CN114312768B - Unmanned vehicle automatic braking system based on linear dynamic prediction model - Google Patents
Unmanned vehicle automatic braking system based on linear dynamic prediction model Download PDFInfo
- Publication number
- CN114312768B CN114312768B CN202210034059.7A CN202210034059A CN114312768B CN 114312768 B CN114312768 B CN 114312768B CN 202210034059 A CN202210034059 A CN 202210034059A CN 114312768 B CN114312768 B CN 114312768B
- Authority
- CN
- China
- Prior art keywords
- brake
- speed
- braking
- data
- prediction model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 230000009471 action Effects 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 16
- 230000006837 decompression Effects 0.000 claims abstract description 11
- 238000012417 linear regression Methods 0.000 claims description 39
- 238000000034 method Methods 0.000 claims description 34
- 238000004364 calculation method Methods 0.000 claims description 21
- 238000010276 construction Methods 0.000 claims description 10
- 230000003044 adaptive effect Effects 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 9
- 230000001419 dependent effect Effects 0.000 claims description 9
- 238000005070 sampling Methods 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 4
- 238000011076 safety test Methods 0.000 claims description 3
- 230000003068 static effect Effects 0.000 claims description 3
- 238000013480 data collection Methods 0.000 claims 1
- 230000000875 corresponding effect Effects 0.000 description 23
- 238000012544 monitoring process Methods 0.000 description 4
- 230000002596 correlated effect Effects 0.000 description 2
- 230000004907 flux Effects 0.000 description 2
- 238000004804 winding Methods 0.000 description 2
- 206010013911 Dysgeusia Diseases 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000008034 disappearance Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Landscapes
- Regulating Braking Force (AREA)
Abstract
An unmanned vehicle automatic braking system based on a linear dynamic prediction model relates to the field of intelligent driving of vehicles. The system comprises an information acquisition module, a data processing module and an execution module. The information acquisition module comprises a millimeter wave radar, a vision sensor and a GPS module, and acquires the current speed, the relative speed with the obstacle and the relative distance; the data processing module constructs a brake linear prediction model through collision time TTC, vehicle speed, relative distance and an executing end brake voltage ratio to obtain brake time, brake speed and decompression speed, and defines the brake output time within a threshold of the brake time TTC to obtain dynamic brake duration; the brake execution module executes a brake command to control the action time, the action speed, the action duration and the action direction of the brake motor, acts on the hydraulic cylinder to realize braking, and the pressure sensor feeds back the pressure of the hydraulic cylinder to the brake ECU in real time to realize locked-rotor protection of the motor, so that the brake is more intelligent, comfortable and stable.
Description
Technical Field
The invention relates to the field of intelligent driving of automobiles, in particular to an automatic braking system of an unmanned vehicle based on a linear dynamic prediction model.
Background
With the unmanned technology of vehicles becoming an important development trend of intelligent travel in the future, the safety and comfort of vehicle driving become important problems in the field of automobile driving. In particular, the advantages and disadvantages of the automatic braking system directly affect the safety and riding comfort of the vehicle. The automatic braking system of the unmanned vehicle needs to make real-time accurate information acquisition and calculation on the speed, the relative distance and the collision time TTC of the unmanned vehicle, and accordingly makes corresponding dynamic automatic braking strategies according to the structure, the braking performance and different application scenes of the unmanned vehicle, so that the intelligent braking is achieved, the safety of the unmanned vehicle is improved, and the riding comfort is improved. The prior art scheme generally uses the monitoring of the vehicle distance and the limitation of the collision time TTC as the basis of braking and decelerating. Physical and chemical data limiting is carried out on the braking time and the braking force. There is no dynamic braking strategy. Particularly, braking in a high-speed running state should be performed, and insufficient braking may directly affect running safety and comfort. The automatic braking system should execute different braking time, braking speed and braking force according to different vehicle distance, relative speed and collision time TTC, and a dynamic automatic braking prediction model is constructed, so that the braking system is more intelligent, and the safety and comfort of the vehicle are improved.
Disclosure of Invention
The invention aims to provide an automatic braking system of an unmanned vehicle based on a linear dynamic prediction model.
The unmanned vehicle automatic braking system based on the linear dynamic prediction model comprises an information acquisition module, a data processing module and an execution module. The information acquisition module comprises a millimeter wave radar, a vision sensor and a GPS module, and acquires the current speed of the vehicle, the relative speed between the information acquisition module and an obstacle and the relative distance between the information acquisition module and the obstacle; the data processing module utilizes the current speed, the relative speed and the relative distance acquired by the signal acquisition module to firstly calculate collision time TTC, then constructs a brake linear prediction model by matching with the acquired relative speed, the relative distance and output brake voltage ratio data of an execution end, obtains the brake time, the brake speed and the decompression speed (release brake speed) of the brake through the constructed brake linear prediction model, defines that the collision time threshold executed by the current brake time is a brake interval, and returns the brake (the motor rotates reversely) when the collision time exceeds the threshold, so that dynamic brake duration is obtained, and then outputs a brake signal to the brake execution module; the brake executing module comprises a brake ECU, a brake motor and a pressure sensor, wherein the brake ECU of the executing module receives brake executing data of the data processing module in real time and performs forward and reverse rotation control on the brake motor, and the control actions comprise action time, action speed and action duration, namely action direction, namely braking time, braking speed, decompression speed (brake return speed) and braking duration of the system, the motor acts on the hydraulic cylinder to realize braking, the pressure sensor feeds back the pressure of the hydraulic cylinder to the brake ECU in real time, and when the pressure exceeds a preset pressure value, the braking is stopped, so that the locked-rotor protection of the motor is realized.
The unmanned vehicle automatic braking system based on the linear dynamic prediction model comprises the following steps:
the method comprises the steps that an information acquisition module millimeter wave radar and a vision sensor are assembled in front of a vehicle, a GPS module is assembled in the vehicle, real-time monitoring is carried out on obstacles in front of the vehicle, and vehicle speed, relative speed and relative distance information between the GPS module and the obstacles are acquired;
The data processing module calculates the speed of the vehicle, the relative speed and the relative distance information of the obstacle, which are obtained by the information acquisition module, and firstly calculates the real-time collision time TTC of the obstacle based on the relative speed and the relative distance between the vehicle and the obstacle in front, namely the collision time TTC=relative distance/relative speed, and the data is accurate to 0.01S;
The data processing module respectively builds three different linear regression prediction models according to the calculated collision time TTC, the acquired relative speed and relative distance and output brake voltage ratio data of the execution end, wherein a linear prediction model 1 is built according to the vehicle speed and the collision time TTC to determine brake time, a linear prediction model 2 is built according to the relative speed and the brake voltage ratio to determine brake speed, a linear prediction model 3 is built according to the relative speed and the brake voltage ratio to determine decompression speed (brake return speed), meanwhile, the collision time threshold executed by the current brake time is defined to be a brake interval, and the brake return (motor reversal) is performed when the threshold is exceeded, so that the dynamic brake duration is obtained;
The linear prediction model 1 construction method comprises the following steps: defining a certain collision time TTC threshold value as a braking signal, namely a braking time, when the TTC threshold value is in a braking state, namely a braking time length, different vehicle speeds correspond to different braking time, the higher the vehicle speed is, the larger the variable is, the higher the danger coefficient of the running vehicle is, such as sudden braking or front emergency accident of the front vehicle in a high-speed state, and the longer the braking time required by the user is, so that a dynamic linear regression prediction model is constructed according to the vehicle speed and the pre-collision time TTC, and the braking time is dynamic;
The data acquisition method required by the linear prediction model 1 comprises the following steps: when the vehicle moves forwards longitudinally, the collision time is calculated according to the collision time calculation formula It can be known that the relative speed is inversely proportional to the collision time, and the greater the relative speed, the shorter the collision time, i.e. the higher the collision risk; let the speed be V1 and the obstacle speed of the front vehicle be V2, the longitudinal relative speed v=v 1-v2, and the maximum value of the relative speed V be V 2 =0 and the relative speed=v 1, so that the most practical and effective data acquisition method is to acquire the braking time corresponding to the manual pedal stepping when the driving vehicle approaches to the stationary obstacle at different speeds, namely the pre-collision time TTC, and the following table data are obtained according to the data:
| Vehicle speed (X) | X1、X2..........Xn |
| Time to pre-crash TTC (Y) | Y1、Y2..........Yn |
According to the data, the speed and TTC are independent variable x and dependent variable y respectively, so that a TTC value corresponding to the speed change is needed to be obtained as the braking time of the system, a linear regression prediction model is needed to be constructed, a and b are undetermined parameters, b is a regression coefficient, n is the number of data groups,Average numbers of x and y are calculated by a regression coefficient formula:
we know that the calculated results of parameters b and a are
Therefore, the specific calculation formula of the linear regression equation is
Wherein the method comprises the steps ofAverage number of x, y, respectively, i.e./>A. b is a parameter to be solved, n is the number of data sampling groups, the solved parameters a and b are substituted into a linear regression equation to obtain a linear regression equation required by us, namely, a linear prediction model 1 of us, and corresponding data can be acquired by combining the structure, the braking performance and different application scenes of a vehicle in actual application, so that an adaptive linear prediction model is constructed;
The construction method of the linear prediction model 2 and the required data acquisition method are as follows: when the vehicle moves forwards longitudinally, the collision time is calculated according to a pre-collision time calculation formula It can be seen that the relative speed is inversely proportional to the collision time, the greater the relative speed, the shorter the collision time, i.e. the higher the collision risk, so that the braking speed is faster, and the relation between the motor speed and the voltage is known according to the relation u=c eΦn+IaRa+2△Us (where n is the speed, U is the motor end voltage, Δus is the brush voltage drop, ia is the armature current, ra is the motor armature winding resistance, ce is the motor constant, Φ is the motor air gap flux) of the current, the speed and the voltage of the dc motor The method is characterized in that the data acquisition method is to acquire the working voltage required by the manual pedal speed of a driving vehicle when the driving vehicle approaches a static obstacle at different speeds and the manual pedal speed corresponds to the braking motor reaching the speed, the voltage data is determined according to the percentage value of the braking voltage, and in addition, the braking voltage ratio is 100% at most as long as the relative speed is greater than zero, so that the linear regression model 2 is required to be limited with a reasonable interval, namely X1 is less than or equal to X is less than or equal to Xn, and the following data are acquired according to the data:
| relative velocity (X) | X1、X2..........Xn |
| Brake voltage ratio (Y) | Y1、Y2..........Yn |
According to the data, the relative speed and the braking voltage ratio are independent variable x and dependent variable y respectively, a braking voltage ratio corresponding to the relative speed change in the interval x 1≤x≤xn is obtained, so that the corresponding system braking speed is obtained, according to the data, a linear prediction model 2 is constructed, a linear regression equation is set as y=bx+a, wherein a and b are undetermined parameters, b is a regression coefficient, n is a data set number, and the regression coefficient is calculated according to the formula:
we know that the calculated results of parameters b and a are
Therefore, the specific calculation formula of the linear regression equation is
Wherein the method comprises the steps ofAverage number of x, y, respectively, i.e./>A. b is a parameter to be solved, n is the number of data sampling groups, the solved parameters a and b are substituted into a linear regression equation to obtain a linear regression equation required by us, namely a linear prediction model 2 of us, in addition, y 1 and y n are respectively corresponding to x < delta 1,x>xn, corresponding data can be acquired by combining the structure, the braking performance and different application scenes of a vehicle in actual application, and thus an adaptive data prediction model is constructed;
The construction method of the linear prediction model 3 and the required data acquisition method are as follows: when the vehicle moves forwards longitudinally, pre-collision is performed according to a pre-collision time calculation formula It can be known that the relative speed is inversely proportional to the collision time, the greater the relative speed, the shorter the collision time, i.e. the higher the collision risk, and the more according to the object kinetic energy formula/>The kinetic energy of the vehicle is positively correlated with the speed, the larger the speed is, the larger the kinetic energy is, the larger the inertia of the vehicle is, the slower the speed of the disappearance of the braking force opposite to the kinetic energy is, the higher the braking efficiency of the vehicle is, the more stable the vehicle is, in addition, the direct current motor rotating speed is verified to be in direct proportion to the voltage from the construction of the linear prediction model 2, according to the direct current motor rotating speed and voltage, the linear prediction model 3 is constructed according to the relative speed of the vehicle and the obstacle and the braking voltage ratio, and the data acquisition method comprises the following steps: selecting a safety test road section, driving forwards and backwards, and acquiring the working voltage required by the brake motor when the return speed of the pedal of the rear vehicle is braked at different relative speeds to reach the speed, so as to obtain the following table data:
| Relative velocity (X) | X1、X2..........Xn |
| Brake voltage ratio (Y) | Y1、Y2..........Yn |
According to the data, the relative speed and the braking voltage ratio are independent variable x and dependent variable y respectively, a braking voltage ratio corresponding to the relative speed change in the interval x 1≤x≤xn is obtained, so that the corresponding system braking speed is obtained, according to the data, a linear prediction model 3 is constructed, a linear regression equation is set as y=bx+a, wherein a and b are undetermined parameters, b is a regression coefficient, n is a data set number, and a regression coefficient calculation formula is as follows:
we know that the calculated results of parameters b and a are
Therefore, the specific calculation formula of the linear regression equation is
Wherein the method comprises the steps ofAverage number of x, y, respectively, i.e./>A. b is a parameter to be solved, n is the number of data sampling groups, the solved parameters a and b are substituted into a linear regression equation to obtain a linear regression equation required by us, namely a linear prediction model 3, and in addition, when x is less than x 1,x>xn, the parameters are respectively defined to correspond to y 1 and y n, and in actual application, the corresponding data can be collected by combining the structure, the braking performance and different application scenes of the vehicle, so that an adaptive linear prediction model is constructed;
the brake execution module comprises a brake ECU, a brake motor and a pressure sensor, wherein the brake ECU of the execution module receives brake execution data of the data processing module in real time and performs forward and reverse rotation control on the brake motor, and the control actions comprise action time, action speed, action duration and action direction, namely, the braking time, the braking speed, the braking duration and the decompression speed (brake return speed) of the system, and the motor acts on the hydraulic cylinder to realize braking.
Based on the linear dynamic prediction model, the invention performs real-time accurate information acquisition and calculation on the speed, the relative distance and the collision TTC of the vehicle, and accordingly combines the structure, the braking performance and different application scenes of the vehicle to make a corresponding dynamic automatic braking strategy so as to achieve more intelligent braking, improve the safety of the vehicle and improve the riding comfort. The prior art scheme generally uses the monitoring of the vehicle distance and the limitation of the collision time TTC as the basis of braking and decelerating. Physical and chemical data limiting is carried out on the braking time and the braking force. There is no dynamic braking strategy execution. Particularly, braking in a high-speed running state should be performed, and insufficient braking may directly affect running safety and comfort. The automatic braking system should execute different braking time, braking speed and braking force according to different vehicle distance, relative speed and collision time TTC, and a dynamic automatic braking prediction model is constructed, so that the braking system is more intelligent, and the safety and comfort of the vehicle are improved.
Drawings
FIG. 1 is a block diagram of a brake system
FIG. two is a flow chart of the brake system
Detailed Description
The following examples will provide those skilled in the art with a more complete understanding of the present invention and are not intended to limit the invention to the embodiments described.
Firstly, selecting a conventional household automobile, assembling an information acquisition module millimeter wave radar and a vision sensor in front of the automobile, assembling a GPS module in the automobile, monitoring obstacles in front of the automobile in real time, and acquiring the speed of the automobile, the relative speed of the automobile and the obstacles and the relative distance information;
Secondly, the data processing module calculates the speed of the vehicle, the relative speed of the obstacle and the relative distance information obtained by the information acquisition module, firstly calculates the real-time collision time TTC with the obstacle based on the relative speed and the relative distance between the vehicle and the obstacle in front, namely the collision time TTC=relative distance/relative speed, and the data is accurate to 0.01S;
Thirdly, respectively constructing three different linear regression prediction models by the data processing module according to the calculated pre-collision time TTC, the acquired relative speed and relative distance and the output brake voltage ratio data of the execution end, wherein a linear prediction model 1 is constructed according to the vehicle speed and the collision time TTC to determine the brake time, a linear prediction model 2 is constructed according to the relative speed and the brake voltage ratio to determine the brake speed, a linear prediction model 3 is constructed according to the relative speed and the brake voltage ratio to determine the decompression speed (brake return speed), meanwhile, the collision time threshold executed by the current brake time is defined to be a brake interval, and the brake return (motor reversal) is performed when the threshold is exceeded, so that the dynamic brake duration is obtained;
step four, collecting data required by three prediction models and constructing corresponding prediction models;
The method for constructing the linear prediction model 1 comprises the following steps: defining a certain pre-collision time TTC threshold value as a braking signal, namely a braking time, when the TTC threshold value is in a braking state, namely a braking time length, different vehicle speeds correspond to different braking time, the higher the vehicle speed is, the larger the variable is, the higher the dangerous coefficient of the running vehicle is, such as sudden braking of the front vehicle in a high-speed state or front emergency, and the longer the braking time required by the user is, so that a dynamic linear regression prediction model is constructed according to the vehicle speed and the collision time TTC, and the braking time is dynamic;
The data acquisition method required by the linear prediction model 1 comprises the following steps: when the vehicle moves forwards longitudinally, pre-collision is performed according to a pre-collision time calculation formula It can be known that the relative speed is inversely proportional to the collision time, and the greater the relative speed, the shorter the collision time, i.e. the higher the collision risk; let the speed be V1 and the obstacle speed of the front vehicle be V2, the longitudinal relative speed v=v 1-v2, and the maximum value of the relative speed V be V 2 =0 and the relative speed=v 1, so that the most practical and effective data acquisition method is to acquire the braking time TTC corresponding to the manual pedal stepping of the driving vehicle when approaching the stationary obstacle at different speeds, and the following table data are obtained according to the data:
according to the data, the speed and TTC are independent variable x and dependent variable y respectively, so that a TTC value corresponding to the speed change is needed to be obtained as the braking time of the system, a linear regression prediction model is needed to be constructed, a and b are undetermined parameters, b is a regression coefficient, n is the number of data groups, Average numbers of x and y are calculated by a regression coefficient formula:
We can get
Substituting the approximated values of the parameters a and b into the linear regression equation to obtain y=0.01x+1.09, namely, the linear prediction model 1; in actual application, the structure, braking performance and different scenes of the vehicle can be combined to collect corresponding data for constructing a linear model, so that an adaptive linear prediction model is constructed;
the method for constructing the linear prediction model 2 and the required data acquisition method are as follows: when the vehicle moves forwards longitudinally, pre-collision is performed according to a pre-collision time calculation formula It can be seen that the relative speed is inversely proportional to the collision time, the greater the relative speed, the shorter the collision time, i.e. the higher the collision risk, so that the braking speed is faster, and the relation between the motor speed and the voltage is known according to the relation u=c eΦn+IaRa+2△Us (where n is the speed, U is the motor end voltage, Δus is the brush voltage drop, ia is the armature current, ra is the motor armature winding resistance, ce is the motor constant, Φ is the motor air gap flux) of the current, the speed and the voltage of the dc motorThe method is characterized in that the data acquisition method is to acquire the working voltage required by the manual pedal speed of a driving vehicle when the driving vehicle approaches a static obstacle at different speeds and the manual pedal speed corresponds to the braking motor reaching the speed, the voltage data is determined according to the percentage value of the braking voltage, and in addition, the braking voltage ratio is 100% at most as long as the relative speed is greater than zero, so that the linear regression model 2 is required to define a reasonable interval, namely, x is not less than 20 and not more than 40, and the following data are acquired according to the data:
According to the data, the relative speed and the braking voltage ratio are independent variable x and dependent variable y respectively, a braking voltage ratio corresponding to the relative speed change in the interval of 20-40 is obtained, so that the corresponding system braking speed is obtained, a linear prediction model 2 is constructed according to the braking voltage ratio, a linear regression equation is set as y=bx+a, a and b are undetermined parameters, b is a regression coefficient, n is the number of data sets, and a regression coefficient is calculated according to the formula:
We can get
Substituting the solved parameters a and b into a linear regression equation to obtain y=1.52x+40, namely, a linear prediction model 2, wherein when x is less than 20 and x is greater than 40, 70% and 100% respectively are defined, and data of a corresponding construction linear model can be acquired by combining the structure, the braking performance and different scenes of a vehicle in actual application, so that an adaptive data prediction model is constructed;
The data acquisition method required by the linear prediction model 3 comprises the following steps: when the vehicle moves forwards longitudinally, pre-collision is performed according to a pre-collision time calculation formula It can be known that the relative speed is inversely proportional to the collision time, the greater the relative speed, the shorter the collision time, i.e. the higher the collision risk, and the more according to the object kinetic energy formula/>The kinetic energy of the vehicle is positively correlated with the speed, the larger the speed is, the larger the kinetic energy is, the larger the inertia of the vehicle is, the slower the speed of the brake braking force opposite to the kinetic energy direction is, the more stable the vehicle is decelerated, in addition, the direct current motor rotating speed and voltage proportional relation is verified from the construction of the linear prediction model 2, according to the direct current motor rotating speed and voltage proportional relation, the linear prediction model 3 is constructed according to the relative speed of the vehicle and the obstacle and the braking voltage ratio, and the data acquisition method comprises the following steps: selecting a safety test road section, running back and forth, collecting working voltage required by a rear vehicle when the aftertaste speed of a pedal is braked at different relative speeds and the speed of a brake motor is corresponding to the speed, and obtaining the following table data:
According to the data, the relative speed and the braking voltage ratio are independent variable x and dependent variable y respectively, a braking voltage ratio corresponding to the relative speed change in the interval of 20-40 is obtained, so that the corresponding decompression speed (braking return speed) is obtained, a linear prediction model 3 is constructed according to the braking voltage ratio, a linear regression equation is set as y=bx+a, a and b are undetermined parameters, b is a regression coefficient, n is the number of data sets, and a regression coefficient calculation formula is adopted:
We can get
Substituting the solved parameters a and b into a linear regression equation to obtain y= -1.52x+131.2, namely, a linear prediction model 3, wherein when x is less than 20 and x is greater than 40, 100% and 70% respectively are defined, and data of a corresponding construction linear model can be acquired by combining the structure, braking performance and different scenes of a vehicle in actual application, so that an adaptive data prediction model is constructed.
And fifthly, the executing module comprises a brake ECU, a brake motor and a pressure sensor, the brake ECU receives brake executing data of the data processing module in real time and performs forward and reverse rotation control on the brake motor, wherein the control actions comprise action time, action speed and action duration, and the action direction comprises action time, action speed, decompression speed (brake return speed) and action duration of the system, namely, the motor acts on the hydraulic cylinder to realize braking, the pressure sensor feeds back the pressure of the hydraulic cylinder to the brake ECU in real time, and when the pressure exceeds a preset pressure value, the braking is stopped, so that the locked-rotor protection of the motor is realized.
While the basic principles and main features of the present invention and advantages of the present invention have been shown and described, it will be understood by those skilled in the art that the present invention is not limited by the foregoing embodiments, which are described in the foregoing specification merely illustrate the principles of the present invention, and various changes and modifications may be made therein without departing from the spirit and scope of the invention as defined in the appended claims and their equivalents.
Claims (1)
1. An automatic braking method of an unmanned vehicle based on a linear dynamic prediction model is characterized in that an automatic braking system of the unmanned vehicle based on the linear dynamic prediction model is adopted, and the automatic braking system comprises an information acquisition module, a data processing module and a braking execution module; the information acquisition module comprises a millimeter wave radar, a vision sensor and a GPS module and is used for acquiring the current speed, the relative speed and the relative distance between the current speed and an obstacle; the data processing module constructs three brake linear prediction models by calculating collision time TTC, vehicle speed, relative distance and output brake voltage ratio of an executing end, respectively obtains brake time, brake speed and decompression speed (brake return speed), and defines that all the collision time thresholds executed by the current brake time are brake intervals to obtain dynamic brake duration; the brake execution module executes a brake command to control the action time, the action speed, the action duration and the action direction of the brake motor, acts on the hydraulic cylinder to realize braking, and the pressure sensor feeds back the pressure of the hydraulic cylinder to the brake ECU in real time to realize locked-rotor protection of the motor;
The automatic braking method comprises the following steps:
(1) The millimeter wave radar and the vision sensor of the information acquisition module are assembled in front of the automobile, the GPS module is assembled in the automobile, the obstacle in front of the automobile is monitored in real time, and the speed, the relative speed and the relative distance information of the obstacle are acquired;
(2) The data processing module calculates the speed of the vehicle, the relative speed and the relative distance information of the obstacle, which are obtained by the information acquisition module, and firstly calculates the real-time collision time TTC of the obstacle based on the relative speed and the relative distance between the vehicle and the obstacle in front, namely the collision time TTC=relative distance/relative speed, and the data is accurate to 0.01S;
(3) The data processing module respectively builds three different linear regression prediction models according to the calculated collision time TTC, the acquired relative speed and relative distance and the brake voltage ratio of the executing end, wherein a linear prediction model 1 is built according to the vehicle speed and the collision time TTC to determine the brake time, a linear prediction model 2 is built according to the relative speed and the brake voltage ratio to determine the brake speed, a linear prediction model 3 is built according to the relative speed and the brake voltage ratio to determine the decompression speed, and meanwhile, the pre-collision time threshold executed by the current brake time is defined to be a brake interval, and the brake returns when the pre-collision time threshold exceeds the threshold, so that the dynamic brake duration is obtained;
(4) The execution module comprises a brake ECU, a brake motor and a pressure sensor, wherein the brake ECU receives brake execution data of the data processing module in real time and performs forward and reverse rotation control on the brake motor, the control actions comprise action time, action speed and action duration, and the action direction comprises action time, braking speed and decompression speed of the system, braking duration, braking or return, the motor acts on the hydraulic cylinder to realize braking, the pressure sensor feeds back the pressure of the hydraulic cylinder to the brake ECU in real time, and when the pressure exceeds a preset pressure value, the braking is stopped to realize locked-rotor protection of the motor;
In the step (3), the construction method and the data acquisition method of the three linear prediction models are as follows:
(1) The method for constructing the linear prediction model 1 and the required data acquisition method are as follows: defining a certain pre-collision time TTC threshold value as a braking signal, namely a braking time, and when the braking state, namely the braking time length, is in the TTC threshold value, different vehicle speeds correspond to different braking time, and the vehicle speeds and the pre-collision time are in a direct proportion relation, so that a dynamic linear regression prediction model is built according to the vehicle speeds and the pre-collision time TTC, and the braking time is dynamic; the data acquisition method comprises the following steps: the corresponding braking time when the driver steps on the pedal by the manual foot when the driver approaches to a stationary obstacle at different speeds, namely the pre-collision time TTC, is acquired, and the following data are obtained according to the braking time:
Vehicle speed (X): x 1、x2...xn; time to pre-crash TTC (Y): y 1、y2...yn;
According to the data, the vehicle speed and TTC are independent variable x and dependent variable y respectively, a TTC value corresponding to the speed change is obtained as braking time of the system, so a linear regression prediction model needs to be constructed, a and b are undetermined parameters, b is a regression coefficient, n is the number of data groups, Average numbers of x and y are calculated by a regression coefficient formula:
The calculation result of the parameters b and a is known as
Therefore, the specific calculation formula of the linear regression equation is
Wherein the method comprises the steps ofAverage number of x, y, respectively, i.e./>A. b is a parameter to be solved, n is the number of data sampling groups, a linear regression equation substituted by the solved parameters a and b can be used for obtaining a required linear regression equation, namely a linear prediction model 1, and corresponding data are collected by combining the structure, the braking performance and different application scenes of a vehicle in actual application, so that an adaptive linear prediction model is constructed;
(2) The construction method of the linear prediction model 2 and the required data acquisition method are as follows: the method comprises the steps of constructing a linear prediction regression model 2 according to the ratio of the relative speed to the braking voltage of a collected vehicle, wherein the relative speed and the braking voltage are in a direct proportion relation, the data collection method is to collect the working voltage required by a brake motor when the speed of an artificial pedal is close to a static obstacle under different speeds of the driven vehicle, the voltage data is determined according to the percentage value of the braking voltage, and in addition, the braking voltage ratio is 100% at most as long as the relative speed is greater than zero when the vehicle runs, a reasonable interval, namely x 1≤x≤xn, is required to be defined for the linear regression model 2, and the following data are obtained according to the collection:
Relative velocity (X): x 1、x2...xn; a brake voltage ratio (Y) Y 1、y2...yn;
according to the data, the relative speed and the braking voltage ratio are independent variable x and dependent variable y respectively, a braking voltage ratio corresponding to the relative speed change in the interval x 1≤x≤xn is obtained, so that the corresponding system braking speed is obtained, a linear prediction model 2 is constructed according to the braking speed, a linear regression equation is set as y=bx+a, a and b are undetermined parameters, b is a regression coefficient, n is the number of data sets, and a regression coefficient calculation formula is used for:
The calculation result of the parameters b and a is known as
Therefore, the specific calculation formula of the linear regression equation is
Wherein the method comprises the steps ofAverage number of x, y, respectively, i.e./>A. b is a parameter to be solved, n is the number of data sampling groups, a required linear regression equation, namely a linear prediction model 2, can be obtained by substituting the solved parameters a and b into the linear regression equation, and in addition, y 1 and y n which respectively correspond to x < x 1,x>xn are defined, and corresponding data are collected by combining the structure, the braking performance and different application scenes of the vehicle in actual application, so that an adaptive data prediction model is constructed;
(3) The construction method of the linear prediction model 3 and the required data acquisition method are as follows: constructing a linear prediction model 3 according to the relative speed of the vehicle and the obstacle and the braking voltage ratio, wherein the relative speed and the braking voltage ratio are in inverse proportion; the data acquisition method comprises the following steps: selecting a safety test road section, driving forwards and backwards, and acquiring the working voltage required by a rear vehicle when the return speed of a rear pedal braked at different relative speeds corresponds to the speed of a brake motor, so as to obtain the following data:
Relative velocity (X): x 1、x2...xn; brake voltage ratio (Y): y 1、y2...yn;
according to the data, the relative speed and the braking voltage ratio are independent variable x and dependent variable y respectively, a braking voltage ratio corresponding to the relative speed change in the interval x 1≤x≤xn is obtained, so that the corresponding system braking speed is obtained, a linear prediction model 3 is constructed according to the braking speed, a linear regression equation is set as y=bx+a, a and b are undetermined parameters, b is a regression coefficient, n is the number of data sets, and a regression coefficient calculation formula is used for:
The calculation result of the parameters b and a is known as
Therefore, the specific calculation formula of the linear regression equation is
Wherein the method comprises the steps ofAverage number of x, y, respectively, i.e./>A. b is a parameter to be solved, n is the number of data sampling groups, a required linear regression equation, namely a linear prediction model 3, can be obtained by substituting the solved parameters a and b into the linear regression equation, and in addition, y 1 and y n are respectively corresponding to x < x 1,x>xn, and corresponding data are acquired by combining the structure, the braking performance and different application scenes of the vehicle in actual application, so that an adaptive data prediction model is constructed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210034059.7A CN114312768B (en) | 2022-01-12 | 2022-01-12 | Unmanned vehicle automatic braking system based on linear dynamic prediction model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210034059.7A CN114312768B (en) | 2022-01-12 | 2022-01-12 | Unmanned vehicle automatic braking system based on linear dynamic prediction model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114312768A CN114312768A (en) | 2022-04-12 |
| CN114312768B true CN114312768B (en) | 2024-06-11 |
Family
ID=81027012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210034059.7A Active CN114312768B (en) | 2022-01-12 | 2022-01-12 | Unmanned vehicle automatic braking system based on linear dynamic prediction model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114312768B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115359682B (en) * | 2022-08-09 | 2024-02-13 | 上汽通用五菱汽车股份有限公司 | Collision early warning method and device |
| CN116691623A (en) * | 2023-08-07 | 2023-09-05 | 北京中科慧眼科技有限公司 | Brake force generation method and device based on self-adaptive vehicle distance maintaining system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6850252B1 (en) * | 1999-10-05 | 2005-02-01 | Steven M. Hoffberg | Intelligent electronic appliance system and method |
| CN107697045A (en) * | 2017-09-07 | 2018-02-16 | 厦门大学 | A kind of pilotless automobile automatic brake controller and method |
| CN107867288A (en) * | 2016-09-26 | 2018-04-03 | 现代自动车美国技术研究所 | For detecting the preceding method to collision |
| CN112596393A (en) * | 2020-12-24 | 2021-04-02 | 武汉理工大学 | Control method, system and storage medium for ship path tracking |
| CN112882474A (en) * | 2021-01-26 | 2021-06-01 | 北京科技大学 | Integrated unmanned automobile lane change control method based on time-varying model predictive control |
| CN113408062A (en) * | 2021-07-09 | 2021-09-17 | 中国石油大学(华东) | Automatic driving full-working-condition road surface self-adaptive MPC (MPC) trajectory tracking control and evaluation method |
-
2022
- 2022-01-12 CN CN202210034059.7A patent/CN114312768B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6850252B1 (en) * | 1999-10-05 | 2005-02-01 | Steven M. Hoffberg | Intelligent electronic appliance system and method |
| CN107867288A (en) * | 2016-09-26 | 2018-04-03 | 现代自动车美国技术研究所 | For detecting the preceding method to collision |
| CN107697045A (en) * | 2017-09-07 | 2018-02-16 | 厦门大学 | A kind of pilotless automobile automatic brake controller and method |
| CN112596393A (en) * | 2020-12-24 | 2021-04-02 | 武汉理工大学 | Control method, system and storage medium for ship path tracking |
| CN112882474A (en) * | 2021-01-26 | 2021-06-01 | 北京科技大学 | Integrated unmanned automobile lane change control method based on time-varying model predictive control |
| CN113408062A (en) * | 2021-07-09 | 2021-09-17 | 中国石油大学(华东) | Automatic driving full-working-condition road surface self-adaptive MPC (MPC) trajectory tracking control and evaluation method |
Non-Patent Citations (3)
| Title |
|---|
| 基于Logistic回归的驾驶人跟车危险认知模型;林庆峰;成波;;中国公路学报(06);107-112 * |
| 林庆峰 ; 成波 ; .基于Logistic回归的驾驶人跟车危险认知模型.中国公路学报.2011,(06),107-112. * |
| 车辆安全距离智能控制与自刹车系统;吴闯;张雄;卢亚玲;;电子设计工程(07);100-103 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114312768A (en) | 2022-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114312768B (en) | Unmanned vehicle automatic braking system based on linear dynamic prediction model | |
| CN110435623B (en) | Automatic hierarchical automatic emergency braking control system of electric motor car of adjustment | |
| CN102490705B (en) | Electronic mechanical brake system and automobile adopting same | |
| CN102490706B (en) | Electromechanical brake control system and automobile | |
| EP1409311B1 (en) | Method and device for initiating and executing a deceleration of a vehicle | |
| CN103935264B (en) | A kind of electric vehicle driver demand torque calculation method | |
| DE102008033026B4 (en) | Method and system for controlling a drive system of a motor vehicle | |
| DE69836951T2 (en) | Electric braking device for controlling a brake drive current | |
| DE602004012196T2 (en) | Driver assistance system for a motor vehicle | |
| DE102013222880A1 (en) | Active safety system and method of operating the same | |
| CN203920440U (en) | A kind of Anti-false stepping system for automobile throttle | |
| CN109131306B (en) | Brake control method and brake control system of electric automobile and automobile | |
| CN106891890A (en) | A kind of automobile active collision-avoidance system | |
| DE102010027896A1 (en) | Control unit for an electric vehicle | |
| CN110962614A (en) | Energy recovery method for pure electric vehicle | |
| CN107600070A (en) | Control method and device of automatic emergency braking system, controller and automobile | |
| CN110530384B (en) | Vehicle path planning method adopting hybrid potential field ant colony algorithm | |
| DE10393181B4 (en) | Device for estimating a road surface state change, corresponding method and motor vehicle with this device | |
| CN111391675B (en) | Variable braking torque energy recovery control method for electric automobile | |
| CN114802166A (en) | Control method for keeping constant pedal feeling of automobile | |
| CN112896127A (en) | Vehicle deceleration control method based on electro-hydraulic brake system | |
| CN111959469A (en) | Wheel anti-lock control method and device and vehicle | |
| CN113928284A (en) | Self-adaptive multistage braking control method for automatic emergency braking system | |
| CN114475541B (en) | Automatic emergency braking method considering passenger safety | |
| CN205554112U (en) | Miss intelligence anticollision initiative control system that steps on gas |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |