A method and device for selecting a key target of a pre-crash warning function

Abstract

The application relates to a key target selection method and device of a front collision warning function, and belongs to the technical field of auxiliary driving. The target selection method comprises the following steps: acquiring self-vehicle condition data of a self-vehicle and first target vehicle condition data of all first target vehicles in a preset range around the self-vehicle; obtaining a first avoidance target acceleration of each first target vehicle through an avoidance acceleration algorithm according to the self-vehicle condition data and the first target vehicle condition data; and obtaining a key target vehicle from all the first target vehicles according to the first avoidance target acceleration. In the method, the driving scene is combined in the key target selection, the key target vehicle is determined through the avoidance target acceleration of the self-vehicle in different driving scenes, and the accuracy of the key target selection is effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of assisted driving, in particular to a key target selection method and device for a front collision warning function. BACKGROUND

[0002] The front collision warning (FCW) function is a warning function that reminds the driver of potential collision risks. It provides visual, sound or tactile warning signals to the driver when detecting potential collision risks through monitoring the situation in front of the vehicle, so that the driver can take risk-avoiding measures. In the front collision warning technology, target selection is a key link; only by selecting the most dangerous target in time and correctly can the potential collision risk be identified in advance and the driver be notified in time, so as to give the driver enough reaction time to avoid collision or reduce the severity of collision.

[0003] The target selection method commonly used in the industry at present is to calculate the collision time based on the relative distance, relative speed and relative acceleration at the current moment, and to select the key target within this collision time. However, the key target selected based on this collision time has the problem of false triggering in different scenes due to the lack of comprehensive consideration of the driving scene. SUMMARY

[0004] Therefore, the present application provides a key target selection method and device for a front collision warning function to solve the problem of false triggering in different scenes caused by the lack of comprehensive consideration of the driving scene in the prior art.

[0005] In a first aspect, a key target selection method for a front collision warning function is provided, which comprises:

[0006] obtaining self-vehicle condition data of a self-vehicle and first target vehicle condition data of all first target vehicles within a preset range around the self-vehicle;

[0007] obtaining a first avoidance target acceleration of each first target vehicle according to the self-vehicle condition data and the first target vehicle condition data through an avoidance acceleration algorithm; the first avoidance target acceleration comprises a lateral avoidance acceleration and a longitudinal avoidance acceleration; the avoidance acceleration algorithm is a method for calculating an avoidance target acceleration in combination with a driving scene;

[0008] obtaining a key target vehicle from all first target vehicles according to the first avoidance target acceleration.

[0009] Further, the avoidance acceleration algorithm comprises:

[0010] selecting one vehicle from the first target vehicles as a second target vehicle;

[0011] obtaining a driver reaction time according to the ego vehicle condition data and second target vehicle condition data of the second target vehicle;

[0012] adding the driver reaction time and a preset brake delay time to obtain a collision delay time of the second target vehicle;

[0013] obtaining a target motion trajectory and an ego vehicle motion trajectory according to the collision delay time; the target motion trajectory is a motion trajectory of the second target vehicle within a corresponding collision delay time; the ego vehicle motion trajectory is a motion trajectory of the ego vehicle within the corresponding collision delay time;

[0014] obtaining a second target avoidance target acceleration corresponding to the second target vehicle according to the ego vehicle motion trajectory and the target motion trajectory.

[0015] Further, the obtaining of the driver reaction time according to the ego vehicle condition data and the second target vehicle condition data of the second target vehicle comprises:

[0016] obtaining a triggered driving scene level according to the ego vehicle condition data; the driving scene level comprises level one, level two, level three, level four and level five, and each driving scene level is provided with a reaction adjustment time;

[0017] obtaining a minimum reaction time according to the triggered driving scene level;

[0018] obtaining the driver reaction time according to the minimum reaction time and a current collision sensitivity in the second target vehicle condition data.

[0019] Further, each driving scene level comprises at least one driving scene; each driving scene is provided with at least one triggering parameter item and a corresponding triggering parameter threshold value, and the obtaining of the triggered driving scene level according to the ego vehicle condition data comprises:

[0020] extracting a key parameter item and a corresponding key parameter value from the ego vehicle condition data; the key parameter item is a triggering parameter item of all the driving scenes;

[0021] determining the triggered driving scene according to a comparison result of the key parameter value and the triggering parameter threshold value;

[0022] obtaining the triggered driving scene level according to the triggered driving scene.

[0023] Further, the obtaining of the minimum reaction time according to the triggered driving scene level comprises:

[0024] obtaining the minimum reaction time through a reaction time formula; the reaction time formula is:

[0025] T1 = min((t1 + t2), (t3 + t4), t5);

[0026] Wherein, T1 is the minimum reaction time; t1 is the reaction adjustment time of level one, level one is a reserved adjustment level, and the default reaction adjustment time is 0; t2 is the reaction adjustment time of level two, the driving scene of level two includes frequent operation of function buttons, and t2 is 0 if level two is not triggered; t3 is the reaction adjustment time of level three, the driving scene of level three includes high-curvature scene, high-speed driving scene on a curve and low-speed acceleration scene, and t3 is 0 if level three is not triggered; t4 is the reaction adjustment time of level four, the driving scene of level four includes driver acceleration, driver steering, adaptive cruise acceleration and deceleration, city driving, low-speed driving, driver overtaking and driver braking, and t4 is 0 if level four is not triggered; t5 is the reaction adjustment time of level five, the driving scene of level five includes driver deceleration, and t5 is 0 if level five is not triggered.

[0027] Further, the collision sensitivity includes low sensitivity, normal sensitivity, high sensitivity and unused collision warning; each of the collision sensitivity is provided with a corresponding preset reaction time; the driver reaction time obtained according to the minimum reaction time and the current collision sensitivity in the second target vehicle condition data includes:

[0028] Obtaining the collision sensitivity currently set by the ego vehicle and the corresponding preset reaction time as the warning reaction time;

[0029] Adding the minimum reaction time and the warning reaction time to obtain the driver reaction time.

[0030] Further, the ego vehicle condition data further includes ego vehicle speed, ego vehicle direction and ego vehicle acceleration; the second target vehicle condition data further includes target vehicle speed, target vehicle direction and target acceleration; the target motion trajectory and the ego motion trajectory obtained according to the collision delay time include:

[0031] Obtaining the ego motion trajectory according to the ego vehicle speed, the ego vehicle direction, the ego vehicle acceleration and the collision delay time;

[0032] Obtaining the target motion trajectory according to the target vehicle speed, the target vehicle direction, the target acceleration and the collision delay time.

[0033] Further, the second target vehicle corresponding second avoidance target acceleration obtained according to the ego motion trajectory and the target motion trajectory includes:

[0034] Inputting the ego motion trajectory and the target motion trajectory as input data into a kinematics model to obtain the lateral avoidance acceleration and the longitudinal avoidance acceleration of the ego vehicle.

[0035] The lateral avoidance acceleration and the longitudinal avoidance acceleration are taken as the second avoidance target acceleration.

[0036] Further, the key target vehicle is obtained from all the first target vehicles according to the first avoidance target acceleration, and the method comprises the following steps:

[0037] The lateral avoidance acceleration and the longitudinal avoidance acceleration corresponding to the first avoidance target acceleration are processed by averaging to obtain the average acceleration of each first target vehicle.

[0038] The first target vehicle with the maximum absolute value of the average acceleration is taken as the key target vehicle.

[0039] In a second aspect, a key target selection device for a pre-crash warning function is provided, and the device comprises:

[0040] An acquisition module is configured to acquire vehicle condition data of a host vehicle and first target condition data of all first target vehicles within a preset range around the host vehicle.

[0041] An avoidance acceleration module is configured to obtain a first avoidance target acceleration of each first target vehicle by an avoidance acceleration algorithm according to the vehicle condition data and the first target condition data, wherein the first avoidance target acceleration comprises a lateral avoidance acceleration and a longitudinal avoidance acceleration, and the avoidance acceleration algorithm is a method for calculating an avoidance target acceleration in combination with a driving scenario.

[0042] A selection module is configured to obtain a key target vehicle from all the first target vehicles according to the first avoidance target acceleration.

[0043] The above technical solution has at least the following beneficial effects:

[0044] The key target selection method and device for a pre-crash warning function are provided, the vehicle condition data of the host vehicle and the first target condition data of all the first target vehicles within a preset range around the host vehicle are acquired, the first avoidance target acceleration of each first target vehicle is obtained by an avoidance acceleration algorithm according to the vehicle condition data and the first target condition data, and the key target vehicle is obtained from all the first target vehicles according to the first avoidance target acceleration, wherein the driving scenario is combined in the key target selection in the method, and the key target vehicle is determined by the avoidance target acceleration of the host vehicle in different driving scenarios, thereby effectively improving the accuracy of the key target selection.

[0045] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0047] Figure 1 is a flow chart of a key target selection method of a pre-crash warning function according to an exemplary embodiment of the present application;

[0048] Figure 2 is a schematic block diagram of a key target selection device of a pre-crash warning function according to an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to make the purposes, technical solutions and advantages of the present application more clear, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0050] In the pre-crash warning technology, target selection is a key link; only if the most dangerous target is selected in time and correctly, the potential collision risk can be identified in advance and the driver can be notified in time, so as to give the driver enough reaction time to avoid collision or reduce the severity of collision. The target selection method commonly used in the industry at present is to calculate the collision time based on the relative distance, relative speed and relative acceleration at the current moment, and to select the key target within the collision time. The key target selected based on this time is not necessarily the target of potential collision, and there is a certain risk of false triggering, causing the driver to have a bad driving experience.

[0051] The embodiments of the present application provide a key target selection method and device of a pre-crash warning function. First, the driving initiative of the driver is estimated through the data of the vehicle, and the driving initiative is classified to set different reaction times, and the avoidance target acceleration is calculated based on the reaction time to select the key target vehicle. This scheme can reduce the risk of false triggering of the pre-crash function caused by ignoring the driver's action, can more accurately identify the potential collision risk and notify the driver in time, so as to improve the accuracy of target selection of the pre-crash warning function and the driving experience of the passenger.

[0052] The method and device of the present application will be described below through specific embodiments.

[0053] Please refer to Figure 1 , Figure 1is a flowchart of a key target selection method of a pre-crash warning function according to an exemplary embodiment of the present application, see Figure 1 The method comprises:

[0054] In step S11, vehicle condition data of the ego vehicle and first target vehicle condition data of all first target vehicles within a preset range around the ego vehicle are acquired.

[0055] In step S12, first avoidance target acceleration of each first target vehicle is obtained according to the ego vehicle condition data and the first target vehicle condition data through an avoidance acceleration algorithm; the first avoidance target acceleration comprises lateral avoidance acceleration and longitudinal avoidance acceleration.

[0056] In step S13, the key target vehicle is obtained from all the first target vehicles according to the first avoidance target acceleration.

[0057] It should be noted that the technical solution provided in the embodiment can be added to an existing system in the form of an applet in specific practice, or an interface can be provided to the outside in the form of an independent application program to complete the key target vehicle selection function; applicable scenarios include but are not limited to: scenarios involving collision warning such as assisted driving, constant speed cruise, etc.

[0058] Specifically, the preset range around the ego vehicle comprises an effective distance; the larger the effective distance, the more the first target vehicles, and the higher the trigger accuracy of the key target vehicle; the effective distance needs to be set according to the specific collision safety level.

[0059] It can be understood that the method provided in the embodiment acquires the ego vehicle condition data of the ego vehicle and the first target vehicle condition data of all the first target vehicles within a preset range around the ego vehicle, obtains the first avoidance target acceleration of each first target vehicle according to the ego vehicle condition data and the first target vehicle condition data through an avoidance acceleration algorithm, and obtains the key target vehicle from all the first target vehicles according to the first avoidance target acceleration, wherein the method combines a driving scene in key target selection, and the key target vehicle is determined through the avoidance target acceleration of the ego vehicle in different driving scenes, thereby effectively improving the accuracy of key target selection.

[0060] In specific practice, the step S11 of acquiring the ego vehicle condition data of the ego vehicle and the first target vehicle condition data of all the first target vehicles within a preset range around the ego vehicle comprises: receiving ego vehicle condition data Host_data at a current time and target vehicle condition data Obj_data of target vehicles within a path in front of the ego vehicle at the current time, wherein the ego vehicle condition data comprises basic vehicle condition data such as ego vehicle steering wheel change rate, ego vehicle brake state, function activation condition, and vehicle function button pressing state, and the target vehicle condition data comprises vehicle speed, driving direction, acceleration, and steering light state of the target vehicle and data reflecting the driving state of the target vehicle.

[0061] In the specific practice, the step S12 "avoidance acceleration algorithm" comprises: selecting one vehicle from the first target vehicle as the second target vehicle; obtaining the driver reaction time according to the vehicle condition data of the ego vehicle and the second target vehicle condition data of the second target vehicle; adding the driver reaction time and the preset brake delay time to obtain the collision delay time of the second target vehicle; obtaining the target motion trajectory and the ego motion trajectory according to the collision delay time; the target motion trajectory is the motion trajectory of the second target vehicle within the corresponding collision delay time; the ego motion trajectory is the motion trajectory of the ego vehicle within the corresponding collision delay time; obtaining the second avoidance target acceleration corresponding to the second target vehicle according to the ego motion trajectory and the target motion trajectory.

[0062] It should be noted that the preset brake delay time is the brake reaction time obtained according to multiple brake experiments, which can be set according to the ego vehicle collision safety level.

[0063] Specifically, obtaining the driver reaction time according to the vehicle condition data of the ego vehicle and the second target vehicle condition data of the second target vehicle comprises: obtaining the triggered driving scene level according to the vehicle condition data of the ego vehicle; the driving scene level comprises level one, level two, level three, level four and level five, and each driving scene level is provided with a reaction adjustment time; obtaining the minimum reaction time according to the triggered driving scene level; obtaining the driver reaction time according to the minimum reaction time and the current collision sensitivity in the second target vehicle condition data.

[0064] It should be noted that the reaction adjustment time of level one, level two, level three, level four and level five decreases in turn, and the specific reaction adjustment time is set according to the ego vehicle collision safety level. The five levels are specifically: level one is a reserved driver state adjustment level, which is used to preset the driver reaction level, not used, and the default adjustment value is 0; level two is suitable for driver frequent operation function scene, such as operating button at last moment; level three is suitable for high curvature scene, high speed driving scene on curve, low speed acceleration scene; level four is suitable for driver acceleration, driver steering, adaptive cruise acceleration and deceleration, city driving, low speed driving, driver overtaking, driver braking and other scenes; level five is suitable for driver deceleration scene.

[0065] Specifically, each driving scene level comprises at least one driving scene; each driving scene is provided with at least one trigger parameter item and corresponding trigger parameter threshold value, and the triggered driving scene level is obtained according to the vehicle condition data of the ego vehicle, comprising: extracting the key parameter item and the corresponding key parameter value from the vehicle condition data of the ego vehicle; the key parameter item is the trigger parameter item of all driving scenes; determining the triggered driving scene according to the comparison result of the key parameter value and the trigger parameter threshold value; obtaining the triggered driving scene level according to the triggered driving scene.

[0066] It should be noted that when all trigger parameter items in a driving scene are met, that is, all key parameter values meet the expectations after comparison with the trigger parameter threshold, the driving scene is triggered, and then the driving scene level of the driving scene is triggered.

[0067] Specifically, the minimum reaction duration is obtained according to the triggered driving scene level, including: the minimum reaction duration is obtained by a reaction duration formula, and the reaction duration formula is:

[0068] T1 = min ((t1 + t2), (t3 + t4), t5);

[0069] Wherein, T1 is the minimum reaction duration; t1 is the reaction adjustment duration of level one, level one is a reserved adjustment level, and the default reaction adjustment duration is 0; t2 is the reaction adjustment duration of level two, level two driving scene includes frequently operated function buttons, and t2 is 0 if level two is not triggered; t3 is the reaction adjustment duration of level three, level three driving scene includes high curvature scene, high speed driving scene on curve and low speed acceleration scene, and t3 is 0 if level three is not triggered; t4 is the reaction adjustment duration of level four, level four driving scene includes driver acceleration, driver steering, adaptive cruise acceleration and deceleration, city driving, low speed driving, driver overtaking and driver braking, and t4 is 0 if level four is not triggered; t5 is the reaction adjustment duration of level five, level five driving scene includes driver deceleration, and t5 is 0 if level five is not triggered.

[0070] Specifically, the collision sensitivity includes low sensitivity, normal sensitivity, high sensitivity and unused collision warning; each collision sensitivity is provided with a corresponding preset reaction duration; the driver reaction duration is obtained according to the minimum reaction duration and the current collision sensitivity in the second target vehicle condition data, including: obtaining the collision sensitivity currently set by the ego vehicle and the corresponding preset reaction duration as the warning reaction duration; the minimum reaction duration and the warning reaction duration are added to obtain the driver reaction duration.

[0071] It should be noted that the unused collision warning, low sensitivity, normal sensitivity and high sensitivity are respectively provided with corresponding preset reaction durations, and the preset reaction durations decrease in turn.

[0072] Specifically, the ego vehicle condition data further includes ego vehicle speed, ego vehicle direction and ego vehicle acceleration; the second target vehicle condition data further includes target vehicle speed, target vehicle direction and target acceleration; the target motion trajectory and the ego vehicle motion trajectory are obtained according to the collision delay duration, including: the ego vehicle motion trajectory is obtained according to the ego vehicle speed, the ego vehicle direction, the ego vehicle acceleration and the collision delay duration; the target motion trajectory is obtained according to the target vehicle speed, the target vehicle direction, the target acceleration and the collision delay duration.

[0073] It should be noted that the self-vehicle motion trajectory and the target motion trajectory can be obtained according to the self-vehicle speed, the self-vehicle direction, the self-vehicle acceleration, the target speed, the target direction, the target acceleration, and the collision delay time length.

[0074] Specifically, the second target vehicle corresponding second avoidance target acceleration is obtained according to the self-vehicle motion trajectory and the target motion trajectory, including: inputting the self-vehicle motion trajectory and the target motion trajectory into a kinematic model as input data to obtain a lateral avoidance acceleration and a longitudinal avoidance acceleration of the self-vehicle; and taking the lateral avoidance acceleration and the longitudinal avoidance acceleration as the second avoidance target acceleration.

[0075] It should be noted that the kinematic model can select an existing motion model according to the input parameters.

[0076] In specific practice, the step S13 of obtaining the key target vehicle from all the first target vehicles according to the first avoidance target acceleration includes: performing mean value processing on the lateral avoidance acceleration and the longitudinal avoidance acceleration corresponding to the first avoidance target acceleration to obtain a mean value acceleration of each first target vehicle; and taking the first target vehicle with the largest absolute value of the mean value acceleration as the key target vehicle.

[0077] It should be noted that the mean value acceleration can be a negative number, and the larger the absolute value of the mean value acceleration, the more dangerous the corresponding target vehicle is, so the target vehicle corresponding to the mean value acceleration with the largest absolute value is taken as the key target vehicle.

[0078] In one specific embodiment, the mean value acceleration calculation process of the target vehicle is as follows:

[0079] The self-vehicle condition data includes: the function sensitivity is normal; the steering wheel angle is 0; the steering wheel speed is 0; the self-vehicle acceleration is -0.1; the self-vehicle speed is 18.9; the self-vehicle acceleration pedal change rate is 0; and the self-vehicle pedal position is 0.

[0080] The driving scene trigger judgment process is: the current self-vehicle acceleration pedal position (0) is less than the pedal position threshold value (5) = true; the acceleration pedal position (30) 10s ago is greater than the pedal position value (5) = true; the comparison result: the driver's acceleration pedal is released, the driving scene trigger, the driving scene reaction time adjustment value is -0.6, and the level is 4; the minimum reaction time length calculation T1 = min ((0+0), (0+-0.6), 0);

[0081] The collision delay time length calculation process is: the function sensitivity default time, the function sensitivity is normal, the default reaction time is 1.18s; the driver's reaction time length is: 1.18-0.6 = 0.58; the preset brake delay time length is 0.2s; and the collision delay time length is: 0.58+0.2 = 0.78;

[0082] The motion trajectory of the ego vehicle is obtained based on the collision delay duration and the ego vehicle speed, ego vehicle direction, ego vehicle acceleration and the collision delay duration; the target motion trajectory is obtained based on the collision delay duration and the target speed, target direction and target acceleration and the collision delay duration; the motion trajectory of the ego vehicle and the target motion trajectory are input into the kinematic model as input data to obtain the lateral avoidance acceleration and the longitudinal avoidance acceleration of the ego vehicle, the longitudinal avoidance acceleration is -5 m / s2, and the lateral avoidance acceleration is -1 m / s2; and the mean acceleration of the target vehicle is | -5 + -1 | = 6.

[0083] Please refer to Figure 2 , Figure 2 is a schematic block diagram of a key target selection device of a forward collision warning function according to an example embodiment of the present application, and please refer to Figure 2 The key target selection device 100 of the forward collision warning function includes:

[0084] The acquisition module 101 is configured to acquire ego vehicle condition data of an ego vehicle and first target condition data of all first target vehicles within a preset range around the ego vehicle.

[0085] The avoidance acceleration module 102 is configured to obtain, by means of an avoidance acceleration algorithm, a first avoidance target acceleration of each first target vehicle according to the ego vehicle condition data and the first target condition data; the first avoidance target acceleration includes a lateral avoidance acceleration and a longitudinal avoidance acceleration; the avoidance acceleration algorithm is a method of calculating the avoidance target acceleration in combination with a driving scene.

[0086] The selection module 103 is configured to obtain a key target vehicle from all the first target vehicles according to the first avoidance target acceleration.

[0087] It should be noted that the technical solution provided in the embodiment is applicable to scenes including but not limited to collision warning scenes such as auxiliary driving and constant speed cruise.

[0088] Specifically, the preset range around the ego vehicle includes an effective distance, and the greater the effective distance is, the more the first target vehicles are, and the higher the triggering precision of the key target vehicle is; the effective distance needs to be set according to a specific collision safety level.

[0089] It can be understood that the device provided by the embodiment acquires the self-vehicle condition data of the self-vehicle and the first target vehicle condition data of all the first target vehicles in a preset range around the self-vehicle, obtains the first avoidance target acceleration of each first target vehicle according to the self-vehicle condition data and the first target vehicle condition data through the avoidance acceleration algorithm, and obtains the key target vehicle from all the first target vehicles according to the first avoidance target acceleration. In the method, the driving scene is combined in the selection of the key target, the key target vehicle is determined through the avoidance target acceleration of the self-vehicle in different driving scenes, and the accuracy of the selection of the key target is effectively improved.

[0090] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it cannot be understood as a limitation on the patent scope of the present application. It should be pointed out that, for ordinary skilled persons in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

[0091] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiments. Any reference to memory, storage, database or other medium used in the embodiments provided by the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM) and the like.

[0092] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0093] The above-described embodiments are merely illustrative of several embodiments of the present application, which are described in more detail and in a more specific and detailed manner, but should not be construed as limiting the scope of the patent. It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method of selecting a key target for a pre-crash warning function, characterized by, The method comprises: obtaining self-vehicle condition data of a self-vehicle and first target vehicle condition data of all first target vehicles within a preset range around the self-vehicle; obtaining first avoidance target acceleration of each first target vehicle according to the self-vehicle condition data and the first target vehicle condition data through an avoidance acceleration algorithm; the first avoidance target acceleration comprises lateral avoidance acceleration and longitudinal avoidance acceleration; the avoidance acceleration algorithm is a method of calculating avoidance target acceleration in combination with a driving scene; obtaining a key target vehicle from all first target vehicles according to the first avoidance target acceleration; the avoidance acceleration algorithm comprises: selecting a vehicle from the first target vehicles as a second target vehicle; obtaining a driver reaction time according to the self-vehicle condition data and second target vehicle condition data of the second target vehicle; adding the driver reaction time to a preset brake delay time to obtain a collision delay time of the second target vehicle; obtaining a target motion trajectory and a self-vehicle motion trajectory according to the collision delay time; the target motion trajectory is a motion trajectory of the second target vehicle within the corresponding collision delay time; the self-vehicle motion trajectory is a motion trajectory of the self-vehicle within the corresponding collision delay time; obtaining second avoidance target acceleration corresponding to the second target vehicle according to the self-vehicle motion trajectory and the target motion trajectory; the driver reaction time obtained according to the self-vehicle condition data and the second target vehicle condition data of the second target vehicle comprises: obtaining a triggered driving scene level according to the self-vehicle condition data; the driving scene level comprises level one, level two, level three, level four and level five, and each driving scene level is provided with a reaction adjustment time; obtaining a minimum reaction time according to the triggered driving scene level; obtaining the driver reaction time according to the minimum reaction time and current collision sensitivity in the second target vehicle condition data.

2. The method of claim 1, wherein, Each driving scene level comprises at least one driving scene; each driving scene is provided with at least one trigger parameter item and a corresponding trigger parameter threshold value; the triggered driving scene level obtained according to the self-vehicle condition data comprises: extracting a key parameter item and a corresponding key parameter value from the self-vehicle condition data; the key parameter item is a trigger parameter item of all driving scenes; determining the triggered driving scene according to a comparison result of the key parameter value and the trigger parameter threshold value; obtaining the triggered driving scene level according to the triggered driving scene.

3. The method of claim 2, wherein, the minimum reaction time obtained according to the triggered driving scene level comprises: obtaining the minimum reaction time through a reaction time formula; the reaction time formula is: T1 = min((t1+t2),(t3+t4),t5) T1 is the minimum reaction time; t1 is the reaction adjustment time of level one, level one is a reserved adjustment level, and the default reaction adjustment time is 0; t2 is the reaction adjustment time of level two, the driving scene of level two includes frequently operating function buttons, and t2 is 0 if level two is not triggered; t3 is the reaction adjustment time of level three, the driving scene of level three includes high-curvature scenes, high-speed driving scenes on curves, and low-speed acceleration scenes, and t3 is 0 if level three is not triggered; t4 is the reaction adjustment time of level four, the driving scene of level four includes driver acceleration, driver steering, adaptive cruise acceleration and deceleration, city driving, low-speed driving, driver overtaking, and driver braking, and t4 is 0 if level four is not triggered; t5 is the reaction adjustment time of level five, the driving scene of level five includes driver deceleration, and t5 is 0 if level five is not triggered.

4. The method of claim 1, wherein, The collision sensitivity includes low sensitivity, normal sensitivity, high sensitivity, and unused collision warning; each of the collision sensitivities is provided with a corresponding preset reaction time; the driver reaction time obtained according to the minimum reaction time and the current collision sensitivity in the second target vehicle condition data includes: obtaining the collision sensitivity currently set for the ego vehicle and the corresponding preset reaction time as a warning reaction time; adding the minimum reaction time and the warning reaction time to obtain the driver reaction time.

5. The method of claim 1, wherein, The ego vehicle condition data further includes ego vehicle speed, ego vehicle direction, and ego vehicle acceleration; the second target vehicle condition data further includes target vehicle speed, target vehicle direction, and target acceleration; the target motion trajectory and the ego vehicle motion trajectory obtained according to the collision delay time include: obtaining the ego vehicle motion trajectory according to the ego vehicle speed, the ego vehicle direction, the ego vehicle acceleration, and the collision delay time; obtaining the target motion trajectory according to the target vehicle speed, the target vehicle direction, the target acceleration, and the collision delay time.

6. The method of claim 1, wherein, The second avoidance target acceleration corresponding to the second target vehicle is obtained according to the ego vehicle motion trajectory and the target motion trajectory, including: inputting the ego vehicle motion trajectory and the target motion trajectory as input data into a kinematics model to obtain lateral avoidance acceleration and longitudinal avoidance acceleration of the ego vehicle; the lateral avoidance acceleration and the longitudinal avoidance acceleration are taken as the second avoidance target acceleration.

7. The method of claim 1, wherein, The key target vehicle is obtained from all the first target vehicles according to the first avoidance target acceleration, including: performing mean value processing on the lateral avoidance acceleration and the longitudinal avoidance acceleration corresponding to the first avoidance target acceleration to obtain a mean value acceleration of each of the first target vehicles; the first target vehicle with the maximum absolute value of the mean value acceleration is taken as the key target vehicle.

8. A key target selection device of a pre-crash warning function, characterized by, The device includes: an acquisition module configured to acquire ego vehicle condition data of an ego vehicle and first target vehicle condition data of all first target vehicles within a preset range around the ego vehicle; An evasive acceleration module is configured to obtain a first evasive target acceleration of each of the first target vehicles by an evasive acceleration algorithm according to the self-vehicle condition data and the first target vehicle condition data; the first evasive target acceleration comprises a lateral evasive acceleration and a longitudinal evasive acceleration; the evasive acceleration algorithm is a method of calculating an evasive target acceleration in combination with a driving scene; the evasive acceleration algorithm is as follows: selecting one vehicle from the first target vehicles as a second target vehicle; obtaining a driver reaction time according to the self-vehicle condition data and second target vehicle condition data of the second target vehicle; adding the driver reaction time and a preset brake delay time to obtain a collision delay time of the second target vehicle; obtaining a target motion trajectory and a self-vehicle motion trajectory according to the collision delay time; the target motion trajectory is a motion trajectory of the second target vehicle within the corresponding collision delay time; the self-vehicle motion trajectory is a motion trajectory of the self-vehicle within the corresponding collision delay time; obtaining a second evasive target acceleration corresponding to the second target vehicle according to the self-vehicle motion trajectory and the target motion trajectory; wherein obtaining a driver reaction time according to the self-vehicle condition data and the second target vehicle condition data of the second target vehicle comprises: obtaining a triggered driving scene level according to the self-vehicle condition data; the driving scene level comprises level one, level two, level three, level four and level five, each driving scene level is provided with a reaction adjustment time; obtaining a minimum reaction time according to the triggered driving scene level; obtaining the driver reaction time according to the minimum reaction time and a current collision sensitivity in the second target vehicle condition data; A selection module is configured to obtain a key target vehicle from all the first target vehicles according to the first evasive target acceleration.

Images