corresponding to the sub-mode, at least one perception performance index value is determined; based on each perception performance index value, the first perception result of the target object is updated to obtain the second perception result of the target object corresponding to each perception performance index value.

[0181] In some optional embodiments, the fourth processing unit 5034 is specifically configured to:

[0182] For any sub-mode, based on the second perception results of the target object corresponding to each perception performance indicator value, the second perception result information corresponding to each perception performance indicator value is determined.

[0183] The third processing module 504 is specifically used for:

[0184] Based on the second perception result information corresponding to each perception performance index value, the functional response state of the preset vehicle function corresponding to each perception performance index value is determined. Based on the functional response state of the preset vehicle function corresponding to each perception performance index value, the perception performance index threshold of the preset vehicle function in the sub-mode is determined. Based on the perception performance index threshold corresponding to each sub-mode of the preset vehicle function, the perception performance index threshold of the preset vehicle function in the operating scenario is determined, and the perception performance index threshold of the preset vehicle function in the operating scenario is used as the verification result.

[0185] Fig.11 4 is a schematic diagram of the structure of a vehicle function verification device provided by yet another exemplary embodiment of the present disclosure.

[0186] In some optional embodiments, the third processing module 504 includes:

[0187] The fifth determining unit 5041 is used to determine, for any operation scenario, the map data of the operation scenario based on the operation scenario data of the operation scenario.

[0188] The sixth determination unit 5042 is used to determine the control information of the vehicle for any second perception result information of the operating scenario, based on the second perception result information and the map data of the operating scenario, using the function algorithm corresponding to the preset vehicle function.

[0189] The seventh determination unit 5043 is used to determine the driving state information of the ego vehicle under the control of the control information based on the control information of the ego vehicle and the vehicle dynamics model.

[0190] The eighth determination unit 5044 is used to determine the functional response state of the preset vehicle function under the second perception result information based on the driving state information and the operating scenario data corresponding to the second perception result information.

[0191] The ninth determination unit 5045 is used to determine the verification result based on the functional response state of the preset vehicle function under each second perception result information.

[0192] In some optional embodiments, the eighth determining unit 5044 is specifically configured to:

[0193] Based on the driving status information and the operating scenario data corresponding to the second perception result information, determine the state of the expected result of the vehicle under the preset vehicle function; in response to the state of the expected result being achieved, determine that the functional response state of the preset vehicle function under the second perception result information is successful; or, in response to the state of the expected result being not achieved, determine that the functional response state of the preset vehicle function under the second perception result information is failed.

[0194] The beneficial technical effects corresponding to the exemplary embodiment of the present device can be found in the corresponding beneficial technical effects of the above exemplary method section, which will not be repeated here.

[0195] Exemplary Electronic Devices

[0196] Fig.12 It is a structural diagram of an electronic device provided by an embodiment of the present disclosure, including at least one processor 11 and a memory 12.

[0197] The processor 11 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.

[0198] The memory 12 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 11 may execute one or more computer program instructions to implement the methods and / or other desired functions of the various embodiments of the present disclosure described above.

[0199] In one example, the electronic device 10 may further include: an input device 13 and an output device 14, and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0200] The input device 13 may also include, for example, a keyboard, a mouse, etc.

[0201] The output device 14 can output various information to the outside, and may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0202] Of course, to simplify, Fig.12 Only some of the components related to the present disclosure in the electronic device 10 are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, according to specific application situations, the electronic device 10 may also include any other appropriate components.

[0203] Exemplary computer program products and computer-readable storage media

[0204] In addition to the above methods and devices, embodiments of the present disclosure may also provide a computer program product, including computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the methods of various embodiments of the present disclosure described in the above “Exemplary Methods” section.

[0205] The computer program product may be written in any combination of one or more programming languages ​​to write program code for performing the operations of the disclosed embodiments, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0206] In addition, an embodiment of the present disclosure may also be a computer-readable storage medium having computer program instructions stored thereon. When the computer program instructions are executed by a processor, the processor executes the steps in the method of various embodiments of the present disclosure described in the above “Exemplary Method” section.

[0207] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium is, for example, but not limited to, a system, device or device including electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0208] The basic principles of the present disclosure are described above in conjunction with specific embodiments. However, the advantages, strengths, effects, etc. mentioned in the present disclosure are only examples and not limitations, and cannot be considered as necessary for each embodiment of the present disclosure. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, rather than limitation, and the above details do not limit the present disclosure to being implemented by adopting the above specific details.

[0209] Those skilled in the art may make various changes and modifications to the present disclosure without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present disclosure claims and their equivalents, the present disclosure is also intended to include these modifications and variations.

Claims

1. A method for verifying vehicle functions, comprising: Obtaining operation scenario data corresponding to each of the operation scenarios in at least one operation scenario; The operation scenario data includes the state information of the vehicle and the surrounding environment information of the vehicle; Based on the operation scenario data corresponding to each of the operation scenarios, determining first perception result information corresponding to the preset vehicle function of the vehicle in each of the operation scenarios; Based on the perception failure simulation rule, performing perception failure simulation processing on each of the first perception result information to obtain second perception result information corresponding to each of the operation scenarios; Based on each of the second perception result information, the preset vehicle function is verified to obtain a verification result.

2. The method according to claim 1, wherein: The first perception result information includes a first perception result of at least one object around the vehicle; The performing perception failure simulation processing on each of the first perception result information based on the perception failure simulation rule to obtain the second perception result information corresponding to each of the operation scenarios respectively includes: For any of the operation scenarios, determining a target object that meets a first condition based on the first perception result information in the operation scenario; Based on the perceived failure simulation rule, determining a failure mode corresponding to the target object; Based on the failure mode, performing a perception failure simulation process on the first perception result of the target object to obtain a second perception result of the perception failure of the target object; Based on the second perception result of the target object, determine the second perception result information corresponding to the operating scenario.

3. The method according to claim 2, wherein: The failure mode includes at least one type of sub-mode; The performing a perception failure simulation process on the first perception result of the target object based on the failure mode to obtain a second perception result of the perception failure of the target object includes: For any of the sub-modes, based on the simulation sub-rule corresponding to the sub-mode, the first perception result of the target object is subjected to a perception failure simulation process to obtain the second perception result of the target object in the sub-mode; The determining, based on the second perception result of the target object, the second perception result information corresponding to the operation scenario includes: Based on the second perception results respectively corresponding to the target object in each of the sub-modes, the second perception result information respectively corresponding to the operation scene in each of the sub-modes is determined.

4. The method according to claim 3, wherein: The verifying the preset vehicle function based on each of the second perception result information to obtain a verification result includes: Determining a functional response state of the preset vehicle function in each of the sub-modes of the operating scenario based on the second perception result information respectively corresponding to the operating scenario in each of the sub-modes; Determining a perception performance index threshold of the preset vehicle function in the operating scenario based on the function response state of the preset vehicle function in each of the sub-modes of the operating scenario; Based on the perception performance index threshold of the preset vehicle function in each of the operating scenarios, the target perception performance index threshold corresponding to the preset vehicle function is determined, and the target perception performance index threshold is used as the verification result.

5. The method according to claim 3, wherein: The performing perception failure simulation processing on the first perception result of the target object based on the simulation sub-rule corresponding to the sub-mode to obtain the second perception result of the target object under the sub-mode includes: Determining at least one perception performance indicator value based on the simulation sub-rule corresponding to the sub-mode; Based on each of the perception performance indicator values, the first perception result of the target object is updated to obtain the second perception result of the target object corresponding to each of the perception performance indicator values.

6. The method according to claim 5, wherein: The determining, based on the second perception results respectively corresponding to the target object in each of the sub-modes, the second perception result information respectively corresponding to the operation scene in each of the sub-modes includes: For any of the sub-modes, determining the second perception result information corresponding to each of the perception performance indicator values ​​based on the second perception result of the target object corresponding to each of the perception performance indicator values; The verifying the preset vehicle function based on each of the second perception result information to obtain a verification result includes: Based on the second perception result information respectively corresponding to each of the perception performance index values, determining the functional response state of the preset vehicle function respectively corresponding to each of the perception performance index values; Determining a perception performance index threshold of the preset vehicle function in the sub-mode based on the function response states of the preset vehicle function corresponding to each of the perception performance index values; Based on the perception performance index thresholds corresponding to the preset vehicle function in each of the sub-modes, the perception performance index threshold of the preset vehicle function in the operating scenario is determined, and the perception performance index threshold of the preset vehicle function in the operating scenario is used as the verification result.

7. The method according to any one of claims 2 to 6, wherein: The determining, based on the operation scenario data corresponding to each of the operation scenarios, first perception result information corresponding to the preset vehicle function of the vehicle in each of the operation scenarios, includes: For any of the operating scenarios, determining first perception result information of the ego vehicle at at least one trajectory point in the operating scenario based on the operating scenario data corresponding to the operating scenario; The determining, based on the first perception result information in the operation scenario, a target object that meets a first condition includes: Based on the first perception result information respectively corresponding to each of the trajectory points, the target objects in different areas around the vehicle are determined.

8. The method according to claim 7, wherein: The verifying the preset vehicle function based on each of the second perception result information to obtain a verification result includes: Based on each of the second perception result information, determining the perception performance index thresholds corresponding to the preset vehicle function in different areas around the vehicle; The perception performance index thresholds corresponding to the preset vehicle function in different areas around the vehicle are used as the verification results.

9. The method according to any one of claims 1 to 6, wherein: The verifying the preset vehicle function based on each of the second perception result information to obtain a verification result includes: For any of the operation scenarios, determining map data of the operation scenario based on the operation scenario data of the operation scenario; For any second perception result information of the operation scenario, based on the second perception result information and the map data of the operation scenario, and using the function algorithm corresponding to the preset vehicle function, determine the control information of the vehicle; Determining driving state information of the vehicle under the control of the control information based on the control information of the vehicle and a vehicle dynamics model; Determining a functional response state of the preset vehicle function under the second perception result information based on the driving state information and the operating scenario data corresponding to the second perception result information; The verification result is determined based on the functional response status of the preset vehicle function under each of the second perception result information.

10. The method according to claim 9, wherein: The determining, based on the driving state information and the operating scenario data corresponding to the second perception result information, a functional response state of the preset vehicle function under the second perception result information, includes: Determining a state of an expected result of the vehicle under the preset vehicle function based on the driving state information and the operating scenario data corresponding to the second perception result information; In response to the expected result being achieved, determining that the function response state of the preset vehicle function under the second perception result information is successful; or, In response to the status of the expected result being not achieved, it is determined that the functional response status of the preset vehicle function under the second perception result information is failure.

11. A vehicle function verification device, comprising: An acquisition module, used to acquire operation scenario data corresponding to each of the operation scenarios in at least one operation scenario; The operation scenario data includes the state information of the vehicle and the surrounding environment information of the vehicle; A first processing module, configured to determine first perception result information corresponding to a preset vehicle function of the vehicle in each of the operating scenarios based on the operating scenario data corresponding to each of the operating scenarios; A second processing module is used to perform perception failure simulation processing on each of the first perception result information based on the perception failure simulation rule to obtain second perception result information corresponding to each of the operation scenarios; The third processing module is used to verify the preset vehicle function based on each of the second perception result information to obtain a verification result.

12. A computer-readable storage medium storing a computer program, wherein the computer program is used to execute the vehicle function verification method according to any one of claims 1 to 10.

13. An electronic device, comprising: processor; a memory for storing instructions executable by the processor; The processor is used to read the executable instructions from the memory and execute the instructions to implement the vehicle function verification method described in any one of claims 1-10 above.

Images