CN116901949A - Sensitivity self-adaptive adjustment lane keeping auxiliary method and system - Google Patents

Abstract

The application provides a sensitivity self-adaptive adjustment lane keeping auxiliary method and a system, wherein an intelligent driving domain controller identifies the identity of a driver through a face recognition system, analyzes the identity of the current driver, and performs ID archiving on identity characteristic information of the driver after analysis; confirming current driver ID information, and establishing driver driving habit data for the driver ID information; according to the ID information of the driver, the intelligent driving domain controller carries out alarm display on the vehicle through an early warning prompt system and carries out intervention control on the vehicle through a chassis system; the lane keeping auxiliary method comprises two-stage sensitivity self-adaptive adjustment of a first loop body and a second loop body. According to the application, the vehicle steady-state data obtained in the driving process of the driver are combined to carry out neural network deep learning and data recharging, different alarm intervention control thresholds are output, more comfortable driving experience is provided for the driver, the sensitivity self-adaptive adjustment of the lane auxiliary system is achieved, and the technological sense of the whole vehicle is improved.

Description

Sensitivity self-adaptive adjustment lane keeping auxiliary method and system

Technical Field

The application relates to the field of vehicle control, in particular to a sensitivity self-adaptive adjustment lane keeping auxiliary method and system.

Background

Based on the statistics of traffic department, about 50% of traffic accidents are caused by vehicles deviating from the lane of normal driving during driving. Lane departure is also one of the main causes of vehicle rollover accidents. According to incomplete statistics, 23% of motorists have signs of drowsiness at least on the steering wheel within one month; 66% of commercial vehicle drivers have the problem of getting trapped during driving; and one driver among every four drivers experiences an casualty accident caused by a deviation of the vehicle. The main reasons are that the driver is not concentrated in attention, tired driving, distracted mind and the like in the process of driving the vehicle, and the safety of the driver in driving the vehicle is seriously affected.

With the development of the intelligence of automobiles, an auxiliary driving function applied to solving lane departure is generated, wherein a lane departure early warning system and a lane departure intervention system are main auxiliary functions for preventing lane departure, the functions prevent the occurrence of lane departure accidents to a certain extent, and the driving safety of drivers is improved. Lane keeping system related functions are becoming one of the factors considered when people purchase vehicles.

At present, vehicles with lane keeping auxiliary systems are arranged on the market, and the sensitivity of the system is adjusted by using three gear switches fixed on a host machine, so that the lane keeping auxiliary systems are given an alarm threshold value, namely, the lane keeping auxiliary systems are given advance alarms at different distances from lane lines. However, because the driving habit and the driving intensity of the driver are different, the requirement of each person on the sensitivity of the lane keeping auxiliary system is also different, the alarm threshold values of different lane keeping auxiliary systems are distinguished only by three high-medium-low sensitivity switches on the vehicle host, so that the driving comfort cannot be brought to the driver, and the system alarms or intervenes to bring discomfort to the driver even in the opposite sense when frequent false alarms or conscious gentle lane changes are carried out on the driver on a high-speed road section, thereby causing distraction to the driver and serious traffic accidents.

Disclosure of Invention

The application aims to provide a sensitivity self-adaptive adjustment lane keeping auxiliary method and system, which are used for solving the technical problems in the background technology.

In order to achieve the above purpose, the specific technical scheme of the sensitivity self-adaptive adjustment lane keeping auxiliary method and system of the application is as follows:

a sensitivity adaptive adjustment lane keeping assist method, comprising: the intelligent driving domain controller identifies the identity of the driver through a face recognition system, analyzes the identity of the current driver, and performs ID archiving on the identity characteristic information of the driver after analysis;

confirming current driver ID information, and establishing driver driving habit data for the driver ID information;

according to the ID information of the driver, the intelligent driving domain controller carries out alarm display on the vehicle through an early warning prompt system and carries out intervention control on the vehicle through a chassis system;

the lane keeping auxiliary method comprises two stages of sensitivity self-adaptive adjustment of a first loop body and a second loop body: the first loop body is used for collecting the ID information of the driver for a single time and obtaining a predicted lane keeping first sensitivity coefficient by combining the parameters of the road sensing system;

the second loop body is used for collecting the ID information of the driver for a plurality of times and obtaining a second estimated sensitivity coefficient of the lane keeping by combining the parameters of the road sensing system.

Further, the judging operation of the first loop body is to identify the ID information of the driver once, and a first threshold curve is formed by combining the driving habit and the road perception system parameters of the driver in a real-time state;

the judgment operation of the second loop body is to identify the ID information of the driver for a plurality of times, and a second threshold curve is formed by combining the driving habit and the road perception system parameters of the driver in a real-time state.

Further, when the intelligent driving control system does not contain the current driver ID information, the first circulating body collects the current driver information;

and according to a first threshold curve output by the first circulating body, carrying out alarm display and intervention control on the vehicle.

Further, when the intelligent driving control system contains the current driver ID information, the second circulating body collects the current driver information;

driver information of the first loop body may be transmitted to the second loop body;

and according to a second threshold curve output by the second circulating body, carrying out alarm display and intervention control on the vehicle.

Further, the driving habit data comprises signals of accelerator opening, brake opening, steering wheel angle speed, yaw rate, longitudinal acceleration and transverse acceleration received by the intelligent driving domain controller from the CAN bus of the whole vehicle.

The application also provides a sensitivity adaptive adjustment lane keeping assist system, comprising:

face recognition system: the method comprises the steps of identifying the identity of a driver to form driver ID information;

road perception system: the method is used for lane line detection and provides information for vehicle alarm display and vehicle intervention control;

chassis system: the input and output system is used for the intelligent driving domain controller and is used for identifying a driver and performing intervention control on a vehicle;

early warning prompt system: the vehicle warning device is used for warning and prompting the vehicle;

an intelligent driving domain controller in a lane keeping assist system includes a first loop body and a second loop body two-stage sensitivity adaptive adjustment.

Further, the first circulating body completes calculation and judgment in the intelligent driving domain controller according to the related parameters of the single face recognition system, the road perception system and the chassis system to form a first threshold curve;

and the second circulating body completes calculation and judgment in the intelligent driving domain controller according to the related parameters of the face recognition system, the road perception system and the chassis system for multiple times to form a second threshold curve.

Further, the vehicle parameters input by the first circulation body and the second circulation body include an accelerator opening, a brake opening, a steering wheel angle rate, a yaw rate, a longitudinal acceleration, a lateral acceleration, and the like.

Further, according to the first threshold curve and the second threshold curve, the early warning prompt system gives an alarm to the vehicle, and the chassis system performs intervention control on the vehicle.

The application also provides a vehicle comprising the sensitivity self-adaptive adjustment lane keeping auxiliary system.

The sensitivity self-adaptive adjustment lane keeping auxiliary method and system have the advantages and beneficial effects that:

in the prior art, the warning threshold value of the vehicle distance lane line is set to be a fixed value through the high, medium and low sensitivity adjusting switches of the lane keeping auxiliary system on the vehicle host, but because different drivers have different driving habits, a lot of drivers can possibly bring unreasonable setting of the warning intervention control threshold value, so that the vehicles frequently give warning interventions, and uncomfortable driving experience is brought to the drivers.

According to the lane keeping assisting method and system with the sensitivity self-adaptive adjustment, different driver ID information is adopted, and the vehicle steady state data obtained in the driving process of the driver are combined to perform neural network deep learning and big data recharging, so that different optimal alarm intervention control thresholds are output for different drivers, more comfortable driving experience is provided for the drivers, the lane assisting system self-adaptive sensitivity adjustment is achieved, and the technological sense of the whole vehicle is improved.

Drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present application, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a lane keeping assist method for adaptive adjustment of sensitivity according to an embodiment of the present application;

fig. 2 is a schematic flow chart of a sensitivity adaptive adjustment lane keeping assist system according to an embodiment of the present application.

Detailed Description

The application is further described in conjunction with specific embodiments of a co-pilot entertainment screen video chat control system and vehicle to assist those skilled in the art in a more complete, accurate and thorough understanding of the present application. The specific embodiments described herein are merely illustrative of the application and are not intended to limit the application.

Example 1

As shown in the flow chart of the sensitivity adaptive lane keeping assist method of fig. 1, the overall overview principle of the present application is as follows:

when the whole vehicle is electrified, the sensitivity self-adaptive switch is turned on the large screen of the host, the intelligent driving domain controller firstly receives the facial feature information of the driver identified by the face recognition controller (DMS), comprehensively analyzes the current driver, carries out ID filing on the corresponding driver feature information through analysis, is convenient for quickly confirming the ID information of the current driver when starting next time, and definitely matches with sensitivity threshold values of different lane keeping auxiliary systems. Secondly, after confirming the current driver ID information again, a series of data of driving habits of the driver are established for the driver with the ID identity.

The intelligent driving domain controller receives signals such as accelerator opening, brake opening, steering wheel angle speed, yaw rate (YawRate), longitudinal acceleration, transverse acceleration and the like from the CAN bus of the whole vehicle at the moment. The intelligent driving domain controller performs memory learning on the data acquired by the bus, and the measurement process can be regarded as a cycle body.

The intelligent driving domain controller memorizes the whole vehicle driving data in the ignition period, and the corresponding deep learning result is memorized in the ID file of the current driver by matching with the ID driver information, so as to provide a calculation basis for the sensitivity self-adaptive adjustment of the follow-up lane keeping auxiliary system.

The position of the loop body for judging operation is in the intelligent driving domain controller, and the technical scheme is that corresponding calculation and judgment are completed in the intelligent driving domain controller (ADCU). The intelligent driving domain controller receives the CAN signal accelerator opening degree, the brake opening degree, the steering wheel corner speed, the yaw angle speed, the longitudinal acceleration and the transverse acceleration of the whole vehicle, deep learning is carried out on the obtained data, the information is collected every time the vehicle is electrified, and the vehicle is circularly memorized and learned, namely a first circulation body. And fitting the best matched curve of the 7 parameter data through a neural network and big data recharging, wherein the numerical value of the curve is the optimal curve of the warning intervention control threshold value of the lane keeping auxiliary system vehicle from the lane line. The threshold curve considers all parameters of the steady state of the whole vehicle, and provides an optimal judgment basis for the sensitivity adjustment of the alarm intervention of the lane keeping auxiliary system. The tamper alert curve output at this time is defined as a first threshold curve.

The intelligent driving domain controller receives driver ID information of a face recognition system (DMS), stores the calculated optimal alarm intervention curve in a first circulation body under a corresponding driver ID address, and in each vehicle ignition period, the intelligent driving domain controller recognizes the driver ID and calculates the current optimal threshold curve through the actions, and can circularly and dynamically calculate the optimal threshold curve of the lane keeping auxiliary system through the actions. And meanwhile, the driving habit data of different driver IDs are combined to obtain an optimized alarm intervention control threshold curve of the lane keeping auxiliary system, namely the second loop body. The tamper control alarm curve output at this time is defined as a second threshold curve.

And (3) controlling an optimal second threshold curve through the warning intervention of the lane keeping auxiliary system calculated in the second circulating body, and combining the current speed of the vehicle to obtain an optimal warning distance DLC1 and an optimal intervention distance DLC2 from the vehicle to the lane line at the current moment. By the action of the system, the sensitivity self-adaptive adjustment of the lane keeping auxiliary system is achieved.

The loop body can adopt two modes of calibration and deep memory learning, and depends on specific projects.

The application aims to identify different drivers through a face recognition system and combine signals of a steering gear, a brake, a clutch and other vehicle-related vehicle body stabilizing systems. The intelligent driving domain controller performs memory and cyclic learning so as to obtain driving habits of a corresponding driver, including information such as braking depth, accelerator depth, steering angle, steering rate and the like.

The sensitivity of the lane keeping auxiliary system is based on the driving condition of each person, an alarm intervention control threshold value which is most suitable for the current driving of the driver is made, false alarm of the system is reduced to a certain extent, driving comfort of the driver is improved, more drivers are willing to accept the functions of the lane keeping auxiliary system, and driving safety is improved.

As shown in the schematic flow chart of the sensitivity adaptive adjustment lane keeping assist system of fig. 2, the sensitivity adaptive adjustment lane keeping assist system includes: the face recognition system comprises an in-cabin infrared camera (DMS); the road sensing system comprises a front-view camera; the chassis system comprises a brake (ESP), a steering gear (EPS) and an acceleration sensor; the early warning prompt system comprises a host machine (IVI), an Instrument (ICU), a seat and a Body Controller (BCM);

wherein, face recognition system: the device comprises an in-cabin infrared camera, a camera module and a control module, wherein the in-cabin infrared camera is used for identifying the identity of a driver to form driver ID information;

road perception system: the system comprises a forward-looking camera, a control module and a control module, wherein the forward-looking camera is used for detecting lane lines and providing information for vehicle alarm display and vehicle intervention control;

chassis system: the system comprises a brake, a steering gear and an acceleration sensor, wherein the brake, the steering gear and the acceleration sensor are used for performing interference control on a vehicle;

early warning prompt system: the automobile body control system comprises a control system host, an automobile body controller, a prompt system instrument and a vehicle seat;

an intelligent driving domain controller in a lane keeping assist system includes a first loop body and a second loop body two-stage sensitivity adaptive adjustment.

The concrete explanation is as follows:

face recognition system (DMS): identifying the current driver characteristics, and matching the corresponding driver ID in the system library, so that the intelligent driving domain controller can be matched with the sensitivity threshold curve of the lane keeping auxiliary system conveniently;

front-view camera (IFC): identifying the lane line edge of the road where the current vehicle is located, and calculating the lane line distance between the left side and the right side of the current vehicle based on the lane line edge identified by the IFC by the intelligent driving domain controller;

brake (ESP): providing vehicle speed, master cylinder pressure and brake deceleration information;

steering gear (EPS): and providing the yaw rate (Yawrate) information of the whole vehicle, the steering wheel angle and the steering wheel angle rate information. The output end is used for preventing the vehicle from deviating from the lane line by controlling the steering wheel to turn;

acceleration sensor: providing vehicle lateral acceleration and longitudinal acceleration information;

host (IVI): a lane keeping auxiliary system switch setting and a system sensitivity self-adaptive adjusting switch;

an Instrument (ICU) for displaying an audible and visual alarm display of the lane keeping aid system;

steering wheel: as one of the modes of vibration alarm display;

a Body Controller (BCM) for lighting the double flashing lights to prompt the safe running of the rear vehicle;

and the intelligent driving domain controller (ADCU) is used for receiving the related data signals of the whole vehicle on the stability of the vehicle body, calculating the distance from the current position of the vehicle to the lane line where the vehicle is positioned, and alarming and/or intervening when the vehicle deviates from the lane line by a certain threshold value based on the transverse distance. Meanwhile, the local domain controller outputs a corresponding alarm intervention control threshold curve by combining the current driver ID through deep memory learning.

In the technical scheme of the patent, a host (IVI) is a large screen in the middle of a vehicle, and the large screen of the general host (IVI) is provided with switch options, entertainment display and the like for displaying various functions. Including early warning prompt facility in this patent among the early warning prompt system, the instrument board can carry out sound warning, voice prompt, vision and reminds, and steering wheel, seat can vibrate, realize the sense of touch and remind.

The chassis system realizes the intervention control of the vehicle, and the chassis system performs the intervention control of the torque of the vehicle through a steering wheel, a brake, a steering gear, an acceleration sensor and the like according to the information provided by the circulating body. A vehicle intervention control method in a chassis system comprises the steps of providing vehicle speed information, braking information and steering information. For example: the chassis system controls the torque of the steering wheel, when the distance between the edge of the vehicle and the lane line is smaller than the threshold value of the threshold curve, the vehicle can give an alarm at the moment, and meanwhile, the steering wheel is controlled to perform torque intervention control, so that the effect that the vehicle does not deviate from the lane is achieved.

In this patent technical scheme, chassis system also is as the input parameter of circulation body intelligent driving area controller (ADCU), for example, driver's driving habit, through the parameter such as chassis system's steering wheel, stopper, steering gear, acceleration sensor, input to the circulation body, define driver's identity, form partial driver ID information. And then outputting a corresponding alarm intervention control threshold curve, and performing alarm display and/or intervention control on the vehicle.

The road sensing system recognizes the front road state through the front-view camera, and the road sensing system comprises information such as lane line type, lane width, lane confidence and the like, and lane line information is needed. And judging whether the vehicle deviates from the lane or not according to the lane line information.

In the technical scheme of the patent, the sensitivity coefficient can be endowed with a coefficient of the threshold curve, namely the sensitivity coefficient multiplied by the threshold curve, and can be used for distinguishing and adjusting the time of early warning and late warning of the vehicle. The alarm display time can also be controlled directly according to the driving habit of the driver.

The first loop body is characterized in that when the vehicle is electrified, the in-cabin camera recognizes the ID of the driver, and when the driver drives the vehicle, parameters representing the driving habit of the driver, such as the accelerator speed, the brake speed, the steering wheel rotation angle speed and the like of the driver are recorded in real time. And combining the lane related parameters such as the curvature, the lane line type, the lane width and the like of the current road; obtaining a first sensitivity coefficient of the estimated lane keeping system, namely, the warning distance from the boundary of the vehicle to the lane line;

the second loop body refers to data of the first loop body with the same driver ID, and the second sensitivity coefficient of the lane keeping system which can more reflect the driving habit of the driver is obtained through long-time deep learning. The second circulation is relative to the first circulation body, records multiple data of the first circulation body, and continuously optimizes in the subsequent driving process, is more accurate relative to the first circulation body, and can synchronously optimize according to the improvement of the driving proficiency of a driver.

The first loop body is data that is integrated at the time of single driving and that is valid only when the driver ID is recognized for the first time. When the same driver ID is recognized a plurality of times, the data of the second loop body is used. The value is determined by the distance between the left and right edges of the vehicle and the lane line, and ranges from-1 m to 1m, namely the value is negative when the edges of the vehicle exceed the lane line.

The alarm level can embody an alarm level, and the multiple layers of alarm levels can cause the driver to be bored, and only the driver can be reminded to use. The intervention control torque is that when the vehicle deviates from the lane, the vehicle provides a gradient torque to enable the vehicle to return to the right position, and the vehicle is prevented from deviating from the lane. When the included angle between the vehicle and the lane line is in a controllable range, the intervention control torque decays to 0;

the first threshold curve is the alarm distance obtained by the first cycle body at a time, and the second threshold curve is the alarm distance obtained by the second cycle body. The intervention control torques corresponding to the first threshold curve and the second threshold curve can be the same, and are all torques with gradient descent; the intervention control torques corresponding to the first threshold curve and the second threshold curve can also be different, and the specific situation needs to be designed according to the actual project.

According to the scheme, different driver ID information is identified through the face recognition system, signals related to vehicle body stability such as accelerator opening, brake opening, steering wheel rotation angle speed, yaw rate, longitudinal acceleration and transverse acceleration are subjected to neural network deep learning and big data recharging when each ID driver drives a vehicle, a first threshold curve of alarm intervention of a lane keeping auxiliary system of different driver IDs is obtained, the threshold curve is circularly optimized in different ignition periods, circular learning calculation is performed aiming at the same ID information, and minimum-precision threshold setting is achieved to the greatest extent. And outputting an optimized second threshold curve in combination with the current vehicle speed, so as to obtain the optimal warning distance DLC1 and the optimal intervention distance DLC2 of the lane keeping auxiliary system.

The personalized ID database of the driver is created by the creatively combined face recognition system, the optimal alarm distance threshold value can be output according to the driving habits of different drivers, and the comfort of different drivers when using the lane keeping auxiliary system is improved.

It should be understood that, although the steps in the flowcharts of the embodiments of the present application are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in various embodiments may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor do the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of the sub-steps or stages of other steps or other steps.

It will be understood that the application has been described in terms of several embodiments, and that various changes and equivalents may be made to these features and embodiments by those skilled in the art without departing from the spirit and scope of the application. In addition, many modifications may be made to adapt a particular situation or material to the teachings of the application without departing from the essential scope thereof. Therefore, it is intended that the application not be limited to the particular embodiment disclosed, but that the application will include all embodiments falling within the scope of the appended claims.

The foregoing description of the preferred embodiments of the application is not intended to be limiting, but rather is intended to cover all modifications, equivalents, and alternatives falling within the spirit and principles of the application.

Claims (10)

1. A sensitivity adaptive adjustment lane keeping assist method, comprising:

the intelligent driving domain controller identifies the identity of the driver through a face recognition system, analyzes the identity of the current driver, and performs ID archiving on the identity characteristic information of the driver after analysis;

confirming current driver ID information, and establishing driver driving habit data for the driver ID information;

according to the ID information of the driver, the intelligent driving domain controller carries out alarm display on the vehicle through an early warning prompt system and carries out intervention control on the vehicle through a chassis system;

the lane keeping auxiliary method comprises two stages of sensitivity self-adaptive adjustment of a first loop body and a second loop body:

wherein,,

the first loop body is used for collecting the ID information of the driver for a single time and obtaining a predicted lane keeping first sensitivity coefficient by combining the parameters of the road sensing system;

the second loop body is used for collecting the ID information of the driver for a plurality of times and obtaining a second estimated sensitivity coefficient of the lane keeping by combining the parameters of the road sensing system.

2. The sensitivity adaptive adjusting lane keeping assist method according to claim 1, wherein,

the judgment operation of the first loop body is to identify the ID information of the driver once and form a first threshold curve by combining the driving habit and the road perception system parameters of the driver in a real-time state;

the judgment operation of the second loop body is to identify the ID information of the driver for a plurality of times, and a second threshold curve is formed by combining the driving habit and the road perception system parameters of the driver in a real-time state.

3. The sensitivity adaptive adjusting lane keeping assist method according to claim 2, wherein,

when the intelligent driving control system does not contain the current driver ID information, the first circulating body collects the current driver information;

and according to a first threshold curve output by the first circulating body, carrying out alarm display and intervention control on the vehicle.

4. A sensitivity adaptive adjusting lane keeping assist method as claimed in claim 3, wherein,

when the intelligent driving control system contains the current driver ID information, the second circulating body collects the current driver information;

driver information of the first loop body may be transmitted to the second loop body;

and according to a second threshold curve output by the second circulating body, carrying out alarm display and intervention control on the vehicle.

5. The sensitivity adaptive adjusting lane keeping assist method of claim 4 wherein the driving habit data comprises an intelligent driving domain controller receiving accelerator opening, brake opening, steering wheel angle rate, yaw rate, longitudinal acceleration, lateral acceleration signals from a vehicle CAN bus.

6. A sensitivity adaptive adjustment lane keeping assist system, comprising:

face recognition system: the method comprises the steps of identifying the identity of a driver to form driver ID information;

road perception system: the method is used for lane line detection and provides information for vehicle alarm display and vehicle intervention control;

chassis system: the input and output system is used for the intelligent driving domain controller and is used for identifying a driver and performing intervention control on a vehicle;

early warning prompt system: the vehicle warning device is used for warning and prompting the vehicle;

an intelligent driving domain controller in a lane keeping assist system includes a first loop body and a second loop body two-stage sensitivity adaptive adjustment.

7. The sensitivity adaptive adjusting lane keeping assist system according to claim 6, wherein,

the first circulating body completes calculation and judgment in the intelligent driving domain controller according to the related parameters of the single face recognition system, the road perception system and the chassis system to form a first threshold curve;

and the second circulating body completes calculation and judgment in the intelligent driving domain controller according to the related parameters of the face recognition system, the road perception system and the chassis system for multiple times to form a second threshold curve.

8. The sensitivity adaptive roadway lane keeping assist system of claim 7, wherein the vehicle parameters input by the first and second endless bodies comprise accelerator opening, brake opening, steering wheel angle rate, yaw rate, longitudinal acceleration, lateral acceleration.

9. The adaptive-sensitivity lane-keeping assist system of claim 8 wherein the warning alert system alerts the vehicle and the chassis system provides an intervention control to the vehicle based on the first threshold curve and the second threshold curve.

10. A vehicle comprising a sensitivity adaptive roadway maintenance assistance system according to any one of claims 6 to 9.