Traffic environment safety situation assessment method, device, vehicle and storage medium

Abstract

The present application relates to a method, device, vehicle and storage medium for evaluating the safety situation of a traffic environment, wherein the method includes: combining the road data and relative perception data of the road on which the vehicle is currently traveling to construct a basic field strength model of the vehicle's safety field, and determining the field strength generated by any single risk source at multiple target positions of the vehicle based on the basic field strength model of the safety field, calculating the actual field strength after attenuation of any single risk source, and in the case of multiple risk sources, superimposing the actual field strengths generated by multiple risk sources at each target position respectively, to obtain the risk field strength corresponding to each target position, so as to evaluate the safety situation of the vehicle in the current traffic environment. Thus, the technical problems in the related technologies, such as the incomplete consideration of risk factors, the single applicable scenario, and the inability to cope with complex and changeable traffic environments, which make it difficult to support vehicles in making intelligent decisions and controls and affect the driving experience, are solved.

Description

Technical Field

[0001] The present application relates to the field of vehicle safety technology, and in particular to a method, device, vehicle and storage medium for evaluating the safety situation of a traffic environment. Background Art

[0002] The development of intelligent driving technology has brought convenience to people's travel and transportation. To ensure driving safety, intelligent vehicles need to have the ability to assess the current environmental risks. However, it is challenging to conduct reliable risk assessments given the uncertainty and complexity of the human-vehicle-road traffic environment. It is difficult for vehicles to safely reach their destinations based solely on their judgment of the current environmental status, but they need to combine predictions of future changes in the surrounding traffic environment to ensure driving safety.

[0003] In normal driving scenarios, several different types of risks may occur simultaneously caused by multi-dimensional and multi-risk sources, that is, each risk source will lead to one or more possible risk events, and each event is characterized by a specific risk indicator. The main sources of risk are the driver, the vehicle, the road environment and other traffic participants. The risk types can be roughly divided into collisions with other vehicles (such as too close relative distance, intersection of trajectory conflict points, too small speed difference, etc.), collisions with vulnerable groups such as pedestrians, collisions with the road environment, and risks caused by the system's own loss of control.

[0004] In order to solve the above problems, relevant technologies can be classified into four categories: methods based on macro-indicator statistics, methods based on vehicle kinematics, methods based on potential energy fields and methods based on data-driven.

[0005] 1. Statistical methods based on macro indicators usually refer to the use of macro traffic data parameters to analyze driving risk factors, including absolute number method, accident rate method and accident severity analysis method. In order to study the relationship between accident probability and cause, traffic accident data needs to be extracted and analyzed based on macro indicator statistical methods. For example, the use of logistic regression model can effectively combine accident factors with possible influencing characteristic variables, and analyze the factors affecting accidents through characteristic scales. Based on the obtained traffic accident data statistics, the operating status of each vehicle in the macro traffic flow can be further evaluated; another example is the use of multi-source mobile data to study driving risk influencing factors, specifically collecting vehicle dynamics information (including speed, acceleration and deceleration), user trajectory and traffic flow information, etc., providing a unique perspective to complete the interactive correlation analysis of driving behavior and collision possibility; another example is to explore the relationship between traffic accidents and traffic volume or traffic flow. For example, the possibility of accidents under congested conditions is high, but on non-congested roads, due to the relatively high average driving speed, it is easy to cause more serious casualties. Although macro-data analysis can reflect the overall traffic system risk, the evaluation results output by statistical data cannot reflect the dynamic driving risk of each driver. Therefore, there are migration difficulties when applied to the steady-state analysis of micro-traffic systems.

[0006] 2. The risk assessment indicators determined based on the vehicle kinematics method can be divided into distance logic indicators and time logic indicators. The vehicle models used can be divided into vehicle kinematics (Kinematic Models) and dynamic models (DynamicModels). In the distance logic indicators, the spatial safety distance is often used as a risk assessment indicator. Typical representatives include the MAZDA model, the HONDA model, the Berkeley model, the NHSTA model, the fixed vehicle distance model, the acceleration-based critical safety distance model and the kinematic model. Most of these methods only describe conflict scenarios in a single dimension, taking into account the kinematic and dynamic characteristics of traffic participants, and can be applied to intelligent vehicles to play a warning role. When the relative distance between the vehicle and the target vehicle is compared with the calculated safety distance, different warnings or intervention measures are taken according to the results, and accidents and collisions are used to measure the risk of the system.

[0007] In the time logic index, the relevant technology can quantitatively evaluate the driving risk by describing the parameters of the collision. Common parameters include TTC (Time to Collision), THW (Time Headway), Extended TTC (Extended TTC), TTS (Time to Steer), PET (Post Encroachment Time), TDTC (Time to Lane Cross), TLC (Time to Lane Cross), and comprehensive indicators expressed in the form of THW and TTC functions. The above improved parameter indicators can realize the quantitative identification of driving risks and are applicable to different traffic scenarios. They are important reference indicators for evaluating the degree of conflict between two targets. Most of these methods are based on the theory of vehicle kinematics and dynamics. The expression of vehicle driving safety is mostly achieved by using vehicle state information (speed, acceleration, yaw rate, etc.) and the relative motion relationship information between the two vehicles (relative speed, relative distance, etc.). This type of method has been applied to a certain extent because of its simple parameters, fixed hazard judgment principles, criteria and model parameters, and its physical meaning is consistent with people's subjective feelings. However, the limitation of this type of method is that it can usually only perform deterministic risk assessment on one-dimensional traffic risks, and it is difficult to achieve uncertain risk assessment of two-dimensional risks in actual traffic scenarios, which leads to greater limitations in practical applications of this method.

[0008] 3. APF (Artificial Potential Field) refers to a class of methods that quantify the impact of drivers, lane lines, traffic signs and obstacles on vehicle driving risks through potential field theory. The artificial potential field method is a common method for handling situation assessment. The main idea is to assume that the vehicle is driving in the potential field formed by the target point and the obstacle, where the target point generates a gravitational field to attract the vehicle and the obstacle generates a repulsive field to prevent collision. Since the artificial potential field method was proposed, the research on quantitative characterization of driving risks based on the potential field has continued to develop. For example, based on the idea of ​​gravitational field, a path planning controller based on fusion model prediction is constructed by taking the gravitational force of the safety gap with the vehicle and the repulsive force of other surrounding vehicles; another example is an electric field model based on an improved artificial potential field, which can output the real-time driving risk distribution of the vehicle in the traffic environment, realize the navigation tracking and collision avoidance application of intelligent vehicles, and thus assist the intelligent vehicle to reach the destination safely; another example is to establish an integrated artificial potential field model of road, vehicle and speed potential field to achieve risk avoidance between the vehicle and other obstacles. These methods can achieve high-dimensional risk assessment in traffic scenarios and obtain better risk sensitivity and accuracy in complex traffic environments. However, the above-mentioned artificial potential field methods ignore the influence of uncertain factors such as driver characteristics, vehicle dynamics, road conditions, and weather.

[0009] Fourth, data-driven methods mainly use data learning and mining to predict the future driving intention of the vehicle based on historical trajectory data. This type of method has been widely used in the field of intention recognition of intelligent vehicles in recent years. Many studies use learning methods such as machine learning, deep reinforcement learning, and deep neural networks to extract potential risk features in traffic events. Through a large amount of training and verification of risk behaviors, data-driven risk quantification and identification are achieved. The end-to-end model of autonomous driving is based on DNN (Deep Neural Network) which can directly map the original input to driving actions (such as acceleration and steering angle), and finally output the action behavior that can avoid collision. In addition, the reasoning process of driver intention can be realized based on machine learning methods, such as the current mainstream time series neural network and deep recurrent neural network, which are often used for intention recognition. The deep recurrent neural network has a deeper network structure and can effectively capture the characteristics of the driver's manipulation behavior, and then characterize the dependencies between different sequences to achieve better performance in intention recognition.

[0010] Intention recognition models based on deep learning are usually based on RNN (Recurrent Neural Network) to identify vehicle behaviors or intentions. Among them, LSTM (Long Short-Term Memory) can control the information flow between input, output and cell memory through a gating mechanism, effectively overcoming the problem of gradient vanishing in RNN. In addition, LSTM has good information mining and deep representation capabilities in time series problem processing. Applying LSTM for intention recognition and trajectory prediction can achieve good results. The recently proposed Transformer model adopts a self-attention mechanism, automatically distributes and adjusts weights, and allows parallel computing. It has excellent performance in motion prediction and can effectively perform joint prediction of a large number of fast-moving multi-agent trajectories. However, in pursuit of the best prediction performance, the parameters and computational costs of the model are relatively more challenging. Compared with traditional methods, data-driven vehicle behavior prediction and intention recognition methods have good performance in complex environments and can input more information for prediction or recognition. However, they rarely consider or directly ignore factors such as road geometry or driver characteristics, making it difficult to clearly describe the coupling mechanism between drivers, vehicles and roads, limiting the good assessment of real traffic environment risks and making it difficult to cope with complex and changing traffic environments.

[0011] To sum up, although the relevant technologies can realize risk situation assessment in traffic environments to a certain extent, they usually have problems such as incomplete consideration of risk factors, single applicable scenarios, and inability to cope with complex and changing traffic environments. They are difficult to support vehicles in making intelligent decisions and control, and need to be improved. Summary of the invention

[0012] The present application provides a method, device, vehicle and storage medium for evaluating the safety situation of a traffic environment, in order to solve the technical problems in related technologies, such as incomplete risk factors, single applicable scenarios, and inability to cope with complex and changing traffic environments, which make it difficult to support vehicles in making intelligent decisions and controls, and thus affect the driving experience.

[0013] A first aspect of the present application provides a method for evaluating the safety situation of a traffic environment, comprising the following steps: obtaining road data of a vehicle's current driving road and relative perception data of any single risk source on the current driving road; constructing a basic field strength model of the vehicle's safety field in combination with the road data and the relative perception data, and determining the field strength generated by any single risk source at multiple target positions of the vehicle based on the basic field strength model of the safety field; calculating the actual field strength of any single risk source after attenuation based on a preset attenuation coefficient, the road data, the relative perception data and the number of risk sources, and in the case of multiple risk sources, superimposing the actual field strengths generated by the multiple risk sources at each target position respectively to obtain the risk field strength corresponding to each target position, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength.

[0014] Optionally, in one embodiment of the present application, after obtaining the relative perception data of any single risk source on the current driving road, it also includes: respectively judging whether the perception contour data and the perception speed data of any single risk source in the relative perception data meet preset uncertainty conditions; if the perception contour data meets the preset uncertainty conditions, then obtaining the expanded contour data of any single risk source according to a preset expansion coefficient and the perception contour data, and using the expanded contour data as the actual contour data of any single risk source; if the perception speed data meets the preset uncertainty conditions, then obtaining the estimated speed data according to the preset uncertainty range and the perception data, and using the estimated speed data as the actual speed data of any single risk source.

[0015] Optionally, in one embodiment of the present application, the construction of the basic field strength model of the safety field of the vehicle in combination with the road data and the relative perception data includes: identifying the actual motion state of any single risk source based on the relative perception data; when the actual motion state is a stationary state, combining the road data and the virtual mass of any single risk source to construct a potential field model, and using the potential field model as the basic field strength model of the safety field; when the actual motion state is a moving state, constructing a kinetic field model based on the current driving data of the vehicle, the structural data of the vehicle and the contour data of any single risk source, and obtaining the basic field strength model of the safety field based on the kinetic field model and the potential field model.

[0016] Optionally, in one embodiment of the present application, the calculation expression of the virtual mass is:

[0017]

[0018] Among them, M a is the virtual quality, a is any single risk source, T a is the type coefficient of any single risk source, v α is the speed of any single risk source, α, β, γ are all unknown constants;

[0019] The calculation expression of the potential energy field model is:

[0020]

[0021] Among them, E p1 is the potential energy generated by any single risk source, K is the virtual quality coefficient, R a is the environmental condition influencing factor, r is the distance to the center of any single risk source, and k1 is the distance coefficient;

[0022] The calculation expression of the kinetic energy field model is:

[0023] w=p(δ) / p max ,

[0024]

[0025] Where w is the weight of the trajectory selected by the vehicle to travel to any point, δ is the front wheel turning angle required for the vehicle to travel to any point, p(δ) is the probability density, p max is the maximum probability density, E d is the kinetic energy generated by any single risk source, E is the basic kinetic energy, Δr is the arc length from the current position to any point, r stdis the standard arc length value, m is the mass of the vehicle, and v is the current speed of the vehicle.

[0026] Optionally, in one embodiment of the present application, determining the field strength generated by any single risk source on the vehicle based on the basic field strength model of the safety field includes: when the actual motion state is a stationary state, using the potential energy field model to calculate the field strength generated by any single risk source on the vehicle; when the actual motion state is a moving state, using the potential energy field model to calculate the static field strength generated by any single risk source on the vehicle, using the kinetic energy field model to calculate the dynamic field strength generated by any single risk source on the vehicle, and superimposing the static field strength and the dynamic field strength to obtain a superposition result, and using the superposition result as the field strength.

[0027] Optionally, in one embodiment of the present application, the actual field strength of any single risk source after attenuation is calculated based on the preset attenuation coefficient, the road data, the relative perception data and the number of risk sources, including: identifying at least one attenuation factor of any single risk source based on the relative perception data and the road data; when the attenuation factor is a constraint-type attenuation factor, matching the corresponding road attenuation coefficient in a preset coefficient comparison table according to the road data; when the attenuation factor is an occlusion-type attenuation factor, determining at least one occluding object corresponding to any single risk source according to the relative perception data, and using the potential energy field model to obtain the potential energy of each occluding object, so as to obtain the occlusion attenuation coefficient based on the potential energy; and obtaining the preset attenuation coefficient according to the number of attenuation factors, the road attenuation coefficient and the occlusion attenuation coefficient.

[0028] Optionally, in one embodiment of the present application, superimposing actual field strengths generated by multiple risk sources at each target location point to obtain the risk field strength corresponding to each target location point includes: taking the actual field strength with the largest value from multiple actual field strengths at each target location point as the risk field strength.

[0029] The second aspect of the present application provides an evaluation device for the safety situation of a traffic environment, including: an acquisition module, used to acquire road data of the current road on which a vehicle is traveling and relative perception data of any single risk source on the current road on which the vehicle is traveling; a construction module, used to construct a basic field strength model of the safety field of the vehicle in combination with the road data and the relative perception data, and determine the field strength generated by any single risk source at multiple target position points of the vehicle based on the basic field strength model of the safety field; an evaluation module, used to calculate the actual field strength of any single risk source after attenuation based on a preset attenuation coefficient, the road data, the relative perception data and the number of risk sources, and in the case of multiple risk sources, superimpose the actual field strengths generated by multiple risk sources at each target position point respectively to obtain the risk field strength corresponding to each target position point, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength.

[0030] Optionally, in one embodiment of the present application, it also includes: a judgment module, used to respectively judge whether the perception contour data and the perception speed data of any single risk source in the relative perception data meet preset uncertainty conditions; a first calculation module, used to obtain the expanded contour data of any single risk source according to a preset expansion coefficient and the perception contour data when the perception contour data meets the preset uncertainty conditions, and use the expanded contour data as the actual contour data of any single risk source; a second calculation module, used to obtain estimated speed data according to a preset uncertainty range and the perception data when the perception speed data meets the preset uncertainty conditions, and use the estimated speed data as the actual speed data of any single risk source.

[0031] Optionally, in one embodiment of the present application, the construction module includes: a first identification unit, used to identify the actual motion state of any single risk source based on the relative perception data; a first construction unit, used to construct a potential field model in combination with the road data and the virtual mass of any single risk source when the actual motion state is a stationary state, and use the potential field model as the basic field strength model of the safety field; a second construction unit, used to construct a kinetic field model based on the current driving data of the vehicle, the structural data of the vehicle and the contour data of any single risk source when the actual motion state is a moving state, and obtain the basic field strength model of the safety field based on the kinetic field model and the potential field model.

[0032] Optionally, in one embodiment of the present application, the calculation expression of the virtual mass is:

[0033]

[0034] Among them, M a is the virtual quality, a is any single risk source, T a is the type coefficient of any single risk source, v α is the speed of any single risk source, α, β, γ are all unknown constants;

[0035] The calculation expression of the potential energy field model is:

[0036]

[0037] Among them, E p1 is the potential energy generated by any single risk source, K is the virtual quality coefficient, R a is the environmental condition influencing factor, r is the distance to the center of any single risk source, and k1 is the distance coefficient;

[0038] The calculation expression of the kinetic energy field model is:

[0039] w=p(δ) / p max ,

[0040]

[0041] Where w is the weight of the trajectory selected by the vehicle to travel to any point, δ is the front wheel turning angle required for the vehicle to travel to any point, p(δ) is the probability density, p max is the maximum probability density, E d is the kinetic energy generated by any single risk source, E is the basic kinetic energy, Δr is the arc length from the current position to any point, r std is the standard arc length value, m is the mass of the vehicle, and v is the current speed of the vehicle.

[0042] Optionally, in one embodiment of the present application, the construction module includes: a first calculation unit, used to calculate the field strength generated by any single risk source on the vehicle using the potential energy field model when the actual motion state is a stationary state; a second calculation unit, used to calculate the static field strength generated by any single risk source on the vehicle using the potential energy field model when the actual motion state is a moving state, and calculate the dynamic field strength generated by any single risk source on the vehicle using the kinetic energy field model, and superimpose the static field strength and the dynamic field strength to obtain a superposition result, and use the superposition result as the field strength.

[0043] Optionally, in one embodiment of the present application, the evaluation module includes: a second identification unit, used to identify at least one attenuation factor of any single risk source based on the relative perception data and the road data; a matching unit, used to match the corresponding road attenuation coefficient in a preset coefficient comparison table according to the road data when the attenuation factor is a constraint-type attenuation factor; a third calculation unit, used to determine at least one occluding object corresponding to any single risk source according to the relative perception data when the attenuation factor is an occlusion-type attenuation factor, and use the potential energy field model to obtain the potential energy of each occluding object to obtain the occlusion attenuation coefficient based on the potential energy; a fourth calculation unit, used to obtain the preset attenuation coefficient based on the number of the attenuation factors, the road attenuation coefficient and the occlusion attenuation coefficient.

[0044] Optionally, in an embodiment of the present application, the evaluation module includes: an acquisition unit, configured to take the actual field strength with the largest value from a plurality of actual field strengths at each target location point as the risk field strength.

[0045] The third aspect of the present application provides a vehicle, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for evaluating the safety situation of a traffic environment as described in the above embodiment.

[0046] The fourth aspect of the present application provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute the method for evaluating the safety situation of the traffic environment as described in the above embodiment.

[0047] A fifth aspect of the present invention provides a computer program product, including a computer program, which, when executed, is used to implement the above-mentioned method for evaluating the safety situation of a traffic environment.

[0048] The embodiment of the present application can build a basic field strength model of the vehicle's safety field in combination with the road data and relative perception data of the road the vehicle is currently traveling on, and determine the field strength generated by any single risk source at multiple target positions of the vehicle based on the basic field strength model of the safety field, and calculate the actual field strength of any single risk source after attenuation according to the preset attenuation coefficient, road data, relative perception data and the number of risk sources, and in the case of multiple risk sources, superimpose the actual field strengths generated by multiple risk sources at each target position respectively, and obtain the risk field strength corresponding to each target position, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength, and by fully considering multiple risk factors, realize the safety situation assessment of multiple scenarios, especially complex traffic environment scenarios, to support the decision-making and control of the vehicle. Thus, the technical problems in the related technology that the risk factors considered are not comprehensive, the applicable scenarios are single, and the complex and changeable traffic environment cannot be faced, resulting in difficulty in supporting the vehicle to make intelligent decisions and control, affecting the driving experience, are solved.

[0049] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0051] Figure 1 A flow chart of a method for evaluating the safety situation of a traffic environment provided according to an embodiment of the present application;

[0052] Figure 2 A schematic diagram of a potential energy field according to an embodiment of the present application;

[0053] Figure 3 A frequency distribution histogram and a probability fitting diagram of the front wheel turning angle according to an embodiment of the present application;

[0054] Figure 4 is a schematic diagram of a kinetic energy field according to an embodiment of the present application;

[0055] Figure 5 This is a schematic diagram of field intensity superposition of a single risk source according to an embodiment of the present application;

[0056] Figure 6 This is a schematic diagram of the field intensity attenuation effect caused by the lane line according to an embodiment of the present application;

[0057] Figure 7 A schematic diagram of the field intensity attenuation effect caused by other risk sources according to an embodiment of the present application;

[0058] Figure 8A schematic diagram of the principle of a method for evaluating the safety situation of a traffic environment according to an embodiment of the present application;

[0059] Fig. 9 A schematic diagram of field strength effect according to an embodiment of the present application;

[0060] Fig.10 is a schematic diagram of field strength effect according to another embodiment of the present application;

[0061] Fig.11 is a schematic diagram of field strength effect according to yet another embodiment of the present application;

[0062] Fig.12 It is a schematic diagram of simulation effect in a complex environment according to an embodiment of the present application;

[0063] Fig.13 It is a schematic diagram of simulation effect in a complex environment according to another embodiment of the present application;

[0064] Fig.14 It is a schematic diagram of simulation effect in a complex environment according to another embodiment of the present application;

[0065] Fig.15 A schematic diagram of a quantitative analysis scenario according to an embodiment of the present application;

[0066] Fig.16 A schematic diagram of the structure of a traffic environment safety situation assessment device provided according to an embodiment of the present application;

[0067] Fig.17 It is a schematic diagram of the structure of a vehicle provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0068] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0069] The following is a description of the traffic environment safety situation assessment method, device, vehicle and storage medium of the embodiment of the present application with reference to the accompanying drawings. In view of the technical problems that the risk factors considered in the related technologies mentioned in the above background technology are not comprehensive, the applicable scenarios are single, and the complex and changeable traffic environment cannot be faced, which makes it difficult to support the vehicle to make intelligent decisions and control, and affects the driving experience, the present application provides a traffic environment safety situation assessment method, in which the road data and relative perception data of the current driving road of the vehicle can be combined to construct a basic field strength model of the vehicle's safety field, and the field strength generated by any single risk source at multiple target position points of the vehicle is determined based on the basic field strength model of the safety field, and the actual field strength after attenuation of any single risk source is calculated according to the preset attenuation coefficient, road data, relative perception data and the number of risk sources, and when there are multiple risk sources, the actual field strength generated by multiple risk sources at each target position point is superimposed to obtain the risk field strength corresponding to each target position point, so as to assess the safety situation of the vehicle in the current traffic environment according to the risk field strength, and by fully considering multiple risk factors, the safety situation assessment of multiple scenarios, especially complex traffic environment scenarios, is achieved to support the decision-making and control of the vehicle. This solves the technical problems in related technologies, such as incomplete risk factors, single applicable scenarios, and inability to cope with complex and changing traffic environments, which makes it difficult to support vehicles in making intelligent decisions and control, and affects the driving experience.

[0070] Specifically, Figure 1 A flow chart of a method for evaluating the safety situation of a traffic environment provided in an embodiment of the present application.

[0071] like Figure 1 As shown, the evaluation method of the traffic environment safety situation includes the following steps:

[0072] In step S101, the road data of the current driving road of the vehicle and the relative perception data of any single risk source on the current driving road are obtained.

[0073] During the actual implementation process, the embodiments of the present application can obtain road data of the vehicle's current road, such as lane line data, road surface adhesion, road curvature, slope, visibility, etc. The embodiments of the present application can also obtain relative perception data of any single risk source on the current road, such as the relative speed and relative distance between the vehicle and the risk source.

[0074] Optionally, in one embodiment of the present application, after obtaining the relative perception data of any single risk source on the current driving road, it also includes: respectively judging whether the perception contour data and the perception speed data of any single risk source in the relative perception data meet the preset uncertainty conditions; if the perception contour data meets the preset uncertainty conditions, then according to the preset expansion coefficient and the perception contour data, the expanded contour data of any single risk source is obtained, and the expanded contour data is used as the actual contour data of any single risk source; if the perception speed data meets the preset uncertainty conditions, then according to the preset uncertainty range and the perception data, the estimated speed data is obtained, and the estimated speed data is used as the actual speed data of any single risk source.

[0075] It is understandable that due to the large number of participants in the traffic environment, there are many situations in which the perceived data is uncertain. For example, if there is another risk source blocking the vehicle from a certain risk source, the data perceived for a certain risk source may be inaccurate, or the data for a certain risk source may not be perceived.

[0076] In the case of inaccurate perception, for risk sources with mobility, the perceived position and speed may have certain uncertainties, and the uncertainty of position and speed needs to be modeled.

[0077] Among them, the position uncertainty includes the uncertainty of the central position, length and width of the risk source. The modeling method for this type of uncertainty is to expand the outline of the risk source.

[0078] One idea is to expand the shape of a generally rectangular object into an ellipse, but this method cannot reasonably cover the four corners of the obstacle. Therefore, in actual application, the expansion is modeled as a rectangle, that is, the length and width of the risk source are expanded by a certain coefficient, that is:

[0079] L=L d α L ; W = W d α W ,

[0080] Among them, L d , W d are the length and width of the risk source, α L , α W are the expansion coefficients corresponding to the length and width respectively.

[0081] The uncertainty of speed perception includes the uncertainty of speed magnitude and direction. For the uncertainty of speed magnitude, when actually calculating the kinetic energy field, an uncertainty range σ is given, v±σ is used instead of speed v, and the maximum value of the kinetic energy field within this range is taken; and the uncertainty of speed direction is already reflected when modeling the probability of the front wheel turning angle.

[0082] For situations that cannot be perceived, that is, sensor blind spots. To ensure safety, the algorithm can assume that there is an approximate maximum field strength source at the nearest position in the blind spot to prevent danger from occurring in the blind spot.

[0083] In step S102, a basic field strength model of the vehicle's safety field is constructed in combination with road data and relative perception data, and the field strength generated by any single risk source at multiple target positions of the vehicle is determined based on the basic field strength model of the safety field.

[0084] It is understandable that the definition of a safety field is: a physical field that characterizes the impact of various human-vehicle-road elements in the vehicle driving environment on driving safety: (1) In the traffic environment, each object that can cause driving hazards will form a driving safety field; (2) The driving safety fields formed by different objects repel each other; (3) The field strength vector of the driving safety field formed by an object is determined by its own properties, state, and the road conditions at its location, and characterizes the size and direction of the driving risk caused by the object, that is, the risk source.

[0085] The basic field strength model of the safety field used for safety situation assessment in complex traffic environments consists of a potential energy field and a kinetic energy field. The potential energy field represents the impact of stationary objects on road safety, while the kinetic energy field represents the additional impact of moving objects on road safety.

[0086] In summary, the embodiments of the present application can combine road data and relative perception data to construct a basic field strength model of the vehicle's safety field, and determine the field strength generated by any single risk source at multiple target locations of the vehicle.

[0087] Optionally, in one embodiment of the present application, a basic field strength model of the vehicle's safety field is constructed in combination with road data and relative perception data, including: identifying the actual motion state of any single risk source based on relative perception data; when the actual motion state is a stationary state, a potential field model is constructed in combination with road data and the virtual mass of any single risk source, and the potential field model is used as the basic field strength model of the safety field; when the actual motion state is a moving state, a kinetic field model is constructed based on the vehicle's current driving data, the vehicle's structural data and the contour data of any single risk source, and the basic field strength model of the safety field is obtained based on the kinetic field model and the potential field model.

[0088] The calculation expression of virtual mass is:

[0089]

[0090] Among them, M a is the virtual quality, a is any single risk source, T a is the type coefficient of any single risk source, v αis the speed of any single risk source, α, β, γ are all unknown constants;

[0091] The calculation expression of the potential energy field model is:

[0092]

[0093] Among them, E p1 is the potential energy generated by any single risk source, K is the virtual quality coefficient, R a is the environmental condition influencing factor, r is the distance to the center of any single risk source, and k1 is the distance coefficient;

[0094] The calculation expression of the kinetic energy field model is:

[0095] w=p(δ) / p max ,

[0096]

[0097] Among them, w is the weight of the trajectory selected by the vehicle to travel to any point, δ is the front wheel turning angle required for the vehicle to travel to any point, p(δ) is the probability density, p max is the maximum probability density, E d is the kinetic energy generated by any single risk source, E is the basic kinetic energy, Δr is the arc length from the current position to any point, r std is the standard arc length value, m is the mass of the vehicle, and v is the current speed of the vehicle.

[0098] As mentioned above, the potential energy field represents the impact of the risk source on the road in a static state on driving safety. The magnitude and direction of the potential energy field strength are mainly determined by the attributes of the risk source. For risk sources such as vehicles and pedestrians on the road, it has the following characteristics:

[0099] First, the severity of a collision between a moving vehicle and the object is related to the “mass” of the risk source. The greater the “mass”, the more serious the damage caused by the collision between the vehicle and the object. In particular, the “mass” here is not the physical mass in the usual sense, but a kind of “virtual mass”.

[0100] Second, the closer a vehicle is to the risk source, the greater the possibility of a collision with it. And as the vehicle approaches, this possibility and the danger of a collision do not increase linearly, but as the distance between the two decreases, the speed at which the driving risk increases accelerates;

[0101] Third, no matter from which direction the vehicle approaches this type of risk source, the changing pattern of the driving danger posed by this type of stationary risk source to the vehicle is the same.

[0102] Based on the above analysis, the potential energy field is modeled as follows:

[0103]

[0104] Among them, R a is the environmental condition influencing factor, and its expression is:

[0105] R a =φ μ (μ a )*φ ρ (ρ a )*φ τ (τ a )*φ δ (δ a )

[0106] where φ μ ,φ ρ ,φ τ ,φ δ They are road adhesion, road curvature, slope, and visibility, and the values ​​of these functions are greater than or equal to 1.

[0107] M a is the virtual mass of risk source a, expressed as:

[0108]

[0109] r is the distance from the risk source to the calculated field strength point, K>0 and k1>1 are correlation factors, where, for example, the embodiment of the present application can analyze the relationship between traffic accident losses and speed and calibrate them as follows:

[0110] α=1.566×10 -14 , β=6.687, γ=0.3345.

[0111] For example, based on calculation, the embodiment of the present application can draw the following Figure 2 Schematic diagram of the potential energy field effect shown.

[0112] The magnitude and direction of the kinetic energy field strength are mainly determined by the risk source attributes (type, mass and speed), motion state and road conditions, i.e., road data. The risk sources in motion state here refer to those risk sources in motion state that can cause substantial collision with the vehicle and cause significant losses. Any such risk source in motion state has the following characteristics:

[0113] (1) The severity of a collision between a moving vehicle and the moving vehicle is related to the virtual mass of the risk source in the moving state. The larger the virtual mass, the more serious the loss caused by the collision between the moving vehicle and the moving vehicle.

[0114] (2) The closer the vehicle is to the risk source in the moving state, the greater the possibility of a collision with it. And as the vehicle approaches, this possibility and the danger of collision do not increase linearly, but as the distance between the two decreases, the speed at which the driving risk increases accelerates;

[0115] (3) The law of how the driving risk posed by a moving risk source to surrounding vehicles changes with the distance between the two is related to the speed of the risk source and the position direction vector of the vehicle relative to the risk source. At the same distance, the smaller the angle between the direction vector of the vehicle relative to the risk source and the speed direction of the risk source, the greater the driving risk caused by the risk source. In particular, at the same distance, the driving risk in front of the moving risk source is the greatest, and the driving risk behind it is the least.

[0116] Since the size of the kinetic energy field is related to the direction of movement of the risk source, the embodiment of the present application can first use the Laplace distribution to fit the probability of the front wheel turning angle of the moving vehicle, and obtain the following: Figure 3 The schematic diagram shown has the following probability density:

[0117]

[0118] Among them, δ is the front wheel steering angle, the parameters are λ=0.27, μ=0.

[0119] According to vehicle dynamics, the relationship between the front wheel turning angle and the turning radius when the vehicle turns is:

[0120]

[0121] Among them, R is the turning radius, K is the stability factor, v is the vehicle speed, l is the vehicle wheelbase, and δ is the vehicle front wheel turning angle.

[0122] The kinetic energy field is finally modeled as:

[0123] w=p(δ) / p max ,

[0124]

[0125] Among them, δ is the front wheel turning angle required for the vehicle to travel to a certain point, w is the weight of the trajectory selected by the vehicle to travel to this point, and Δr is the arc length from the current position to the point, which is calculated by the physical model.

[0126] Furthermore, considering the shapes of vehicles and pedestrians, the final kinetic energy field effect is as follows: Figure 4 shown.

[0127] Optionally, in one embodiment of the present application, the field strength generated by any single risk source on the vehicle is determined based on the basic field strength model of the safety field, including: when the actual motion state is a static state, using the potential field model to calculate the field strength generated by any single risk source on the vehicle; when the actual motion state is a moving state, using the potential field model to calculate the static field strength generated by any single risk source on the vehicle, using the kinetic field model to calculate the dynamic field strength generated by any single risk source on the vehicle, and superimposing the static field strength and the dynamic field strength to obtain a superimposed result, and using the superimposed result as the field strength.

[0128] In some embodiments, for a single motion risk source, it will produce kinetic energy field and potential energy field effects at the same time. In the embodiment of the present application, the two field strengths can be superimposed by scalar superposition to obtain the following: Figure 5 The schematic diagram shown in the figure, where the superposition expression can be:

[0129] E dsf，i =(E p，i +E d，i )

[0130] Among them, E p,i is the potential energy field strength at point i, E d,i is the kinetic energy field strength at point i, E dsf,i is the comprehensive field strength value after superposition.

[0131] In step S103, based on the preset attenuation coefficient, road data, relative perception data and the number of risk sources, the actual field strength of any single risk source after attenuation is calculated, and if there are multiple risk sources, the actual field strengths generated by multiple risk sources at each target position are superimposed to obtain the risk field strength corresponding to each target position, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength.

[0132] After completing the modeling of the basic field strength model of the safety field, the embodiment of the present application can perform complex traffic environment situation modeling.

[0133] The embodiment of the present application can calculate the field intensity shielding and attenuation caused by virtual and real risk sources on the road, and superimpose multiple risk sources to obtain the risk field intensity corresponding to each target position point of the vehicle, thereby realizing safety situation assessment in a complex environment.

[0134] Optionally, in one embodiment of the present application, the actual field strength of any single risk source after attenuation is calculated based on a preset attenuation coefficient, road data, relative perception data and the number of risk sources, including: identifying at least one attenuation factor of any single risk source based on the relative perception data and the road data; when the attenuation factor is a constraint-type attenuation factor, matching the corresponding road attenuation coefficient in a preset coefficient comparison table according to the road data; when the attenuation factor is an occlusion-type attenuation factor, determining at least one occluding object corresponding to any single risk source according to the relative perception data, and using a potential energy field model to obtain the potential energy of each occluding object, so as to obtain the occlusion attenuation coefficient based on the potential energy; and obtaining a preset attenuation coefficient according to the number of attenuation factors, the road attenuation coefficient and the occlusion attenuation coefficient.

[0135] In the actual implementation process, there are two principles for the energy generated by a certain risk source to decay at a certain location:

[0136] (1) Figure 6 As shown in the figure, other risk sources have an impact on the behavior of this risk source. For example, the lane line constrains the lane-changing behavior of the vehicle, and the vehicles in other lanes will also constrain the lane-changing behavior of the vehicle. Therefore, the risk generated by the vehicle should be attenuated after crossing the lane line.

[0137] (2) Figure 7 As shown, other risk sources have an impact on the behavior of this risk source. For example, when a collision occurs, the existence of other risk sources will absorb part of the collision energy brought by this risk source, thus causing energy attenuation.

[0138] For both types of attenuation, an attenuation coefficient φ is used to characterize the attenuation. For the first type of attenuation, different attenuation coefficients designed based on lane line types can be obtained from a preset coefficient comparison table, for example, as shown in Table 1, where Table 1 is an attenuation coefficient table corresponding to lane line types.

[0139] Table 1

[0140] Lane Type Attenuation coefficient Yellow dotted line / half dotted line / double dotted line 0.7 White dotted line / half dotted line / double dotted line 0.9 Solid yellow line / double solid line 0.6 Solid white line / double solid line 0.8 Road edge / median strip 0.2

[0141] As for the second type of attenuation, its attenuation effect mainly depends on the potential energy of the shielding risk source. Therefore, φ=K1 / log(E p_max )

[0142] Among them, K1 is the parameter to be calibrated, E p_max Represents the maximum value in the potential energy field of the virtual object.

[0143] It should be noted that if there are multiple attenuation factors from the risk source to a certain location, the α of these attenuation factors need to be multiplied.

[0144] Optionally, in one embodiment of the present application, actual field strengths generated by multiple risk sources at each target location point are superimposed to obtain the risk field strength corresponding to each target location point, including: taking the actual field strength with the largest value from multiple actual field strengths at each target location point as the risk field strength.

[0145] As a possible implementation method, the most intuitive way to superimpose multiple safety fields generated by multiple risk sources is to add the field strength scalars. However, considering that in actual driving, the risk to the vehicle is not the superposition of multiple dangerous objects, but the most dangerous risk source, in actual application, the maximum value of the field strength generated by multiple risk sources at each position in space is taken as the risk field strength at that point, rather than directly adding the field strengths. That is:

[0146] E all =max(E1,E2,...E n ).

[0147] Furthermore, in the embodiments of the present application, there are two main approaches to calculating the field force on the vehicle in the safety field: one is to directly use the value of the field strength at any position and multiply it by the virtual mass of the vehicle.

[0148] F=(E p +E d )*M r ,

[0149]

[0150] Another more complex approach is to combine the trajectory prediction of the obstacle, multiply the absolute speed of the vehicle by the warning time to translate the position of the vehicle (take the maximum risk in the trajectory from this moment to the warning time), and multiply it by the virtual mass of the vehicle, that is:

[0151]

[0152] Considering that the early warning task requires strong real-time performance, the first method with lower computational complexity can be adopted.

[0153] For example, when conducting risk assessment in the embodiment of the present application, for the assisted driving warning function required by the project, since the decision is still controlled by the driver, there is no need to consider the global risk. Instead, based on the position and size of the vehicle, a total of 8 points are determined, including the front, rear, left, right, left front, left rear, right front, and right rear of the vehicle, and the 8 points are used as target position points. The impact of each risk source on these 8 points is calculated, and medium-risk and high-risk field strength thresholds are set. If any point exceeds the threshold, an alarm is issued, thereby reducing the amount of calculation and improving the resolution.

[0154] If the above method is applied to areas such as autonomous driving decision-making and planning, other comprehensive considerations need to be adopted, such as screening out trajectories with smaller risk field values ​​from the sampled trajectories.

[0155] And because there are many uncertainties and complexities in the actual application process, when calculating the field strength, whether to use the absolute speed or relative speed of the object needs to be judged according to the actual situation. The risk trend generated by the absolute speed must be consistent with the driving direction, which is a global perspective; while the risk trend generated by the relative speed is more in line with the situation faced by the vehicle. For example, when the front vehicle is traveling at a slower speed, although its actual speed direction is still forward, the impact on the vehicle should be directed backward to the vehicle. In other words, the risk situation of the vehicle is actually inconsistent with the risk situation of the road. Therefore, in actual applications, other relative speeds are used to calculate the field energy, and multiplied by the virtual mass of the vehicle (absolute speed) to obtain the field force.

[0156] Combination Figures 8 to 15 As shown, a method for evaluating the safety situation of a traffic environment of an embodiment of the present application is described in detail using an embodiment.

[0157] The method for evaluating the traffic environment safety situation in the embodiment of the present application can be based on the following: Figure 8 The structure shown is implemented.

[0158] like Figure 8 As shown, the functions of each module of the structure involved in the embodiment of the present application are as follows:

[0159] BaseNode: The common base class of all modules, a virtual base class, contains a pure virtual function init, which is the initialization function. This class is used as an interface and cannot be instantiated.

[0160] AdderNode: The base class of the adder, inherited from BaseNode, is a virtual base class with a new pure virtual function add, which is the most important interface of this class and is used to add two fields together. This class is used as an interface and cannot be instantiated.

[0161] AdaptorNode: The base class of the adapter, inherited from BaseNode, is a virtual base class, which adds a pure virtual function get_data, which is the most important interface of this class and is used to return a frame of data. This class is used as an interface and cannot be instantiated.

[0162] AlgorithmNode: The base class of the subfield algorithm, inherited from BaseNode, is a virtual base class, which adds a pure virtual function run and a pure virtual function energy, which are used to calculate the entire field of an object and the field strength at a certain point respectively. This class is used as an interface and cannot be instantiated.

[0163] ReprocessorNode: The base class of the postprocessor, inherited from BaseNode, is a virtual base class, which has a new pure virtual function run, which is the most important interface of this class and is used to post-process a field (such as the maximum limit mentioned above). This class is used as an interface and cannot be instantiated.

[0164] LaneNode: The base class of the potential field algorithm, inherited from BaseNode, is a virtual base class with a new pure virtual function generate, which is used to calculate the potential field generated by the lane line. This class is used as an interface and cannot be instantiated.

[0165] Adder1: An implementation of the adder, inherited from AdderNode (there may be other implementations, this is just an example). The virtual functions in AdderNode are implemented, and the functions of this module are realized with an algorithm.

[0166] Adaptor1: An implementation of the adapter, inherited from AdaptorNode. The virtual functions in AdaptorNode are implemented, and the functions of the adapter are realized by an algorithm.

[0167] Algorithm_default: An implementation of the subfield algorithm module, inherited from AlgorithmNode. The virtual functions in AlgorithmNode are implemented, and the functions of the subfield algorithm are realized with an algorithm.

[0168] Reprocessor1: An implementation of a postprocessor, inherited from ReprocessorNode. Implements the virtual functions in ReprocessorNode and implements the functions of the postprocessor with an algorithm

[0169] LanesDefault: An implementation of the potential field algorithm, inherited from LanesNode. The virtual function in LanesNode is implemented, and the function of the potential field algorithm module is realized by an algorithm.

[0170] Module Manager:

[0171] AdderManager: Adder manager. It defines an important function get_adder, which is used to read the adder name from the configuration file, import the adder module, and instantiate it. It returns a pointer to an adder instance.

[0172] AdaptorManager: Adapter Manager. Defines an important function get_adaptor, which is used to read the adapter name from the configuration file, import the adapter module, and instantiate it. Returns a pointer to an adapter instance.

[0173] ReprocessorManager: Postprocessor manager. Defines an important function get_reprocessor, which is used to read the postprocessor from the configuration file, import the adapter module, and instantiate it. Returns a pointer to a postprocessor instance.

[0174] AlgorithmManager: Subfield algorithm manager. Defines an important function get_algorithm, which is used to read the subfield algorithm from the configuration file, import the algorithm module, and instantiate it. Returns a map whose key is the label of the obstacle and whose value is the pointer to the subfield algorithm instance.

[0175] LanesManager: Potential field algorithm manager. Defines an important function get_lanes, which is used to read the potential field algorithm from the configuration file, import the algorithm module, and instantiate it. Returns a pointer to a potential field instance.

[0176] When the program starts, the embodiment of the present application can first initialize each main module, including the adapter module, the superimposer module, the subfield algorithm module and the post-processing algorithm module. One of the important purposes of initialization is to select a suitable replacement component for each module. During the initialization process, we select components for these four modules in turn and configure their corresponding parameters.

[0177] The adapter reads the data and then calculates the kinetic energy field and potential energy field in sequence. Each time a field is calculated, the new field is superimposed on the old field until all the kinetic energy fields and potential energy fields have been calculated and superimposed.

[0178] Then, post-processing is performed, for example, the field strength of points exceeding a certain threshold is set to the maximum value.

[0179] Finally, the calculated results are output to the downstream modules. At the same time, a visualization module can be designed to conveniently observe the overall field strength state.

[0180] like Figure 9-11 As shown, this is the field strength effect when the relative distance Δd and relative speed Δv from the rear vehicle to the front vehicle, that is, the vehicle itself, take different values.

[0181] in, Fig. 9 The field strength effect is Δd=60m, Δv=10m / s.

[0182] Fig.10 The field strength effect is Δd=60m, Δv=20m / s.

[0183] Fig.11 The field strength effect is Δd=60m, Δv=30m / s.

[0184] It can be seen that when the relative distance remains unchanged, as the relative speed of the rear vehicle increases, the field strength at the location of the vehicle also increases.

[0185] like Figure 12-14 As shown, in the embodiment of the present application, it can be seen that the complex environment simulation effect is achieved by using actual road data.

[0186] like Fig.12 As shown in the figure, there is an electric vehicle moving forward on the left side of the ego vehicle. Since the ego vehicle is turning left at this time, the relative speed between it and the electric vehicle is relatively large, and it poses a greater risk to the left front of the ego vehicle. Since the relative speed direction of the electric vehicle is not completely parallel to the relative position, the algorithm calculates the electric vehicle as a medium-risk object, which is displayed in dark yellow in the visualization interface.

[0187] like Fig.13 As shown in the figure, there is a vehicle with a horizontal speed direction on the left side of the ego vehicle. The distance between the vehicle and the ego vehicle is relatively close, and the relative speed direction is also basically parallel to the relative position. However, due to the low absolute speed of the vehicle, the algorithm calculates the vehicle as a medium-risk object, which is displayed in dark yellow in the visualization interface.

[0188] like Fig.14 As shown, there is an electric car quite close to the right rear. Since the electric car still has a certain speed at this time, it can be considered to be very risky. The algorithm calculates the electric car as a high-risk object and presents it in dark red in the visualization interface.

[0189] Furthermore, if Fig.15 As shown, the embodiment of the present application can take the scene CPFA-25 as an example, perform a quantitative analysis on the algorithm, and compare the field force with TTC (Time To Collision), 1 / TTC, relative distance, etc. to achieve quantitative analysis of the scene.

[0190] According to the evaluation method of the traffic environment safety situation proposed in the embodiment of the present application, the basic field strength model of the vehicle's safety field can be constructed in combination with the road data and relative perception data of the road the vehicle is currently traveling on, and the field strength generated by any single risk source at multiple target positions of the vehicle can be determined based on the basic field strength model of the safety field. According to the preset attenuation coefficient, road data, relative perception data and the number of risk sources, the actual field strength after attenuation of any single risk source is calculated, and in the case of multiple risk sources, the actual field strength generated by multiple risk sources at each target position is superimposed to obtain the risk field strength corresponding to each target position, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength, and by fully considering multiple risk factors, the safety situation evaluation of multiple scenarios, especially complex traffic environment scenarios, is realized to support the decision-making and control of the vehicle. Thus, the technical problems in the related technology that the risk factors considered are not comprehensive, the applicable scenarios are single, and the complex and changeable traffic environment cannot be faced, resulting in difficulty in supporting the vehicle to make intelligent decisions and control, affecting the driving experience, are solved.

[0191] Next, a traffic environment safety situation assessment device proposed according to an embodiment of the present application is described with reference to the accompanying drawings.

[0192] Fig.16 It is a block diagram of a device for evaluating the safety situation of a traffic environment according to an embodiment of the present application.

[0193] like Fig.16 As shown, the traffic environment safety situation assessment device 10 includes: an acquisition module 100, a construction module 200 and an assessment module 300.

[0194] Specifically, the acquisition module 100 is used to acquire the road data of the current driving road of the vehicle and the relative perception data of any single risk source on the current driving road.

[0195] Construction module 200 is used to construct a basic field strength model of the vehicle's safety field in combination with road data and relative perception data, and determine the field strength generated by any single risk source at multiple target position points of the vehicle based on the basic field strength model of the safety field.

[0196] The evaluation module 300 is used to calculate the actual field strength of any single risk source after attenuation based on the preset attenuation coefficient, road data, relative perception data and the number of risk sources, and when there are multiple risk sources, superimpose the actual field strengths generated by multiple risk sources at each target position point to obtain the risk field strength corresponding to each target position point, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength.

[0197] Optionally, in one embodiment of the present application, the traffic environment safety situation assessment device 10 further includes: a judgment module, a first calculation module and a second calculation module.

[0198] Among them, the judgment module is used to judge whether the perception profile data and perception speed data of any single risk source in the relative perception data meet the preset uncertainty conditions.

[0199] The first calculation module is used to obtain the expanded contour data of any single risk source according to the preset expansion coefficient and the perceived contour data when the perceived contour data meets the preset uncertainty condition, and use the expanded contour data as the actual contour data of any single risk source.

[0200] The second calculation module is used to obtain estimated speed data according to a preset uncertainty range and the perception data when the perception speed data meets a preset uncertainty condition, and use the estimated speed data as the actual speed data of any single risk source.

[0201] Optionally, in one embodiment of the present application, the construction module 200 includes: a first identification unit, a first construction unit and a second construction unit.

[0202] Among them, the first identification unit is used to identify the actual movement state of any single risk source based on the relative perception data.

[0203] The first construction unit is used to construct a potential energy field model by combining road data and the virtual mass of any single risk source when the actual motion state is a static state, and use the potential energy field model as a basic field strength model of the safety field.

[0204] The second construction unit is used to construct a kinetic field model according to the current driving data of the vehicle, the structural data of the vehicle and the contour data of any single risk source when the actual motion state is a motion state, and obtain a basic field strength model of the safety field according to the kinetic field model and the potential field model.

[0205] Optionally, in one embodiment of the present application, the calculation expression of the virtual mass is:

[0206]

[0207] Among them, M a is the virtual quality, a is any single risk source, T a is the type coefficient of any single risk source, v α is the speed of any single risk source, α, β, γ are all unknown constants;

[0208] The calculation expression of the potential energy field model is:

[0209]

[0210] Among them, E p1is the potential energy generated by any single risk source, K is the virtual quality coefficient, R a is the environmental condition influencing factor, r is the distance to the center of any single risk source, and k1 is the distance coefficient;

[0211] The calculation expression of the kinetic energy field model is:

[0212] w=p(δ) / p max ,

[0213]

[0214] Among them, w is the weight of the trajectory selected by the vehicle to travel to any point, δ is the front wheel turning angle required for the vehicle to travel to any point, p(δ) is the probability density, p max is the maximum probability density, E d is the kinetic energy generated by any single risk source, E is the basic kinetic energy, Δr is the arc length from the current position to any point, r std is the standard arc length value, m is the mass of the vehicle, and v is the current speed of the vehicle.

[0215] Optionally, in one embodiment of the present application, the construction module 200 includes: a first computing unit and a second computing unit.

[0216] Wherein, the first calculation unit is used to calculate the field strength generated by any single risk source on the vehicle using the potential energy field model when the actual motion state is a stationary state.

[0217] The second calculation unit is used to calculate the static field strength generated by any single risk source on the vehicle using the potential field model when the actual motion state is a motion state, and to calculate the dynamic field strength generated by any single risk source on the vehicle using the kinetic field model, and to superimpose the static field strength and the dynamic field strength to obtain a superposition result, and use the superposition result as the field strength.

[0218] Optionally, in one embodiment of the present application, the evaluation module 300 includes: a second recognition unit, a matching unit, a third calculation unit and a fourth calculation unit.

[0219] The second identification unit is used to identify at least one attenuation factor of any single risk source based on the relative perception data and the road data.

[0220] The matching unit is used to match the corresponding road attenuation coefficient in a preset coefficient comparison table according to the road data when the attenuation factor is a constraint-type attenuation factor.

[0221] The third calculation unit is used to determine at least one occluding object corresponding to any single risk source according to the relative perception data when the attenuation factor is an occlusion-type attenuation factor, and obtain the potential energy of each occluding object using the potential energy field model to obtain the occlusion attenuation coefficient based on the potential energy.

[0222] The fourth calculation unit is used to obtain a preset attenuation coefficient according to the number of attenuation factors, the road attenuation coefficient and the occlusion attenuation coefficient.

[0223] Optionally, in one embodiment of the present application, the evaluation module includes: an acquisition unit.

[0224] The acquisition unit is used to take the actual field strength with the largest value from multiple actual field strengths at each target location point as the risk field strength.

[0225] It should be noted that the above explanation of the embodiment of the method for evaluating the safety situation of a traffic environment is also applicable to the device for evaluating the safety situation of a traffic environment of this embodiment, and will not be repeated here.

[0226] According to the traffic environment safety situation assessment device proposed in the embodiment of the present application, the basic field strength model of the vehicle's safety field can be constructed in combination with the road data and relative perception data of the vehicle's current driving road, and the field strength generated by any single risk source at multiple target position points of the vehicle can be determined based on the basic field strength model of the safety field. According to the preset attenuation coefficient, road data, relative perception data and the number of risk sources, the actual field strength after attenuation of any single risk source is calculated, and in the case of multiple risk sources, the actual field strengths generated by multiple risk sources at each target position point are superimposed to obtain the risk field strength corresponding to each target position point, so as to assess the safety situation of the vehicle in the current traffic environment according to the risk field strength, and by fully considering multiple risk factors, the safety situation assessment of multiple scenarios, especially complex traffic environment scenarios, is achieved to support the decision-making and control of the vehicle. Thus, the technical problems in the related technology that the risk factors considered are not comprehensive, the applicable scenarios are single, and the complex and changeable traffic environment cannot be faced, resulting in difficulty in supporting the vehicle to make intelligent decisions and control, affecting the driving experience, are solved.

[0227] Fig.17 A schematic diagram of the structure of a vehicle provided in an embodiment of the present application. The vehicle may include:

[0228] A memory 1701 , a processor 1702 , and a computer program stored in the memory 1701 and executable on the processor 1702 .

[0229] When the processor 1702 executes the program, the method for evaluating the traffic environment safety situation provided in the above embodiment is implemented.

[0230] Furthermore, the vehicle also includes:

[0231] The communication interface 1703 is used for communication between the memory 1701 and the processor 1702 .

[0232] The memory 1701 is used to store computer programs that can be executed on the processor 1702 .

[0233] The memory 1701 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0234] If the memory 1701, the processor 1702 and the communication interface 1703 are implemented independently, the communication interface 1703, the memory 1701 and the processor 1702 can be connected to each other through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.17 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0235] Optionally, in a specific implementation, if the memory 1701, the processor 1702 and the communication interface 1703 are integrated on a chip, the memory 1701, the processor 1702 and the communication interface 1703 can communicate with each other through an internal interface.

[0236] The processor 1702 may be a central processing unit (CPU), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0237] This embodiment also provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned method for evaluating the safety situation of a traffic environment is implemented.

[0238] An embodiment of the present invention further provides a computer program product, including a computer program. When the computer program is executed, it is used to implement the method for evaluating the safety situation of a traffic environment provided by an embodiment of the present invention.

[0239] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or N embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0240] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of this application, "N" means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0241] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, fragment or portion of code comprising one or N executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present application belong.

[0242] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or N wirings (electronic devices), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways as necessary and then storing it in a computer memory.

[0243] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiment, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0244] A person skilled in the art may understand that all or part of the steps in the method for implementing the above-mentioned embodiment may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0245] In addition, each functional unit in each embodiment of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0246] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present application. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present application.

Claims

1. A method for evaluating the safety situation of a traffic environment, characterized in that: The following steps are involved: Acquire road data of a current driving road of the vehicle and relative perception data of any single risk source on the current driving road; Building a basic field strength model of the safety field of the vehicle in combination with the road data and the relative perception data, and determining the field strength generated by any single risk source at multiple target position points of the vehicle based on the basic field strength model of the safety field; Based on the preset attenuation coefficient, the road data, the relative perception data and the number of risk sources, the actual field strength after attenuation of any single risk source is calculated, and when there are multiple risk sources, the actual field strengths generated by the multiple risk sources at each target location are superimposed to obtain the risk field strength corresponding to each target location, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength; Among them, after obtaining the relative perception data of any single risk source on the current driving road, it also includes: respectively judging whether the perception contour data and the perception speed data of any single risk source in the relative perception data meet the preset uncertainty conditions; if the perception contour data meets the preset uncertainty conditions, then according to the preset expansion coefficient and the perception contour data, the expanded contour data of any single risk source is obtained, and the expanded contour data is used as the actual contour data of any single risk source; if the perception speed data meets the preset uncertainty conditions, then according to the preset uncertainty range and the perception data, the estimated speed data is obtained, and the estimated speed data is used as the actual speed data of any single risk source.

2. The method according to claim 1, characterized in that The step of combining the road data and the relative perception data to construct a basic field strength model of the safety field of the vehicle includes: identifying the actual motion state of any single risk source based on the relative perception data; In the case where the actual motion state is a stationary state, a potential energy field model is constructed by combining the road data and the virtual mass of any single risk source, and the potential energy field model is used as the basic field strength model of the safety field; When the actual motion state is a motion state, a kinetic field model is constructed according to the current driving data of the vehicle, the structural data of the vehicle and the contour data of any single risk source, and the basic field strength model of the safety field is obtained according to the kinetic field model and the potential field model.

3. The method according to claim 2, characterized in that The calculation expression of the virtual mass is: Among them, M a is the virtual quality, a is any single risk source, T a is the type coefficient of any single risk source, v α is the speed of any single risk source, α, β, γ are all unknown constants; The calculation expression of the potential energy field model is: Among them, E p1 is the potential energy generated by any single risk source, K is the virtual quality coefficient, R a is the environmental condition influencing factor, r is the distance to the center of any single risk source, and k1 is the distance coefficient; The calculation expression of the kinetic energy field model is: w=p(δ) / p max , Where w is the weight of the trajectory selected by the vehicle to travel to any point, δ is the front wheel turning angle required for the vehicle to travel to any point, p(δ) is the probability density, p max is the maximum probability density, E d is the kinetic energy generated by any single risk source, E is the basic kinetic energy, Δr is the arc length from the current position to any point, r std is the standard arc length value, m is the mass of the vehicle, and v is the current speed of the vehicle.

4. The method according to claim 2, characterized in that: The determining the field strength generated by any single risk source on the vehicle based on the basic field strength model of the safety field includes: When the actual motion state is a stationary state, using the potential energy field model, calculating the field strength generated by any single risk source on the vehicle; When the actual motion state is a motion state, the potential energy field model is used to calculate the static field strength generated by any single risk source on the vehicle, and the kinetic energy field model is used to calculate the dynamic field strength generated by any single risk source on the vehicle, and the static field strength and the dynamic field strength are superimposed to obtain a superposition result, and the superposition result is used as the field strength.

5. The method according to claim 2, characterized in that: The calculating the actual field strength of any single risk source after attenuation based on the preset attenuation coefficient, the road data, the relative perception data and the number of risk sources includes: identifying at least one attenuation factor of any single risk source based on the relative perception data and the road data; In the case where the attenuation factor is a constrained attenuation factor, matching a corresponding road attenuation coefficient in a preset coefficient comparison table according to the road data; In the case where the attenuation factor is an occlusion-type attenuation factor, determining at least one occluding object corresponding to any single risk source according to the relative perception data, and obtaining the potential energy of each occluding object using the potential energy field model, so as to obtain an occlusion attenuation coefficient based on the potential energy; The preset attenuation coefficient is obtained according to the number of the attenuation factors, the road attenuation coefficient and the occlusion attenuation coefficient.

6. The method according to claim 1, characterized in that The superposition of the actual field strengths generated by the multiple risk sources at each target location point to obtain the risk field strength corresponding to each target location point includes: The actual field strength with the largest value among the multiple actual field strengths at each target location point is taken as the risk field strength.

7. A traffic environment safety situation assessment device, characterized in that: include: An acquisition module, used to acquire road data of a current driving road of the vehicle and relative perception data of any single risk source on the current driving road; A construction module, used to construct a basic field strength model of the safety field of the vehicle in combination with the road data and the relative perception data, and determine the field strength generated by any single risk source at multiple target position points of the vehicle based on the basic field strength model of the safety field; An evaluation module, for calculating the actual field strength after attenuation of any single risk source based on a preset attenuation coefficient, the road data, the relative perception data and the number of risk sources, and in the case of multiple risk sources, superimposing the actual field strengths generated by the multiple risk sources at each target location point to obtain the risk field strength corresponding to each target location point, so as to evaluate the safety situation of the vehicle in the current traffic environment according to the risk field strength; A judgment module, used to judge whether the perception profile data and the perception speed data of any single risk source in the relative perception data meet a preset uncertainty condition; A first calculation module is used for obtaining the expansion contour data of any single risk source according to a preset expansion coefficient and the perception contour data when the perception contour data satisfies the preset uncertainty condition, and using the expansion contour data as the actual contour data of any single risk source; The second calculation module is used to obtain estimated speed data according to a preset uncertainty range and the perceived data when the perceived speed data meets the preset uncertainty condition, and use the estimated speed data as the actual speed data of any single risk source.

8. A vehicle, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method for evaluating the safety situation of a traffic environment as described in any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method for evaluating the safety situation of a traffic environment as described in any one of claims 1 to 6.

10. A computer program product, characterized in that It comprises a computer program, which, when executed, is used to implement the method for evaluating the safety situation of a traffic environment as described in any one of claims 1-6.

Images