A driving risk identification and early warning method, system, device and medium

Abstract

The application relates to a driving risk identification and early warning method, system, device and medium, which comprises the following steps: collecting driving environment and driving data of a vehicle; based on the driving environment and driving data of the vehicle, fitting real-time parameters of a high-risk driving model through a genetic algorithm to obtain model parameters representing driver operation characteristics and vehicle motion characteristics; based on the obtained model parameters, classifying and identifying high-risk driving behaviors through an SVM classification algorithm, and early warning of driving risks. The application is oriented to traffic safety, uses driving data to carry out comprehensive evaluation and early warning of driving risks under the coupling influence of factors, that is, the coupling influence relationship of the 'driver-vehicle' system under the influence of road environment, individual characteristics of a driver and driving state is comprehensively considered to identify and early warn driving risks. The application can be widely applied to the field of driving safety.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of driving safety, and particularly relates to a driving risk identification and early warning method, system, device and medium based on driving data. BACKGROUND

[0002] In order to ensure road traffic driving safety, it is crucial to give early warning of possible risks in driving. At present, a key driving risk identification and early warning method is to monitor the state of a driver and evaluate the safety, and to give early warning of driving risks. Since traffic accidents are largely related to drivers, the influence of a driver on driving risks is closely related to individual characteristics, perception and decision-making characteristics and driving state of the driver. Therefore, the focus of driving safety protection is to evaluate the safety of the driver and give early warning of driving risks. Therefore, the core challenge of driving safety protection and risk early warning is to comprehensively mine and quantitatively identify the indicators of driver safety characteristics and driving risks, which is not only a key technical challenge faced by the current research field, but also a hot issue of common concern in the academic and industrial circles.

[0003] With the technological progress of sensors such as radar and camera, the auxiliary driving perception ability of a driving vehicle and the driver monitoring system have also been further developed. Existing researches have begun to evaluate the safety of the state of a driver from the perspectives of driver facial video, driver operation characteristics, driver physiological characteristics and in-cabin voice. For example, by monitoring driver facial feature points and extracting ROI features, indexes such as eye closure time percentage, continuous eye closure duration, blink frequency and mouth aspect ratio are obtained, and in combination with heart rate, electromyography and other indexes, the fatigue state of the driver is monitored; by monitoring the pitch, tone envelope and frequency spectrum of the sound emitted by the driver, the emotional state of the driver is monitored; by monitoring the position of the focal point of the line of sight of the driver, in combination with indexes such as the movement of the driver in the cabin, the distraction state of the driver is monitored.

[0004] In recent years, with the development of artificial intelligence technology, machine vision and affective computing algorithms have made significant progress in monitoring the risk states of driver fatigue, emotion and distraction. However, these methods often monitor and evaluate and give early warning of a single state of the driver, and rarely consider the influence of personal characteristics of the driver and the traffic environment, so it is difficult to unify the early warning thresholds of various risk states, and there is also a lack of evaluation methods for coupled risks under the influence of multiple factors. SUMMARY

[0005] In view of the above problems, the present application aims to provide a driving risk identification and early warning method, system, device and medium based on driving data, which is oriented to traffic safety and uses driving data to carry out comprehensive evaluation and early warning of driving risk under the coupling influence of factors, that is, the coupling influence relationship of the "driver-vehicle" system under the influence of road environment, individual characteristics of the driver and driving state is comprehensively considered to identify and early warn the driving risk.

[0006] To achieve the above-mentioned purpose, the present application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a driving risk identification and early warning method, comprising the following steps:

[0008] Collecting driving environment and driving data of the vehicle;

[0009] Based on the driving environment and driving data of the vehicle, fitting the real-time parameters of the high-risk driving model by using genetic algorithm to obtain model parameters representing the driving operation characteristics and vehicle motion characteristics;

[0010] Based on the model parameters obtained by fitting, classifying and identifying the high-risk driving behavior by using SVM classification algorithm, and early warning the driving risk.

[0011] Further, the collecting of the driving environment and driving data of the vehicle comprises:

[0012] Collecting the driving data of the host vehicle in road traffic by using GPS and IMU sensors, and modeling the vehicle motion in the Frenet coordinate system;

[0013] Collecting the road environment data and the position and speed information of other traffic participants around the host vehicle by using joint perception of radar and camera;

[0014] Fusing the driving data, road environment data and position and speed data of other traffic participants of the host vehicle within a preset time period to form the continuous driving time series data of the host vehicle.

[0015] Further, the fitting of the real-time parameters of the high-risk driving model by using genetic algorithm based on the driving environment and driving data of the vehicle to obtain model parameters representing the driving operation characteristics and vehicle motion characteristics comprises:

[0016] Modeling the vehicle motion process by considering the risk perception and expected trade-off decision process of the driving subject in view of the vehicle motion characteristics and risk characteristics to obtain a high-risk driving model;

[0017] Based on the driving environment and driving data of the vehicle and the high-risk driving model, fitting the parameters of the high-risk driving model by using genetic algorithm to obtain model parameters representing the driving operation characteristics and vehicle motion characteristics.

[0018] Further, the vehicle motion characteristics and risk characteristics are considered, the risk perception and expected trade-off decision process of the driver are comprehensively considered, the vehicle motion process is modeled, a high-risk driving model is obtained, including:

[0019] The risk field theory is used to model the perception risk distribution of the driver to the current road space, and the comprehensive risk of the road space is quantified based on the risk distribution result;

[0020] The expectation of the driver to the current driving task is quantified by the ratio of target displacement to expected maximum displacement;

[0021] Based on the driving risk perceived by the driver and the expectation of the current driving task, the driving decision process is modeled, and the risk preference and decision characteristics of the driver are quantified by model parameters, so as to obtain the decision target point of the driver at each time step and realize the simulation of driving.

[0022] Further, the risk field theory is used to model the perception risk distribution of the driver to the current road space, which is expressed as:

[0023]

[0024] In the formula, R represents risk, (d, s) respectively represent the position coordinates of the host vehicle in the Frenet reference frame, v j,s and v j,d respectively represent the speed of object j in the Frenet longitudinal direction and transverse direction in the road space, T0 represents the reaction time of the driver; k s , k d represent constant parameters suitable for the risk perception characteristics of the driver; β is a risk influence factor parameter considering the severity of the collision consequences; Δs j , Δd j respectively represent the longitudinal and transverse distances from object j to the host vehicle in the road space.

[0025] Further, the expectation of the driver to the current driving task is quantified by the ratio of target displacement to expected maximum displacement, which is expressed as:

[0026]

[0027] In the formula, Δs represents the longitudinal displacement of the target point relative to the current position; P exp is the maximum expectation of the driver to the motion distance; v s represents the current longitudinal motion speed of the vehicle, a exp represents the expected acceleration of the driver, T pre represents the preview time of the driver.

[0028] Further, the driving decision-making process is modeled based on the driving risk R perceived by the driving subject and the expectation p of the driving subject for the current driving task, and the risk preference and decision-making characteristics of the driving subject are quantified by model parameters to obtain the decision-making target point of the driving subject at each time step, and the simulation of driving is realized, including:

[0029] The driving decision-making process is modeled as a trade-off between the driving risk R perceived by the driving personnel and the expectation p of the driving personnel for the driving task, and the relationship between the two is described by a decision curve.

[0030] According to the utility theory, the decision-making characteristics of the driver are modeled as a decision curve, which is divided into four types: neutral, aggressive, conservative and mixed.

[0031] The decision-making feasible region of the driving risk expectation trade-off is determined, and the risk preference and decision-making characteristics of the driving subject are quantified by the high-risk driving model parameter set.

[0032] In a second aspect, the present application provides a driving risk identification and early warning system, comprising:

[0033] A real-time data acquisition module is used to acquire driving environment and driving data of the vehicle;

[0034] A high-risk driving model real-time parameter fitting module is used to fit the high-risk driving model real-time parameters based on the driving environment and driving data of the vehicle by genetic algorithm, and obtain model parameters representing the driving operation characteristics and vehicle motion characteristics;

[0035] A driving risk identification and early warning module is used to classify and identify high-risk driving behaviors by SVM classification algorithm based on the high-risk driving model parameters fitted from the driving data, and to warn the driving risk.

[0036] In a third aspect, the present application provides a computer readable storage medium storing one or more programs, the one or more programs including instructions that, when executed by a computing device, cause the computing device to perform any of the methods.

[0037] In a fourth aspect, the present application provides a computing device, comprising one or more processors and a memory, the memory storing one or more programs and being configured to execute the one or more programs with the one or more processors, the one or more programs including instructions for executing any of the methods.

[0038] The present application has the following advantages due to the above technical solutions:

[0039] 1、The application can recognize and warn the comprehensive driving risk under the influence of the coupling state of the driver-vehicle system by combining the driving environment data with the driving data, fitting the high-risk driving model digitally, and parameterizing the driving risk characteristics. This method makes full use of the motion characteristics of the vehicle as a direct influencing factor of driving risk, quantifies the influence of individual characteristics, decision characteristics, operation characteristics, and driving state of the driver on driving risk, and extracts the parameters of driving risk through high-risk driving model parameter fitting. The driver and the vehicle can be regarded as a whole, and compared with the monitoring and warning method for the single state of the driver, the coupling influence of different factors on driving risk can be more systematically and fully reflected, so that the overall driving risk can be more accurately evaluated.

[0040] 2、The risk recognition and warning based on model fitting parameters can improve the operation speed, parameterize and warn the driving risk in real time, thereby reducing the human factors in traffic accidents and improving the road traffic safety level.

[0041] Therefore, the application can be widely applied in the field of driving safety. BRIEF DESCRIPTION OF DRAWINGS

[0042] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not intended to limit the scope of the application. Throughout the drawings, the same reference designates the same elements. In the drawings:

[0043] Figure 1 is a flow chart of the driving risk recognition and warning method provided in the embodiments of the application;

[0044] Figure 2 is a quantitative schematic diagram of modeling the comprehensive risk of road space provided in the embodiments of the application;

[0045] Figure 3 is a different decision curve schematic diagram provided in the embodiments of the application;

[0046] Figure 4 is a decision feasible region schematic diagram of driving risk expected trade-off provided in the embodiments of the application. DETAILED DESCRIPTION

[0047] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the described embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art belong to the scope of protection of the present application.

[0048] It should be noted that the terms used herein are only intended to describe specific embodiments, and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form, unless the context clearly indicates otherwise, and it should also be understood that when the terms "comprise" and / or "include" are used in the specification, there is a feature, step, operation, device, component and / or combination thereof.

[0049] The driving risk monitoring and early warning system is in a period of rapid development. At present, the research is mostly focused on the monitoring index mining and early warning algorithm development for the risk state, and few studies propose a comprehensive early warning method for the coupling state, that is, how to carry out risk assessment and early warning for the "driver-vehicle" complex system under the influence of various factors such as driver characteristics and different driving states, which has not been fully researched and technically supported.

[0050] Therefore, in some embodiments of the present application, a driving risk identification and early warning method is provided, which starts from the motion and interaction characteristics of the "driver-vehicle" system and other traffic participants, which are direct influencing factors of driving risk, characterizes the risk perception and decision-making characteristics of the driver through parameterization, quantifies the driving risk, carries out coupling state safety assessment, and carries out early warning for high-risk driving behavior. The simulation results of the natural driving data set verify the above problems and provide a new idea for the safety monitoring of the driver and the technical support for the guarantee of traffic safety.

[0051] Correspondingly, in some other embodiments of the present application, a driving risk identification and early warning system, device and medium are provided.

[0052] Embodiment 1

[0053] As shown in Figure 1 The present application provides a driving risk identification and early warning method, which collects road traffic and driving data, fits high-risk driving model parameters, and carries out driving risk identification and early warning based on feature parameters, which can effectively carry out driving risk comprehensive assessment and safety early warning under the influence of multiple factors such as individual characteristics, decision-making characteristics and driving state of the driver. Specifically, the method comprises the following steps:

[0054] Step 1: Collecting driving environment and driving data of the vehicle;

[0055] Step 2: Based on the driving environment and driving data of the vehicle, fitting the real-time parameters of the high-risk driving model through genetic algorithm to obtain model parameters representing the driving operation characteristics and vehicle motion characteristics;

[0056] Step 3: Based on the model parameters obtained by fitting, classifying and identifying the high-risk driving behavior through SVM classification algorithm, and warning the driving risk.

[0057] Further, in the above step 1, when collecting the driving environment and driving data of the vehicle, the following steps are included:

[0058] Step 1.1: Collecting driving data such as position, speed and acceleration of the host vehicle (i.e. the vehicle driven by the driver) in road traffic through GPS, IMU and other sensors, and modeling the vehicle motion in the Frenet coordinate system;

[0059] Step 1.2: Collecting road environment data such as lanes, obstacles and traffic signals around the host vehicle, as well as position and speed information of other traffic participants such as motor vehicles, non-motor vehicles and pedestrians through joint perception of radar and cameras;

[0060] Step 1.3: Fusing the host vehicle driving data, road environment data and position and speed data of other traffic participants within a preset time period to form continuous driving time series data of the host vehicle.

[0061] Further, in the above step 2, based on the driving environment and driving data of the vehicle, the real-time parameters of the high-risk driving model are fitted through genetic algorithm to obtain model parameters representing the driving operation characteristics and vehicle motion characteristics, including the following steps:

[0062] Step 2.1: Modeling the vehicle motion process by considering the risk perception and expected trade-off decision process of the driving subject for vehicle motion characteristics and risk characteristics, obtaining a high-risk driving model, and realizing simulation and emulation of high-risk vehicle motion driving behavior;

[0063] Step 2.2: Based on the driving environment and driving data of the vehicle and the high-risk driving model, fitting the parameters of the high-risk driving model using genetic algorithm to obtain model parameters representing the driving operation characteristics and vehicle motion characteristics.

[0064] Further, in the above step 2.1, this embodiment proposes a high-risk driving model to parameterize the risk of the driving process. The high-risk driving model consists of risk perception and driving decision.

[0065] Specifically, the following steps are included:

[0066] Step 2.1.1: Model the driving subject's perception of the risk distribution R of the current road space using the risk field theory, and quantify the comprehensive risk of the road space based on the perceived risk distribution.

[0067] In this embodiment, the driving risk perception process models the risk distribution of the current road space using the risk field theory:

[0068]

[0069] where R represents the risk, (d, s) respectively represent the position coordinates of the host vehicle in the Frenet reference frame, v j,s and v j,d respectively represent the longitudinal and lateral velocities of object j in the road space along the Frenet reference frame, T0 represents the reaction time of the driver; k s , k d represent constant parameters adapted to the risk perception characteristics of the driver; β is a risk influence factor parameter considering the severity of the collision consequences; Δs j , Δd j are respectively the longitudinal and lateral distances from object j to the host vehicle in the road space. The modeling and quantification of the comprehensive risk of the road space based on the above formula is shown in Figure 2 . The perception of driving risk by the driving subject has a certain time delay compared to the actual road situation, which is determined by the reaction time T0 of the driver.

[0070] Step 2.1.2: Quantify the driving subject's expectation p of the current driving task by the ratio of the target displacement to the expected maximum displacement.

[0071] Specifically, if the driver travels in a way that best meets his or her driving expectations, the maximum expectation of the driver's movement distance after a preview time T pre can be quantified by P exp :

[0072]

[0073] where v s represents the current longitudinal movement speed of the vehicle, a exp represents the expected acceleration of the driver, and T pre represents the preview time of the driver.

[0074] The driving subject has a certain expectation p of the current driving task, and this embodiment quantifies it by the ratio of the target displacement to the expected maximum displacement:

[0075]

[0076] wherein Δs represents the road longitudinal displacement of the target point relative to the current position.

[0077] Step 2.1.3: Based on the driving risk R perceived by the driving subject and the expectation p of the driving subject to the current driving task, the driving decision-making process is modeled, and the risk preference and decision-making characteristics of the driving subject are quantified by the model parameters, to obtain the decision-making target point of the driving subject at each time step, and to realize the simulation of driving.

[0078] Specifically, the following steps are included:

[0079] ①The driving decision-making process is modeled as a trade-off between the driving risk R perceived by the driving personnel and the expectation p of the driving personnel to the driving task, and the relationship between the two is described by a decision-making curve:

[0080] p = R γ

[0081] wherein γ is a decision-making characteristic parameter of the driver.

[0082] ②According to the utility theory, the decision-making characteristics of the driver are modeled as a decision-making curve, which is divided into four types: neutral type, aggressive type, conservative type and mixed type.

[0083] The acceptance of the driving risk and the driving expectation level of the driver can be modeled by the decision-making curve, and the acceptable driving risk of the driver is different corresponding to different driving expectation levels. Different drivers have different trade-off characteristics between driving risk and driving expectation implementation, and according to the utility theory, the decision-making style of the driver can be divided into four types: neutral type, aggressive type, conservative type and mixed type, and the decision-making curves of different decision-making styles have different characteristics. The decision-making curve is divided into four types: neutral type, aggressive type, conservative type and mixed type, as shown in FIG. 1. The risk preference and decision-making characteristics of the driving personnel are quantified by the decision-making curve parameters, and the decision-making target point of the driving personnel at each time step is obtained, and the simulation of driving is realized. Figure 3

[0084] ③The decision-making feasible region of the driving risk expectation trade-off is determined, and the risk preference and decision-making characteristics of the driving subject are quantified by the high-risk driving model parameter set.

[0085] Considering that the driver has a certain acceptable risk level R a Therefore, the lower bound of the decision-making feasible region of the driver can be represented by the following formula:

[0086]

[0087] Based on the risk balance theory, it is known that the driver will not blindly pursue the maximum expected level during driving, and the comfort degree is also one of the influencing factors of the driver's behavior, therefore, the decision-making curve p L ​In the upper region, the driver tends to choose to operate the vehicle to make its driving state within an acceptable range, so the upper bound of the decision feasible region can be quantified by the following formula:

[0088]

[0089] In the formula, ω1 represents the weight of the conservative component in the mixed decision curve, 0≤ω1≤1, and the greater ω1 is, the greater the proportion of the risk aversion component in the driver's decision preference, and the more conservative the driver's driving style is;

[0090] ω2 represents the weight of the aggressive component in the mixed decision curve, and ω1+ω2=1; the greater ω2 is, the greater the proportion of the risk preference component in the driver's decision preference, and the more aggressive the driver's driving style is;

[0091] γ1 is a characteristic parameter of the conservative decision curve, representing the risk aversion degree of the conservative component in the decision curve, and γ1≥1; and the greater γ1 is, the farther the conservative decision curve is from the neutral decision curve, indicating that the higher the risk aversion degree in the conservative decision curve, the more conservative the driver's style is;

[0092] γ2 is a characteristic parameter of the aggressive decision curve, representing the risk aversion degree of the aggressive component in the decision curve, and 0<γ2≤1; and the greater γ2 is, the closer the aggressive decision curve is to the neutral decision curve, indicating that the higher the risk aversion degree in the aggressive decision curve, the more conservative the driver's style is;

[0093] R a represents the maximum risk level acceptable to the driver when the expected realization level is 0, and 0<R a <1; and the greater R a , the greater the risk acceptable to the driver under zero expectation, and the more aggressive the driver's style is;

[0094] m represents the maximum expected realization level that the driver is willing to pursue under zero risk, and 0<m<1; and the greater m is, the higher the pursuit of realization by the driver, and the more aggressive the driving style of the driver is;

[0095] R m represents the maximum risk level acceptable to the driver under any condition, and R a <R m <1; and the greater R m , the higher the risk tolerance of the driver, and the more aggressive the driver's style is;

[0096] T delay represents the delay time of the driver from the perception of the current state to the completion of the decision and the response to the change in vehicle acceleration; T delayThe larger, the longer the reaction time of the driver.

[0097] Accordingly, the driving risk and the decision-making feasible region of the expected trade-off can be represented as Figure 4 the middle gray area. The embodiment sets the high-risk driving model parameter set as {ω1, ω2, γ1, γ2, R a ,m,R m ,R delay}.

[0098] Further, in the actual driving process, the road traffic timing information and the host vehicle driving data in the driving process can be collected by the sensor in step 1, combined with the high-risk driving behavior model construction and parameter fitting, and the driving segment risk features and parameters can be extracted.

[0099] Based on the high-risk driving model constructed in step 2.1 and the road environment and host vehicle driving data collected in step 1, the parameters of the driving model are fitted using a genetic algorithm, so as to obtain the real-time parameter set {ω1, ω2, γ1, γ2, R a ,m,R m ,T delay} of the current “driver-vehicle” motion.

[0100] Further, in the above step 3, based on the parameter set {ω1, ω2, γ1, γ2, R a ,m,R m ,T delay} of the high-risk driving model, the SVM classification method can be used to classify the driving risk by taking the model parameter values obtained by fitting the high-risk driving model in the actual driving process as input. Different driving model parameters will be classified as “low-risk” or “high-risk”, so as to provide a warning for high-risk driving. Considering that the number of high-risk samples is much smaller than the number of low-risk samples, a higher weight is set for the high-risk samples to balance the influence of the uneven sample quantity. The driving risk identification and warning method is verified, and the identification accuracy of the high-risk driving behavior is 94.6%.

[0101] Embodiment 2

[0102] The embodiment further introduces the driving risk identification and warning method based on driving data provided by the application.

[0103] The driving data used in the embodiment is derived from CitySim aerial vehicle trajectory data.

[0104] CitySim is a large-scale, scene-rich, high-precision unmanned aerial vehicle trajectory open data set launched by researchers from Tongji University and the University of Central Florida in 2022. The data covers driving scenes in multiple countries, various weather conditions, and multiple locations, and collects driving data under various road types, including basic sections of expressways, expressway merging and exit ramps, urban expressways, signalized intersections, roundabouts, etc. The sampling frame rate is 30fps, and the total recording time is more than 1200 minutes. Moreover, the CitySim data set provides 7 key point positioning of the vehicle, which can help researchers understand the position and boundary information of the vehicle on the road and reconstruct the driving scene in the simulation driving scene, providing valuable data support for natural driving vehicle motion simulation research.

[0105] By simulating and verifying the high-risk driving model through the natural driving time sequence data in the CitySim data set, the accuracy of the model in describing the driving subject and vehicle motion characteristics can be obtained, and the quantitative comparison of the accuracy between different models is shown in Table 1.

[0106] Table 1 Error of each driving model on CitySim data set

[0107] Driving model Model error Driving model Model error BL-OV 8.966 BLVD 10.255 FVD 4.810 GHR 1.776 Gipps 0.346 IBDM 3.914 IDM 3.647 OVM 5.250 RRDM 0.065 The model 0.0199

[0108] From the table, it can be known that the high-risk driving model used in the present application can better simulate the motion of natural driving vehicles, and the corresponding parameters such as reaction time can well represent the individual characteristics, decision-making characteristics and operation characteristics of the driver.

[0109] Based on the above high-risk driving model and the collected road environment and host vehicle driving data, the parameters of the driving model are fitted using a genetic algorithm, thereby obtaining the real-time parameter set {ω1, ω2, γ1, γ2, R a ,m,R m ,T delay} of the current "driver-vehicle" motion.

[0110] Based on the parameter set {ω1, ω2, γ1, γ2, R a ,m,R m ,T delayUsing the SVM classification method, the model parameters obtained by fitting a high-risk driving model to actual driving experiences are used as input to classify driving risks. Different driving model parameter inputs are classified as "medium-low risk" or "high risk," thereby providing early warnings for high-risk driving. Given that the number of high-risk samples is significantly smaller than that of medium- and low-risk samples, a higher weight is assigned to high-risk samples to balance the impact of the sample size imbalance. This driving risk identification and early warning method has been validated, achieving an accuracy rate of 94.6% for identifying high-risk driving behaviors.

[0111] Example 3

[0112] The above-mentioned embodiment 1 provides a driving risk identification and warning method based on driving data. Correspondingly, this embodiment provides a driving risk identification and warning system based on driving data. The system provided in this embodiment can implement the driving risk identification and warning method based on driving data in embodiment 1. The system can be implemented through software, hardware, or a combination of software and hardware. For example, the system may include integrated or separate functional modules or functional units to execute the corresponding steps in each method of embodiment 1. Since the system of this embodiment is basically similar to the method embodiment, the description process of this embodiment is relatively simple. For relevant matters, please refer to the partial description of embodiment 1. The embodiment of the system provided in this embodiment is merely illustrative.

[0113] The driving risk identification and warning system provided in this embodiment includes:

[0114] Real-time data acquisition module, used to collect vehicle driving environment and driving data;

[0115] The high-risk driving model real-time parameter fitting module is used to fit the real-time parameters of the high-risk driving model based on the vehicle driving environment and driving data through a genetic algorithm to obtain model parameters that characterize the driver's operating characteristics and vehicle motion characteristics;

[0116] The driving risk identification and warning module is used to classify and identify high-risk driving behaviors based on the high-risk driving model parameters obtained by fitting driving data through the SVM classification algorithm, and to issue warnings for driving risks.

[0117] Example 4

[0118] This embodiment provides a processing device corresponding to the driving risk identification and warning method based on driving data provided in this embodiment 1. The processing device can be a processing device for a client, such as a mobile phone, laptop computer, tablet computer, desktop computer, etc., to execute the method of embodiment 1.

[0119] The processing device includes a processor, a memory, a communication interface and a bus, the processor, the memory and the communication interface are connected through the bus to complete the communication between each other. The memory stores a computer program capable of running on the processor, and the processor executes the computer program to perform the driving risk identification and early warning method based on driving data provided in Embodiment 1.

[0120] Preferably, the memory can be a high-speed random access memory (RAM: Random Access Memory), and can also include a non-volatile memory, for example, at least one disk memory.

[0121] Preferably, the processor can be a central processing unit (CPU), a digital signal processor (DSP) and various types of general-purpose processors, which are not limited here.

[0122] Embodiment 5

[0123] The driving risk identification and early warning method based on driving data of Embodiment 1 can be specifically implemented as a computer program product, and the computer program product can include a computer readable storage medium, which is loaded with computer readable program instructions for executing the driving risk identification and early warning method based on driving data described in Embodiment 1.

[0124] The computer readable storage medium can be a tangible device that maintains and stores instructions for use by an instruction execution device. The computer readable storage medium can be, for example but not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any combination of the above.

[0125] Those skilled in the art will understand that the embodiments of the present application can be provided as a method, a system or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0126] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0127] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0128] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 one or more flow or blocks

[0129] Finally, it should be noted that the above-mentioned embodiments are merely used to illustrate the technical solutions of the present application, rather than limit the technical solutions of the present application. Although the present application is described in detail with reference to the above embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced, and any modification or replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A driving risk identification and warning method, characterized in that: The following steps are involved: Collect the vehicle's driving environment and driving data; Based on the vehicle driving environment and driving data, a genetic algorithm is used to fit the real-time parameters of the high-risk driving model to obtain model parameters that characterize the driver's operating characteristics and vehicle motion characteristics; The following steps are involved: Based on the vehicle's motion characteristics and risk characteristics, the vehicle motion process is modeled by comprehensively considering the driver's risk perception and expected trade-off decision-making process to obtain a high-risk driving model. This includes: using risk field theory to model the driver's perceived risk distribution of the current road space, and quantifying the comprehensive risk of the road space based on the risk distribution results; quantifying the driver's expectations for the current driving task through the ratio of target displacement to expected maximum displacement; based on the driver's perceived driving risk and his expectations for the current driving task, the driving decision-making process is modeled, and the driver's risk preference and decision-making characteristics are quantified using model parameters. The decision target point of the driver at each time step is determined to achieve driving simulation. Based on the vehicle driving environment, driving data and high-risk driving model, a genetic algorithm is used to fit the parameters of the high-risk driving model to obtain model parameters that characterize the driver's operating characteristics and vehicle motion characteristics; Based on the fitted model parameters, the SVM classification algorithm is used to classify and identify high-risk driving behaviors and issue early warnings for driving risks.

2. A driving risk identification and warning method according to claim 1, characterized in that: The collection of the vehicle's driving environment and driving data includes: The driving data of the main vehicle in road traffic is collected through GPS and IMU sensors, and the Frenet coordinate system is constructed; Through the combined perception of radar and cameras, it collects road environment data around the main vehicle and the position and speed information of other traffic participants; The main vehicle's driving data, road environment data, and the position and speed data of other traffic participants within a preset time period are integrated to form the main vehicle's continuous driving time series data.

3. The driving risk identification and warning method according to claim 1, characterized in that: The risk field theory is used to model the driver's perceived risk distribution of the current road space, and the calculation formula is: Where, R Indicates risk, Respectively represent the position coordinates of the main vehicle in the Frenet reference system, and Represent objects in the road space j The longitudinal and transverse velocities along the Frenet reference frame, Indicates the driver's reaction time; , It represents a constant parameter that is adapted to the driver's risk perception characteristics; It is a risk factor parameter that considers the severity of the collision consequences; , are objects in the road space. Longitudinal and lateral distance to the host vehicle.

4. The driving risk identification and warning method according to claim 1, characterized in that: The driving subject's expectation of the current driving task is quantified by the ratio of the target displacement to the expected maximum displacement, which is expressed as: Where, Indicates the longitudinal displacement of the target point relative to the current position; The driver's maximum expectation of his own movement distance; Indicates the current longitudinal speed of the vehicle. represents the driver's expected acceleration, Indicates the driver's preview time.

5. The driving risk identification and warning method according to claim 1, characterized in that: The driving decision-making process is modeled based on the driving risk R perceived by the driving subject and his / her expectations for the current driving task, and the risk preference and decision-making characteristics of the driving subject are quantified using model parameters to obtain the decision target point of the driving subject at each time step, thereby realizing driving simulation, including: The driving decision process is modeled as the driver's perceived driving risk R and drivers' expectations of driving tasks p The trade-off between the two, and the relationship between the two is described by a decision curve; According to the utility theory, the driver's decision characteristics are modeled as decision curves, which are divided into four types: neutral, aggressive, conservative and mixed; The feasible domain of decision making for the expected trade-off of driving risk is determined, and the risk preference and decision-making characteristics of the driving subject are quantified using a high-risk driving model parameter set.

6. A driving risk identification and warning system, characterized in that: include: Real-time data acquisition module, used to collect vehicle driving environment and driving data; The high-risk driving model real-time parameter fitting module is used to fit the real-time parameters of the high-risk driving model based on the vehicle driving environment and driving data through a genetic algorithm to obtain model parameters that characterize the driver's operating characteristics and vehicle motion characteristics, including: Based on the vehicle's motion characteristics and risk characteristics, the vehicle motion process is modeled by comprehensively considering the driver's risk perception and expected trade-off decision-making process to obtain a high-risk driving model. This includes: using risk field theory to model the driver's perceived risk distribution of the current road space, and quantifying the comprehensive risk of the road space based on the risk distribution results; quantifying the driver's expectations for the current driving task through the ratio of target displacement to expected maximum displacement; based on the driver's perceived driving risk and his expectations for the current driving task, the driving decision-making process is modeled, and the driver's risk preference and decision-making characteristics are quantified using model parameters. The decision target point of the driver at each time step is determined to achieve driving simulation. Based on the vehicle driving environment, driving data and high-risk driving model, a genetic algorithm is used to fit the parameters of the high-risk driving model to obtain model parameters that characterize the driver's operating characteristics and vehicle motion characteristics; The driving risk identification and warning module is used to classify and identify high-risk driving behaviors based on the high-risk driving model parameters obtained by fitting driving data through the SVM classification algorithm, and to issue warnings for driving risks.

7. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions that, when executed by a computing device, cause the computing device to perform any one of the methods of claims 1 to 5 .

8. A computing device, characterized in that include: One or more processors and a memory, wherein the memory stores one or more programs and is configured to be executed by the one or more processors, wherein the one or more programs include instructions for executing any one of the methods according to claims 1 to 5.

Images