CN114407847B - Single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning - Google Patents
Single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning Download PDFInfo
- Publication number
- CN114407847B CN114407847B CN202210179569.3A CN202210179569A CN114407847B CN 114407847 B CN114407847 B CN 114407847B CN 202210179569 A CN202210179569 A CN 202210179569A CN 114407847 B CN114407847 B CN 114407847B
- Authority
- CN
- China
- Prior art keywords
- steering wheel
- maximum value
- longitudinal control
- control pedal
- automobile
- 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
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000010801 machine learning Methods 0.000 title claims abstract description 13
- 230000001133 acceleration Effects 0.000 claims abstract description 79
- 230000006399 behavior Effects 0.000 claims abstract description 8
- 239000002245 particle Substances 0.000 claims abstract description 8
- 230000003044 adaptive effect Effects 0.000 claims abstract description 7
- 230000007246 mechanism Effects 0.000 claims abstract description 5
- 238000012360 testing method Methods 0.000 claims description 24
- 230000006870 function Effects 0.000 claims description 21
- 238000012549 training Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 12
- 238000013507 mapping Methods 0.000 claims description 9
- 230000001172 regenerating effect Effects 0.000 claims description 4
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 claims description 3
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000009977 dual effect Effects 0.000 claims description 3
- 230000001788 irregular Effects 0.000 claims description 3
- 238000012417 linear regression Methods 0.000 claims description 3
- 238000012544 monitoring process Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 206010039203 Road traffic accident Diseases 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T7/00—Brake-action initiating means
- B60T7/12—Brake-action initiating means for automatic initiation; for initiation not subject to will of driver or passenger
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60K—ARRANGEMENT OR MOUNTING OF PROPULSION UNITS OR OF TRANSMISSIONS IN VEHICLES; ARRANGEMENT OR MOUNTING OF PLURAL DIVERSE PRIME-MOVERS IN VEHICLES; AUXILIARY DRIVES FOR VEHICLES; INSTRUMENTATION OR DASHBOARDS FOR VEHICLES; ARRANGEMENTS IN CONNECTION WITH COOLING, AIR INTAKE, GAS EXHAUST OR FUEL SUPPLY OF PROPULSION UNITS IN VEHICLES
- B60K26/00—Arrangements or mounting of propulsion unit control devices in vehicles
- B60K26/02—Arrangements or mounting of propulsion unit control devices in vehicles of initiating means or elements
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60T—VEHICLE BRAKE CONTROL SYSTEMS OR PARTS THEREOF; BRAKE CONTROL SYSTEMS OR PARTS THEREOF, IN GENERAL; ARRANGEMENT OF BRAKING ELEMENTS ON VEHICLES IN GENERAL; PORTABLE DEVICES FOR PREVENTING UNWANTED MOVEMENT OF VEHICLES; VEHICLE MODIFICATIONS TO FACILITATE COOLING OF BRAKES
- B60T7/00—Brake-action initiating means
- B60T7/02—Brake-action initiating means for personal initiation
- B60T7/04—Brake-action initiating means for personal initiation foot actuated
- B60T7/06—Disposition of pedal
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/08—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to drivers or passengers
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/72—Electric energy management in electromobility
Landscapes
- Engineering & Computer Science (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
The application discloses a single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning, which comprises the following steps: sample data acquisition; constructing a driver behavior model based on a support vector regression algorithm and an adaptive particle swarm algorithm; the system monitors whether the driver operates the longitudinal control pedal by mistake in real time, after the system judges that the driver operates the longitudinal control pedal by mistake, the automobile is changed into a braking mode from a single pedal driving mode, when the driver downwards steps on the longitudinal control pedal, the automobile receives a braking signal, a braking executing mechanism starts executing, and meanwhile, an accelerating signal is cut off, so that the safety of the automobile is ensured, and traffic accidents caused by unexpected rapid acceleration are avoided.
Description
Technical Field
The application relates to the technical field of automobiles, in particular to a single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning.
Background
With the development of automobile technology, a drive-by-wire chassis automobile with a single pedal driving mode is developed, and a longitudinal control pedal is adopted to replace a traditional accelerator pedal. The longitudinal control pedal has the characteristic of acceleration and braking integration, and the working principle is that when a driver steps on the longitudinal control pedal, the motor outputs driving torque to realize acceleration of the automobile; after the driver releases the longitudinal control pedal, the motor outputs braking torque to realize the speed reduction of the automobile through regenerative braking.
However, when emergency braking is required in an emergency after the driver has adapted to the setting of stepping on the longitudinal control pedal for acceleration and releasing the longitudinal control pedal for braking, it is likely that the longitudinal control pedal is erroneously operated due to the presence of the brake pedal, which is neglected due to driving habits. At present, a plurality of single-pedal automobiles are accidentally accelerated at home and abroad, and the accidents are all investigated to be caused by improper use of pedals by drivers. How to minimize the occurrence of such severe accidents and further improve the safety of automobiles is unprecedented. It is necessary to develop a single pedal chassis-by-wire vehicle auxiliary braking method that can accurately sense the dangerous behavior of the driver operating the pedal by mistake and help the driver take braking measures to avoid or mitigate casualty accidents.
Disclosure of Invention
The application aims to provide a single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning to solve the technical problems.
The auxiliary braking method of the single-pedal drive-by-wire chassis automobile comprises the following steps of:
s1: sample data acquisition:
100 male testers and 100 female testers with the driver license of the motor vehicle of the level of C2 and above and the ages of 18-60 are selected to carry out real vehicle tests in a driving field, the real vehicle adopts a longitudinal control pedal integrated with acceleration and braking to replace the traditional acceleration pedal, the driver steps on the longitudinal control pedal for accelerating, the automobile is decelerated by regenerative braking when the longitudinal control pedal is loosened, and a brake pedal consistent with the traditional automobile is arranged on the left side of the longitudinal control pedal for emergency braking; collecting age, driving age, height and weight information of each tester before the test starts, and then respectively testing each tester under three working conditions of stable acceleration, rapid acceleration and error operation of a longitudinal control pedal when encountering obstacles, wherein 10 tests are arranged on each working condition, and the three working conditions appear randomly in the test process; in the test process, the speed of the vehicle and the distance between the vehicle and the obstacle in front are monitored in real time, and the maximum longitudinal control pedal angular acceleration, the maximum longitudinal control pedal pressure, the maximum steering wheel hand holding force, the maximum steering wheel torque, the maximum steering wheel corner, the maximum steering wheel rotational angular velocity and the maximum steering wheel rotational angular acceleration of all testers under the working conditions of rapid acceleration and obstacle misoperation of the longitudinal control pedal are collected.
S2: constructing a driver behavior model:
and constructing an end-to-end prediction model of the age, the driving age, the height, the weight, the speed and the distance between the automobile and the obstacle in front of the driver on the basis of the test data, wherein the end-to-end prediction model is established through a support vector regression model, and the end-to-end prediction model is related to the maximum value of the longitudinal control pedal angular acceleration, the maximum value of the longitudinal control pedal pressure, the maximum value of the steering wheel hand holding force, the maximum value of the steering wheel torque, the maximum value of the steering wheel corner, the maximum value of the steering wheel rotational angular speed and the maximum value of the steering wheel rotational angular acceleration.
The age, the driving age, the height, the weight, the speed and the distance between the automobile and the front obstacle in the sample data are taken as input variables of a support vector regression model, and the longitudinal control pedal angular acceleration maximum value, the longitudinal control pedal pressure maximum value, the steering wheel hand holding force maximum value, the steering wheel torque maximum value, the steering wheel corner maximum value, the steering wheel rotation angular speed maximum value and the steering wheel rotation angular acceleration maximum value in the sample data are taken as output variables of the support vector regression model. And sending the normalized training data set into a support vector regression model for training, and predicting the maximum longitudinal control pedal angular acceleration, the maximum longitudinal control pedal pressure, the maximum steering wheel hand holding force, the maximum steering wheel torque, the maximum steering wheel corner, the maximum steering wheel rotational angular velocity and the maximum steering wheel rotational angular acceleration by using the learned high-dimensional mapping relation, so as to provide a data basis for establishing a judgment logic of mistaken stepping and mistaken acceleration of a driver.
The method comprises the following specific steps:
the sample data is normalized according to the following equation:
wherein X is min Is the minimum value of the sample data; x is X max Is the maximum value of the sample data; x is sample data; x' is normalized data, ranging from [0,1]。
After the normalization processing of the sample data is completed, the normalized data is divided into two parts, wherein 80% of the normalized data are classified into a training data set, and 20% of the normalized data are classified into a test data set.
Integrating sample data into a data set d= { (x) 1 ,y 1 ),(x 2 ,y 2 )...,(x n ,y n )}，The training samples are mapped from the low-dimensional space to the high-dimensional space through nonlinear mapping, and a linear regression model established in the high-dimensional space can be expressed as the following equation:
f(x)＝w·Φ(x)+b
where x is the input variable, Φ (x) is a nonlinear function mapping x to a high-dimensional linear space, w is a weight vector, and b is a bias.
To minimize regression errors, the objective function of the support vector regression algorithm may be expressed as follows:
wherein C is p For punishment coefficient, represent punishment degree of model to sample with error greater than epsilon in training process, l ε For the epsilon-insensitive loss function, epsilon represents the insensitive loss coefficient, epsilon is smaller to represent the smaller error requirement of the regression function, l ε The expression can be expressed as the following equation:
where z represents the error of the fitted value and the true value of the support vector regression algorithm.
In case of data disagreement with l ε When the constraint of (z) is satisfied, a relaxation variable delta is introduced i ，δ i * To correct the irregular factor, after which the following equation can be obtained:
by introducing Lagrangian multiplier alpha i 、α i * Simplifying the calculation, converting the above formula into alpha i ，α i * Is a dual problem:
wherein K (x i ,x j ) The application selects RBF kernel function, which is defined as the following equation:
K(x i ,x j )＝exp(-γ||x i -x j || 2 )
wherein γ is a nuclear parameter.
The solution to the regression function f (x) according to the karman-coulen-tak condition can be expressed as:
based on the above method, the driver behavior model may be abstracted as:
y＝f(x|(C p ,ε,γ))。
then, three super parameters of the support vector regression model, namely penalty coefficient C, are calculated by using an adaptive particle swarm algorithm p And optimizing the core parameter gamma and the insensitive loss coefficient epsilon. Selecting an average absolute percentage error MAPE capable of directly reflecting regression performance as a fitness function fitness of the adaptive particle swarm algorithm, namely:
where n is the number of sample data, y i Is a predicted value, f (x i ) Is an experimental value.
Predicting the maximum value of longitudinal control pedal angular acceleration, the maximum value of longitudinal control pedal pressure, the maximum value of steering wheel hand holding force, the maximum value of steering wheel torque, the maximum value of steering wheel angle, the maximum value of steering wheel rotational angular speed and the maximum value of steering wheel rotational angular acceleration by using a support vector regression model obtained through training, and adopting a mean square error MSE and a decision coefficient R 2 Evaluating the prediction result of the model:
wherein,is the average of the predicted values,/>Is the average of the experimental values.
S3: monitoring whether a driver operates the longitudinal control pedal by mistake in real time:
the method comprises the steps that the age, the driving age, the height and the weight of a driver are collected before the driver starts to drive an automobile, the speed of the automobile and the distance between the automobile and a front obstacle are monitored in real time in the running process of the automobile, and the longitudinal control pedal angular acceleration a, the longitudinal control pedal pressure h, the steering wheel hand holding force f, the steering wheel torque m, the steering wheel rotation angle theta, the steering wheel rotation angular speed omega and the steering wheel rotation angular acceleration beta are obtained; simultaneously, calculating the maximum value a of the longitudinal control pedal angular acceleration of the driver under the sudden acceleration working condition through a support vector regression model 1 Maximum value h of longitudinal control pedal pressure 1 Maximum value f of hand grip force of steering wheel 1 Maximum steering wheel torque m 1 Maximum value theta of steering wheel angle 1 Maximum value omega of steering wheel rotation angular velocity 1 And steering wheel rotational angular acceleration maximum value beta 1 And a maximum value a of the angular acceleration of the longitudinal control pedal under the condition of encountering an obstacle and misoperation of the longitudinal control pedal 2 Maximum value h of longitudinal control pedal pressure 2 Maximum value f of hand grip force of steering wheel 2 Maximum steering wheel torque m 2 Maximum value theta of steering wheel angle 2 Maximum value omega of steering wheel rotation angular velocity 2 And steering wheel rotational angular acceleration maximum value beta 2 。
The system determines that the driver has misoperated the longitudinal control pedal and activates the auxiliary brake module when either:
case one: the following seven conditions are satisfied simultaneously for four or more:
①a 1 ≤a＜a 2 ；
②h 1 ≤h＜h 2 ；
③f 1 ≤f＜f 2 ；
④m 1 ≤m＜m 2 ；
⑤θ 1 ≤θ＜θ 2 ；
⑥ω 1 ≤ω＜ω 2 ；
⑦β 1 ≤β＜β 2 ；
and a second case: the following seven conditions are satisfied:
①a≥a 2 ；
②h≥h 2 ；
③f≥f 2 ；
④m≥m 2 ；
⑤θ≥θ 2 ；
⑥ω≥ω 2 ；
⑦β≥β 2 。
the auxiliary braking module starts working after the system judges that a driver operates the longitudinal control pedal by mistake, the automobile is changed into a braking mode from a single pedal driving mode integrating acceleration and braking, the automobile receives a braking signal when the driver downwards steps on the longitudinal control pedal, the braking executing mechanism starts executing, meanwhile, the acceleration signal is cut off, when the pressure of the longitudinal control pedal and the speed of the automobile are both 0, the auxiliary braking module is restored, the automobile is restored to the single pedal driving mode, and then the driver can normally drive the automobile, the automobile is accelerated when stepping on the longitudinal control pedal, and the automobile is decelerated when the longitudinal control pedal is released.
The beneficial effects of the application are as follows:
1. and establishing a proxy model of the age, the driving age, the height, the weight, the speed and the distance between the automobile and the obstacle in front of the automobile with respect to the maximum acceleration value of the accelerator pedal, the maximum pressure value of the accelerator pedal, the maximum hand holding force of the steering wheel, the maximum torque of the steering wheel, the maximum steering wheel angle, the maximum steering wheel rotation angular speed and the maximum steering wheel rotation angular acceleration based on the self-adaptive particle swarm algorithm and the support vector regression algorithm.
2. After the system judges that the driver erroneously operates the longitudinal control pedal, the automobile is changed from a single pedal driving mode to a braking mode, when the driver downwards presses the longitudinal control pedal, the automobile receives a braking signal, a braking executing mechanism starts to execute, and the accelerating signal is cut off, so that traffic accidents caused by unexpected sudden acceleration are avoided.
3. The method is high in universality and is generally suitable for automobiles produced by different manufacturers.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
Fig. 1 is a system frame diagram of a single pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning.
Detailed Description
The following embodiments of the present application will be described in detail with reference to the accompanying drawings, which are only used to more clearly illustrate the technical solution of the present application, and therefore are only used as examples, and are not to be construed as limiting the scope of the present application.
It is noted that unless otherwise indicated, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs.
A single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning is shown in a system frame diagram in figure 1.
The auxiliary braking method of the single-pedal drive-by-wire chassis automobile comprises the following steps of:
the auxiliary braking method of the single-pedal drive-by-wire chassis automobile comprises the following steps of:
s1: sample data acquisition:
100 male testers and 100 female testers with the driver license of the motor vehicle of the level of C2 and above and the ages of 18-60 are selected to carry out real vehicle tests in a driving field, the real vehicle adopts a longitudinal control pedal integrated with acceleration and braking to replace the traditional acceleration pedal, the driver steps on the longitudinal control pedal for accelerating, the automobile is decelerated by regenerative braking when the longitudinal control pedal is loosened, and a brake pedal consistent with the traditional automobile is arranged on the left side of the longitudinal control pedal for emergency braking; collecting age, driving age, height and weight information of each tester before the test starts, and then respectively testing each tester under three working conditions of stable acceleration, rapid acceleration and error operation of a longitudinal control pedal when encountering obstacles, wherein 10 tests are arranged on each working condition, and the three working conditions appear randomly in the test process; in the test process, the speed of the vehicle and the distance between the vehicle and the obstacle in front are monitored in real time, and the maximum longitudinal control pedal angular acceleration, the maximum longitudinal control pedal pressure, the maximum steering wheel hand holding force, the maximum steering wheel torque, the maximum steering wheel corner, the maximum steering wheel rotational angular velocity and the maximum steering wheel rotational angular acceleration of all testers under the working conditions of rapid acceleration and obstacle misoperation of the longitudinal control pedal are collected.
S2: constructing a driver behavior model:
and constructing an end-to-end prediction model of the age, the driving age, the height, the weight, the speed and the distance between the automobile and the obstacle in front of the driver on the basis of the test data, wherein the end-to-end prediction model is established through a support vector regression model, and the end-to-end prediction model is related to the maximum value of the longitudinal control pedal angular acceleration, the maximum value of the longitudinal control pedal pressure, the maximum value of the steering wheel hand holding force, the maximum value of the steering wheel torque, the maximum value of the steering wheel corner, the maximum value of the steering wheel rotational angular speed and the maximum value of the steering wheel rotational angular acceleration.
The age, the driving age, the height, the weight, the speed and the distance between the automobile and the front obstacle in the sample data are taken as input variables of a support vector regression model, and the longitudinal control pedal angular acceleration maximum value, the longitudinal control pedal pressure maximum value, the steering wheel hand holding force maximum value, the steering wheel torque maximum value, the steering wheel corner maximum value, the steering wheel rotation angular speed maximum value and the steering wheel rotation angular acceleration maximum value in the sample data are taken as output variables of the support vector regression model. And sending the normalized training data set into a support vector regression model for training, and predicting the maximum longitudinal control pedal angular acceleration, the maximum longitudinal control pedal pressure, the maximum steering wheel hand holding force, the maximum steering wheel torque, the maximum steering wheel corner, the maximum steering wheel rotational angular velocity and the maximum steering wheel rotational angular acceleration by using the learned high-dimensional mapping relation, so as to provide a data basis for establishing a judgment logic of mistaken stepping and mistaken acceleration of a driver.
The method comprises the following specific steps:
the sample data is normalized according to the following equation:
wherein X is min Is the minimum value of the sample data; x is X max Is the maximum value of the sample data; x is sample data; x' is normalized data, ranging from [0,1]。
After the normalization processing of the sample data is completed, the normalized data is divided into two parts, wherein 80% of the normalized data are classified into a training data set, and 20% of the normalized data are classified into a test data set.
Integrating sample data into a data set d= { (x) 1 ,y 1 ),(x 2 ,y 2 )...,(x n ,y n )}，The training samples are mapped from the low-dimensional space to the high-dimensional space through nonlinear mapping, and a linear regression model established in the high-dimensional space can be expressed as the following equation:
f(x)＝w·Φ(x)+b
where x is the input variable, Φ (x) is a nonlinear function mapping x to a high-dimensional linear space, w is a weight vector, and b is a bias.
To minimize regression errors, the objective function of the support vector regression algorithm may be expressed as follows:
wherein C is p For punishment coefficient, represent punishment degree of model to sample with error greater than epsilon in training process, l ε For the epsilon-insensitive loss function, epsilon represents the insensitive loss coefficient, epsilon is smaller to represent the smaller error requirement of the regression function, l ε The expression can be expressed as the following equation:
where z represents the error of the fitted value and the true value of the support vector regression algorithm.
In case of data disagreement with l ε When the constraint of (z) is satisfied, a relaxation variable delta is introduced i ，δ i * To correct the irregular factor, after which the following equation can be obtained:
by introducing Lagrangian multiplier alpha i 、α i * Simplifying the calculation, converting the above formula into alpha i ，α i * Is a dual problem:
wherein K (x i ,x j ) The application selects RBF kernel function, which is defined as the following equation:
K(x i ,x j )＝exp(-γ||x i -x j || 2 )
wherein γ is a nuclear parameter.
The solution to the regression function f (x) according to the karman-coulen-tak condition can be expressed as:
based on the above method, the driver behavior model may be abstracted as:
y＝f(x|(C p ,ε,γ))。
then, three super parameters of the support vector regression model, namely penalty coefficient C, are calculated by using an adaptive particle swarm algorithm p And optimizing the core parameter gamma and the insensitive loss coefficient epsilon. Selecting an average absolute percentage error MAPE capable of directly reflecting regression performance as a fitness function fitness of the adaptive particle swarm algorithm, namely:
where n is the number of sample data, y i Is a predicted value, f (x i ) Is an experimental value.
Predicting the maximum value of longitudinal control pedal angular acceleration, the maximum value of longitudinal control pedal pressure, the maximum value of steering wheel hand holding force, the maximum value of steering wheel torque, the maximum value of steering wheel angle, the maximum value of steering wheel rotational angular speed and the maximum value of steering wheel rotational angular acceleration by using a support vector regression model obtained through training, and adopting a mean square error MSE and a decision coefficient R 2 Evaluating the prediction result of the model:
wherein,is the average of the predicted values,/>Is the experimental valueAverage value of (2).
S3: monitoring whether a driver operates the longitudinal control pedal by mistake in real time:
the method comprises the steps that the age, the driving age, the height and the weight of a driver are collected before the driver starts to drive an automobile, the speed of the automobile and the distance between the automobile and a front obstacle are monitored in real time in the running process of the automobile, and the longitudinal control pedal angular acceleration a, the longitudinal control pedal pressure h, the steering wheel hand holding force f, the steering wheel torque m, the steering wheel rotation angle theta, the steering wheel rotation angular speed omega and the steering wheel rotation angular acceleration beta are obtained; simultaneously, calculating the maximum value a of the longitudinal control pedal angular acceleration of the driver under the sudden acceleration working condition through a support vector regression model 1 Maximum value h of longitudinal control pedal pressure 1 Maximum value f of hand grip force of steering wheel 1 Maximum steering wheel torque m 1 Maximum value theta of steering wheel angle 1 Maximum value omega of steering wheel rotation angular velocity 1 And steering wheel rotational angular acceleration maximum value beta 1 And a maximum value a of the angular acceleration of the longitudinal control pedal under the condition of encountering an obstacle and misoperation of the longitudinal control pedal 2 Maximum value h of longitudinal control pedal pressure 2 Maximum value f of hand grip force of steering wheel 2 Maximum steering wheel torque m 2 Maximum value theta of steering wheel angle 2 Maximum value omega of steering wheel rotation angular velocity 2 And steering wheel rotational angular acceleration maximum value beta 2 。
The system determines that the driver has misoperated the longitudinal control pedal and activates the auxiliary brake module when either:
case one: the following seven conditions are satisfied simultaneously for four or more:
①a 1 ≤a＜a 2 ；
②h 1 ≤h＜h 2 ；
③f 1 ≤f＜f 2 ；
④m 1 ≤m＜m 2 ；
⑤θ 1 ≤θ＜θ 2 ；
⑥ω 1 ≤ω＜ω 2 ；
⑦β 1 ≤β＜β 2 ；
and a second case: the following seven conditions are satisfied:
①a≥a 2 ；
②h≥h 2 ；
③f≥f 2 ；
④m≥m 2 ；
⑤θ≥θ 2 ；
⑥ω≥ω 2 ；
⑦β≥β 2 。
the auxiliary braking module starts working after the system judges that a driver operates the longitudinal control pedal by mistake, the automobile is changed into a braking mode from a single pedal driving mode integrating acceleration and braking, the automobile receives a braking signal when the driver downwards steps on the longitudinal control pedal, the braking executing mechanism starts executing, meanwhile, the acceleration signal is cut off, when the pressure of the longitudinal control pedal and the speed of the automobile are both 0, the auxiliary braking module is restored, the automobile is restored to the single pedal driving mode, and then the driver can normally drive the automobile, the automobile is accelerated when stepping on the longitudinal control pedal, and the automobile is decelerated when the longitudinal control pedal is released.
The preferred embodiments of the present application have been described in detail above with reference to the accompanying drawings, but the present application is not limited to the specific details of the above embodiments, and various simple modifications can be made to the technical solution of the present application within the scope of the technical concept of the present application, and all the simple modifications belong to the protection scope of the present application.
In addition, the specific features described in the above embodiments may be combined in any suitable manner, and in order to avoid unnecessary repetition, various possible combinations are not described further.
Claims (6)
1. The single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning is characterized by comprising the following steps of:
s1: sample data acquisition: 100 male testers and 100 female testers with the driver license of the motor vehicle of the level of C2 and above and the ages of 18-60 are selected to carry out real vehicle tests in a driving field, the real vehicle adopts a longitudinal control pedal integrated with acceleration and braking to replace the traditional acceleration pedal, the driver steps on the longitudinal control pedal for accelerating, the automobile is decelerated by regenerative braking when the longitudinal control pedal is loosened, and a brake pedal consistent with the traditional automobile is arranged on the left side of the longitudinal control pedal for emergency braking; collecting age, driving age, height and weight information of each tester before the test starts, and then respectively testing each tester under three working conditions of stable acceleration, rapid acceleration and error operation of a longitudinal control pedal when encountering obstacles, wherein 10 tests are arranged on each working condition, and the three working conditions appear randomly in the test process; in the test process, the speed of the vehicle and the distance between the vehicle and a front obstacle are monitored in real time, and the maximum value of the angular acceleration of the longitudinal control pedal, the maximum value of the pressure of the longitudinal control pedal, the maximum value of the hand holding force of the steering wheel, the maximum value of the torque of the steering wheel, the maximum value of the steering wheel angle, the maximum value of the rotational angular speed of the steering wheel and the maximum value of the rotational angular acceleration of the steering wheel under the working conditions that all testers rapidly accelerate and meet the obstacle misoperation of the longitudinal control pedal are collected;
s2: constructing a driver behavior model: constructing an end-to-end prediction model of the age, the driving age, the height, the weight, the speed and the distance between the automobile and a front obstacle of the driver with respect to a longitudinal control pedal angular acceleration maximum value, a longitudinal control pedal pressure maximum value, a steering wheel hand-held force maximum value, a steering wheel torque maximum value, a steering wheel corner maximum value, a steering wheel rotation angular velocity maximum value and a steering wheel rotation angular acceleration maximum value based on test data, wherein the end-to-end prediction model is established through a support vector regression model;
taking the age, the driving age, the height, the weight, the speed and the distance between the automobile and a front obstacle in the sample data as input variables of a support vector regression model, and taking the longitudinal control pedal angular acceleration maximum value, the longitudinal control pedal pressure maximum value, the steering wheel hand holding force maximum value, the steering wheel torque maximum value, the steering wheel corner maximum value, the steering wheel rotation angular speed maximum value and the steering wheel rotation angular acceleration maximum value in the sample data as output variables of the support vector regression model; the normalized sample data is sent into a support vector regression model for training, and the learned high-dimensional mapping relation is used for predicting the maximum longitudinal control pedal angular acceleration, the maximum longitudinal control pedal pressure, the maximum steering wheel hand holding force, the maximum steering wheel torque, the maximum steering wheel corner, the maximum steering wheel rotational angular velocity and the maximum steering wheel rotational angular acceleration, so that a data basis is provided for establishing a judgment logic for mistaken stepping and mistaken acceleration of a driver;
s3: monitoring whether a driver operates the longitudinal control pedal by mistake in real time: the method comprises the steps that the age, the driving age, the height and the weight of a driver are collected before the driver starts to drive an automobile, the speed of the automobile and the distance between the automobile and a front obstacle are monitored in real time in the running process of the automobile, and the longitudinal control pedal angular acceleration a, the longitudinal control pedal pressure h, the steering wheel hand holding force f, the steering wheel torque m, the steering wheel rotation angle theta, the steering wheel rotation angular speed omega and the steering wheel rotation angular acceleration beta are obtained; simultaneously, calculating the maximum value a of the longitudinal control pedal angular acceleration of the driver under the sudden acceleration working condition through a support vector regression model 1 Maximum value h of longitudinal control pedal pressure 1 Maximum value f of hand grip force of steering wheel 1 Maximum steering wheel torque m 1 Maximum value theta of steering wheel angle 1 Maximum value omega of steering wheel rotation angular velocity 1 And steering wheel rotational angular acceleration maximum value beta 1 And a maximum value a of the angular acceleration of the longitudinal control pedal under the condition of encountering an obstacle and misoperation of the longitudinal control pedal 2 Maximum value h of longitudinal control pedal pressure 2 Maximum value f of hand grip force of steering wheel 2 Maximum steering wheel torque m 2 Maximum value theta of steering wheel angle 2 Maximum value omega of steering wheel rotation angular velocity 2 And steering wheel rotational angular acceleration maximum value beta 2 ；
The system determines that the driver has misoperated the longitudinal control pedal and activates the auxiliary brake module when either:
case one: the following seven conditions are satisfied simultaneously for four or more:
①a 1 ≤a＜a 2 ；
②h 1 ≤h＜h 2 ；
③f 1 ≤f＜f 2 ；
④m 1 ≤m＜m 2 ；
⑤θ 1 ≤θ＜θ 2 ；
⑥ω 1 ≤ω＜ω 2 ；
⑦β 1 ≤β＜β 2 ；
and a second case: the following seven conditions are satisfied:
①a≥a 2 ；
②h≥h 2 ；
③f≥f 2 ；
④m≥m 2 ；
⑤θ≥θ 2 ；
⑥ω≥ω 2 ；
⑦β≥β 2 ；
the auxiliary braking module starts working after the system judges that a driver operates the longitudinal control pedal by mistake, the automobile is changed into a braking mode from a single-pedal driving mode, the automobile receives a braking signal when the driver downwards steps on the longitudinal control pedal, the braking executing mechanism starts executing and cuts off an accelerating signal at the same time, when the longitudinal control pedal pressure and the automobile speed are both 0, the auxiliary braking module is restored, the automobile is restored to the single-pedal driving mode, and then the driver can normally drive the automobile, the automobile is accelerated when stepping on the longitudinal control pedal, and the automobile is decelerated when the longitudinal control pedal is released.
2. The machine learning-based single pedal chassis-by-wire vehicle auxiliary braking method according to claim 1, characterized in that in said step S2, it comprises the steps of:
integrating sample data into a data set d= { (x) 1 ,y 1 ),(x 2 ,y 2 )...,(x n ,y n )}，The training samples are mapped from the low-dimensional space to the high-dimensional space through nonlinear mapping, and a linear regression model established in the high-dimensional space can be expressed as the following equation:
f(x)＝w·Φ(x)+b
where x is the input variable, Φ (x) is a nonlinear function mapping x to a high-dimensional linear space, w is a weight vector, b is a bias,
to minimize regression errors, the objective function of the support vector regression algorithm may be expressed as follows:
wherein C is p For punishment coefficient, represent punishment degree of model to sample with error greater than epsilon in training process, l ε For the epsilon-insensitive loss function, epsilon represents the insensitive loss coefficient, epsilon is smaller to represent the smaller error requirement of the regression function, l ε The expression can be expressed as the following equation:
wherein z represents the error of the fitting value and the true value of the support vector regression algorithm;
in case of data disagreement with l ε When the constraint of (z) is satisfied, a relaxation variable delta is introduced i ，δ i * To correct the irregular factor, after which the following equation can be obtained:
by introduction ofLagrangian multiplier alpha i 、α i * Simplifying the calculation, converting the above formula into alpha i ，α i * Is a dual problem:
wherein K (x i ,x j ) The application selects RBF kernel function, which is defined as the following equation:
K(x i ,x j )＝exp(-γ||x i -x j || 2 )
wherein, gamma is a nuclear parameter;
the solution to the regression function f (x) according to the karman-coulen-tak condition can be expressed as:
based on the above method, the driver behavior model may be abstracted as:
y＝f(x|(C p ,ε,γ))。
3. the machine learning-based single pedal drive-by-wire chassis automobile auxiliary braking method according to claim 2, wherein three super parameters of a support vector regression model, namely penalty coefficients C, are calculated by using an adaptive particle swarm algorithm p And optimizing the core parameter gamma and the insensitive loss coefficient epsilon.
4. The machine learning-based single pedal drive-by-wire chassis automobile auxiliary braking method of claim 3, wherein the average absolute percentage error MAPE capable of directly reflecting regression performance is selected as the fitness function fitness of the adaptive particle swarm algorithm, namely:
where n is the number of sample data, y i Is a predicted value, f (x i ) Is an experimental value.
5. The machine learning based single pedal chassis-by-wire vehicle auxiliary braking method according to claim 1, wherein in step S2, the sample data is normalized according to the following equation:
wherein X is min Is the minimum value of the sample data; x is X max Is the maximum value of the sample data; x is sample data; x' is normalized data, ranging from [0,1]；
After the normalization processing of the sample data is completed, the normalized data is divided into two parts, wherein 80% of the normalized data are classified into a training data set, and 20% of the normalized data are classified into a test data set.
6. The machine learning based single pedal chassis-by-wire vehicle auxiliary braking method of claim 1, wherein in said step S2, a trained support vector regression model is used to predict a longitudinal control pedal angular acceleration maximum, a longitudinal control pedal pressure maximum, a steering wheel hand grip maximum, a steering wheel torque maximum, a steering wheel angle maximum, a steering wheel rotational angular velocity maximum, and a steering wheel rotational angular acceleration maximum, using a mean square error MSE and a decision coefficient R 2 Evaluating the prediction result of the model:
wherein,is the average of the predicted values,/>Is the average of the experimental values.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210179569.3A CN114407847B (en) | 2022-02-25 | 2022-02-25 | Single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210179569.3A CN114407847B (en) | 2022-02-25 | 2022-02-25 | Single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114407847A CN114407847A (en) | 2022-04-29 |
| CN114407847B true CN114407847B (en) | 2023-11-17 |
Family
ID=81261177
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210179569.3A Active CN114407847B (en) | 2022-02-25 | 2022-02-25 | Single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114407847B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102010045694A1 (en) * | 2010-09-16 | 2012-03-22 | Daimler Ag | Method for preventing collision of e.g. motor car with obstacle during parking in tight parking lot, involves determining whether collision of vehicle is threatened in predeterminable period based on application of simulation evaluation |
| CN106274483A (en) * | 2016-11-10 | 2017-01-04 | 合肥工业大学 | The Vehicular automatic driving switching device differentiated based on driving behavior of diverting one's attention and method |
| KR20170137427A (en) * | 2016-06-03 | 2017-12-13 | 현대자동차주식회사 | Safety apparatus and method for operating pedal of autonomous driving vehicle |
| CN108909526A (en) * | 2018-06-26 | 2018-11-30 | 北京新能源汽车股份有限公司 | A kind of single pedal driving mode method for slowing-down control, device and electric car |
| CN109334564A (en) * | 2018-09-11 | 2019-02-15 | 南京航空航天大学 | A kind of automobile active safety early warning system of anticollision |
| CN112477865A (en) * | 2020-12-02 | 2021-03-12 | 南京航空航天大学 | Regenerative braking control method based on driver braking behavior prediction |
| CN112622614A (en) * | 2020-12-18 | 2021-04-09 | 的卢技术有限公司 | Single-pedal control system and control method thereof |
| CN113386768A (en) * | 2021-08-02 | 2021-09-14 | 合肥工业大学 | Single-pedal nonlinear model prediction control method for pure electric vehicle |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP7458141B2 (en) * | 2018-05-22 | 2024-03-29 | ヤンマーパワーテクノロジー株式会社 | Work vehicles |
-
2022
- 2022-02-25 CN CN202210179569.3A patent/CN114407847B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102010045694A1 (en) * | 2010-09-16 | 2012-03-22 | Daimler Ag | Method for preventing collision of e.g. motor car with obstacle during parking in tight parking lot, involves determining whether collision of vehicle is threatened in predeterminable period based on application of simulation evaluation |
| KR20170137427A (en) * | 2016-06-03 | 2017-12-13 | 현대자동차주식회사 | Safety apparatus and method for operating pedal of autonomous driving vehicle |
| CN106274483A (en) * | 2016-11-10 | 2017-01-04 | 合肥工业大学 | The Vehicular automatic driving switching device differentiated based on driving behavior of diverting one's attention and method |
| CN108909526A (en) * | 2018-06-26 | 2018-11-30 | 北京新能源汽车股份有限公司 | A kind of single pedal driving mode method for slowing-down control, device and electric car |
| CN109334564A (en) * | 2018-09-11 | 2019-02-15 | 南京航空航天大学 | A kind of automobile active safety early warning system of anticollision |
| CN112477865A (en) * | 2020-12-02 | 2021-03-12 | 南京航空航天大学 | Regenerative braking control method based on driver braking behavior prediction |
| CN112622614A (en) * | 2020-12-18 | 2021-04-09 | 的卢技术有限公司 | Single-pedal control system and control method thereof |
| CN113386768A (en) * | 2021-08-02 | 2021-09-14 | 合肥工业大学 | Single-pedal nonlinear model prediction control method for pure electric vehicle |
Non-Patent Citations (2)
| Title |
|---|
| 基于双层隐式马尔科夫模型的驾驶意图辨识;宗长富;王畅;何磊;郑宏宇;张泽星;;汽车工程(第08期);全文 * |
| 驾驶员对汽车方向的自适应控制行为建模;段立飞;高振海;王德平;机械工程学报;第47卷(第8期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114407847A (en) | 2022-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP6214730B2 (en) | Vehicle control system | |
| EP2177413B1 (en) | Vehicle control system | |
| JP4478037B2 (en) | Vehicle control device | |
| CN108297877A (en) | Control method for vehicle, system and device | |
| CN106364365A (en) | Processing method and device for double-way accelerator pedal signals | |
| CN113071326A (en) | Method and system for distributing regenerated energy recovery torque of four-wheel drive new energy automobile | |
| US11858525B2 (en) | Chassis-by-wire cyber physical system in intelligent traffic environment, and control method | |
| CN109606460B (en) | Steering wheel connection decision method of double-rudder driving automobile based on steer-by-wire | |
| CN112519793A (en) | Intelligent drive-by-wire chassis architecture based on digital twin and active fault tolerance method thereof | |
| CN106864436A (en) | A kind of automatic driving vehicle brakes and braking method | |
| US20190144027A1 (en) | Control method for electric power steering and control system for electric power steering | |
| CN107585207A (en) | A kind of vehicle line traffic control four-wheel steering system and its control method | |
| CN114407847B (en) | Single-pedal drive-by-wire chassis automobile auxiliary braking method based on machine learning | |
| CN114802205A (en) | System and method for vehicle turning radius reduction | |
| CN106394668A (en) | Automobile steering mode switching method and system and automobile | |
| US11724739B2 (en) | Vehicle actuation commands to affect transient handling | |
| CN110412951A (en) | A kind of pure electric vehicle controller application layer platform software model framework | |
| CN115421473A (en) | Simulation verification method and system for automobile driving function | |
| CN103253274B (en) | First driver expects the method and control system of the credibility Analysis of sensor | |
| CN114670668B (en) | Torque control method and device and electric automobile | |
| Zhu et al. | Research on energy regenerative braking of electric vehicle based on functional safety analysis | |
| CN113561783B (en) | Energy recovery control method and device of distributed driving system and electric automobile | |
| US11625959B2 (en) | System and method for determining backlash in driving system of vehicle | |
| CN115092108B (en) | Integral architecture auxiliary brake control system | |
| CN115742754B (en) | Torque detection and fault processing method and system for pure electric automobile |
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 |