A method, device, electronic device and storage medium for quantifying collision risk

Abstract

The present invention relates to the field of autonomous driving technology, and in particular to a method, device, electronic device and storage medium for quantifying collision risk, the method comprising: predicting collision scenarios that meet preset collision scenario rules and target state position information corresponding to target vehicles and traffic participants in the collision time period corresponding to the collision scenarios based on initial state position information corresponding to target vehicles and traffic participants in the future time period; calculating the influence index of traffic participants on the driving path of the target vehicle in each collision scenario based on the type of things corresponding to the traffic participants and the target state position information corresponding to the target vehicles and traffic participants in the collision time period corresponding to each collision scenario; calculating the collision risk of the target vehicle in each collision scenario based on the influence index of all traffic participants on the driving path of the target vehicle in each collision scenario. Through the method of the present application, the collision risk of a self-driving car can be effectively quantified in a real scenario.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a collision risk quantification method, device, electronic device and storage medium. Background Art

[0002] With the rapid development of autonomous driving technology, ensuring the safety of autonomous vehicles in various complex traffic environments has become a focus of industry attention. Collisions of autonomous vehicles involve a variety of dynamic factors, such as vehicles traveling in different directions, pedestrians, bicycles, and changes in traffic lights.

[0003] Therefore, it is particularly important to effectively quantify the collision risk of autonomous vehicles in real scenarios. Summary of the invention

[0004] In view of this, the purpose of the present application is to provide a collision risk quantification method, device, electronic device and storage medium, which can effectively quantify the collision risk of a driving car in a real scenario.

[0005] In a first aspect, an embodiment of the present application provides a method for quantifying collision risk, the method comprising:

[0006] Acquire initial state position information corresponding to the target vehicle and traffic participants in a future time period; the traffic participants refer to objects that have a potential collision risk with the target vehicle in the current driving scenario;

[0007] According to the initial state position information, predicting a collision scene that meets a preset collision scene rule and target state position information corresponding to the target vehicle and the traffic participant respectively within a collision time period corresponding to the collision scene; the preset collision scene rule combines the interaction time and interaction position between the target vehicle and the traffic participant;

[0008] Calculate the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene;

[0009] The collision risk of the target vehicle in each collision scenario is calculated according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle.

[0010] In a possible implementation, predicting, based on the initial state position information, a collision scene that meets a preset collision scene rule and target state position information corresponding to the target vehicle and the traffic participant within a collision time period corresponding to the collision scene, includes:

[0011] Predicting an initial collision scene that meets a preset collision scene rule, and target state position information corresponding to the target vehicle and the traffic participant respectively within a collision time period corresponding to the initial collision scene, based on the initial state position information;

[0012] The initial collision scenes with overlapping or continuous collision time periods are merged to obtain the target collision scene and the corresponding collision time period.

[0013] In a possible implementation, predicting, based on the initial state position information, an initial collision scene that meets a preset collision scene rule and target state position information corresponding to the target vehicle and the traffic participant within a collision time period corresponding to the initial collision scene, includes:

[0014] For each traffic participant whose thing type is an entity type, the earliest time point corresponding to the initial state position information of the target vehicle and the traffic participant at the same time that meets the preset collision scene rule is determined as the starting time point of the initial collision scene;

[0015] According to the initial state position information of the target vehicle and the traffic participant respectively corresponding to the starting time point, predicting the target state position information of the target vehicle and the traffic participant respectively corresponding to the target vehicle and the traffic participant within a preset time period starting from the starting time point;

[0016] The earliest time point corresponding to the target state position information of the target vehicle and the traffic participant at the same time that does not conform to the preset collision scene rule is determined as the end time point of the initial collision scene;

[0017] A collision time period corresponding to the initial collision scenario is determined according to the starting time point and the ending time point of the initial collision scenario.

[0018] In a possible implementation manner, the initial state position information includes initial position information, and the target state position information includes target position information.

[0019] In a possible implementation manner, the transaction type includes an entity type and a non-entity type; the entity type includes a static entity type and a dynamic entity type.

[0020] In a possible implementation, the calculation of the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene includes:

[0021] For traffic participants whose object type is a dynamic entity type in each collision scene, the target state position information corresponding to the target vehicle and the traffic participant at each collision time point within the collision time period corresponding to the collision scene is substituted into the following formula to obtain the travel intention potential field function value of the collision scene at each collision time point;

[0022] ;

[0023] ;

[0024] ;

[0025] ;

[0026] ;

[0027] in, is the potential field function of the moving intention, is the static potential field function, is the mass of the traffic participant, is the attenuation coefficient of the moving intention potential field, is the relative distance between the target vehicle and the traffic participant calculated based on the position information in the target state position information, is the relative angle between the target vehicle and the traffic participant calculated based on the steering angle information in the target state position information, is the directional function, is the longitudinal potential field strength of the collision scene, is the transverse potential field strength of the collision scene, is the ordinate of the target vehicle's position information, is the horizontal coordinate of the target vehicle’s position information, is the ordinate of the position information of the traffic participant, is the horizontal coordinate of the position information of the traffic participant, is the steep drop coefficient in the x direction of the potential field space, is the steep drop coefficient in the y direction of the potential field space.

[0028] The travel intention potential field function value of the collision scene at each collision time point is determined as the influence index of the traffic participant on the driving path of the target vehicle at each collision time point in the collision scene.

[0029] In a possible implementation, the method further includes:

[0030] Obtain the traffic light status and stop line position information at each collision time point in each collision scenario;

[0031] For each collision time point in each collision scenario, judging whether the target vehicle can pass according to the corresponding traffic light state, the position information of the stop line, and the position information in the target state position information of the target vehicle;

[0032] If the target vehicle can pass, the fixed potential field value corresponding to the passable state is determined as the influence index of the pass constraint on the driving path of the target vehicle;

[0033] If the target vehicle is prohibited from passing, the fixed potential field value corresponding to the prohibited passing state is determined as the influence index of the passing constraint on the driving path of the target vehicle;

[0034] The collision risk of the target vehicle in each collision scenario is calculated according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle and the influence index of the traffic constraint on the driving path of the target vehicle.

[0035] In a second aspect, an embodiment of the present application further provides a collision risk quantification device, the device comprising:

[0036] An acquisition module is used to acquire initial state position information corresponding to the target vehicle and traffic participants in a future time period; the traffic participants refer to objects that have a potential collision risk with the target vehicle in the current driving scenario;

[0037] A prediction module, configured to predict, based on the initial state position information, a collision scene that complies with a preset collision scene rule and target state position information corresponding to the target vehicle and the traffic participant within a collision time period corresponding to the collision scene; the preset collision scene rule combines the interaction time and interaction position between the target vehicle and the traffic participant;

[0038] A calculation module, used to calculate the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene;

[0039] The calculation module is further used to calculate the collision risk of the target vehicle in each collision scenario according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle.

[0040] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor communicates with the storage medium through the bus, and the processor executes the machine-readable instructions to perform the steps of the collision risk quantification method as described in any one of the first aspects.

[0041] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for quantifying collision risk as described in any one of the first aspects are executed.

[0042] The embodiment of the present application provides a method, device, electronic device and storage medium for quantifying collision risk, the method comprising: obtaining initial state position information corresponding to the target vehicle and traffic participants in the future time period; the traffic participant refers to an object with potential collision risk with the target vehicle in the current driving scene; based on the initial state position information, predicting the collision scene that meets the preset collision scene rules and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to the collision scene; the preset collision scene rules combine the interaction time and interaction position between the target vehicle and the traffic participant; according to the object type corresponding to the traffic participant, the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene, calculating the influence index of the traffic participant on the driving path of the target vehicle in each collision scene; according to the influence index of all traffic participants on the driving path of the target vehicle in each collision scene, calculating the collision risk of the target vehicle in each collision scene. Through the method of the present application, the collision risk of a driving car can be effectively quantified in a real scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] 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.

[0044] Figure 1 A flowchart of a collision risk quantification method provided in an embodiment of the present application is shown;

[0045] Figure 2 A flowchart of another collision risk quantification method provided in an embodiment of the present application is shown;

[0046] Figure 3 A schematic diagram of the structure of a collision risk quantification device provided in an embodiment of the present application is shown;

[0047] Figure 4 A schematic structural diagram of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0048] 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. It should be understood that the drawings in the present application only serve the purpose of explanation and description and are not used to limit the scope of protection of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in this application shows the operations implemented according to some embodiments of the present application. It should be understood that the operations of the flowchart can be implemented out of sequence, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the present application, or remove one or more operations from the flowchart.

[0049] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various 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, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present application.

[0050] In order to enable those skilled in the art to use the contents of this application, the following implementation is provided in conjunction with a specific application scenario, the "field of autonomous driving technology". For those skilled in the art, the general principles defined herein may be applied to other embodiments and application scenarios without departing from the spirit and scope of this application. Although this application is mainly described around the "field of autonomous driving technology", it should be understood that this is only an exemplary embodiment.

[0051] It should be noted that the term "comprising" will be used in the embodiments of the present application to indicate the existence of the features declared thereafter, but does not exclude the addition of other features.

[0052] A method for quantifying collision risk provided in an embodiment of the present application is described in detail below.

[0053] Reference Figure 1As shown, it is a flowchart of a collision risk quantification method provided in an embodiment of the present application. The exemplary steps of the embodiment of the present application are described below:

[0054] S101. Obtain initial state position information corresponding to target vehicles and traffic participants in a future time period.

[0055] In the implementation mode of the present application, the target vehicle refers to a vehicle for which collision risk quantification is required, which may be a human-driven vehicle equipped with an environmental sensing device or a smart car equipped with an automatic driving function. Traffic participants refer to objects that have a potential collision risk with the target vehicle in the current driving scenario; the object types corresponding to traffic participants include entity types (such as motor vehicles, non-motor vehicles, pedestrians, road facilities, etc.) and non-entity types (such as road boundaries, non-compliant lanes); the entity types include static entity types and dynamic entity types. The initial state position information includes initial state information and initial position information; the initial state information may specifically include speed, acceleration, steering angle, etc.; the initial position information includes abscissa and ordinate. The initial state position information is used to describe the position and motion state of the target vehicle and traffic participants. The initial state position information is predicted based on the current state position information in the current driving scenario. The initial state position information corresponding to the target vehicle and the initial state position information corresponding to the traffic participants in the future time period are obtained.

[0056] The method for predicting the initial state position information can be based on a single-vehicle model, or other prediction techniques can be used to obtain similar prediction results, such as machine learning-based methods (such as recursive neural networks (RNN) or long short-term memory networks (LSTM)), nonlinear system state estimation methods such as particle filters, etc.

[0057] Specifically, obtaining the initial state position information corresponding to the target vehicle or traffic participant in the future time period T includes: Substitute the initial state position information of the target vehicle or traffic participant at the time into the first bicycle model below to obtain The initial state position information corresponding to the target vehicle or traffic participant at the moment; The initial value is ; = , is the preset sampling step (such as 0.1 second); jump to Substitute the initial state position information of the target vehicle or traffic participant at the time into the bicycle model and obtain The execution continues until the initial state position information corresponding to the target vehicle or traffic participant at the moment Execution stops when for the current moment; The initial state position information corresponding to the target vehicle or traffic participant is the current state position information corresponding to the target vehicle or traffic participant.

[0058] Furthermore, the expression of the first bicycle model is as follows:

[0059] ;

[0060] ;

[0061] in, for The horizontal coordinate in the initial position information at the moment, for The horizontal coordinate in the initial position information corresponding to the target vehicle or traffic participant at the moment, for The speed of the target vehicle or traffic participant in the initial state information corresponding to the time, for The steering angle in the initial state information corresponding to the target vehicle or traffic participant at the moment, for The acceleration in the initial state information corresponding to the target vehicle or traffic participant at the moment, for The ordinate in the initial position information corresponding to the target vehicle or traffic participant at the moment, for The ordinate in the initial position information corresponding to the target vehicle or traffic participant at the moment.

[0062] Here, the traditional collision risk quantification method only focuses on the collision risk between vehicles, ignoring the risk differences of other types of traffic participants such as pedestrians and bicycles. The embodiment of the present application takes into account all traffic participants who have potential collision risks with the target vehicle in the current driving scenario.

[0063] S102. Predicting, based on the initial state position information, a collision scene that meets a preset collision scene rule and target state position information corresponding to a target vehicle and a traffic participant within a collision time period corresponding to the collision scene.

[0064] In the implementation mode of the present application, the preset collision scene rules combine the interaction time and interaction position between the target vehicle and the traffic participant, and can accurately identify the collision scene. The collision scene is a scene where the target vehicle has a certain collision risk. It exists in the real world but the probability of occurrence is not high. It can quickly retain the functional deficiencies of the autonomous driving car and is of great value for the testing of the autonomous driving system. The target state position information includes target state information and target position information; the target state information may specifically include speed, acceleration, steering angle, etc.; the target position information includes abscissa and ordinate.

[0065] Specifically, the following steps are used to predict the collision scene that meets the preset collision scene rules and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to the collision scene according to the initial state position information:

[0066] Step 1: predict an initial collision scene that meets the preset collision scene rules based on the initial state position information, as well as the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to the initial collision scene.

[0067] In the implementation manner of the present application, the initial collision scenario refers to a scenario in which a traffic participant whose object type is an entity type has a risk of collision with a target vehicle; there may be one initial collision scenario or multiple initial collision scenarios.

[0068] Furthermore, the following steps are used to predict an initial collision scene that meets the preset collision scene rules, and target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to the initial collision scene:

[0069] a. For each traffic participant whose thing type is an entity type, the earliest time point corresponding to the initial state position information of the target vehicle and the traffic participant at the same time that meets the preset collision scene rules is determined as the starting time point of the initial collision scene.

[0070] b. Based on the initial state position information of the target vehicle and the traffic participant respectively corresponding to the starting time point, predict the target state position information of the target vehicle and the traffic participant respectively corresponding to the preset time length starting from the starting time point.

[0071] In the implementation manner of the present application, the step of predicting the target state position information corresponding to the target vehicles and traffic participants within a preset time period starting from the starting time point is the same as the step of predicting the initial state position information corresponding to the target vehicles and traffic participants in the future time period, so they will not be repeated here.

[0072] c. The earliest time point corresponding to the target state position information of the target vehicle and the traffic participant at the same time that does not conform to the preset collision scenario rules is determined as the end time point of the initial collision scenario.

[0073] d. Determine the collision time period corresponding to the initial collision scene according to the starting time point and the ending time point of the initial collision scene.

[0074] For example, the starting time point of the initial collision scene is 3:00:00 and the ending time point is 3:00:10, then the collision time period corresponding to the initial collision scene is 3:00:00 to 3:00:10.

[0075] Further, the initial state position information includes initial position information, and the target state position information includes target position information.

[0076] Here, when the rectangular frame corresponding to the traffic participant and the rectangular frame corresponding to the target vehicle overlap at the same time, the preset collision scene rule is satisfied, that is, the traffic participant and the target vehicle collide.

[0077] Step 2: Merge the initial collision scenes with overlapping or continuous collision time periods to obtain the target collision scene and the corresponding collision time period.

[0078] In an embodiment of the present application, if the collision time periods corresponding to different initial collision scenes overlap or are continuous, it means that these initial collision scenes are the same collision scene. Therefore, the initial collision scenes with overlapping or continuous collision time periods are merged to obtain the target collision scene and the corresponding collision time period.

[0079] For example, the collision time period corresponding to the initial collision scene 1 is 3:00:00 to 3:00:10, the collision time period corresponding to the initial collision scene 2 is 3:00:11 to 3:00:13, the collision time period corresponding to the initial collision scene 3 is 3:00:15 to 3:00:20, and the collision time period corresponding to the initial collision scene 4 is 3:00:19 to 3:00:21. Since the collision time periods corresponding to the initial collision scene 1 and the initial collision scene 2 are continuous, the initial collision scene 1 and the initial collision scene 2 are merged to obtain the target collision scene with a collision time period of 3:00:00 to 3:00:13. Since the collision time periods corresponding to the initial collision scene 3 and the initial collision scene 4 overlap, the initial collision scene 3 and the initial collision scene 4 are merged to obtain the target collision scene with a collision time period of 3:00:15 to 3:00:21.

[0080] S103, calculating the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant, the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene.

[0081] In the implementation manner of the present application, the greater the influence index of the traffic participant on the driving path of the target vehicle, the higher the collision risk of the target vehicle in the collision scenario. Different types of traffic participants correspond to different calculation formulas.

[0082] Specifically, the influence index of the traffic participants on the driving path of the target vehicle in each collision scenario is calculated, including:

[0083] Step 1: For traffic participants whose object type is a dynamic entity type in each collision scene, the target state position information corresponding to the target vehicle and the traffic participant at each collision time point in the collision time period corresponding to the collision scene is substituted into the following formula to obtain the travel intention potential field function value of the collision scene at each collision time point;

[0084] ;

[0085] ;

[0086] ;

[0087] ;

[0088] ;

[0089] in, is the potential field function of the moving intention, is the static potential field function, is the quality of traffic participants (reflecting the risk weight of traffic participants), is the attenuation coefficient of the moving intention potential field, is the relative distance between the target vehicle and the traffic participant calculated based on the position information in the target state position information, is the relative angle between the target vehicle and the traffic participant calculated based on the steering angle information in the target state position information, is the directivity function (used to adjust the distribution of field strength in different directions), is the longitudinal potential field strength of the collision scene, is the lateral potential field strength of the collision scene, is the ordinate of the target vehicle’s location information, is the horizontal coordinate of the target vehicle’s location information, is the ordinate of the location information of the traffic participant, is the horizontal coordinate of the location information of the traffic participant, is the steep drop coefficient in the x direction of the potential field space, is the steep drop coefficient in the y direction of the potential field space.

[0090] It is specifically used to simulate the directional diffusion characteristics of the travel intention potential field. For example, in the direction of travel, The larger the value of is, the slower the field strength drops in that direction. On the side or in other directions, the field strength is relatively weak and drops quickly. When it is close to 0 (i.e. the direction of the traffic participant is consistent with that of the target vehicle), the field strength is maximum; As the direction of the target vehicle increases (i.e., the traffic participant deviates from the target vehicle), the field strength gradually decreases.

[0091] Here, it can be seen from the above formula that the moving intention potential field reflects the impact of the static potential field on the target vehicle, and also reflects the potential risks of the target vehicle and other traffic participants in the direction of travel. The moving intention potential field is different from the static potential field and has obvious directionality. The driving direction of the target vehicle and the traffic participants determines the diffusion form of the moving intention potential field. The moving intention potential field mainly expands along the direction of travel and weights the direction where the conflict may occur. In order to simulate this asymmetry, the field strength of the moving intention potential field will be larger in front of the target vehicle, and relatively weaker on the side and rear.

[0092] The prior art usually adopts an artificial potential field method based on a position field or a velocity field. Although this method can provide a certain risk assessment, it only considers the position and velocity information of traffic participants and ignores the important factor of travel intention. Therefore, the artificial potential field method cannot accurately reflect the real-time risk situation. This single risk quantification method may lead to underestimation or overestimation of certain potential risks, thereby affecting the decision-making quality of the autonomous driving vehicle. The embodiment of the present application takes into account the travel intention between the dynamic entity type of traffic participants and the target vehicle when calculating the influence index of the dynamic entity type of traffic participants on the driving path of the target vehicle, which can improve the accuracy of collision risk quantification and improve the decision-making quality of the autonomous driving vehicle.

[0093] Step 2: Determine the travel intention potential field function value of the collision scene at each collision time point as the influence index of the traffic participant on the driving path of the target vehicle at each collision time point in the collision scene.

[0094] Specifically, calculating the influence index of the traffic participants on the driving path of the target vehicle in each collision scenario also includes:

[0095] Step 1: For traffic participants whose object type is a dynamic entity type in each collision scene, the target state position information corresponding to the target vehicle and the traffic participant at each collision time point in the collision time period corresponding to the collision scene is substituted into the following formula to obtain the static potential field function value of the collision scene at each collision time point;

[0096] ;

[0097] ;

[0098] .

[0099] Step 2: Determine the static potential field function value of the collision scene at each collision time point as the influence index of the traffic participant on the driving path of the target vehicle at each collision time point in the collision scene.

[0100] Specifically, calculating the influence index of the traffic participants on the driving path of the target vehicle in each collision scenario also includes:

[0101] Step 1: For the traffic participants whose object type is non-entity type in each collision scene, substitute the target state position information of the target vehicle and the traffic participant at each collision time point in the collision time period corresponding to the collision scene into the following formula to obtain the influence index of the traffic participant on the driving path of the target vehicle at each collision time point in the collision scene. ;

[0102]

[0103] in, represents the relative distance between the target vehicle and the traffic participants, To adjust the parameters of the curve steepness, is the coefficient that controls the decay speed.

[0104] S104. Calculate the collision risk of the target vehicle in each collision scenario according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle.

[0105] In an implementation manner of the present application, the sum of the influence indexes of all traffic participants in each collision scenario on the driving path of the target vehicle is determined as the collision risk of the target vehicle in each collision scenario; or the weighted sum of the influence indexes of all traffic participants in each collision scenario on the driving path of the target vehicle is determined as the collision risk of the target vehicle in each collision scenario.

[0106] In addition, in addition to the influence index of traffic participants on the target vehicle's driving path affecting the collision risk of the target vehicle in each collision scenario, traffic constraints will also affect the collision risk of the target vehicle in each collision scenario.

[0107] The present application provides a method for quantifying collision risk, which includes: predicting collision scenarios that meet preset collision scenario rules and target state position information corresponding to target vehicles and traffic participants in the collision time period corresponding to the collision scenarios based on the initial state position information corresponding to the target vehicles and traffic participants in the future time period; calculating the influence index of traffic participants on the driving path of the target vehicle in each collision scenario based on the type of things corresponding to the traffic participants and the target state position information corresponding to the target vehicles and traffic participants in the collision time period corresponding to each collision scenario; calculating the collision risk of the target vehicle in each collision scenario based on the influence index of all traffic participants on the driving path of the target vehicle in each collision scenario. The method of the present application can effectively quantify the collision risk of a driving car in a real scenario.

[0108] Specifically, refer to Figure 2 As shown, it is a flow chart of another collision risk quantification method provided in an embodiment of the present application. The exemplary steps of the embodiment of the present application are described below:

[0109] S201: Obtain the traffic light status and the position information of the stop line at each collision time point in each collision scenario.

[0110] In the implementation mode of the present application, the traffic light status includes a vehicle passable status (such as a green light) and a vehicle prohibited communication status (such as a red light). The stop line is an important concept in traffic rules. It refers to the horizontal solid line before the zebra crossing at the intersection, which is used to indicate the reference line where vehicles should stop when waiting for traffic lights. When the red light is on, the vehicle needs to stop before the stop line. Crossing or exceeding the line is a violation of traffic rules.

[0111] S202: for each collision time point in each collision scenario, determine whether the target vehicle can pass according to the corresponding traffic light state, the position information of the stop line, and the position information in the target state position information of the target vehicle.

[0112] In an embodiment of the present application, based on the position information in the target state position information of the target vehicle and the position information of the stop line, it is determined whether the target vehicle has passed the stop line; if the traffic light state is a vehicle passable state, the target vehicle can pass; if the traffic light state is a vehicle communication prohibited state and the target vehicle has passed the stop line, the target vehicle can pass; if the traffic light state is a vehicle communication prohibited state and the target vehicle has not passed the stop line, the target vehicle is prohibited from passing.

[0113] S203: If the target vehicle can pass, determine the fixed potential field value corresponding to the passable state as the influence index of the pass constraint on the driving path of the target vehicle.

[0114] In the implementation manner of the present application, the fixed potential field value is set in advance.

[0115] S204: If the target vehicle is prohibited from passing, the fixed potential field value corresponding to the prohibited passing state is determined as the influence index of the passing constraint on the driving path of the target vehicle.

[0116] S205. Calculate the collision risk of the target vehicle in each collision scenario according to the influence index of all traffic participants on the driving path of the target vehicle and the influence index of the traffic constraint on the driving path of the target vehicle in each collision scenario.

[0117] In the implementation mode of the present application, the sum of the impact index of all traffic participants on the target vehicle's driving path and the impact index of traffic constraints on the target vehicle's driving path in each collision scenario is determined as the collision risk of the target vehicle in each collision scenario. Alternatively, the weighted sum of the impact index of all traffic participants on the target vehicle's driving path and the impact index of traffic constraints on the target vehicle's driving path in each collision scenario is performed to obtain the collision risk of the target vehicle in each collision scenario.

[0118] The present application provides another collision risk quantification method, which can effectively quantify the collision risk of a driving car in a real scenario.

[0119] Based on the same inventive concept, an embodiment of the present application also provides a collision risk quantification device corresponding to the collision risk quantification method. Since the principle of solving the problem by the device in the embodiment of the present application is similar to the above-mentioned collision risk quantification method in the embodiment of the present application, the implementation of the device can refer to the implementation of the method, and the repeated parts will not be repeated.

[0120] Reference Figure 3 FIG. 1 is a schematic diagram of a collision risk quantification device provided in an embodiment of the present application, wherein the collision risk quantification device includes:

[0121] The acquisition module 301 is used to acquire the initial state position information corresponding to the target vehicle and the traffic participants in the future time period; the traffic participants refer to objects that have a potential collision risk with the target vehicle in the current driving scene;

[0122] A prediction module 302 is used to predict, based on the initial state position information, a collision scene that meets a preset collision scene rule and target state position information corresponding to the target vehicle and the traffic participant within a collision time period corresponding to the collision scene; the preset collision scene rule combines the interaction time and interaction position between the target vehicle and the traffic participant;

[0123] The calculation module 303 is used to calculate the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene;

[0124] The calculation module 303 is further used to calculate the collision risk of the target vehicle in each collision scenario according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle.

[0125] In one possible implementation, the prediction module 302 is specifically used to predict an initial collision scene that meets preset collision scene rules based on the initial state position information, as well as the target state position information corresponding to the target vehicle and the traffic participant within the collision time period corresponding to the initial collision scene; and merge initial collision scenes with overlapping or continuous collision time periods to obtain a target collision scene and a corresponding collision time period.

[0126] In a possible implementation, the prediction module 302 is specifically used to, for each traffic participant whose thing type is an entity type, determine the earliest time point corresponding to the initial state position information of the target vehicle and the traffic participant at the same time that meets the preset collision scene rules as the starting time point of the initial collision scene; predict the target state position information of the target vehicle and the traffic participant respectively corresponding to the initial state position information at the starting time point within a preset time length starting from the starting time point; determine the earliest time point corresponding to the target state position information of the target vehicle and the traffic participant at the same time that does not meet the preset collision scene rules as the end time point of the initial collision scene; determine the collision time period corresponding to the initial collision scene according to the starting time point and the end time point of the initial collision scene.

[0127] In a possible implementation manner, the initial state position information includes initial position information, and the target state position information includes target position information.

[0128] In a possible implementation manner, the transaction type includes an entity type and a non-entity type; the entity type includes a static entity type and a dynamic entity type.

[0129] In a possible implementation, the calculation module 303 is specifically used for, for traffic participants whose object type is a dynamic entity type in each collision scene, substituting the target state position information corresponding to the target vehicle and the traffic participant at each collision time point within the collision time period corresponding to the collision scene into the following formula to obtain the travel intention potential field function value of the collision scene at each collision time point;

[0130] ;

[0131] ;

[0132] ;

[0133] ;

[0134] ;

[0135] in, is the potential field function of the moving intention, is the static potential field function, is the mass of the traffic participant, is the attenuation coefficient of the moving intention potential field, is the relative distance between the target vehicle and the traffic participant calculated based on the position information in the target state position information, is the relative angle between the target vehicle and the traffic participant calculated based on the steering angle information in the target state position information, is the directional function, is the longitudinal potential field strength of the collision scene, is the transverse potential field strength of the collision scene, is the ordinate of the target vehicle's position information, is the horizontal coordinate of the target vehicle’s position information, is the ordinate of the position information of the traffic participant, is the horizontal coordinate of the position information of the traffic participant, is the steep drop coefficient in the x direction of the potential field space, is the steep drop coefficient in the y direction of the potential field space.

[0136] The travel intention potential field function value of the collision scene at each collision time point is determined as the influence index of the traffic participant on the driving path of the target vehicle at each collision time point in the collision scene.

[0137] In a possible implementation manner, the device further includes: a determination module 304;

[0138] The acquisition module 301 is further used to acquire the traffic light status and the position information of the stop line at each collision time point in each collision scenario;

[0139] A judgment module 304 is used to judge whether the target vehicle can pass according to the corresponding traffic light state, the position information of the stop line, and the position information in the target state position information of the target vehicle for each collision time point in each collision scenario;

[0140] The judgment module 304 is further configured to determine, if the target vehicle is passable, a fixed potential field value corresponding to the passable state as an influence index of the pass constraint on the driving path of the target vehicle;

[0141] The judgment module 304 is further configured to determine, if the target vehicle is prohibited from passing, a fixed potential field value corresponding to the prohibited passing state as an influence index of the passing constraint on the driving path of the target vehicle;

[0142] The calculation module 303 is specifically used to calculate the collision risk of the target vehicle in each collision scenario according to the influence index of all traffic participants on the driving path of the target vehicle in each collision scenario and the influence index of the traffic constraint on the driving path of the target vehicle.

[0143] The present application provides a device for quantifying collision risk, which includes: an acquisition module 301, which is used to obtain the initial state position information corresponding to the target vehicle and the traffic participants in the future time period; the traffic participants refer to the things with potential collision risk with the target vehicle in the current driving scene; the prediction module 302, which is used to predict the collision scene that meets the preset collision scene rules and the target state position information corresponding to the target vehicle and the traffic participants in the collision time period corresponding to the collision scene according to the initial state position information; the preset collision scene rules combine the interaction time and interaction position between the target vehicle and the traffic participants; the calculation module 303, which is used to calculate the influence index of the traffic participants on the driving path of the target vehicle in each collision scene according to the type of things corresponding to the traffic participants, the target state position information corresponding to the target vehicle and the traffic participants in the collision time period corresponding to each collision scene; the calculation module 303 is also used to calculate the collision risk of the target vehicle in each collision scene according to the influence index of all traffic participants on the driving path of the target vehicle in each collision scene. Through the method of the present application, the collision risk of a driving car can be effectively quantified in a real scene.

[0144] like Figure 4As shown, an electronic device 400 provided in an embodiment of the present application includes: a processor 401, a memory 402 and a bus, wherein the memory 402 stores machine-readable instructions executable by the processor 401. When the electronic device is running, the processor 401 communicates with the memory 402 through the bus, and the processor 401 executes the machine-readable instructions to perform the steps of the collision risk quantification method as described above.

[0145] Specifically, the memory 402 and the processor 401 can be general-purpose memories and processors, which are not specifically limited here. When the processor 401 runs the computer program stored in the memory 402, the collision risk quantification method can be executed.

[0146] Corresponding to the above-mentioned collision risk quantification method, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned collision risk quantification method are executed.

[0147] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the method embodiment, and will not be repeated in this application. In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules 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, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0148] The modules described as separate components may or may not be physically separated, and the components shown as modules 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.

[0149] 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.

[0150] 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, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the information processing method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard drives, ROM, RAM, magnetic disks, or optical disks.

[0151] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for quantifying collision risk, characterized in that: The method comprises: Acquire initial state position information corresponding to the target vehicle and traffic participants in a future time period; the traffic participants refer to objects that have a potential collision risk with the target vehicle in the current driving scenario; According to the initial state position information, predicting a collision scene that meets the preset collision scene rules and the target state position information of the target vehicle and the traffic participant respectively corresponding to the collision time period corresponding to the collision scene, including: according to the initial state position information, predicting an initial collision scene that meets the preset collision scene rules and the target state position information of the target vehicle and the traffic participant respectively corresponding to the collision time period corresponding to the initial collision scene; merging initial collision scenes with overlapping or continuous collision time periods to obtain a target collision scene and a corresponding collision time period; the preset collision scene rules combine the interaction time and interaction position of the target vehicle and the traffic participant; Calculate the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene; The collision risk of the target vehicle in each collision scenario is calculated according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle.

2. The method for quantifying collision risk according to claim 1, characterized in that: The method of predicting, based on the initial state position information, an initial collision scene that meets a preset collision scene rule and target state position information corresponding to the target vehicle and the traffic participant within a collision time period corresponding to the initial collision scene, includes: For each traffic participant whose thing type is an entity type, the earliest time point corresponding to the initial state position information of the target vehicle and the traffic participant at the same time that meets the preset collision scene rule is determined as the starting time point of the initial collision scene; According to the initial state position information of the target vehicle and the traffic participant respectively corresponding to the starting time point, predicting the target state position information of the target vehicle and the traffic participant respectively corresponding to the target vehicle and the traffic participant within a preset time period starting from the starting time point; The earliest time point corresponding to the target state position information of the target vehicle and the traffic participant at the same time that does not conform to the preset collision scene rule is determined as the end time point of the initial collision scene; A collision time period corresponding to the initial collision scenario is determined according to the starting time point and the ending time point of the initial collision scenario.

3. The method for quantifying collision risk according to claim 2, characterized in that: The initial state position information includes initial position information, and the target state position information includes target position information.

4. The method for quantifying collision risk according to claim 1, characterized in that: The thing types include entity types and non-entity types; the entity types include static entity types and dynamic entity types.

5. The method for quantifying collision risk according to claim 4, characterized in that: The method of calculating the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene comprises: For traffic participants whose object type is a dynamic entity type in each collision scene, the target state position information corresponding to the target vehicle and the traffic participant at each collision time point within the collision time period corresponding to the collision scene is substituted into the following formula to obtain the travel intention potential field function value of the collision scene at each collision time point; ； ； ； ； ； in, is the potential field function of the moving intention, is the static potential field function, is the mass of the traffic participant, is the attenuation coefficient of the moving intention potential field, is the relative distance between the target vehicle and the traffic participant calculated based on the position information in the target state position information, is the relative angle between the target vehicle and the traffic participant calculated based on the steering angle information in the target state position information, is the directional function, is the longitudinal potential field strength of the collision scene, is the transverse potential field strength of the collision scene, is the ordinate of the target vehicle's position information, is the horizontal coordinate of the target vehicle’s position information, is the ordinate of the position information of the traffic participant, is the horizontal coordinate of the position information of the traffic participant, is the steep drop coefficient in the x direction of the potential field space, is the steep drop coefficient in the y direction of the potential field space; The travel intention potential field function value of the collision scene at each collision time point is determined as the influence index of the traffic participant on the driving path of the target vehicle at each collision time point in the collision scene.

6. The method for quantifying collision risk according to claim 1, characterized in that: The method further comprises: Obtain the traffic light status and stop line position information at each collision time point in each collision scenario; For each collision time point in each collision scenario, judging whether the target vehicle can pass according to the corresponding traffic light state, the position information of the stop line, and the position information in the target state position information of the target vehicle; If the target vehicle can pass, the fixed potential field value corresponding to the passable state is determined as the influence index of the pass constraint on the driving path of the target vehicle; If the target vehicle is prohibited from passing, the fixed potential field value corresponding to the prohibited passing state is determined as the influence index of the passing constraint on the driving path of the target vehicle; The collision risk of the target vehicle in each collision scenario is calculated according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle and the influence index of the traffic constraint on the driving path of the target vehicle.

7. A device for quantifying collision risk, characterized in that: The device comprises: An acquisition module is used to acquire initial state position information corresponding to the target vehicle and traffic participants in a future time period; the traffic participants refer to objects that have a potential collision risk with the target vehicle in the current driving scenario; A prediction module, used for predicting a collision scene that meets the preset collision scene rules and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to the collision scene according to the initial state position information, including: predicting the initial collision scene that meets the preset collision scene rules and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to the initial collision scene according to the initial state position information; merging the initial collision scenes with overlapping or continuous collision time periods to obtain the target collision scene and the corresponding collision time period; the preset collision scene rules combine the interaction time and interaction position of the target vehicle and the traffic participant; A calculation module, used to calculate the influence index of the traffic participant on the driving path of the target vehicle in each collision scene according to the object type corresponding to the traffic participant and the target state position information corresponding to the target vehicle and the traffic participant in the collision time period corresponding to each collision scene; The calculation module is further used to calculate the collision risk of the target vehicle in each collision scenario according to the influence index of all traffic participants in each collision scenario on the driving path of the target vehicle.

8. An electronic device, characterized in that: include: A processor, a storage medium and a bus, wherein the storage medium stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the storage medium communicate via the bus, and the processor executes the machine-readable instructions to perform the steps of the collision risk quantification method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method for quantifying collision risk according to any one of claims 1 to 6 are executed.

Images