Dynamic evaluation method and system under vehicle risk scene state based on real-time dynamic state

Abstract

The invention provides a dynamic evaluation method and system under a vehicle risk scene state based on real-time dynamics, and the method comprises the following steps: S1, detecting a risk scene where a vehicle is located according to the road information collected in real time, the risk scene comprising the lane cut-in of a side vehicle, the sudden braking of a front vehicle, and the lane cut-out of the front vehicle; s2, acquiring driving state data of the vehicle; s3, calling a feature braking model corresponding to the risk scene according to the acquired risk scene, and predicting the collision risk of the vehicle in combination with the driving state data and the feature braking model to obtain a collision risk prediction result; and S4, executing a corresponding intelligent strategy according to a collision risk prediction result. The intelligent level of intelligent driving assistance and intelligent driving control of the vehicle in the risk scene state is improved, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving assistance technology, and in particular to a dynamic assessment method and system based on real-time dynamic vehicle risk scenario states. Background Art

[0002] At present, intelligent driving technologies based on L1 and L2 levels have prompt functions for risk scenarios, but the current prompt functions are usually based on fixed braking models when performing collision warnings, and the collision time or collision distance is calculated based on the fixed braking model to achieve collision warnings. However, in actual real-life scenarios, the factors that affect collisions are not static. Therefore, if collision warnings or collision assessments are performed solely in a fixed manner, they cannot meet the needs of modern intelligent driving assistance or intelligent control, nor can they meet the needs of users for intelligent driving assistance. Summary of the invention

[0003] In view of the above problems, the present invention aims to provide a dynamic assessment method and system based on real-time dynamic vehicle risk scenario states.

[0004] The purpose of the present invention is achieved by the following technical solutions:

[0005] In a first aspect, the present invention provides a dynamic assessment method based on real-time dynamic vehicle risk scenario state, comprising the following steps:

[0006] S1 detects the risk scenarios of the vehicle based on the road information collected in real time, where the risk scenarios include the vehicle cutting into the lane, the vehicle in front braking suddenly, and the vehicle in front cutting out of the lane;

[0007] S2 obtains the driving status data of the vehicle;

[0008] S3 retrieves a characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, and predicts the collision risk of the vehicle in combination with the driving state data and the characteristic braking model to obtain a collision risk prediction result;

[0009] S4 executes corresponding intelligent strategies based on the collision risk prediction results.

[0010] Preferably, step S1 comprises:

[0011] S11 acquires and collects road information in front of and to the side of the vehicle through vehicle-mounted sensors, wherein the road information includes position information, azimuth information, relative speed information, and relative acceleration information of the vehicle in front or on the side being measured;

[0012] S12: When a vehicle is detected in front of the vehicle, the relative position information, azimuth information, relative speed information and relative acceleration information of the front vehicle A1 and the vehicle are further obtained; when the distance between the front vehicle and the vehicle is less than a preset first threshold distance, the azimuth is within the first standard azimuth range, the relative speed is greater than the first threshold speed, and / or the relative acceleration is greater than the first threshold acceleration, the front vehicle is judged to be a risk target vehicle due to sudden braking, and the first risk scenario is outputted;

[0013] S13 When a vehicle is detected in front of the vehicle, the relative position information and azimuth information of the front vehicle A1 and the vehicle are further obtained; when the range of the lane covered by the front vehicle is less than the preset second threshold range, and the azimuth is within the second standard azimuth range, it is judged that the front vehicle is cutting out of the lane, and the relative position information, azimuth information, relative speed information and relative acceleration information of the vehicle A2 further ahead and the vehicle are further obtained; when the distance between the vehicle further ahead and the vehicle is less than the preset first threshold distance, the azimuth is within the first standard azimuth range, and the relative speed is greater than the first threshold speed, and / or the relative acceleration is greater than the first threshold acceleration, it is judged that the vehicle further ahead is a risk target vehicle, and the current entry into the second risk scenario is output;

[0014] S14 When a vehicle is detected on the side of the vehicle, the relative position information, azimuth information, relative speed information and relative acceleration information of the side vehicle B1 and the vehicle are further obtained; when it is obtained according to the relative position that the distance between the front vehicle and the vehicle is less than the preset third threshold distance, the azimuth angle is within the third standard azimuth angle range, and the relative speed is greater than the third threshold speed, and / or the relative acceleration is greater than the third threshold acceleration, it is judged that the front vehicle has cut into the lane and is a risk target vehicle, and the current third risk scenario is output.

[0015] Preferably, in step S11, the vehicle-mounted sensors include radar, camera, and lidar; by fusing and feature extracting data collected by different sensors, feature information corresponding to the vehicle in front or the vehicle on the side is extracted, including vehicle speed, distance, relative speed, and acceleration.

[0016] Preferably, step S2 comprises:

[0017] S21 obtains the current driving state data of the vehicle, wherein the driving state data includes the vehicle speed V c (t), acceleration / deceleration a c (t), maximum deceleration Bs max And standard braking performance improvement time bT.

[0018] 5. The dynamic assessment method based on real-time dynamic vehicle risk scenario state according to claim 4, characterized in that step S3 comprises:

[0019] S31 retrieves a characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, wherein the characteristic braking model is:

[0020]

[0021] In the formula, Bs n (t) represents the braking deceleration of the vehicle at time t, where T0≤t≤T S , T0 represents the moment when the risk scenario is detected, T S Indicates the time when the vehicle is expected to decelerate to a standstill, Bs0 indicates the deceleration data of the vehicle at time T0, Bs0 = a c (T0), rT n represents the standard reaction time corresponding to the current nth risk scenario, RF represents the reaction force factor, and Bs max Indicates the maximum deceleration of the vehicle, bbT n Indicates the standard braking efficiency improvement time corresponding to the current nth risk scenario;

[0022] S32 detects the current vehicle speed V at the time when the risk scenario occurs c (T0), combined with the corresponding characteristic braking model, the collision risk prediction between the vehicle and the target risk vehicle is performed, including: calculating the current collision characteristic value Cru(t), where the calculation function of the collision characteristic value is:

[0023]

[0024] In the formula, Cru(t) represents the collision characteristic quantity, dis AB (T0) represents the relative distance between the vehicle and the risk target vehicle when the risk scene is detected, dis y Indicates the preset reserved safety distance, ΔV ABc (T0) represents the relative speed between the vehicle and the risk target vehicle when the risk scenario is detected; Δa ABc (T0) represents the relative acceleration between the host vehicle and the risk target vehicle when the risk scenario is detected; T S Indicates the estimated time of deceleration to a standstill;

[0025] When the collision characteristic value Cru(t) is less than the second standard distance disT2, the output risk prediction result is a second-level collision risk;

[0026] When the collision characteristic value Cru(t) is less than the first standard distance disT1, the output risk prediction result is a first-level collision risk;

[0027] When the collision characteristic value Cru(t) is not less than the second standard distance disT2, the risk prediction result is output as no risk.

[0028] Preferably, step S4 comprises:

[0029] When the risk prediction result is a level 2 collision risk, a collision warning prompt message is issued;

[0030] When the risk prediction result is a level 1 collision risk, an emergency braking command is initiated.

[0031] In the second aspect, the present invention shows a dynamic assessment system based on real-time dynamic vehicle risk scenario state, including a road information acquisition module, a driving information acquisition module, an analysis and processing module and a strategy control module; wherein,

[0032] The road information acquisition module is used to detect the risk scenario of the vehicle based on the road information collected in real time, where the risk scenario includes the side vehicle cutting into the lane, the front vehicle braking suddenly, and the front vehicle cutting out of the lane;

[0033] The driving information acquisition module is used to obtain the driving status data of the vehicle;

[0034] The analysis and processing module is used to retrieve the characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, and predict the collision risk of the vehicle in combination with the driving state data and the characteristic braking model to obtain the collision risk prediction result;

[0035] The strategy control module is used to execute corresponding intelligent strategies according to the collision risk prediction results.

[0036] The beneficial effects of the present invention are as follows: a dynamic assessment method and system based on real-time dynamic vehicle risk scenario status is proposed, firstly, the road information during the vehicle driving process is collected, and the risk scenario of the current vehicle is detected based on the road information; combined with the vehicle's own driving state data, when a risk scenario is detected, the corresponding characteristic braking model is adaptively called according to the risk scenario, and collision risk analysis is performed for the risk scenario based on the vehicle driving state data and the characteristic braking model. By using a targeted characteristic braking model based on risk scenarios as the basis for collision analysis, the pertinence and adaptability of the collision analysis results can be improved, the analysis effect can be improved, and at the same time, the intelligence level of the vehicle's intelligent driving assistance and intelligent driving control in risk scenario states is improved, and the user experience is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0038] Figure 1 A method flow chart of a dynamic assessment method based on real-time dynamic vehicle risk scenario state shown in an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of an example of road information of the present invention;

[0040] Figure 3 for Figure 1 A schematic diagram of the process of step S1 shown in the embodiment;

[0041] Figure 4 for Figure 1 A schematic flow chart of step S3 shown in the embodiment. DETAILED DESCRIPTION

[0042] The present invention is further described in conjunction with the following application scenarios.

[0043] See also Figure 1 , which shows a dynamic assessment method based on real-time dynamic vehicle risk scenario state, including the following steps:

[0044] S1 detects the risk scenarios of the vehicle based on the road information collected in real time, where the risk scenarios include the vehicle cutting into the lane, the vehicle in front braking suddenly, and the vehicle in front cutting out of the lane;

[0045] S2 obtains the driving status data of the vehicle;

[0046] S3 retrieves a characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, and predicts the collision risk of the vehicle in combination with the driving state data and the characteristic braking model to obtain a collision risk prediction result;

[0047] S4 executes corresponding intelligent strategies based on the collision risk prediction results.

[0048] The above-mentioned embodiment of the present invention proposes a dynamic assessment method based on real-time dynamic vehicle risk scenario status. First, the road information during the vehicle driving process is collected, and the risk scenario of the current vehicle is detected based on the road information; combined with the vehicle's own driving state data, when a risk scenario is detected, the corresponding characteristic braking model is adaptively called according to the risk scenario, and the collision risk analysis is performed for the risk scenario based on the vehicle driving state data and the characteristic braking model. By using a targeted characteristic braking model based on risk scenarios as the basis for collision analysis, the pertinence and adaptability of the collision analysis results can be improved, the analysis effect can be improved, and the intelligent level of intelligent driving assistance and intelligent driving control of the vehicle in risk scenario states can also be improved, thereby improving the user experience.

[0049] The method steps shown in the above embodiments can be executed based on the vehicle-mounted intelligent control terminal, or the cloud server, etc., and can obtain relevant data collected by the vehicle-mounted sensor or the vehicle itself by connecting to the vehicle-mounted information acquisition system, and dynamically analyze and evaluate the acquired data, and make corresponding feedback when the collision risk is analyzed. It can adapt to the application of most L1 or L2 level intelligent driving models.

[0050] Preferably, the intelligent strategy includes emergency braking or risk reminder.

[0051] Among them, as attached Figure 2 As shown, the front vehicle referred to in the present invention is a vehicle in the front area of ​​the same lane as the vehicle, and the side vehicle is a vehicle in the side or side front area of ​​the lanes adjacent to the lane on the left and right of the vehicle.

[0052] Preferably, see Figure 3 , step S1 comprises:

[0053] S11 acquires and collects road information in front of and to the side of the vehicle through vehicle-mounted sensors, wherein the road information includes position information, azimuth information, relative speed information, and relative acceleration information of the vehicle in front or on the side being measured;

[0054] S12: when a vehicle is detected in front of the vehicle, further obtain the relative position information, azimuth information, relative speed information and relative acceleration information of the front vehicle A1 and the vehicle; when the distance dis between the front vehicle and the vehicle is obtained according to the relative position OA1 is smaller than the preset first threshold distance disT1, azimuth angle θ A1 Within the first standard azimuth angle θT1, and the relative speed ΔV OA1 Greater than the first threshold speed ΔVT1, and / or relative acceleration Δa OA1When the acceleration is greater than the first threshold acceleration ΔaT1, the vehicle ahead is judged to be a risk target vehicle due to sudden braking, and the first risk scenario is output;

[0055] S13: When a vehicle is detected in front of the vehicle, the relative position information and azimuth information of the front vehicle A1 and the vehicle are further obtained; when the range of the lane covered by the front vehicle according to the relative position is less than the preset second threshold range, and the azimuth angle θ A1 When the vehicle ahead is judged to be cutting out of the lane within the second standard azimuth angle θT2, the relative position information, azimuth information, relative speed information and relative acceleration information of the vehicle ahead A2 and the vehicle ahead are further obtained; when the distance dis between the vehicle ahead and the vehicle ahead is obtained according to the relative position OA2 is less than the preset first threshold distance disT1, the azimuth angle θ2 is within the first standard azimuth angle θT1 range, and the relative speed ΔV OA2 Greater than the first threshold speed ΔVT1, and / or relative acceleration Δa OA When the acceleration is greater than the first threshold acceleration ΔaT1, the vehicle ahead is judged to be a risk target vehicle, and the second risk scenario is outputted;

[0056] S14: when a vehicle is detected on the side of the vehicle, further obtain the relative position information, azimuth information, relative speed information and relative acceleration information of the side vehicle B1 and the vehicle; when the distance dis between the front vehicle and the vehicle is obtained according to the relative position OB1 is smaller than the preset third threshold distance disT3, azimuth angle θ B1 Within the third standard azimuth angle θT3, and the relative speed ΔV OB1 Greater than a third threshold speed ΔVT3, and / or a relative acceleration Δa OB1 When the acceleration is greater than the third threshold acceleration ΔaT3, it is determined that the vehicle ahead has cut into the lane and is a risk target vehicle, and the third risk scenario is output.

[0057] In an exemplary implementation scenario, the settings of the first threshold distance disT1, the first standard azimuth angle θT1, the first threshold speed ΔVT1, the first threshold acceleration ΔaT1, the second standard azimuth angle θT2, the third threshold distance disT3, the third standard azimuth angle θT3, the third threshold speed ΔVT3, and the third threshold acceleration ΔaT3 can all be set according to experience or historical data. For example, disT1∈[5,30] / m, θT1:[1,10]~[10,15] / °, ΔVT1∈[5,10] / (km / h), ΔaT1∈[3,5] / (m / s 2), θT2: [15,20]~[20,25] / °, disT3∈[2,20] / m, θT3: [15,20]~[20,25] / °, ΔVT3∈[5,10] / (km / h), ΔaT3∈[3,5] / (m / s 2 ).

[0058] The above-mentioned implementation mode of the present invention collects the status data of the front and side vehicles in the lane of the vehicle and the adjacent lanes during the driving process of the vehicle, and can compare and judge with the corresponding threshold value based on the driving status data of the surrounding vehicles to detect the risk scene of the vehicle. Among them, by detecting the risk scene of the vehicle, on the one hand, it is used as a judgment condition to start the subsequent further collision risk analysis; on the other hand, through the identification of the risk scene, it can provide a basis for the characteristic braking model called in the subsequent collision risk analysis process, thereby improving the intelligent level of dynamic collision risk prediction.

[0059] Among them, considering that the driver's attention or vigilance is different in different risk scenarios (for example, the vehicle in front brakes suddenly, or the vehicle suddenly cuts in from the adjacent lane, etc.), therefore, the conditions that ultimately cause the collision are also different for different risk scenarios. The above-mentioned embodiment of the present invention performs corresponding collision risk analysis according to different risk scenarios, which can help improve the accuracy and reliability of collision risk prediction.

[0060] In an exemplary implementation scenario, the relative speed can be decomposed based on the lateral speed and the longitudinal speed, and compared with the corresponding thresholds based on the relative lateral speed and the relative longitudinal speed, as a judgment condition for the risk scenario; similarly, the relative acceleration can also be decomposed based on the lateral acceleration and the longitudinal acceleration, and compared with the corresponding thresholds based on the relative lateral acceleration and the relative longitudinal acceleration, as a judgment condition for the risk scenario; wherein the longitudinal direction refers to the direction along the lane, and the lateral direction refers to the tangent direction along the lane.

[0061] Optionally, the on-board sensors include radars, cameras, lidars, etc.; by fusing and extracting features from the data collected by different sensors, it is possible to accurately extract feature information corresponding to the vehicle in front or the vehicle on the side, such as vehicle speed, distance, relative speed, acceleration, etc.

[0062] Preferably, step S2 comprises:

[0063] S21 obtains the current driving state data of the vehicle, wherein the driving state data includes the vehicle speed V c (t), acceleration / deceleration a c (t), maximum deceleration Bsmax And standard braking performance improvement time bT.

[0064] In an exemplary scenario, the standard braking performance improvement time is measured based on vehicle test data or historical braking data, which can be a preset standard value or an empirical value.

[0065] The acceleration / deceleration data can be obtained based on the throttle / brake opening slope of the current vehicle, and the maximum deceleration is obtained based on the performance parameters or test data of the vehicle.

[0066] Preferably, see Figure 4 , step S3 comprises:

[0067] S31 retrieves a characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, wherein the characteristic braking model is:

[0068] In the formula, Bs n (t) represents the braking deceleration of the vehicle at time t, where T0≤t≤T S , T0 represents the moment when the risk scenario is detected, T S Indicates the time when the vehicle is expected to decelerate to a standstill, Bs0 indicates the deceleration data of the vehicle at time T0, Bs0 = a c (T0), rT n represents the standard reaction time corresponding to the current nth risk scenario, RF represents the reaction force factor, and Bs max Indicates the maximum deceleration of the vehicle, bT n Indicates the standard braking efficiency improvement time corresponding to the current nth risk scenario;

[0069] S32 detects the current vehicle speed V at the time when the risk scenario occurs c (T0), combined with the corresponding characteristic braking model, the collision risk prediction between the vehicle and the target risk vehicle is performed, including: calculating the current collision characteristic value Cru(t), where the calculation function of the collision characteristic value is:

[0070]

[0071] In the formula, Cru(t) represents the collision characteristic quantity, dis AB (T0) represents the relative distance between the vehicle and the risk target vehicle when the risk scene is detected, dis y Indicates the preset reserved safety distance, ΔV ABc (T0) represents the relative speed between the vehicle and the risk target vehicle when the risk scenario is detected; Δa ABc (T0) represents the relative acceleration between the host vehicle and the risk target vehicle when the risk scenario is detected; TS Indicates the estimated time of deceleration to a standstill;

[0072] When the collision characteristic value Cru(t) is less than the second standard distance disT2, the output risk prediction result is a second-level collision risk;

[0073] When the collision characteristic value Cru(t) is less than the first standard distance disT1, the output risk prediction result is a first-level collision risk;

[0074] When the collision characteristic value Cru(t) is not less than the second standard distance disT2, the risk prediction result is output as no risk.

[0075] In an exemplary scenario, the characteristic braking model called for different risk scenarios has a standard reaction time rT n , Standard braking performance improvement time bT n The settings are not the same. Based on simulation tests and reality tests, it can be concluded that for the first risk scenario of sudden braking in front, the standard reaction time rT1=0.6s and the standard braking efficiency improvement time bT1=0.43s are obtained; for the second risk scenario of the vehicle in front cutting out, the standard reaction time rT2=0.58s and the standard braking efficiency improvement time bT2=0.52s are obtained; for the third risk scenario of the side vehicle urgently cutting into the vehicle, the standard reaction time rT3=0.52s and the standard braking efficiency improvement time bT3=0.48s are obtained.

[0076] In the above-mentioned embodiment of the present invention, when a risk scenario is detected, the corresponding characteristic braking model is called according to the risk scenario, wherein the key parameters in the characteristic braking model are set accordingly according to the empirical values ​​or test values ​​obtained in different scenarios. Based on different characteristic braking models, the vehicle collision risk is further analyzed, and the braking distance or braking time of the vehicle can be calculated based on the characteristic braking model, thereby accurately estimating the possible collision risk and improving the accuracy of the collision risk estimation.

[0077] At the same time, further decisions can be made based on the results of the collision risk analysis to issue risk warning reminders or vehicle emergency braking, which can improve the intelligence level of intelligent driving assistance or intelligent driving control.

[0078] In an exemplary implementation scenario, the relative speed ΔV between the host vehicle and the risk target vehicle when a risk scenario is detected is ABc(T0) represents the longitudinal relative speed of the two vehicles in the actual calculation process. When the risk target vehicle is in the same lane as the vehicle, the relative speed does not need to be corrected, and the relative speed value measured in step S1 is directly used for calculation; when the risk target vehicle is in a different lane from the vehicle, the relative speed is corrected, that is, the corrected relative speed ΔV A ′ Bc (T0) = cos(θ) × ΔV ABc (T0), where θ represents the angle between the risk target vehicle and the longitudinal direction, that is, the corrected relative speed ΔV A ′ Bc (T0) is calculated.

[0079] Similarly, the relative acceleration Δa between the vehicle and the risk target vehicle when the risk scenario is detected is ABc (T0) In the actual calculation process, the relative acceleration can also be corrected in a similar manner as described above, which will not be repeated in the present invention.

[0080] In an exemplary implementation scenario, the estimated time T for deceleration to rest is S Based on the current vehicle speed V c (T0), combined with the braking deceleration Bs corresponding to the characteristic braking model n (t) is deduced, when the vehicle speed drops to 0, the corresponding time T is obtained S .

[0081] In an exemplary implementation scenario, in step S32, in addition to the method of predicting the collision risk based on the collision feature quantity proposed in the above embodiment, a collision prediction model that has been trained in the prior art can be used to further analyze the collision risk between the vehicle and the target risk vehicle based on the collected parameters and the braking model proposed above.

[0082] Further, on the basis of setting different characteristic braking models based on risk scenarios, considering that in the actual application of collision risk warning, the driver's state cannot be ignored, therefore, in the decision-making and processing process facing the risk scenario, the driver's driving state will also have an important impact on the final collision risk result. In the existing collision risk warning technology, it is usually judged based on a fixed value of distance, for example, if the distance is less than 2 meters, it is judged as a collision risk, or if the estimated collision time is less than 2s, it will be judged as a risk; however, in actual scenarios, different drivers have different psychological expectations of collision risks and the ability to deal with risks. For example, when the driver is in a fatigued state, his reaction force will be reduced, so the reaction time to the risk scenario will also fluctuate. Therefore, using a fixed value of reaction time in the braking model to estimate the final collision risk will definitely result in deviation. Therefore, in the characteristic braking model proposed by the present invention, a reaction force factor is further added as an adjustment parameter to further adjust the characteristic braking model based on the driver's current real driving state, so as to further improve the accuracy and adaptability of the characteristic braking model, thereby improving the reliability of the collision risk.

[0083] The larger the reaction force factor is, the lower the driver's reaction force is, and thus the braking capacity obtained based on the braking model is lower, thereby increasing the sensitivity of the collision risk trigger.

[0084] Preferably, step S3 further includes:

[0085] S30 obtains real-time reaction force factor RF, including:

[0086] S301 obtains the identity information of the driver of the vehicle, and retrieves corresponding driving characteristic data according to the obtained identity information, wherein the driving characteristic data includes a driving style factor Std based on the driver's preset, wherein Std∈[1,10], the larger the Std, the more aggressive the driving style set by the driver, and vice versa, the larger the Std, the more conservative the driving style set by the driver;

[0087] S302 obtains the eye image of the driver of the vehicle based on the infrared camera, and uses the image recognition engine to detect the driver's blinking action based on the obtained eye image, and counts the driver's blinking frequency PE(t) in the current time period GT1, that is, the number of blinks per minute;

[0088] S303 calculates the current reaction force factor RF according to the vehicle speed data in the current time period GT1 and the driver's blink frequency data, wherein the reaction force factor calculation function used is:

[0089]

[0090] In the formula, RF(t) represents the reaction force factor at the current moment, ck represents the preset reaction force adjustment parameter, where ck∈[0.01,1]; ave represents the preset sensitivity factor, ave∈[0.01,100], ω1 and ω2 represent the preset weight factors, PE(t) represents the driver's blinking frequency in the current time period GT1, and PET represents the standard blinking frequency, which is set according to experience; represents the vehicle speed change rate of the i-th sampling point in the current time period, where i = 1, 2, …, ni, ni represents the total number of sampling points, mean{·} represents the mean function, sd{·} represents the standard deviation function, and Std represents the preset driving style factor.

[0091] Among them, the setting method of the current time period GT1 can be a time range of a fixed time length from the current moment to the previous moment; it can be a time range of a fixed time length calculated from a new time node after every fixed time length, and the current period GT1 represents the latest time range at the specific current moment; or it can be set according to other methods, and the present invention does not make specific limitations here.

[0092] In an exemplary implementation scenario, in step S301, the identity information of the driver of the vehicle is obtained by having the driver log in to a corresponding account to obtain the identity information, or by identifying the driver's identity based on face recognition, fingerprint recognition, etc., to obtain the driver's identity information. When the driver account is initially set up, it can provide the driver with parameters such as the corresponding driving style factor for subsequent personalized calls.

[0093] The above-mentioned embodiment of the present invention proposes a method for dynamically acquiring the reaction force factor of the key parameter in the characteristic braking model, wherein the driver's actual reaction force state can be reflected based on the current driver's fatigue state (the fatigue state is reflected by the blinking frequency. During normal driving, the higher the driver's blinking frequency, the higher the fatigue degree reflected. It can be considered that the driver's reaction force will be reduced in the fatigue state) and operation data (reflected by the vehicle's speed change rate. When the driving state is more normal, the vehicle speed usually changes more slowly. When the mental state or reaction force is insufficient, the vehicle usually undergoes timely correction operations such as emergency braking, which leads to an increase in the rate of change of the vehicle speed) to reflect the driver's actual reaction force state, thereby correcting the characteristic braking model, so that the braking analysis and collision analysis results obtained based on the braking model are more in line with the actual driving scenario, thereby improving the accuracy and adaptability of collision risk analysis, and helping to improve the effect and user experience of intelligent driving assistance and intelligent driving control.

[0094] Preferably, step S4 comprises:

[0095] When the risk prediction result is a level 2 collision risk, a collision warning prompt message is issued;

[0096] When the risk prediction result is a level 1 collision risk, an emergency braking command is initiated.

[0097] By setting the secondary collision risk analysis results, corresponding auxiliary prompts or emergency braking functions can be executed based on the different collision risk severities to adapt to the use of different application scenarios.

[0098] Based on the above-mentioned dynamic assessment method based on real-time dynamic vehicle risk scenario state, the present invention further proposes a dynamic assessment system based on real-time dynamic vehicle risk scenario state, which is used to perform the above-mentioned Figure 1 The method shown includes: a road information acquisition module, a driving information acquisition module, an analysis and processing module and a strategy control module; wherein,

[0099] The road information acquisition module is used to detect the risk scenario of the vehicle based on the road information collected in real time, where the risk scenario includes the side vehicle cutting into the lane, the front vehicle braking suddenly, and the front vehicle cutting out of the lane;

[0100] The driving information acquisition module is used to obtain the driving status data of the vehicle;

[0101] The analysis and processing module is used to retrieve the characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, and predict the collision risk of the vehicle in combination with the driving state data and the characteristic braking model to obtain the collision risk prediction result;

[0102] The strategy control module is used to execute corresponding intelligent strategies according to the collision risk prediction results.

[0103] It should be noted that, in the above implementation mode, each functional module is also used to execute the above Figure 1 The methods of the corresponding steps shown, as well as the corresponding specific implementation methods of each step, are not repeated in the present invention.

[0104] It should be noted that each functional unit / module in each embodiment of the present invention may be integrated into one processing unit / module, or each unit / module may exist physically separately, or two or more units / modules may be integrated into one unit / module. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of software functional unit / module.

[0105] Through the description of the above implementation mode, it can be clearly understood by those skilled in the art that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein or their combination. For software implementation, part or all of the process of the embodiment can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein the communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that can be accessed by a computer. Computer-readable media can include, but are not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0106] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should analyze that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. A dynamic assessment method based on real-time dynamic vehicle risk scenario state, characterized in that: The steps include: S1 detects the risk scenarios of the vehicle based on the road information collected in real time, where the risk scenarios include the vehicle cutting into the lane, the vehicle in front braking suddenly, and the vehicle in front cutting out of the lane; S2 obtains the driving status data of the vehicle; S3 retrieves a characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, and predicts the collision risk of the vehicle in combination with the driving state data and the characteristic braking model to obtain a collision risk prediction result; S4 executes corresponding intelligent strategies based on the collision risk prediction results.

2. The dynamic assessment method based on real-time dynamic vehicle risk scenario state according to claim 1 is characterized in that: Step S1 includes: S11 acquires and collects road information in front of and to the side of the vehicle through on-board sensors, wherein the road information includes position information, azimuth information, relative speed information, and relative acceleration information of the vehicle in front or on the side being measured; S12: When a vehicle is detected in front of the vehicle, the relative position information, azimuth information, relative speed information and relative acceleration information of the front vehicle A1 and the vehicle are further obtained; when the distance between the front vehicle and the vehicle is less than a preset first threshold distance, the azimuth is within the first standard azimuth range, the relative speed is greater than the first threshold speed, and / or the relative acceleration is greater than the first threshold acceleration, the front vehicle is judged to be a risk target vehicle due to sudden braking, and the first risk scenario is outputted; S13 When a vehicle is detected in front of the vehicle, the relative position information and azimuth information of the front vehicle A1 and the vehicle are further obtained; when the range of the lane covered by the front vehicle is less than the preset second threshold range, and the azimuth is within the second standard azimuth range, it is judged that the front vehicle is cutting out of the lane, and the relative position information, azimuth information, relative speed information and relative acceleration information of the vehicle A2 further ahead and the vehicle are further obtained; when the distance between the vehicle further ahead and the vehicle is less than the preset first threshold distance, the azimuth is within the first standard azimuth range, and the relative speed is greater than the first threshold speed, and / or the relative acceleration is greater than the first threshold acceleration, it is judged that the vehicle further ahead is a risk target vehicle, and the current entry into the second risk scenario is output; S14 When a vehicle is detected on the side of the vehicle, the relative position information, azimuth information, relative speed information and relative acceleration information of the side vehicle B1 and the vehicle are further obtained; when it is obtained according to the relative position that the distance between the front vehicle and the vehicle is less than the preset third threshold distance, the azimuth angle is within the third standard azimuth angle range, and the relative speed is greater than the third threshold speed, and / or the relative acceleration is greater than the third threshold acceleration, it is judged that the front vehicle has cut into the lane and is a risk target vehicle, and the current third risk scenario is output.

3. The dynamic assessment method based on real-time dynamic vehicle risk scenario state according to claim 2 is characterized in that: In step S11, the vehicle-mounted sensors include radar, camera, and lidar; by fusing and feature extracting the data collected by different sensors, feature information corresponding to the vehicle in front or the vehicle on the side is extracted, including vehicle speed, distance, relative speed, and acceleration.

4. The dynamic assessment method based on real-time dynamic vehicle risk scenario state according to claim 2 is characterized in that: Step S2 includes: S21 obtains the current driving state data of the vehicle, wherein the driving state data includes the vehicle speed V c (t), acceleration / deceleration a c (t), maximum deceleration Bs max And standard braking performance improvement time bT.

5. The dynamic assessment method based on real-time dynamic vehicle risk scenario state according to claim 4 is characterized in that: Step S3 includes: S31 retrieves a characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, wherein the characteristic braking model is: In the formula, Bs n (t) represents the braking deceleration of the vehicle at time t, where T0≤t≤T S , T0 represents the moment when the risk scenario is detected, T S Indicates the time when the vehicle is expected to decelerate to a standstill, Bs0 indicates the deceleration data of the vehicle at time T0, Bs0 = a c (T0), rT n represents the standard reaction time corresponding to the current nth risk scenario, RF represents the reaction force factor, and Bs max Indicates the maximum deceleration of the vehicle, bT n Indicates the standard braking efficiency improvement time corresponding to the current nth risk scenario; S32 detects the current vehicle speed V at the time when the risk scenario occurs c (T0), combined with the corresponding characteristic braking model, the collision risk prediction between the vehicle and the target risk vehicle is performed, including: calculating the current collision characteristic value Cru(t), where the calculation function of the collision characteristic value is: In the formula, Cru(t) represents the collision characteristic quantity, dis AB (T0) represents the relative distance between the vehicle and the risk target vehicle when the risk scene is detected, dis y Indicates the preset reserved safety distance, ΔV ABc (T0) represents the relative speed between the vehicle and the risk target vehicle when the risk scenario is detected; Δa ABc (T0) represents the relative acceleration between the host vehicle and the risk target vehicle when the risk scenario is detected; T S Indicates the estimated time of deceleration to a standstill; When the collision characteristic value Cru(t) is less than the second standard distance disT2, the output risk prediction result is a second-level collision risk; When the collision characteristic value Cru(t) is less than the first standard distance disT1, the output risk prediction result is a first-level collision risk; When the collision characteristic value Cru(t) is not less than the second standard distance disT2, the risk prediction result is output as no risk.

6. The dynamic assessment method based on real-time dynamic vehicle risk scenario state according to claim 5 is characterized in that: Step S4 includes: When the risk prediction result is a level 2 collision risk, a collision warning prompt message is issued; When the risk prediction result is a level 1 collision risk, an emergency braking command is initiated.

7. A dynamic assessment system based on real-time dynamic vehicle risk scenario status, characterized by: It includes a road information acquisition module, a driving information acquisition module, an analysis and processing module and a strategy control module; wherein, The road information acquisition module is used to detect the risk scenario of the vehicle based on the road information collected in real time, where the risk scenario includes the side vehicle cutting into the lane, the front vehicle braking suddenly, and the front vehicle cutting out of the lane; The driving information acquisition module is used to obtain the driving status data of the vehicle; The analysis and processing module is used to retrieve the characteristic braking model corresponding to the risk scenario according to the acquired risk scenario, and predict the collision risk of the vehicle in combination with the driving state data and the characteristic braking model to obtain the collision risk prediction result; The strategy control module is used to execute corresponding intelligent strategies according to the collision risk prediction results.

Images