CN-121989886-A - Scene self-adaptive parameter adjusting method and device and vehicle

CN121989886ACN 121989886 ACN121989886 ACN 121989886ACN-121989886-A

Abstract

The application provides a scene self-adaptive parameter adjusting method, a device and a vehicle, which are applied to an automatic emergency braking system of the vehicle. And based on the driving state information and the environment information, identifying a driving scene where the vehicle is currently located by utilizing a pre-trained scene classification model and evaluating the scene credibility of the driving scene. And determining a scene correction coefficient according to the driving scene, and determining a credibility correction coefficient according to the scene credibility. And dynamically adjusting the basic braking parameters of the automatic emergency braking system based on the scene correction coefficient and the credibility correction coefficient to obtain target braking parameters. According to the application, the basic braking parameters are dynamically adjusted according to different driving scenes, so that the automatic emergency braking system has better adaptability, safety and reliability in various driving environments.

Inventors

Xiang Maiyang
XU HONGXING
CHEN ANJUN

Assignees

北京瞰瞰智能科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260312

Claims (10)

1. A scene adaptive parameter adjustment method applied to an automatic emergency braking system of a vehicle, characterized in that the scene adaptive parameter adjustment method comprises: Acquiring running state information and environment information of the vehicle; Based on the driving state information and the environment information, identifying a driving scene where the vehicle is currently located by utilizing a pre-trained scene classification model and evaluating the scene credibility of the driving scene; Determining a scene correction coefficient according to the driving scene, and determining a credibility correction coefficient according to the scene credibility; And dynamically adjusting the basic braking parameters of the automatic emergency braking system based on the scene correction coefficient and the credibility correction coefficient to obtain target braking parameters.
2. The scene adaptive parameter adjustment method according to claim 1, wherein acquiring the running state information and the environment information of the vehicle includes: Collecting three-axis acceleration, three-axis angular velocity, vehicle speed, global positioning information and positioning signal quality parameters under a vehicle carrier coordinate system as the driving state information, wherein the positioning signal quality parameters comprise the number of visible satellites and a horizontal precision factor; Collecting static characteristic information and real-time traffic flow information of a road as the environment information; Performing time synchronization on the running state information and the environment information to control a time synchronization error within a preset error range; Converting the three-axis acceleration and the three-axis angular velocity from the vehicle-carrier coordinate system to the same navigation coordinate system; Filtering the converted triaxial acceleration and triaxial angular velocity to calculate the standard deviation of the longitudinal acceleration change rate and the vertical acceleration; Dividing the quality level of the positioning signal according to the number of the visible satellites and the horizontal precision factor; And calculating a standard deviation of the vehicle speed in a preset duration window based on the vehicle speed so as to quantify the running stability of the vehicle.
3. The scene adaptive parameter adjustment method according to claim 2, characterized in that, based on the driving state information and the environmental information, identifying a driving scene in which the vehicle is currently located and evaluating a scene reliability of the driving scene, comprises: Fusing the vehicle speed, the longitudinal acceleration change rate, the vertical acceleration standard deviation, the positioning signal quality grade, the vehicle speed standard deviation, the road static characteristic information and the real-time traffic flow information to obtain a multidimensional characteristic vector; and inputting the multidimensional feature vector into the pre-trained scene classification model to identify the driving scene where the vehicle is currently located, and outputting the scene credibility corresponding to the identified driving scene.
4. The method for adjusting scene adaptive parameters according to claim 3, wherein the driving scene at least comprises one of a expressway scene, an urban road scene, a tunnel scene and a complex road condition scene, the pre-trained scene classification model comprises a plurality of preset recognition rules, each preset recognition rule is at least constructed based on the part of the vehicle speed, the longitudinal acceleration change rate, the vertical acceleration standard deviation, the positioning signal quality level, the vehicle speed standard deviation, the road static characteristic information and the real-time traffic flow information, and the preset recognition rules corresponding to different driving scenes are different.
5. The scene adaptive parameter adjustment method according to claim 4, characterized in that determining a scene correction coefficient from the driving scene and determining a reliability correction coefficient from the scene reliability, comprises: Determining a scene correction coefficient corresponding to the driving scene from a first preset mapping relation according to the type of the driving scene, wherein the first preset mapping relation is used for defining the corresponding relation between different driving scenes and the scene correction coefficient; And determining a reliability correction coefficient corresponding to the numerical interval from a second preset mapping relation according to the numerical interval in which the scene reliability is positioned, wherein the second preset mapping relation is used for defining the corresponding relation between the numerical interval of different scene reliability and the reliability correction coefficient.
6. The method for adjusting scene adaptive parameters according to claim 5, wherein the scene correction coefficients are used for characterizing adjustment directions and magnitudes of the foundation brake parameters for different driving scenes, and the first preset mapping relationship is: The expressway scene corresponds to a first scene correction coefficient, and the first scene correction coefficient is larger than a scene reference value so as to improve the response sensitivity of the automatic emergency braking system in the expressway scene; The urban road scene corresponds to a second scene correction coefficient which is smaller than the scene reference value so as to reduce the response sensitivity of the automatic emergency braking system in the urban road scene; the tunnel scene and the complex road condition scene correspond to a third scene correction coefficient, and the absolute value of the difference value between the third scene correction coefficient and the scene reference value is within a preset threshold range so as to maintain the basic braking parameters under the tunnel scene and the complex road condition scene.
7. The method for adjusting scene adaptive parameters according to claim 5, wherein the reliability correction coefficients are used for characterizing adjustment weights of different scene reliabilities on the foundation brake parameters, the second preset mapping relationship is that a value range of the scene reliability is divided into a plurality of continuous value intervals, each value interval corresponds to a preset reliability correction coefficient, and the plurality of continuous value intervals at least comprises: The method comprises the steps that a first numerical value interval corresponds to a scene reliability greater than or equal to a first preset threshold value, and a reliability correction coefficient corresponding to the first numerical value interval is a first weight value; The corresponding scene reliability is greater than or equal to a second preset threshold value and smaller than the first preset threshold value, the reliability correction coefficient corresponding to the second numerical interval is a second weight value, and the second weight value is smaller than the first weight value; And the third numerical value interval corresponds to a scene reliability smaller than the second preset threshold, and the reliability correction coefficient corresponding to the third numerical value interval is a third weight value which is smaller than the second weight value.
8. The scene adaptive parameter adjustment method according to claim 1, wherein dynamically adjusting a foundation braking parameter of the automatic emergency braking system based on the scene correction coefficient and the reliability correction coefficient to obtain a target braking parameter, comprises: Acquiring basic braking parameters of the automatic emergency braking system, wherein the basic braking parameters comprise an early warning collision time threshold value and a safety distance coefficient; Determining the adjusting direction of the early warning collision time threshold according to the magnitude relation between the scene correction coefficient and a preset scene reference value; Determining the adjusting amplitude of the early warning collision time threshold according to the adjusting direction, the reliability correction coefficient and a preset adjusting step length; based on the scene correction coefficient and the adjustment amplitude, dynamically correcting the early warning collision time threshold value to obtain a corrected early warning collision time threshold value, wherein the corrected early warning collision time threshold value is expressed as: , wherein, A pre-warning collision time threshold value is indicated, The scene modifier coefficients are represented as such, Indicating that the preset adjustment step size is to be performed, The reliability correction coefficient is represented by a coefficient of reliability, Representing a sign taking function; dynamically correcting the safety distance coefficient according to the scene correction coefficient to obtain a corrected safety distance coefficient, wherein the corrected safety distance coefficient is expressed as: , wherein, Representing a safety distance coefficient; And taking the corrected early warning collision time threshold and the corrected safety distance coefficient as the target braking parameters.
9. A scene adaptive parameter adjusting device applied to an automatic emergency braking system of a vehicle, characterized in that the scene adaptive parameter adjusting device comprises: The information acquisition module is used for acquiring the running state information and the environment information of the vehicle; The scene recognition module is used for recognizing a driving scene where the vehicle is currently located by utilizing a pre-trained scene classification model based on the driving state information and the environment information and evaluating the scene credibility of the driving scene; the coefficient determining module is used for determining a scene correction coefficient according to the driving scene and determining a credibility correction coefficient according to the scene credibility; And the parameter adjusting module is used for dynamically adjusting the basic braking parameters of the automatic emergency braking system based on the scene correction coefficient and the credibility correction coefficient to obtain target braking parameters.
10. A vehicle, characterized in that the vehicle comprises: Automatic emergency braking system, and The master control equipment, communication connect in automatic emergency braking system, the master control equipment includes: a memory for storing a computer program, and A processor for executing the computer program to implement the scene adaptive parameter tuning method according to any one of claims 1-8.

Description

Scene self-adaptive parameter adjusting method and device and vehicle Technical Field The application relates to the technical field of active safety control of vehicles, in particular to a scene self-adaptive parameter adjusting method and device and a vehicle. Background An automatic emergency braking system (Automatic Emergency Braking Systems, AEBS) is a central component of active safety of a vehicle, which functions to automatically apply braking when a potential collision risk is detected, to mitigate or avoid an accident. At present, most AEBS adopts fixed early warning and braking triggering threshold values, or simply adjusts parameters according to the speed of the vehicle, and cannot adapt to complex and changeable actual driving scenes. For example, when the system is driven on a highway, the system needs higher response sensitivity to cope with high-speed risks, and in special environments such as urban congested road sections or tunnels, too high response sensitivity is easy to cause frequent false triggering, so that the driving comfort and the system reliability are affected. To solve the above problems, the prior art attempts to improve environmental awareness through multi-sensor fusion. For example, part of schemes are introduced into a laser radar to improve the perception precision, but the hardware cost is obviously increased and is difficult to be widely applied to middle-low end vehicle types, and other schemes adopt the combination of the millimeter wave radar, a camera and other perception sensors to trigger braking through a preset collision time threshold, but decision parameters are still relatively fixed, and the fine recognition capability on driving scenes is lacked. In addition, the reliability of the driving scene recognition by the existing system lacks quantitative evaluation, and when a sensor signal is interfered (such as GPS unlocking in a tunnel and urban canyon signal shielding), the system is difficult to accurately judge the current environment, so that the problem of false triggering and missed triggering coexist. Disclosure of Invention The application provides a scene self-adaptive parameter adjusting method, a scene self-adaptive parameter adjusting device and a vehicle, which can enable an automatic emergency braking system to have better adaptability, safety and reliability under different running environments. The embodiment of the application provides a scene self-adaptive parameter adjusting method which is applied to an automatic emergency braking system of a vehicle, and the scene self-adaptive parameter adjusting method comprises the steps of obtaining running state information and environment information of the vehicle, identifying a driving scene where the vehicle is currently located and evaluating scene credibility of the driving scene by utilizing a pre-trained scene classification model based on the running state information and the environment information, determining a scene correction coefficient according to the driving scene and determining a credibility correction coefficient according to the scene credibility, and dynamically adjusting basic braking parameters of the automatic emergency braking system based on the scene correction coefficient and the credibility correction coefficient to obtain target braking parameters. The embodiment of the application provides a scene self-adaptive parameter adjusting device which is applied to an automatic emergency braking system of a vehicle, and comprises an information acquisition module, a scene identification module, a coefficient determination module and a parameter adjusting module, wherein the information acquisition module is used for acquiring running state information and environment information of the vehicle, the scene identification module is used for identifying a driving scene where the vehicle is currently located and evaluating scene reliability of the driving scene by utilizing a pre-trained scene classification model based on the running state information and the environment information, the coefficient determination module is used for determining a scene correction coefficient according to the driving scene and determining the reliability correction coefficient according to the scene reliability, and the parameter adjusting module is used for dynamically adjusting basic braking parameters of the automatic emergency braking system to obtain target braking parameters based on the scene correction coefficient and the reliability correction coefficient. In a third aspect, an embodiment of the present application provides a vehicle, where the vehicle includes an automatic emergency braking system and a master control device, where the master control device is communicatively connected to the automatic emergency braking system, and the master control device includes a memory and a processor, where the memory is configured to store a computer program, and the processor is configured to e