Quantitative evaluation method and device for controllability of potential driving hazards

Abstract

The invention provides a quantitative evaluation method and device for controllability of potential driving hazards. The quantitative evaluation method comprises the following steps: acquiring operation data of a vehicle and other entities in a current driving scene; according to the operation data of the own vehicle, in combination with a kinematic model of the own vehicle, a time function of a control output quantity under the action of influence factors and motion constraints of the own vehicle, determining an own vehicle reachable domain in a preset control period from the current moment; the self-vehicle reachable region refers to a space range to which the self-vehicle can move according to the self-movement capability in the current driving scene; according to the operation data of other entities, determining an interaction reachable domain of the own vehicle in a preset control period from the reachable domains of the own vehicle; the interactive reachable domain refers to a space range to which the self-vehicle can actually move when the motion trails of other entities are considered. Afterwards, an objectively quantified controllability index is defined through the self-vehicle reachable domain and the interactive reachable domain, the ability of the vehicle to avoid potential hazards in a specific driving scene is evaluated, and precise evaluation of the controllability of the self-vehicle is achieved.

Description

Technical Field

[0001] The present application relates to the field of vehicle safety technology, and in particular to a quantitative evaluation method and device for the controllability of potential hazards in driving. Background Art

[0002] With the rapid development of autonomous driving technology, it has shown great potential in improving traffic efficiency, reducing traffic accidents and improving driving experience. However, the safety of autonomous vehicles (AVs) has always been a core issue of concern within and outside the industry. A series of safety issues of autonomous vehicles are defined as Safety of the Intended Functionality (SOTIF), which requires AVs to ensure that unreasonable risks caused by performance limitations can be avoided even in the absence of faults.

[0003] In actual driving road environments, vehicles may encounter various unexpected situations, including emergencies caused by the external environment and unforeseen behaviors that occur during the AVs' own decision-making process. Whether potential hazards are generated actively or passively, they pose a challenge to the safety of autonomous driving systems. In particular, when the behaviors and actions performed by the autonomous driving system involve interactions with the surrounding environment and other traffic participants, the degree of its controllability directly determines whether the system can effectively avoid risks and ensure driving safety. Controllability refers to the ability of the vehicle to avoid hazards when faced with them. It requires the system to be able to identify potential threats in a timely manner and to have sufficient flexibility to adjust strategies to meet the needs of different situations.

[0004] However, current research on the controllability of potential hazards in autonomous vehicle behavior lacks a reasonable method to quantify and evaluate the degree of such controllability, which makes it difficult for developers to objectively evaluate the safety level of their design solutions and increases the difficulty for regulators to formulate relevant standards and specifications. In addition, during the SOTIF analysis phase, test scenarios need to be quantitatively constructed based on identified hazards, which also involves the problem that the controllability of potential hazards during driving is difficult to reasonably quantify and evaluate. Summary of the invention

[0005] In view of this, the purpose of the present application is to provide a method and device for quantitatively evaluating the controllability of potential hazards during driving, so as to solve the problem that the controllability of potential hazards during driving in the prior art is difficult to reasonably quantify and evaluate.

[0006] The present application embodiment provides a quantitative evaluation method for the controllability of potential driving hazards, the method comprising:

[0007] Obtain the operating data of the vehicle and other entities in the current driving scenario;

[0008] According to the operation data of the ego vehicle, in combination with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors and the motion constraints of the ego vehicle, determine the reachable domain of the ego vehicle within the preset control period from the current moment; wherein the reachable domain of the ego vehicle refers to the spatial range to which the ego vehicle can move according to its own motion ability under the current driving scenario; the influencing factors include the control execution delay of the vehicle and / or the inhibition of the control output by the road environment; the time function of the control output is used to characterize the change of the control output over time;

[0009] According to the operation data of the other entities, determine the interactive reachable domain of the ego vehicle within the preset control period from the ego vehicle reachable domain; wherein the interactive reachable domain refers to the spatial range to which the ego vehicle can actually move when considering the motion trajectory of other entities;

[0010] According to the ego-vehicle reachable domain and the interactive reachable domain, an index value of a controllability index of the ego-vehicle at the current moment is determined.

[0011] Further, according to the operation data of the ego vehicle, combined with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors and the motion constraints of the ego vehicle, the reachable domain of the ego vehicle within the preset control period from the current moment is determined, including:

[0012] Constructing an actual kinematic model of the ego vehicle according to the kinematic model of the ego vehicle, the time function of the control output, and the motion constraints of the ego vehicle;

[0013] Substituting the running data of the ego vehicle at the current moment into the actual kinematic model, and calculating the preset control period backward to determine the spatial position of the ego vehicle within the preset control period;

[0014] The reachable domain of the ego vehicle is determined according to the spatial position of the ego vehicle within the preset control period.

[0015] Furthermore, when the influencing factor includes a control execution delay of the vehicle, the method further includes determining a time function of the control output under the influence of the influencing factor in the following manner:

[0016] Determine the driving mode of the ego vehicle;

[0017] When the driving mode is manual driving, the delay time is determined according to the human perception delay time and the human action delay time;

[0018] When the driving mode is automatic driving, determining the delay time according to the operation time of the automatic driving system;

[0019] The control output is corrected according to the delay time to obtain a time function of the control output.

[0020] Furthermore, when the influencing factor includes the inhibition of the control output by the environment, the method further includes determining the time function of the control output under the influence of the influencing factor in the following manner:

[0021] Obtain environmental data in the current driving scenario;

[0022] Determining a suppression coefficient according to the environmental data; wherein the suppression coefficient is used to characterize the suppression effect from the expected control output of the vehicle to the actual control effect of the vehicle due to the road environment;

[0023] The control output is corrected according to the suppression coefficient to obtain a time function of the control output.

[0024] Further, the motion constraint of the ego vehicle includes at least one of the following: a maximum constraint of the control output, an anti-rollover constraint, and a road constraint;

[0025] For the maximum constraint of the control output, the control output is greater than or equal to the maximum deceleration of the vehicle in the current driving scenario, and less than or equal to the maximum acceleration of the vehicle;

[0026] For the anti-rollover constraint, the lateral acceleration of the vehicle is less than or equal to the lateral acceleration threshold of the vehicle in the current driving scenario;

[0027] For the road constraint, if there is a physical obstruction at the road boundary in the current driving scenario, the vehicle position must not exceed the road boundary.

[0028] Further, according to the operation data of the other entities, determining the interactively reachable domain of the vehicle within the preset control period from the reachable domain of the vehicle includes:

[0029] Predicting the operation trajectory of other entities within the preset control period according to the operation data of the other entities;

[0030] The interactively reachable domain is obtained by removing the spatial intersection of the running trajectory and the reachable domain of the own vehicle from the reachable domain of the own vehicle.

[0031] Further, determining the index value of the controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain includes:

[0032] Performing rasterization and discretization processing on the lanes in the current driving scenario;

[0033] Determine a first grid corresponding to the self-vehicle reachable domain and a second grid corresponding to the interactive reachable domain;

[0034] The ratio of the number of the second grids to the number of the first grids is determined as the index value of the controllability index of the vehicle at the current moment; or, the ratio of the area of ​​the second grid to the area of ​​the first grid is determined as the index value of the controllability index of the vehicle at the current moment.

[0035] Further, the preset control cycle includes a plurality of control time nodes; determining the index value of the controllability index of the self-vehicle at the current moment according to the self-vehicle reachable domain and the interactive reachable domain, including:

[0036] For each control time node, according to the vehicle-accessible subdomain and the interactively-accessible subdomain corresponding to the control time node, determine the index value of the controllability index corresponding to the control time node;

[0037] The minimum index value among the index values ​​of the controllability index corresponding to each control time node is determined as the index value of the controllability index of the vehicle at the current moment.

[0038] The embodiment of the present application also provides a quantitative evaluation device for the controllability of potential driving hazards, the device comprising:

[0039] The acquisition module is used to obtain the operation data of the vehicle and other entities in the current driving scenario;

[0040] A first determination module is used to determine the reachable domain of the ego vehicle within a preset control period from the current moment based on the operation data of the ego vehicle, in combination with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors, and the motion constraints of the ego vehicle; wherein the reachable domain of the ego vehicle refers to the spatial range to which the ego vehicle can move according to its own motion ability under the current driving scenario; the influencing factors include the control execution delay of the vehicle and / or the inhibition of the control output by the environment; the time function of the control output is used to characterize the change of the control output over time;

[0041] A second determination module is used to determine the interactive reachable domain of the ego vehicle within the preset control period from the ego vehicle reachable domain according to the operation data of the other entities; wherein the interactive reachable domain refers to the spatial range to which the ego vehicle can actually move when considering the motion trajectory of other entities;

[0042] The third determination module is used to determine the index value of the controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain.

[0043] An embodiment of the present application also provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the quantitative assessment method for the controllability of potential driving hazards as described above are performed.

[0044] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the quantitative evaluation method for the controllability of potential driving hazards as described above are executed.

[0045] The quantitative evaluation method and device for the controllability of potential hazards in driving provided in the embodiment of the present application defines an objective and quantitative controllability index through the self-vehicle reachable domain and the interactive reachable domain to evaluate the vehicle's ability to avoid potential hazards in a specific driving scenario, thereby achieving an accurate evaluation of the controllability of the self-vehicle. Among them, the kinematic characteristics of the vehicle itself are taken into account, and motion constraints are introduced to fully reflect the vehicle's behavioral capabilities; further combined with the actual physical limitations of the vehicle when performing risk avoidance operations, the vehicle's control execution delay and / or the road environment's suppression of the control output are introduced to determine the time function of the control output, simulating the actual control process of the vehicle in risk avoidance operations, and ensuring the reasonable accuracy of the controllability index.

[0046] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0048] Figure 1 A flowchart of a quantitative evaluation method for controllability of potential driving hazards provided by an embodiment of the present application is shown;

[0049] Figure 2 A schematic diagram of the structure of a quantitative evaluation system for controllability of potential driving hazards provided by an embodiment of the present application is shown;

[0050] Figure 3 A schematic diagram of the structure of a device for quantitatively evaluating the controllability of potential driving hazards provided by an embodiment of the present application is shown;

[0051] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0052] To make the purpose, technical scheme and advantages of the embodiments of the present application clearer, the technical scheme in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application usually described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application claimed for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, each other embodiment obtained by those skilled in the art without making creative work belongs to the scope of protection of the present application.

[0053] Research has found that existing solutions for evaluating vehicle safety often divide different levels through a series of standards and rules set by experts, thereby attempting to provide a reference standard for the safety of autonomous driving systems. However, this hierarchical evaluation method based on expert experience has certain problems of unexplainability and non-repeatability in actual application. If applied to controllability research, the knowledge background and personal preferences of experts in different fields may lead to large differences in the understanding of the degree of controllability of vehicle behavior in the same scenario, which makes the final evaluation results lack consistency. In addition, since this method is qualitative rather than quantitative in nature, the explanation of controllability is not sufficient in terms of rationality, and it is difficult to provide design guidance for developers.

[0054] Therefore, current research on the controllability of potential hazards in the behavior of autonomous vehicles lacks a reasonable method to quantify and evaluate the degree of such controllability, which makes it difficult for developers to objectively evaluate the safety level of their design solutions and increases the difficulty for regulators to formulate relevant standards and specifications. In addition, during the SOTIF analysis phase, test scenarios need to be quantitatively constructed based on the identified hazards, which also involves the problem that the controllability of potential hazards during driving is difficult to reasonably quantify and evaluate.

[0055] Based on this, an embodiment of the present application provides a quantitative assessment method for the controllability of potential driving hazards, so as to evaluate the vehicle's ability to avoid potential hazards in specific driving scenarios through an objective and quantifiable controllability index, thereby achieving an accurate assessment of the vehicle's controllability.

[0056] See also Figure 1 , Figure 1A flowchart of a quantitative evaluation method for the controllability of potential driving hazards provided by an embodiment of the present application. Figure 1 As shown in, the evaluation method provided in the embodiment of the present application includes:

[0057] S101. Acquire the operating data of the vehicle and the operating data of other entities in the current driving scenario.

[0058] In this step, the operation data of the vehicle may include the location coordinates of the vehicle and its driving speed, and the operation data of other entities may include the location coordinates and speed information of other surrounding vehicles. In specific implementation, the data can be collected through the vehicle-mounted sensor system and V2X communication technology. The sensor system may include radar, lidar, camera, etc. During data collection, in order to ensure the accuracy and timeliness of data collection, the sensor must have high-precision positioning capabilities and fast response characteristics, and V2X communication should support low-latency information exchange to achieve real-time data updates.

[0059] S102: Determine a reachable domain of the ego vehicle within a preset control period from the current moment based on the operation data of the ego vehicle, in combination with the kinematic model of the ego vehicle, the time function of the control output under the influence of influencing factors, and the motion constraints of the ego vehicle.

[0060] Among them, the control period T can be set based on the computing power of the computing device equipped with the algorithm, but at least T must be ensured to be greater than the duration of the potential hazardous event. The self-vehicle reachable domain refers to the spatial range to which the self-vehicle can move according to its own movement ability in the current driving scenario; more specifically, the boundary of the self-vehicle reachable domain can be determined based on the maximum operation performed by the vehicle in the maximum avoidance capability, that is, the maximum lateral control, steering control, horizontal and longitudinal joint control, etc. The edge that can be moved to. This boundary can represent the upper limit of the operation and avoidance capabilities that the vehicle can perform in the current driving scenario, and thus the spatial range of the self-vehicle reachable domain can be determined based on this boundary.

[0061] The influencing factors include the control execution delay of the vehicle and / or the inhibition of the control output by the road environment; the time function of the control output is used to characterize the change of the control output over time.

[0062] Here, control execution delay refers to the necessary delay before the driver makes the expected response when avoiding risks in potential hazards and hazardous events. It should be noted that the driver here can be a human driver or an autonomous driving system. The inhibition of control output by the road environment refers to the inhibition of the vehicle's action output by the road environment. Generally speaking, the control output of the vehicle mainly includes the longitudinal speed change output, that is, the acceleration or deceleration of the vehicle, denoted as a t , and the lateral velocity change output of the vehicle, that is, the turning rate of the vehicle, denoted as ωt . t is the time.

[0063] S103: Determine, according to the operation data of the other entities, an interactively reachable domain of the own vehicle within the preset control period from the reachable domain of the own vehicle.

[0064] The interactive reachable domain refers to the spatial range to which the vehicle can actually move when considering the motion trajectory of other entities. It should be noted that in addition to the vehicle, the driving environment also includes other moving entities, such as other vehicles (including motor vehicles and non-motor vehicles), pedestrians, etc. The positions of other entities may also change within the preset control period, forming a motion trajectory and occupying the range of the vehicle's reachable domain.

[0065] Therefore, in this step, the motion trajectory can be predicted through the operation data of other entities, and then the spatial range occupied by the motion trajectory can be removed from the reachable domain of the ego vehicle to obtain the spatial range to which the ego vehicle can actually move, that is, the interactive reachable domain.

[0066] Furthermore, the driving environment also includes fixed entities, such as road facilities, obstacles, etc. When calculating the self-vehicle reachable domain and the interactive reachable domain, the space occupied by these fixed entities can also be removed to achieve a more accurate calculation effect.

[0067] S104: Determine an index value of a controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain.

[0068] In this step, the controllability index is defined by the ratio between the spatial range of the interactive reachable domain and the spatial range of the ego vehicle reachable domain. The ratio between the interactive reachable domain and the ego vehicle reachable domain is used to express the quantitative result of the ego vehicle being able to reach the safe area without colliding with other traffic participants under all possible operations, that is, controllability.

[0069] In this way, the quantitative evaluation method for the controllability of potential driving hazards provided in the embodiment of the present application defines an objective and quantitative controllability index through the reachable domain of the vehicle itself and the interactive reachable domain to evaluate the vehicle's ability to avoid potential hazards in specific driving scenarios, thereby achieving an accurate evaluation of the controllability of the vehicle itself.

[0070] In a possible implementation, step S102 may include:

[0071] According to the kinematic model of the ego vehicle, the time function of the control output and the motion constraints of the ego vehicle, an actual kinematic model of the ego vehicle is constructed; the operation data of the ego vehicle at the current moment is substituted into the actual kinematic model, and the preset control period is calculated backward to determine the spatial position of the ego vehicle within the preset control period; according to the spatial position of the ego vehicle within the preset control period, the reachable domain of the ego vehicle is determined.

[0072] Here, the kinematic model of the vehicle is used to describe the kinematic behavior of the vehicle, that is, the change of the vehicle's operating parameters over time. For example, the kinematic behavior of the vehicle can be described by a Markov model, a differential equation, or a random process (such as a Poisson process). Among them, if the vehicle's motion process is a Markov process, the vehicle motion control model can be described by the following formula:

[0073] x t+1 =x t +v t cos(θ t )Δt,

[0074] y t+1 =y t +v t sin(θ t )Δt,

[0075] v t+1 =v t +a t Δt,

[0076] θ t+1 =θ t +ω t Δt,

[0077] Among them, x t ,y t is the position of the vehicle at time t; v t is the speed of the vehicle; θ t is the heading angle of the vehicle (relative to the direction directly ahead; a t is the acceleration of the vehicle; ω t is the angular velocity (turning rate) of the vehicle; Δt is the time step, which represents a time interval in discrete time.

[0078] It is worth noting that when performing risk avoidance operations in the actual control process, the driver begins to perform lateral and longitudinal operations of the vehicle. Since the driver cannot complete the control of the vehicle in an instant, the acceleration and steering of the vehicle do not reach the maximum value immediately but change gradually, so a control time function is needed to simulate the driver's actual control operation. Considering this problem, the time function of the control output under the influence of the influencing factors can be determined in the following way in the embodiment of the present application.

[0079] Method 1: When the influencing factor includes the control execution delay of the vehicle, first determine the driving mode of the vehicle.

[0080] When the driving mode is manual driving, the delay time is determined according to the human perception delay time and the human action delay time. The human perception delay time refers to the delay time for the human body to perceive potential hazards and hazardous events, and the human action delay time refers to the delay time for the human body to react after perception, which can be determined by human experience data.

[0081] When the driving mode is automatic driving, the delay time is determined according to the operation time of the automatic driving system. The operation time of the automatic driving system includes the time for the automatic driving system to receive potential hazards and hazardous event signals, the time for calculating control signals according to the signals, and the time for outputting control signals, etc., which can be determined according to the computing power of the automatic driving system.

[0082] Afterwards, the control output is corrected according to the delay time to obtain a time function of the control output.

[0083] Here, the delay time can be substituted into the time function of the control response to achieve the correction of the control output. The form of the time function may include Sigmoid function, linear, exponential and polynomial function, etc. For example, assuming that the maximum value of the control output acceleration and steering increases with time according to the Sigmoid time function curve, the time function of the control output can be expressed as:

[0084]

[0085] Among them, a max and ω max are the maximum values ​​of acceleration and angular velocity respectively; t0 is the starting point of the response time after delay, that is, the delay time; k controls the steepness of the curve.

[0086] In addition, it is noted that during the actual control process, when performing risk avoidance operations, the control output cannot fully produce the corresponding actual control effect due to the influence of the road environment; for example, the deceleration effect expected by the driver is -4m / s 2 , that is, the acceleration of the expected control output is -4m / s 2 ; However, due to the slippery road surface, the actual deceleration effect of the vehicle is -3.5m / s 2 , at this time, the road environment has an inhibitory effect on the control output. Considering this problem, in the embodiment of the present application, the time function of the control output under the influence of the influencing factors can be determined in the following way.

[0087] Method 2: When the influencing factors include the inhibition of the control output by the environment, first obtain the environmental data in the current driving scenario. The environmental data may include the current weather, road conditions, etc., and may also be obtained based on the vehicle sensor system and V2X communication technology.

[0088] Afterwards, a suppression coefficient is determined according to the environmental data; wherein the suppression coefficient is used to characterize the suppression effect from the expected control output of the vehicle to the actual control effect of the vehicle due to the road environment.

[0089] Finally, the control output is corrected according to the suppression coefficient to obtain a time function of the control output.

[0090] For example, the time function of the control output can be expressed as: a(t)=a′ max =λa max ω(t)=ω′ max =λω max , λ is the friction coefficient, and its value range is 0<λ≤1. On a dry road, λ≈1; on a wet or icy road, λ drops to 0.3 or lower.

[0091] Further, considering the above two points, in the third approach, when the influencing factors include the control execution delay of the vehicle and the inhibition of the control output by the environment, the time function of the control output can be expressed as:

[0092]

[0093] On the other hand, the motion constraint of the ego vehicle includes at least one of the following: a maximum constraint of the control output, an anti-rollover constraint, and a road constraint;

[0094] For the maximum constraint of the control output, the control output is greater than or equal to the maximum deceleration of the vehicle under the current driving scenario, and less than or equal to the maximum acceleration of the vehicle; the formula is expressed as:

[0095] a′ min ≤a′ t ≤a′ max

[0096] -ω′ max ≤ω′ t ≤ω′ max

[0097] where a′ max is the maximum acceleration of the vehicle, a′ min is the maximum deceleration (negative acceleration during emergency braking); ω′ max is the maximum angular velocity that the vehicle can achieve. This constraint takes into account the road friction coefficient λ that suppresses the output.

[0098] The anti-rollover constraint is to ensure that the vehicle itself does not have the risk of rolling over when performing the combined lateral and longitudinal extreme operations to avoid risks. Therefore, the lateral acceleration of the vehicle is less than or equal to the lateral acceleration threshold of the vehicle in the current driving scenario. The critical lateral acceleration threshold can be defined based on the vehicle's center of gravity height h and wheelbase b, and the formula is expressed as where g is the acceleration due to gravity, a′ lat is the lateral acceleration of the vehicle taking output suppression into account.

[0099] The lateral acceleration a′ lat Through a′ lat =v t ·ω t Calculate that if the lateral acceleration exceeds this critical value, the vehicle may roll over, so set the rollover constraint:

[0100] For the road constraint, if there is a physical obstruction at the road boundary in the current driving scenario, the vehicle position must not exceed the road boundary.

[0101] Road constraints consider two situations: one is when there is a solid obstruction at the road boundary, the constraint is a hard constraint; the other is when there is no solid obstruction at the road boundary, the constraint is a soft constraint. Soft constraints mean that the vehicle can choose to exit the road area when performing emergency risk avoidance. At this time, this type of constraint is not involved in the reachable domain solution during the controllability calculation process.

[0102] Here, by obtaining environmental data, it is possible to determine whether there is a physical obstruction at the road boundary in the current driving scenario; if so, the lateral boundary of the road is set to [y min ,y max ], the road constraint means that the position of the vehicle needs to satisfy y min ≤y t ≤y max .

[0103] Optionally, constraints may be adjusted dynamically based on the vehicle state, or more complex physical models may be used to predict and constrain vehicle behavior.

[0104] In one example, the actual kinematic model of the constructed ego vehicle can be expressed as:

[0105]

[0106] The reachable domain refers to the total set of all road areas that a vehicle can reach under constraints. Specifically, the reachable domain is based on the vehicle's motion model. Within the constraints, all possible positions of the vehicle in each frame in the next T time steps are calculated. When time is continuous, the position of the vehicle is also continuous, so the reachable domain of the vehicle will diffuse over time, and there is a continuous boundary of the reachable domain.

[0107] In a possible implementation, step S103 may include:

[0108] According to the operation data of the other entities, the operation trajectory of the other entities within the preset control period is predicted; and the spatial intersection of the operation trajectory and the reachable domain of the own vehicle is removed from the reachable domain of the own vehicle to obtain the interactive reachable domain.

[0109] In this step, the position of other entities in the future preset control period can be predicted based on the prediction model and operation data, and the motion trajectory can be formed. The prediction model can be consistent with the motion model of the vehicle mentioned above. The prediction method infers the future trajectory based on the current position and speed of the entity to obtain an effective time series trajectory and future position.

[0110] Afterwards, the spatial intersection of the running trajectory and the ego-vehicle reachable domain is removed from the ego-vehicle reachable domain to obtain the interactive reachable domain.

[0111] Here, in order to facilitate calculation, an embodiment of the present application provides a road rasterization method.

[0112] Specifically, the roads are gridded, all reachable areas in the future T frames are discretized into grids, and each grid point is calculated whether it is within the constraints. The reachable domain is divided into n×m grids, each grid has a size of Δx and Δy. For each grid point (x, y), if there is a path from the initial state that satisfies all constraints to reach the point, then the point belongs to the ego vehicle reachable domain. Traverse all possible grid points and mark those points that meet the above conditions as part of the ego vehicle reachable domain:

[0113]

[0114] The Grid Space is the gridded road, and the Reachable Set describes a set that contains all grid points (x, y) that can be reached by the vehicle under given constraints. Specifically, the reachable domain of the vehicle is composed of those points that can be reached by a certain acceleration a′. t and angular velocity ω tThe combination of is the set of all (x, y) points that can be reached from the initial state in the next T frames and comply with all given constraints. In other words, if there is a path that allows the vehicle to move from the current state to a certain position (x, y) without violating any constraints in the process, then this position (x, y) is included in the reachable domain of the vehicle.

[0115] At the same time, the reachable domains of all other entities in the scene are represented based on the same grid representation method. According to the future positions of other entities in the scene, after aligning the predicted time frame, the grids in the reachable domain of the ego vehicle are marked as occupied. In a certain frame within the future T time, if a grid space in the frame reachable domain of the ego vehicle is occupied by other entities, the occupied grid is marked as unreachable.

[0116] Accordingly, step S104 may include:

[0117] The lanes in the current driving scenario are subjected to grid-based discretization processing; and a first grid corresponding to the self-vehicle reachable domain and a second grid corresponding to the interactive reachable domain are determined.

[0118] If a grid of uniform size is used for road division, the ratio of the number of the second grid to the number of the first grid can be determined as the index value of the controllability index of the vehicle at the current moment; or, grids of different sizes or an adaptive grid division method can be used, and the ratio of the area of ​​the second grid to the area of ​​the first grid can be determined as the index value of the controllability index of the vehicle at the current moment. In addition, the use of non-uniform grids or other spatial discretization techniques, such as quadtree or octree structures, can also be considered to improve efficiency or accuracy.

[0119] It should be noted that when real-time performance is not considered, non-rasterized methods can also be used to determine the reachable domain of the vehicle, such as using geometric methods (such as convex hulls) or trajectory optimization methods in continuous space. Although such methods are more complex, they can provide higher accuracy and flexibility in some cases.

[0120] Furthermore, the preset control period T includes a plurality of control time nodes t (prediction frames); then step S104 may also include:

[0121] For each control time node, the index value of the controllability index corresponding to the control time node is determined according to the subdomain reachable by the self-vehicle and the interactively reachable subdomain corresponding to the control time node; the minimum index value among the index values ​​of the controllability index corresponding to each control time node is determined as the index value of the controllability index of the self-vehicle at the current moment.

[0122] Assume that the occupied area of ​​the other car in the grid of time frame t is Then the controllability of the current prediction frame is defined as the proportion of the unoccupied reachable domain:

[0123]

[0124] in, It means that when t = T, the subdomain that the vehicle can reach under the prediction frame is is the number of grids in the subdomain that the vehicle can reach in the prediction frame. is the number of grids occupied by other cars in the prediction frame; is the number of grids in the subdomain that is reachable by the vehicle in the prediction frame and is not occupied by other vehicles, that is, the number of grids in the interactively reachable subdomain.

[0125] Similarly, the ratio between the interactive reachable domain and the ego vehicle reachable domain in each prediction frame represents the quantitative result of the ego vehicle being able to reach the safe area without colliding with other traffic participants under all possible operations at that time point, that is, controllability.

[0126] Furthermore, the controllability of the current frame t0 is calculated based on the reachable domain of the next T frames. For each frame t, the proportion of the reachable domain that is not occupied is calculated, and the minimum value among all frames is taken as the controllability of the current frame:

[0127]

[0128] The embodiment of the present application provides a quantitative evaluation method for the controllability of potential driving hazards. On the one hand, it proposes a method for calculating the vehicle's reachable domain that comprehensively considers execution delay, output suppression, control output maximum value constraints, road constraints, and anti-rollover constraints. By introducing a kinematic model and a control time function to simulate the actual control process of the driver or the automatic driving system in the risk avoidance operation, the controllability evaluation is closer to the actual situation. In addition, by considering the influence of the road friction coefficient, the accuracy of the evaluation under different road conditions is ensured. This method overcomes the limitations of the traditional method of relying solely on expert experience for qualitative grading, and provides an objective and quantifiable evaluation standard.

[0129] On the other hand, a gridding method is used to discretize all reachable areas in the future T frames into grids, and calculate whether each grid point is within the constraints. In this way, the calculation of the controllability index can be simplified and the calculation efficiency can be improved. By comparing the reachable domain of the ego vehicle with the predicted positions of other vehicles, the occupied grids are marked, and the controllability index of each frame is calculated accordingly, thereby achieving an accurate evaluation of the controllability of the ego vehicle.

[0130] See also Figure 2 , Figure 2This is a schematic diagram of the structure of a quantitative evaluation system for the controllability of potential driving hazards provided by an embodiment of the present application. The evaluation system can be run on a vehicle controller or a cloud platform. Figure 2 As shown in , the evaluation system provided in the embodiment of the present application includes: a data reading system 201 , a controllability calculation system 202 and an output and recording module 203 .

[0131] The data reading system 201 is responsible for collecting all necessary input information and providing basic data for subsequent calculations. The system consists of three submodules: a road information acquisition module, a vehicle information acquisition module, and a vehicle information acquisition module.

[0132] Among them, the road information acquisition module obtains the road parameters of the current driving environment, including lane width, curvature radius, road coverage, and road boundary type, through on-board sensors (such as cameras, lidar) and V2X communication technology. In addition, high-precision map data can be used to supplement or correct the information collected by the sensor. The other vehicle information acquisition module receives the position coordinates, speed, and other relevant information of other vehicles around through V2V (Vehicle-to-Vehicle) communication technology. In the absence of V2V communication, on-board radar, lidar, and visual sensors can be used to detect the surrounding traffic conditions. The self-vehicle information acquisition module directly reads key parameters such as position coordinates, speed, acceleration, and heading angle from the vehicle's own sensor system. These sensors usually include but are not limited to GPS positioning systems, inertial measurement units (IMUs), wheel speed sensors, etc. The data exchange between the above submodules can be carried out through CAN bus or Ethernet.

[0133] The controllability calculation system 202 performs real-time vehicle controllability calculation based on the data collected by the data reading system 201, including three parts: vehicle reachable domain calculation module, vehicle position prediction and vehicle controllability calculation.

[0134] Among them, the vehicle reachable domain calculation module calculates the reachable domain of the vehicle in the next T time steps based on the vehicle motion control model, taking into account execution delay, output suppression, maximum control output constraints, road boundary constraints, and anti-rollover constraints. The vehicle position prediction module predicts the future trajectory of the ego vehicle and the future positions of other vehicles in the scene. This step relies on the vehicle motion model and infers the future trajectory based on the current given state. The ego vehicle controllability calculation module compares the reachable domain of the ego vehicle with the predicted positions of other vehicles, marks the occupied grids by rasterization, and calculates the controllability index of each frame accordingly. Finally, the minimum value is selected as the controllability of the ego vehicle at the current moment. In this process, the selection of the grid size needs to take into account both computational efficiency and accuracy, and can generally be flexibly adjusted according to the actual application scenario.

[0135] The output and recording module 203 is responsible for outputting the calculated real-time controllability results of the vehicle to the driver or the automatic driving system, and storing the information for subsequent analysis. The output method can be an intuitive dashboard display, a sound prompt, or a more detailed visual information presented through the HMI interface. At the same time, in order to facilitate data analysis and accident investigation, all controllability information will be recorded, and the storage medium can be selected from the vehicle hard disk or cloud service.

[0136] See also Figure 3 , Figure 3 This is a schematic diagram of a device for quantitatively evaluating the controllability of potential driving hazards provided by an embodiment of the present application. Figure 3 As shown in , the quantitative evaluation device 300 includes:

[0137] An acquisition module 310 is used to acquire the operation data of the vehicle and the operation data of other entities in the current driving scenario;

[0138] The first determination module 320 is used to determine the reachable domain of the ego vehicle within a preset control period from the current moment based on the operation data of the ego vehicle, in combination with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors, and the motion constraints of the ego vehicle; wherein the reachable domain of the ego vehicle refers to the spatial range to which the ego vehicle can move according to its own motion ability under the current driving scenario; the influencing factors include the control execution delay of the vehicle and / or the inhibition of the control output by the environment; the time function of the control output is used to characterize the change of the control output over time;

[0139] A second determination module 330 is used to determine the interactive reachable domain of the ego vehicle within the preset control period from the ego vehicle reachable domain according to the operation data of the other entities; wherein the interactive reachable domain refers to the spatial range to which the ego vehicle can actually move when considering the motion trajectory of other entities;

[0140] The third determination module 340 is used to determine the index value of the controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain.

[0141] Further, when the first determination module 320 is used to determine the reachable domain of the ego vehicle within a preset control period from the current moment based on the operation data of the ego vehicle, in combination with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors, and the motion constraints of the ego vehicle, the first determination module 320 is used to:

[0142] Constructing an actual kinematic model of the ego vehicle according to the kinematic model of the ego vehicle, the time function of the control output, and the motion constraints of the ego vehicle;

[0143] Substituting the running data of the ego vehicle at the current moment into the actual kinematic model, and calculating the preset control period backward to determine the spatial position of the ego vehicle within the preset control period;

[0144] The reachable domain of the ego vehicle is determined according to the spatial position of the ego vehicle within the preset control period.

[0145] Further, when the influencing factor includes the control execution delay of the vehicle, the quantitative evaluation device 340 includes: a function determination module; the function determination module is used to determine the time function of the control output under the influence of the influencing factor in the following manner:

[0146] Determine the driving mode of the ego vehicle;

[0147] When the driving mode is manual driving, the delay time is determined according to the human perception delay time and the human action delay time;

[0148] When the driving mode is automatic driving, determining the delay time according to the operation time of the automatic driving system;

[0149] The control output is corrected according to the delay time to obtain a time function of the control output.

[0150] Further, when the influencing factor includes the inhibition of the control output by the environment, the function determination module is further used to determine the time function of the control output under the influence of the influencing factor in the following manner:

[0151] Obtain environmental data in the current driving scenario;

[0152] Determining a suppression coefficient according to the environmental data; wherein the suppression coefficient is used to characterize the suppression effect from the expected control output of the vehicle to the actual control effect of the vehicle due to the road environment;

[0153] The control output is corrected according to the suppression coefficient to obtain a time function of the control output.

[0154] Further, the motion constraint of the ego vehicle includes at least one of the following: a maximum constraint of the control output, an anti-rollover constraint, and a road constraint;

[0155] For the maximum constraint of the control output, the control output is greater than or equal to the maximum deceleration of the vehicle in the current driving scenario, and less than or equal to the maximum acceleration of the vehicle;

[0156] For the anti-rollover constraint, the lateral acceleration of the vehicle is less than or equal to the lateral acceleration threshold of the vehicle in the current driving scenario;

[0157] For the road constraint, if there is a physical obstruction at the road boundary in the current driving scenario, the vehicle position must not exceed the road boundary.

[0158] Further, when the second determination module 330 is used to determine the interactively reachable domain of the vehicle within the preset control period from the reachable domain of the vehicle according to the operation data of the other entity, the second determination module 330 is used to:

[0159] Predicting the operation trajectory of other entities within the preset control period according to the operation data of the other entities;

[0160] The interactively reachable domain is obtained by removing the spatial intersection of the running trajectory and the reachable domain of the own vehicle from the reachable domain of the own vehicle.

[0161] Furthermore, when the third determination module 340 is used to determine the index value of the controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain, the third determination module is used to:

[0162] Performing rasterization and discretization processing on the lanes in the current driving scenario;

[0163] Determine a first grid corresponding to the self-vehicle reachable domain and a second grid corresponding to the interactive reachable domain;

[0164] The ratio of the number of the second grids to the number of the first grids is determined as the index value of the controllability index of the vehicle at the current moment; or, the ratio of the area of ​​the second grid to the area of ​​the first grid is determined as the index value of the controllability index of the vehicle at the current moment.

[0165] Further, the preset control cycle includes a plurality of control time nodes; when the third determination module 340 is used to determine the index value of the controllability index of the self-vehicle at the current moment according to the self-vehicle reachable domain and the interactive reachable domain, the third determination module is used to:

[0166] For each control time node, according to the vehicle-accessible subdomain and the interactively-accessible subdomain corresponding to the control time node, determine the index value of the controllability index corresponding to the control time node;

[0167] The minimum index value among the index values ​​of the controllability index corresponding to each control time node is determined as the index value of the controllability index of the vehicle at the current moment.

[0168] See also Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 4As shown in , the electronic device 400 includes a processor 410 , a memory 420 and a bus 430 .

[0169] The memory 420 stores machine-readable instructions executable by the processor 410. When the electronic device 400 is running, the processor 410 communicates with the memory 420 via the bus 430. When the machine-readable instructions are executed by the processor 410, the above-mentioned Figure 1 The steps of the method for quantitatively evaluating the controllability of potential driving hazards in the method embodiment shown in the embodiment, and the specific implementation method can be referred to the method embodiment, which will not be repeated here.

[0170] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program can execute the above-mentioned Figure 1 The steps of the method for quantitatively evaluating the controllability of potential driving hazards in the method embodiment shown in the embodiment, and the specific implementation method can be referred to the method embodiment, which will not be repeated here.

[0171] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0172] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0173] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0174] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0175] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present application can essentially be embodied in the form of a software product, or in other words, the part that contributes to the prior art or the part of the technical solution. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0176] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present application, which are used to illustrate the technical solutions of the present application, rather than to limit them. The protection scope of the present application is not limited thereto. Although the present application is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed in the present application, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A quantitative evaluation method for the controllability of potential driving hazards, characterized in that: The method comprises: Obtain the operating data of the vehicle and other entities in the current driving scenario; According to the operation data of the ego vehicle, in combination with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors and the motion constraints of the ego vehicle, determine the reachable domain of the ego vehicle within the preset control period from the current moment; wherein the reachable domain of the ego vehicle refers to the spatial range to which the ego vehicle can move according to its own motion ability under the current driving scenario; the influencing factors include the control execution delay of the vehicle and / or the inhibition of the control output by the road environment; the time function of the control output is used to characterize the change of the control output over time; According to the operation data of the other entities, determine the interactive reachable domain of the ego vehicle within the preset control period from the ego vehicle reachable domain; wherein the interactive reachable domain refers to the spatial range to which the ego vehicle can actually move when considering the motion trajectory of other entities; According to the ego-vehicle reachable domain and the interactive reachable domain, an index value of a controllability index of the ego-vehicle at the current moment is determined.

2. The method according to claim 1, characterized in that: According to the operation data of the ego vehicle, combined with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors and the motion constraints of the ego vehicle, the reachable domain of the ego vehicle within the preset control period from the current moment is determined, including: Constructing an actual kinematic model of the ego vehicle according to the kinematic model of the ego vehicle, the time function of the control output, and the motion constraints of the ego vehicle; Substituting the running data of the ego vehicle at the current moment into the actual kinematic model, and calculating the preset control period backward to determine the spatial position of the ego vehicle within the preset control period; The reachable domain of the ego vehicle is determined according to the spatial position of the ego vehicle within the preset control period.

3. The method according to claim 1, characterized in that When the influencing factor includes a control execution delay of the vehicle, the method further includes determining a time function of the control output under the influence of the influencing factor in the following manner: Determine the driving mode of the ego vehicle; When the driving mode is manual driving, the delay time is determined according to the human perception delay time and the human action delay time; When the driving mode is automatic driving, determining the delay time according to the operation time of the automatic driving system; The control output is corrected according to the delay time to obtain a time function of the control output.

4. The method according to claim 1 or 3, characterized in that: When the influencing factor includes the inhibition of the control output by the environment, the method further includes determining the time function of the control output under the influence of the influencing factor in the following manner: Obtain environmental data in the current driving scenario; Determining a suppression coefficient according to the environmental data; wherein the suppression coefficient is used to characterize the suppression effect from the expected control output of the vehicle to the actual control effect of the vehicle due to the road environment; The control output is corrected according to the suppression coefficient to obtain a time function of the control output.

5. The method according to claim 1, characterized in that: The motion constraint of the vehicle includes at least one of the following: a maximum constraint of the control output, an anti-rollover constraint, and a road constraint; For the maximum constraint of the control output, the control output is greater than or equal to the maximum deceleration of the vehicle in the current driving scenario, and less than or equal to the maximum acceleration of the vehicle; For the anti-rollover constraint, the lateral acceleration of the vehicle is less than or equal to the lateral acceleration threshold of the vehicle in the current driving scenario; For the road constraint, if there is a physical obstruction at the road boundary in the current driving scenario, the vehicle position must not exceed the road boundary.

6. The method according to claim 1, characterized in that Determining the interactively reachable domain of the vehicle within the preset control period from the reachable domain of the vehicle according to the operation data of the other entities includes: Predicting the operation trajectory of other entities within the preset control period according to the operation data of the other entities; The interactively reachable domain is obtained by removing the spatial intersection of the running trajectory and the reachable domain of the own vehicle from the reachable domain of the own vehicle.

7. The method according to claim 1, characterized in that Determining an index value of a controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain includes: Performing rasterization and discretization processing on the lanes in the current driving scenario; Determine a first grid corresponding to the self-vehicle reachable domain and a second grid corresponding to the interactive reachable domain; The ratio of the number of the second grids to the number of the first grids is determined as the index value of the controllability index of the vehicle at the current moment; or, the ratio of the area of ​​the second grid to the area of ​​the first grid is determined as the index value of the controllability index of the vehicle at the current moment.

8. The method according to claim 1, characterized in that: The preset control cycle includes a plurality of control time nodes; Determining an index value of a controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain includes: For each control time node, according to the vehicle-accessible subdomain and the interactively-accessible subdomain corresponding to the control time node, determine the index value of the controllability index corresponding to the control time node; The minimum index value among the index values ​​of the controllability index corresponding to each control time node is determined as the index value of the controllability index of the vehicle at the current moment.

9. A quantitative evaluation device for the controllability of potential driving hazards, characterized in that: The device comprises: The acquisition module is used to obtain the operation data of the vehicle and other entities in the current driving scenario; A first determination module is used to determine the reachable domain of the ego vehicle within a preset control period from the current moment based on the operation data of the ego vehicle, in combination with the kinematic model of the ego vehicle, the time function of the control output under the influence of the influencing factors, and the motion constraints of the ego vehicle; wherein the reachable domain of the ego vehicle refers to the spatial range to which the ego vehicle can move according to its own motion ability under the current driving scenario; the influencing factors include the control execution delay of the vehicle and / or the inhibition of the control output by the environment; the time function of the control output is used to characterize the change of the control output over time; A second determination module is used to determine the interactive reachable domain of the ego vehicle within the preset control period from the ego vehicle reachable domain according to the operation data of the other entities; wherein the interactive reachable domain refers to the spatial range to which the ego vehicle can actually move when considering the motion trajectory of other entities; The third determination module is used to determine the index value of the controllability index of the ego vehicle at the current moment according to the ego vehicle reachable domain and the interactive reachable domain.

10. An electronic device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate via the bus. When the processor is running, the machine-readable instructions execute the steps of the method for quantitatively evaluating the controllability of potential driving hazards as described in any one of claims 1 to 8.

Images