Method and device for determining driving right takeover risk and vehicle

Abstract

The invention is suitable for the technical field of road traffic safety, and provides a method and a device for determining a driving right takeover risk, and a vehicle. The method comprises the following steps: acquiring driving experiment data of a plurality of drivers in different driving state scenes; for each driving state scene, determining a driving right takeover risk value of each driver according to the driving experiment data of each driver in the driving state scene; clustering the driving right takeover risk values of each driver in the driving state scene to obtain a plurality of categories, and determining a value range and a risk degree corresponding to each category; and monitoring the driving right takeover risk in actual driving based on the category division in each driving state scene. According to the invention, the accuracy of driving right takeover risk judgment in man-machine co-driving can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of road traffic safety, and in particular relates to a method, device and vehicle for determining the risk of taking over driving rights. Background Art

[0002] Although the intelligence level of current autonomous driving systems is constantly increasing, the intelligence level of most cars is still at a low level, and it is still difficult to upgrade to a fully intelligent level in a short period of time. In the actual driving process, the uncertainty and complexity of various problems faced by driving vehicles determine that the current autonomous driving system cannot completely replace the driver. Automobile intelligence is in the process of gradually developing to full automation, and the transition stage during this period is the human-machine co-driving stage. Human-machine co-driving is an inevitable transition stage from traditional manual driving to the realization of fully autonomous driving. Before the autonomous driving technology is fully mature, the driver must still be prepared to take over the driving rights of the car at any time, and the situation of human-machine collaborative co-driving of vehicles will exist for a long time.

[0003] In the stage of human-machine co-driving, the driver cannot completely leave the driving task. When the car is driving automatically, if it encounters sudden danger on the road or emergency traffic situation, due to the complexity and urgency of the situation, the automatic driving system will issue a takeover request. This is because the system is not competent for the current driving task. In related technologies, most of them use the collision time to decide whether to switch driving rights. Due to the influence of multiple risk factors at the same time, the existing methods are inaccurate in judging the takeover risk. Summary of the invention

[0004] In view of this, an embodiment of the present invention provides a method, device and vehicle for determining the risk of taking over driving rights, so as to improve the accuracy of judging the risk of taking over driving rights in human-machine co-driving.

[0005] A first aspect of an embodiment of the present invention provides a method for determining a driving right takeover risk, comprising:

[0006] Obtain driving test data of multiple drivers in different driving status scenarios; wherein the driving test data includes vehicle driving data, driver physiological data, driver basic information data and driver takeover time data;

[0007] For each driving state scenario, the driving right takeover risk value of each driver is determined based on the driving test data of each driver in the driving state scenario;

[0008] Clustering the driving right takeover risk values ​​of each driver in this driving state scenario to obtain multiple categories, and determining the value range and risk level corresponding to each category;

[0009] Based on the classification of various driving status scenarios, the risk of taking over driving rights in actual driving is monitored.

[0010] In combination with the first aspect, in a possible implementation of the first aspect, for each driving state scenario, determining the driving right takeover risk value of each driver according to the driving test data of each driver in the driving state scenario includes:

[0011] Perform weighted fusion on the driver's physiological data and basic information data to obtain the driver's risk factor;

[0012] Determine the safe time interval for driver takeover based on the driver takeover time data;

[0013] The risk value of taking over driving rights is determined based on the driver's risk factor, the driver's safe time interval for taking over and the vehicle driving data.

[0014] In combination with the first aspect, in a possible implementation of the first aspect, the driver's physiological data and the driver's basic information data each include at least one indicator;

[0015] Before weighted fusion of the driver's physiological data and the driver's basic information data is performed to obtain the driver's risk factor, the following is also included:

[0016] By using factor analysis, the component coefficients of each indicator under the driving proficiency factor, attention distraction factor and driving cognition factor were analyzed;

[0017] Accordingly, the driver's physiological data and basic information data are weighted and fused to obtain the driver's risk factor, including:

[0018] Taking the component coefficient of each indicator under the driving proficiency factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving proficiency factor;

[0019] Taking the component coefficient of each indicator under the attention distraction factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the attention distraction factor;

[0020] Taking the component coefficient of each indicator under the driving awareness factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving awareness factor;

[0021] according to Determine the driver risk factor; where D is the driver risk factor, c 1 、c 2 、c 3 are the component coefficients of driving proficiency factor, attention distraction factor and driving cognition factor, respectively. 1 , D 2 , D3 They are the quantitative values ​​of the driving proficiency factor, the quantitative value of the attention distraction factor, and the quantitative value of the driving cognition factor respectively.

[0022] In combination with the first aspect, in a possible implementation of the first aspect, the driver takeover time data includes the driver takeover reaction time and the driver takeover decision time;

[0023] Based on the takeover time data, determine the safe time interval for the driver to take over, including:

[0024] according to Determine the safe time interval for the driver to take over; where t a Take over the safe distance for the driver; re Take over the driver's reaction time; t de is the driver's decision-making time; μ is the adhesion coefficient between the vehicle tire and the road surface; τ is the road slope of the road section where the vehicle is located.

[0025] In combination with the first aspect, in a possible implementation of the first aspect, the vehicle driving data includes the vehicle speed v of the vehicle i and the vehicle speed v of the vehicle i. i , the road condition influencing factor R of the road where the vehicle i and the preceding vehicle j are located i and R j , and the equivalent mass M of vehicle i and vehicle j i and M j ;

[0026] Based on the driver's risk factor, the driver's safe time to take over and the vehicle's driving data, the risk value of taking over driving rights is determined, including:

[0027] according to Determine the driving right takeover risk value; where T i Take over the risk value for driving rights; D ri and D rj are the driver risk factors of the driver of vehicle i and the driver of vehicle j, respectively; G, k 1 With k 2 is the preset coefficient; r is the safe time interval for the driver to take over.

[0028] In combination with the first aspect, in a possible implementation manner of the first aspect, the road condition influencing factor is determined by at least one of the following indicators:

[0029] Road environment visibility, road adhesion coefficient, road curvature and road slope.

[0030] In combination with the first aspect, in a possible implementation of the first aspect, based on the classification of each driving state scenario, monitoring the risk of taking over the driving right in actual driving includes:

[0031] Acquire driving monitoring data during actual driving; wherein the driving monitoring data includes vehicle travel data, driver physiological data, driver basic information data, and driver takeover time data;

[0032] Determine the current driving right takeover risk value based on driving monitoring data;

[0033] Determine the current driving status scenario;

[0034] According to the value range corresponding to each category in the current driving status scenario, the category to which the current driving right takeover risk value belongs is determined, and the risk level of the driving right takeover risk is obtained.

[0035] A second aspect of an embodiment of the present invention provides a device for determining a driving right takeover risk, comprising:

[0036] The experimental module is used to obtain driving experimental data of multiple drivers in different driving state scenarios; wherein the driving experimental data includes vehicle driving data, driver physiological data, driver basic information data and driver takeover time data; for each driving state scenario, the driving right takeover risk value of each driver is determined according to the driving experimental data of each driver in the driving state scenario; the driving right takeover risk value of each driver in the driving state scenario is clustered to obtain multiple categories, and the value range and risk level corresponding to each category are determined;

[0037] The monitoring module is used to monitor the risk of taking over driving rights in actual driving based on the classification of various driving status scenarios.

[0038] A third aspect of an embodiment of the present invention provides a vehicle, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method in the first aspect or any one of the implementations of the first aspect are implemented.

[0039] A fourth aspect of an embodiment of the present invention provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method in the first aspect or any one of the implementations of the first aspect are implemented.

[0040] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0041] The embodiments of the present invention take into account the different reaction abilities of drivers in different driving status scenarios, and distinguish between scenarios to determine the risk of taking over driving rights. In addition, the impact of multiple risk factors in the human-vehicle-road environment is comprehensively analyzed through vehicle driving data, driver physiological data, driver basic information data, and driver takeover time data, so as to more accurately judge the risk of taking over driving rights. The present invention can accurately monitor the driver's real-time driving rights takeover risk, ensure that the driver has sufficient time and conditions to safely take over the vehicle driving rights, improve the synergy of human-machine co-driving, reduce the risk of automatic driving, and ensure driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0043] Figure 1 The following is a flow chart of a method for determining the risk of taking over driving rights provided by an embodiment of the present invention. Figure 1 ;

[0044] Figure 2 is a box plot of the driver's takeover reaction time distribution in different driving state scenarios provided by an embodiment of the present invention;

[0045] Figure 3 is a box plot of the risk value distribution of driving right takeover in different driving status scenarios provided by an embodiment of the present invention;

[0046] Figure 4 The following is a flow chart of a method for determining the risk of taking over driving rights provided by an embodiment of the present invention. Figure 2 ;

[0047] Figure 5 It is a structural schematic diagram of a device for determining the risk of taking over driving rights provided by an embodiment of the present invention;

[0048] Figure 6 It is a schematic diagram of the structure of a vehicle provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0050] In order to illustrate the technical solution of the present invention, a specific embodiment is provided below for illustration.

[0051] Human-machine co-driving is a common phenomenon in the current autonomous driving vehicles. Due to the influence of risk factors such as drivers, vehicles and road environment, the takeover of driving rights is highly uncertain and there are great driving safety risks.

[0052] At present, most of the studies on the risk of taking over driving rights use the collision time (i.e., the collision time between the obstacle in front of the lane where the vehicle is located and the vehicle), ignoring the influence of multiple risk factors in the human-vehicle-road environment, resulting in inaccurate judgment of the takeover risk. Research on the takeover of driving rights needs to focus on the factors affecting the takeover performance, including the takeover request time, the timing of the takeover request, the driver's influence, non-driving related secondary tasks and the road traffic environment. These factors affect each other and are complex and difficult to express. At the same time, there is a lack of research on the influence of driver's personal factors on the takeover of driving rights in the existing technology. The existing research on the influence of individual differences of drivers on the takeover of driving rights is mainly based on the driver's age, driving skill level and acceptance of takeover. There is no further research on the influence of other driver-related factors. Therefore, it is of great significance to comprehensively consider multiple risk factors and accurately monitor the driver's takeover risk threshold to explore the takeover risk and improve the driving safety of autonomous vehicles.

[0053] An embodiment of the present invention provides a method for determining the risk of taking over driving rights, which includes two stages. The first stage is to determine the driving right takeover risk determination threshold under different driving status scenarios through experiments, and the second stage is to monitor the driving right takeover risk determination in actual driving based on the driving right takeover risk determination threshold.

[0054] See also Figure 1 As shown, the process of phase one is as follows:

[0055] Step S101, obtaining driving test data of multiple drivers in different driving status scenarios.

[0056] In the human-machine co-driving scenario, the driver may be in different sub-task states, such as making a phone call, watching a video, N-back, etc. The takeover reaction time of the same driver in different sub-tasks may also vary. For example, Figure 2The box plot of the driver's takeover reaction time distribution under different driving tasks shows that when there is no driving task, the driver's attention is still in the road traffic environment, and there is no interference from the external environment, so the driver's cognitive level is still in a normal state. Therefore, when there is no driving task, the takeover reaction time is the shortest, with an average value of 2.379s. When the driver performs the driving task, the driver's attention is transferred from the road environment to the driving task as the line of sight moves, and the interference of the sub-task will also make the driver's cognitive level present different levels with the complexity of the sub-task. The takeover reaction time will show obvious changes due to the distraction of attention and the improvement of cognitive level. When the automatic driving control vehicle makes a request, considering that the driver is in the driving task state, it takes a certain amount of time to transfer the line of sight and attention from other tasks to the road. Therefore, in this case, the driver's takeover reaction time will be longer than when the driving task is not performed. For example, the takeover reaction times of making phone calls, watching videos and N-back driving tasks are 2.810, 2.784 and 2.731 respectively.

[0057] Figure 3The box-line diagram of the risk value distribution of driving right takeover under different driving tasks shows that the takeover risk when the driver does not perform the driving task is significantly higher than the takeover risk when the driver performs the driving task. This is because when the driver does not perform the driving task, that is, when there is no driving task, although the driver does not need to perform the dynamic driving task related to the vehicle, the driver's mental state is still in a good state, and the driver's attention is still in the road traffic environment where the vehicle is driving. Therefore, the driver can promptly devote himself to the driving task according to the takeover request issued by the system, so that he can respond to the upcoming emergency and dangerous situation in time. According to the driver's mental state and ability to deal with risks, in order to avoid false reports caused by premature takeover, the driver does not need to take over at a long distance in advance while ensuring the safe takeover time. Therefore, when the system issues a takeover request in the state without driving task takeover, the vehicle's position is closer to the dangerous scene, and the takeover risk value is generally higher, with an average of 510.833, and is significantly higher than the takeover risk values ​​in the other three states with driving tasks, which is in line with the real driving scene. When the driver performs the driving task, the driver's sight is mainly focused on the driving task, such as on the screen or the N-back screen, which causes the sight to deviate from the vehicle's driving direction, and the attention will also shift as the sight moves. And because of the interference effect of the driving task, when the autonomous driving vehicle faces an emergency, the system needs to consider that the driver's takeover reaction time is longer than the takeover reaction time when the driving task is not performed, so it is necessary to issue a takeover request earlier to ensure that the driver can take over smoothly and avoid dangerous scenes within a safe takeover time interval. Therefore, the takeover risk value when the driver performs the driving task is generally lower than the takeover risk value when there is no driving task. The takeover risk values ​​of the three driving tasks are analyzed in detail. Due to the different difficulty and complexity of the task, the driver's concentration and mental state in the task will also be different. As a relatively simple phone call task, since the driver's sight does not need to be shifted too much, the driver's mental state and sight are still partially in the road environment, so the risk generated by the driver itself is low. When the vehicle issues a takeover request, it can take over the vehicle driving rights smoothly and easily. When performing driving tasks such as watching videos and making phone calls, the driver's sight and attention will be attracted by the screen and the content of the picture. The risk factors generated by the driver himself are higher, so the takeover risk value also increases with the driver's state.As can be seen from the analysis of takeover reaction time and takeover safety interval in the previous article, there is no significant difference in takeover reaction time and takeover safety interval when the driver performs the driving task, indicating that the takeover request time issued by the system is similar in the states of watching videos and N-back driving tasks. Therefore, the factor that can affect the difference in takeover risk value is the driver's state, that is, the driver risk factor mentioned in the previous article. Since watching videos and making phone calls have different effects on the driver's spirit and attention, the driver risk factors of the two are different. Therefore, the more complex N-back has a higher takeover risk value, with a mean of 521.643, which is greater than the takeover risk value of watching videos, with a mean of 514.666.

[0058] Therefore, the driving right takeover risk determination thresholds under different sub-tasks should be set separately. Here, according to the sub-task states that the driver may be in, the driving state scenarios are divided into non-driving sub-task scenarios, phone call sub-task scenarios, video watching sub-task scenarios and N-back sub-task scenarios.

[0059] In this embodiment, the driving test data includes vehicle driving data, driver physiological data, driver basic information data, driver takeover time data, etc.

[0060] Among them, the vehicle driving data can be vehicle speed, information about the vehicle in front, the road adhesion coefficient between the vehicle tires and the road surface, the road slope of the road section where the vehicle is located, road environment visibility, road curvature, road slope, etc. The vehicle driving data can be collected through a driving simulator. The driver's physiological data includes gaze information, etc. The driver's visual environment, eye state, and pupil changes can be captured through a wearable eye tracker to achieve the collection of driver's physiological data. The driver's basic information data includes age, gender, vision, etc. The driver's takeover time data includes the driver's takeover reaction time and the driver's takeover decision time.

[0061] Step S102: for each driving state scenario, the driving right takeover risk value of each driver is determined according to the driving test data of each driver in the driving state scenario.

[0062] In an exemplary experimental scenario, different drivers can be asked to drive vehicles on the highway and put the vehicles in automatic driving mode. Then, the driving test data of the drivers in the non-driving sub-task scenario, the phone call sub-task scenario, the video watching sub-task scenario and the N-back sub-task scenario are measured respectively to obtain the driving test data of each driver in each driving state scenario. By integrating the vehicle driving data, driver physiological data, driver basic information data and driver takeover time data in the driving test data, the driving right takeover risk value is accurately calculated. The driving right takeover risk value is used by the vehicle to determine the timing of the driving right switch prompt.

[0063] Step S103, clustering the driving right takeover risk values ​​of each driver in the driving state scenario to obtain multiple categories, and determining the value range and risk level corresponding to each category.

[0064] By clustering the driving right takeover risk values ​​of each driver in each driving state scenario, multiple categories are obtained, and the value range and risk level corresponding to each category are determined, which can guide the vehicle to provide driving right switching prompts. For example, when any driver is driving a vehicle, if the vehicle is in an automatic driving state, the current risk level in the driving state scenario can be judged based on the driver's driving state scenario and the driving right takeover risk value, and then a corresponding risk prompt can be issued. For example, when the risk level is low, a simple prompt is given by a display icon, and when the risk level is high, a strong prompt is given by a display icon combined with a voice broadcast.

[0065] It can be seen that the embodiments of the present invention take into account the different reaction abilities of drivers in different driving status scenarios, and distinguish the scenarios to determine the risk of taking over the driving rights. In addition, the impact of multiple risk factors in the human-vehicle-road environment is comprehensively analyzed through vehicle driving data, driver physiological data, driver basic information data and driver takeover time data, so as to more accurately judge the risk of taking over the driving rights. The present invention can accurately monitor the driver's real-time driving rights takeover risk, ensure that the driver has sufficient time and conditions to safely take over the vehicle driving rights, improve the coordination of human-machine co-driving, reduce the risk of automatic driving, and ensure driving safety.

[0066] As a possible implementation, in step S102, for each driving state scenario, the driving right takeover risk value of each driver is determined according to the driving test data of each driver in the driving state scenario, including:

[0067] Step S1021, weighted fusion of the driver's physiological data and the driver's basic information data to obtain the driver's risk factor;

[0068] Step S1022, determining the safe time interval for the driver to take over according to the driver takeover time data;

[0069] Step S1023, determining the driving right takeover risk value based on the driver risk factor, the driver takeover safety interval and the vehicle driving data.

[0070] Here, the driver's physiological data and the driver's basic information data each include at least one indicator. For example, the indicator of the driver's physiological data may be the percentage of gaze points and the average gaze time. The indicator of the driver's basic information data may be the driving experience.

[0071] Optionally, in step S1021, the weight values ​​of the driver's physiological data and the driver's basic information data may be predetermined in the following manner:

[0072] By using factor analysis, the component coefficients of each index under each common factor (driving proficiency factor, attention distraction factor and driving cognition factor) are analyzed. In this embodiment, factor analysis can reduce the dimension of multiple indicators with correlation and extract the main indicators. This embodiment uses the percentage of fixation points, average fixation time and driving age to represent the driver's risk factor, and obtains the driver's risk factor when the driver is driving on the highway section.

[0073] Correspondingly, in step S1021, the driver's physiological data and the driver's basic information data are weighted and fused to obtain the driver's risk factor, including:

[0074] Taking the component coefficient of each indicator under the driving proficiency factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving proficiency factor;

[0075] Taking the component coefficient of each indicator under the attention distraction factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the attention distraction factor;

[0076] Taking the component coefficient of each indicator under the driving awareness factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving awareness factor;

[0077] according to Determine the driver risk factor; where D is the driver risk factor, c 1 、c 2 、c 3 are the component coefficients of driving proficiency factor, attention distraction factor and driving cognition factor, respectively. 1 , D 2 , D 3 They are the quantitative values ​​of the driving proficiency factor, the quantitative value of the attention distraction factor, and the quantitative value of the driving cognition factor respectively.

[0078] The detailed steps include:

[0079] The values ​​of various indicators of different drivers when driving on the highway section are obtained in advance; according to the average value of each indicator, the variance contribution rate of each common factor is calculated by factor analysis to obtain the first weight value of each common factor; and the component coefficient of each indicator under each common factor is calculated to obtain the second weight value of each indicator under each common factor; the second weight value of each indicator under each common factor is used as the component coefficient of each indicator under each common factor.

[0080] The following is a specific example to illustrate the factor analysis process:

[0081] Factor analysis requires that there is a certain correlation between the indicators. First, the Pearson correlation analysis method is used to test the correlation of the indicators. As shown in Table 1, under the 95% confidence interval, the correlation of the indicators is significant, which meets the conditions for factor analysis.

[0082] Table 1 Pearson correlation analysis

[0083]

[0084] The KMO sampling test and Bartelett sphericity test were performed on the data. The KMO value was 0.637>0.6, and the Bartelett sphericity test was less than 0.05, indicating that there was a correlation between the indicators, so it was suitable for factor analysis. At the same time, the common factor variance was greater than 0.8, indicating that the proportion of information extracted from each variable was greater than 80%, so the variables could be well represented by the common factor.

[0085] Table 2 is the total variance explanation table. According to the variance contribution rate of common factors, it can be seen that the indicators are reasonable. Two common factors (eigenvalues ​​greater than 1) can be extracted before and after the common factor rotation using the maximum variance method. The variance contribution rate of the first common factor reaches 33.421%, the cumulative variance contribution rate of the first two common factors reaches 66.788%, and the cumulative variance contribution rate of the three common factors reaches 100%.

[0086] Table 2 Total variance explained

[0087]

[0088] Table 3 is the component matrix after rotation. Common factor 1 has a larger factor loading coefficient of 0.976 for attention proficiency, common factor 2 has a larger factor loading coefficient of 0.970 for attention distraction, and common factor 3 has a larger factor loading of 0.963 for driving cognition. All three common factors have a strong correlation with the indicators, so the three common factors after rotation all play an explanatory role in the indicators. According to the size of the correlation between the common factors and the indicators, common factor 1 is classified as the driving proficiency factor, common factor 2 is classified as the attention distraction factor, and common factor 3 is classified as the driving cognition factor.

[0089] Table 3 Rotated component matrix

[0090]

[0091] Table 4 Rotated component score coefficient matrix

[0092]

[0093] The score function D of the three common factors can be calculated from the component score coefficient matrix 4 1 , D 2 and D 3 They are:

[0094] D 1 =0.116h-0.166l+1.071g

[0095] D 2 =1.091h+0.207l+0.118g

[0096] D 3 =0.211h+1.113l-0.172g

[0097] Where h, l, and g represent attention distraction, driving awareness, and driving proficiency, respectively.

[0098] From the variance explanation table, we can see that the variance contribution rate of driving proficiency factor is 33.421%, the variance contribution rate of attention distraction factor is 33.368%, and the variance contribution rate of driving cognition factor is 33.212%. Therefore, the driver risk factor model D in the task-free driving state is:

[0099] D=0.334D 1 +0.334D 2 +0.332D 3

[0100] It can be seen that this embodiment obtains the physiological data and basic information data of the driver at the current position when the driver is driving the automatic driving vehicle on the highway, and weightedly fuses the driver's physiological data and the driver's basic information data according to the preset weight values ​​of the driver's physiological data and the driver's basic information data to obtain the driver risk factor of the vehicle driver in the current state, which can accurately, truthfully and comprehensively reflect the driver's risk factor, and further provide a data basis for the subsequent quantification and classification of the risk of taking over the driving right.

[0101] Optionally, the driver takeover time data includes the driver takeover reaction time and the driver takeover decision time. In step S1022, according to the takeover time data, determining the driver takeover safety interval includes:

[0102] according to Determine the safe time interval for the driver to take over; where t a Take over the safe distance for the driver; re Take over the driver's reaction time; t de is the driver's decision-making time; μ is the adhesion coefficient between the vehicle tire and the road surface; τ is the road slope of the road section where the vehicle is located.

[0103] For example, Table 5 is a statistical table of the safe time interval for a driver to take over in the experiment. Since the vehicle has been performing a fixed speed automatic driving at a speed of 100 km / h before the driver takes the decision to take over, v = 100 km / h. The road surface in the experimental design road scene is a dry and high-quality asphalt straight road surface, that is, the adhesion coefficient between the vehicle tire and the dry road surface is = 0.8. Since the road is a straight road surface, the road surface height difference is 0, that is, the road slope is = 0.

[0104] Table 5 Statistics of safe time intervals for drivers to take over

[0105]

[0106]

[0107] Optionally, the vehicle driving data includes the speed v of vehicle i and i , the road condition influencing factor R of the road where the vehicle i and the preceding vehicle j are located i and R j , and the equivalent mass M of vehicle i and vehicle j i and M j In step S1023, based on the driver risk factor, the driver takeover safety interval and the vehicle driving data, the driving right takeover risk value is determined, including:

[0108] according to Determine the driving right takeover risk value; where T i Take over the risk value for driving rights; D ri and D rj are the driver risk factors of the driver of vehicle i and the driver of vehicle j, respectively; G, k 1 With k 2 is the preset coefficient; r is the safe time interval for the driver to take over.

[0109] Among them, the calculation formula of equivalent mass M is M i =M i (m i ,T i ,v i )=m i ·T i ·(a·v i b +c). In the formula, m i is the mass of object i. i is the type of object i. i is the velocity of object i, and the velocity of a stationary object is 0. a, b, c are all unknown constants. m i ·T i ·(a·v ib +c) The formula in brackets is a speed fitting polynomial, which represents the impact of the object's speed on the risk level.

[0110] The calculation formula of road impact factor is R i =R i (δ i ,μ i ,ρ i ,τ i ) = f δ (δ i )f μ (μ i )f ρ (ρ i )f τ (τ i ). In the formula, f δ (δ i ) represents the visibility of the road environment, f μ (μ i ) represents the road adhesion coefficient, f ρ (ρ i ) represents the road curvature, f τ (τ i ) represents the road slope.

[0111] Before conducting a cluster analysis of the driving takeover risk classification, considering that the driving takeover risk value is an indicator with a large range of numerical changes and gaps, if the driving takeover risk value is directly used as the basis for the driving takeover risk threshold, it will be difficult to obtain the takeover risk threshold. Therefore, the driving takeover risk value can be processed to a certain extent before studying the takeover risk threshold. A "relative takeover risk index" can be defined to evaluate the driving takeover risk before and during the takeover. Optionally, the calculation formula for the relative takeover risk index is Where RTRI i is the relative takeover risk index, T * is the average value of takeover risk.

[0112] As a possible implementation method, in step S103, after the driver completes driving a driving state scenario, the driver's driving right takeover risk can be categorized using a K-means clustering algorithm, a K-means clustering algorithm based on the elbow rule, or a DBSCAN clustering algorithm.

[0113] For example, the following steps:

[0114] Step S1031, using the silhouette coefficient method to evaluate the classification results after clustering, to obtain the algorithm with the best clustering effect and the classification results.

[0115] Step S1032, taking the minimum driving right takeover risk value of each category of the driver in each driving state scenario as the driving right takeover risk threshold of the driver in that state.

[0116] Step S1033, using the driving right takeover risk threshold as a takeover request trigger condition during driving, and triggering an early warning when the driving right takeover risk meets the threshold to remind the driver to take over the driving right.

[0117] Exemplarily, according to the above steps, a cluster evaluation table 6 of the risk of taking over the driver's driving right in a scenario without a driving secondary task and a takeover risk category table 7 are obtained.

[0118] Table 6 Clustering effect evaluation table in the scenario without driving sub-task

[0119]

[0120] Table 7 Takeover risk classification table in non-driving task state

[0121]

[0122] The method for classifying takeover risk categories in other driving status scenarios is the same as described above.

[0123] See also Figure 4 As shown, after the takeover risk classification is performed in each driving state scenario, the actual driving right takeover risk determination in stage 2 can be performed according to the classification rules. The process of stage 2 is as follows:

[0124] Step S401, obtaining driving monitoring data during actual driving; wherein the driving monitoring data includes vehicle travel data, driver physiological data, driver basic information data, and driver takeover time data.

[0125] Step S402: determining the current driving right takeover risk value based on the driving monitoring data.

[0126] Step S403, determining the current driving status scene.

[0127] Step S404, judging the category to which the current driving right takeover risk value belongs according to the value range corresponding to each category in the current driving state scenario, and obtaining the risk level of the driving right takeover risk.

[0128] Here, the specific implementation of step S401, step S402 and step S404 can be found in Figure 1 The description of the embodiment will not be repeated in this embodiment.

[0129] In this embodiment, step S403 can obtain the driver's image, identify the driver's behavior through a pre-trained behavior recognition model, and combine the operating status of the vehicle multimedia screen to determine which driving state scenario the current driver is in. This embodiment is mainly improved to determine the threshold value for judging the risk of taking over driving rights, and does not limit the method for judging the current driving state scenario.

[0130] In combination with the above embodiments, the present invention collects vehicle driving data, driver physiological data, and driver basic information data, constructs a driver driving risk factor model based on factor analysis, builds a driving right takeover risk quantification model based on driving risk field theory, and performs driving right takeover risk cluster analysis and category classification. The research results can monitor the driver's real-time driving right takeover risk and provide a reference for the driving safety of human-machine co-driving on highways. For example, when the vehicle driver's takeover risk meets the driving right takeover risk threshold, an early warning can be issued to remind the driver to take over, ensuring that the driver has sufficient time and conditions to safely take over the vehicle driving rights, improving the coordination of human-machine co-driving on highways, reducing the risk of automatic driving, and ensuring driving safety.

[0131] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0132] See also Figure 5 As shown, an embodiment of the present invention provides a device 50 for determining a risk of taking over driving rights, comprising:

[0133] The experimental module 51 is used to obtain driving test data of multiple drivers in different driving state scenarios; wherein the driving test data includes vehicle travel data, driver physiological data, driver basic information data and driver takeover time data; for each driving state scenario, the driving right takeover risk value of each driver is determined according to the driving test data of each driver in the driving state scenario; the driving right takeover risk value of each driver in the driving state scenario is clustered to obtain multiple categories, and the value range and risk level corresponding to each category are determined.

[0134] The monitoring module 52 is used to monitor the risk of taking over the driving right in actual driving based on the classification under each driving state scenario.

[0135] As a possible implementation, the experimental module 51 is used to:

[0136] Perform weighted fusion on the driver's physiological data and basic information data to obtain the driver's risk factor;

[0137] Determine the safe time interval for driver takeover based on the driver takeover time data;

[0138] The risk value of taking over driving rights is determined based on the driver's risk factor, the driver's safe time interval for taking over and the vehicle driving data.

[0139] As a possible implementation method, the driver's physiological data and the driver's basic information data each include at least one indicator; the experimental module 51 is used to:

[0140] By using factor analysis, the component coefficients of each indicator under the driving proficiency factor, attention distraction factor and driving cognition factor were analyzed;

[0141] Accordingly, the driver's physiological data and basic information data are weighted and fused to obtain the driver's risk factor, including:

[0142] Taking the component coefficient of each indicator under the driving proficiency factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving proficiency factor;

[0143] Taking the component coefficient of each indicator under the attention distraction factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the attention distraction factor;

[0144] Taking the component coefficient of each indicator under the driving awareness factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving awareness factor;

[0145] according to Determine the driver risk factor; where D is the driver risk factor, c 1 、c 2 、c 3 are the component coefficients of driving proficiency factor, attention distraction factor and driving cognition factor, respectively. 1 , D 2 , D 3 They are the quantitative values ​​of the driving proficiency factor, the quantitative value of the attention distraction factor, and the quantitative value of the driving cognition factor respectively.

[0146] As a possible implementation method, the driver takeover time data includes the driver takeover reaction time and the driver takeover decision time; the experimental module 51 is used to:

[0147] according to Determine the safe time interval for the driver to take over; where t a Take over the safe distance for the driver; re Take over the driver's reaction time; t de is the driver's decision-making time; μ is the adhesion coefficient between the vehicle tire and the road surface; τ is the road slope of the road section where the vehicle is located.

[0148] As a possible implementation method, the vehicle driving data includes the vehicle speed v of vehicle i and i , the road condition influencing factor R of the road where the vehicle i and the preceding vehicle j are located i and R j , and the equivalent mass M of vehicle i and vehicle j i and M j ; Experimental module 51 is used for:

[0149] according to Determine the driving right takeover risk value; where T i Take over the risk value for driving rights; D ri and D rj are the driver risk factors of the driver of vehicle i and the driver of vehicle j, respectively; G, k 1 With k 2 is the preset coefficient; r is the safe time interval for the driver to take over.

[0150] As a possible implementation method, the road condition influencing factor is determined by at least one of the following indicators: road environment visibility, road adhesion coefficient, road curvature and road slope.

[0151] As a possible implementation, the monitoring module 52 is used to:

[0152] Acquire driving monitoring data during actual driving; wherein the driving monitoring data includes vehicle travel data, driver physiological data, driver basic information data, and driver takeover time data;

[0153] Determine the current driving right takeover risk value based on driving monitoring data;

[0154] Determine the current driving status scenario;

[0155] According to the value range corresponding to each category in the current driving status scenario, the category to which the current driving right takeover risk value belongs is determined, and the risk level of the driving right takeover risk is obtained.

[0156] The embodiments of the present invention take into account the different reaction abilities of drivers in different driving status scenarios, and distinguish between scenarios to determine the risk of taking over driving rights. In addition, the impact of multiple risk factors in the human-vehicle-road environment is comprehensively analyzed through vehicle driving data, driver physiological data, driver basic information data, and driver takeover time data, so as to more accurately judge the risk of taking over driving rights. The present invention can accurately monitor the driver's real-time driving rights takeover risk, ensure that the driver has sufficient time and conditions to safely take over the vehicle driving rights, provide the synergy of human-machine co-driving, reduce the risk of automatic driving, and ensure driving safety.

[0157] Figure 6Schematic diagram of a vehicle 60 provided in an embodiment of the present invention. Figure 6 As shown, the vehicle 60 of this embodiment includes: a processor 61, a memory 62, and a computer program 63 stored in the memory 62 and executable on the processor 61, such as a program for determining the risk of taking over driving authority. When the processor 61 executes the computer program 63, the steps of the above-mentioned various embodiments of the method for determining the risk of taking over driving authority are implemented, such as Figure 1 Alternatively, when the processor 61 executes the computer program 63, the functions of each module in the above-mentioned device embodiments are realized, for example Figure 5 The functions of modules 51 and 52 are shown.

[0158] Exemplarily, the computer program 63 may be divided into one or more modules / units, one or more modules / units are stored in the memory 62, and executed by the processor 61 to implement the present invention. One or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program 63 in the vehicle 60.

[0159] The vehicle 60 may include, but is not limited to, a processor 61 and a memory 62. Those skilled in the art will appreciate that Figure 6 This is only an example of the vehicle 60 and does not constitute a limitation of the vehicle 60 , which may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the vehicle 60 may also include input and output devices, network access devices, buses, etc.

[0160] The processor 61 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0161] The memory 62 may be an internal storage unit of the vehicle 60, such as a hard disk or memory of the vehicle 60. The memory 62 may also be an external storage device of the vehicle 60, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the vehicle 60. Further, the memory 62 may also include both an internal storage unit and an external storage device of the vehicle 60. The memory 62 is used to store computer programs and other programs and data required by the vehicle 60. The memory 62 may also be used to temporarily store data that has been output or is to be output.

[0162] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0163] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0164] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0165] In the embodiments provided by the present invention, it should be understood that the disclosed devices / vehicles and methods can be implemented in other ways. For example, the device / vehicle embodiments described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0166] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0167] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0168] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media can include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0169] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for determining the risk of taking over driving rights, characterized in that: include: Acquire driving test data of multiple drivers in different driving status scenarios; wherein the driving test data includes vehicle driving data, driver physiological data, driver basic information data and driver takeover time data; For each driving state scenario, the driving right takeover risk value of each driver is determined based on the driving test data of each driver in the driving state scenario; Clustering the driving right takeover risk values ​​of each driver in this driving state scenario to obtain multiple categories, and determining the value range and risk level corresponding to each category; Based on the classification of various driving status scenarios, the risk of taking over driving rights in actual driving is monitored.

2. The method for determining the risk of taking over driving rights according to claim 1, characterized in that: For each driving state scenario, the driving right takeover risk value of each driver is determined according to the driving test data of each driver in the driving state scenario, including: Performing weighted fusion on the driver's physiological data and the driver's basic information data to obtain a driver risk factor; Determining a safe time interval for the driver to take over based on the driver takeover time data; The driving right takeover risk value is determined based on the driver risk factor, the driver takeover safety interval and the vehicle driving data.

3. The method for determining the risk of taking over driving rights according to claim 2, characterized in that: The driver's physiological data and the driver's basic information data each include at least one indicator; Before weighted fusion of the driver's physiological data and the driver's basic information data is performed to obtain the driver's risk factor, the method further includes: By using factor analysis, the component coefficients of each indicator under the driving proficiency factor, attention distraction factor and driving cognition factor were analyzed; Accordingly, the driver's physiological data and the driver's basic information data are weighted and fused to obtain the driver's risk factor, including: Taking the component coefficient of each indicator under the driving proficiency factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving proficiency factor; Taking the component coefficient of each indicator under the attention distraction factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the attention distraction factor; Taking the component coefficient of each indicator under the driving awareness factor as the weight, each indicator is weighted and fused to obtain the quantitative value of the driving awareness factor; according to Determine the driver risk factor; wherein D is the driver risk factor, c1, c2, c3 are component coefficients under the driving proficiency factor, the attention distraction factor and the driving cognition factor, respectively, and D1, D2, D3 are the quantitative values ​​of the driving proficiency factor, the quantitative values ​​of the attention distraction factor, and the quantitative values ​​of the driving cognition factor, respectively.

4. The method for determining the risk of taking over driving rights according to claim 2, characterized in that: The driver takeover time data includes the driver takeover reaction time and the driver takeover decision time; Determining the safe time interval for the driver to take over according to the takeover time data includes: according to Determine the safe time interval for the driver to take over; where t a Take over the safe distance for the driver; re Take over the driver's reaction time; t de is the driver's decision-making time; μ is the adhesion coefficient between the vehicle tire and the road surface; τ is the road slope of the road section where the vehicle is located.

5. The method for determining the risk of taking over driving rights according to claim 2, characterized in that: The vehicle driving data includes the vehicle speed v of the vehicle i and the i , the road condition influencing factor R of the road where the vehicle i and the preceding vehicle j are located i and R j , and the equivalent mass M of vehicle i and vehicle j i and M j ; The determining the driving right takeover risk value based on the driver risk factor, the driver takeover safety interval and the vehicle driving data includes: according to Determine the driving right takeover risk value; wherein, T i Take over the risk value for driving rights; D ri and D rj are the driver risk factors of the driver of vehicle i and the driver of vehicle j in front; G, k1 and k2 are the preset coefficients; r is the safe time interval for the driver to take over.

6. The method for determining the risk of taking over driving rights according to claim 5, characterized in that: The road condition influencing factor is determined by at least one of the following indicators: Road environment visibility, road adhesion coefficient, road curvature and road slope.

7. The method for determining the risk of taking over driving rights according to any one of claims 1 to 6, characterized in that: The classification based on each driving state scenario and monitoring the risk of taking over driving rights in actual driving include: Acquire driving monitoring data during actual driving; wherein the driving monitoring data includes vehicle travel data, driver physiological data, driver basic information data, and driver takeover time data; Determining a current driving right takeover risk value based on the driving monitoring data; Determine the current driving status scenario; According to the value range corresponding to each category in the current driving status scenario, the category to which the current driving right takeover risk value belongs is determined, and the risk level of the driving right takeover risk is obtained.

8. A device for determining the risk of taking over driving rights, characterized in that: include: An experimental module is used to obtain driving experimental data of multiple drivers under different driving state scenarios; wherein the driving experimental data includes vehicle travel data, driver physiological data, driver basic information data and driver takeover time data; for each driving state scenario, determine the driving right takeover risk value of each driver according to the driving experimental data of each driver under the driving state scenario; cluster the driving right takeover risk values ​​of each driver under the driving state scenario to obtain multiple categories, and determine the value range and risk level corresponding to each category; The monitoring module is used to monitor the risk of taking over driving rights in actual driving based on the classification of various driving status scenarios.

9. A vehicle comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images