CN107161150A - A kind of method and device of control vehicle start - Google Patents
A kind of method and device of control vehicle start Download PDFInfo
- Publication number
- CN107161150A CN107161150A CN201710455533.2A CN201710455533A CN107161150A CN 107161150 A CN107161150 A CN 107161150A CN 201710455533 A CN201710455533 A CN 201710455533A CN 107161150 A CN107161150 A CN 107161150A
- Authority
- CN
- China
- Prior art keywords
- starting
- driver
- value
- ratio
- opening
- 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.)
- Granted
Links
Classifications
-
- 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
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/18—Propelling the vehicle
- B60W30/18009—Propelling the vehicle related to particular drive situations
- B60W30/18027—Drive off, accelerating from standstill
-
- 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
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/02—Conjoint control of vehicle sub-units of different type or different function including control of driveline clutches
-
- 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
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/04—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units
- B60W10/06—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units including control of combustion engines
-
- 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
- B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- 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
- B60W2540/00—Input parameters relating to occupants
- B60W2540/10—Accelerator pedal position
-
- 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
- B60W2540/00—Input parameters relating to occupants
- B60W2540/14—Clutch pedal position
Abstract
The invention provides a kind of method and device of control vehicle start, wherein, this method is included before vehicle start, and the opening value of driver's bend the throttle is repeatedly obtained in predetermined time period；According to the multiple aperture values pre-set, determine to be more than or equal to the acquisition time length and the ratio of predetermined time period of all opening values of each aperture value respectively；According to the ratio for the multiple time spans determined, vehicle start is controlled.In the present invention, used acquisition time length can reflect that the starting of driver is intended to, therefore, when controlling vehicle start, consider that the starting of driver is intended to, so that the combination of the air throttle and automatic clutch of engine matches with driver to the operation of accelerator pedal during starting, so that the phenomenon for occurring play when reducing vehicle start, fluttering, realizes the smooth-going starting of vehicle.
Description
Technical field
The present invention relates to automobile technical field, in particular to a kind of method and device of control vehicle start.
Background technology
In vehicle start, especially equipped with automatic mechanical transmission (Automated Mechanical
Transmission, AMT) vehicle starting when, phase interworking is wanted in the combination of the throttle opening and automatic clutch of engine
Close, so as to realize smooth-going starting.
In the prior art, in vehicle start, it can only typically be started to walk according to a kind of predetermined pattern, so so that starting
The starting that pattern cannot respond to driver is intended to, i.e. the combination of the throttle opening and automatic clutch of engine cannot respond to drive
Operation of the person of sailing to accelerator pedal, so that the problems such as vehicle easily vehicle play occurs in starting, fluttered, so as to cause
Smooth-going starting can not be realized.
To sum up, in the prior art, vehicle is in starting, the problems such as vehicle play easily occur, flutter, so as to lead to not
Realize the smooth-going starting of vehicle.
The content of the invention
In view of this, the purpose of the embodiment of the present invention is to provide a kind of method and device of control vehicle start, to solve
Certainly or attempt alleviate above mentioned problem.
In a first aspect, the embodiments of the invention provide a kind of method for controlling vehicle start, wherein, methods described includes：
Before vehicle start, the opening value of driver's bend the throttle is repeatedly obtained in predetermined time period；
According to the multiple aperture values pre-set, determine that being more than or equal to each described all of aperture value opens respectively
The acquisition time length of angle value and the ratio of the predetermined time period；
According to the multiple ratios determined, the vehicle start is controlled.
With reference in a first aspect, the embodiments of the invention provide the possible implementation of the first of above-mentioned first aspect, its
In, it is described according to the multiple aperture values pre-set, determine that being more than or equal to each described all of aperture value opens respectively
The acquisition time length of angle value and the ratio of the predetermined time period, including：
According to the multiple aperture values pre-set, determine that being more than or equal to each described all of aperture value opens respectively
The number of angle value；
The ratio of each number and the number of the opening value of acquisition in the predetermined time period is calculated respectively；
The ratio of the number is defined as to the ratio of the acquisition time length and the predetermined time period.
With reference in a first aspect, the embodiments of the invention provide the possible implementation of second of above-mentioned first aspect, its
In, multiple ratios that the basis is determined control the vehicle start, including：
According to the ratio and the weight conversion coefficient of the neural network model pre-established, the starting meaning of driver is calculated
Map values；
According to the starting intention value of the driver, the starting mode of the vehicle is determined；
The vehicle is controlled to be started to walk according to the starting mode.
With reference to second of possible implementation of first aspect, the embodiments of the invention provide the of above-mentioned first aspect
Three kinds of possible implementations, wherein, the starting intention value according to the driver determines the starting mould of the vehicle
Formula, including：
Select the corresponding starting mode of starting intention value with the driver；
Or,
Mapping between being intended to according to the starting intention value of the driver, and starting intention value set in advance and starting
Relation, determines that the starting of the driver is intended to；
Selection and the starting of the driver are intended to the starting mode matched.
With reference to the third possible realization of second of the possible implementation or combination first aspect of first aspect
Mode, the embodiments of the invention provide the possible implementation of the 4th of above-mentioned first aspect kind, wherein, built according to following steps
Found the neural network model：
Set up the input layer, hidden layer and output layer of the neural network model；
The sample set of training in advance is obtained, the sample set of the training in advance includes multiple training samples, each instruction
Practicing sample includes multiple opening values of accelerator pedal and the starting intention value of driver；
According to multiple opening values of the accelerator pedal and the starting intention value of the driver, the output layer is determined extremely
The first of the hidden layer changes weight coefficient and the hidden layer to the second conversion weight coefficient of the input layer；
According to described first conversion weight coefficient, it is described second conversion weight coefficient, the input layer, the hidden layer and
The output layer, sets up the neural network model.
Second aspect, the embodiments of the invention provide a kind of device of control vehicle start, wherein, described device includes：
First acquisition module, before vehicle start, repeatedly obtains the acceleration that driver steps in predetermined time period
The opening value of pedal；
First determining module, for according to the multiple aperture values pre-set, determining to be more than or equal to each institute respectively
State the acquisition time length and the ratio of predetermined time period of all opening values of aperture value；
Control module, for according to the multiple ratios determined, controlling the vehicle start.
With reference to second aspect, the embodiments of the invention provide the possible implementation of the first of above-mentioned second aspect, its
In, first determining module includes：
First determining unit, for according to the multiple aperture values pre-set, determining to be more than or equal to each institute respectively
State the number of all opening values of aperture value；
First computing unit, for the opening value for calculating each number respectively with being obtained in the predetermined time period
Number ratio；
Second determining unit, for the ratio of the number to be defined as into the acquisition time length and the preset time
The ratio of length.
With reference to second aspect, the embodiments of the invention provide the possible implementation of second of above-mentioned second aspect, its
In, the control module includes：
Second computing unit, for the weight conversion coefficient according to the ratio and the neural network model pre-established,
Calculate the starting intention value of driver；
3rd determining unit, for the starting intention value according to the driver, determines the starting mode of the vehicle；
Control unit, for controlling the vehicle to be started to walk according to the starting mode.
With reference to second of possible implementation of second aspect, the embodiments of the invention provide the of above-mentioned first aspect
Three kinds of possible implementations, wherein, the 3rd determining unit specifically for,
Select with the starting of the driver with the corresponding starting mode of drawing；
Or,
Mapping between being intended to according to the starting intention value of the driver, and starting intention value set in advance and starting
Relation, determines that the starting of the driver is intended to；
Selection and the starting of the driver are intended to the starting mode matched.
With reference to the third possible realization of second of the possible implementation or combination second aspect of second aspect
Mode, the embodiments of the invention provide the possible implementation of the 4th of above-mentioned second aspect kind, wherein, described device is also wrapped
Include：
First sets up module, input layer, hidden layer and output layer for setting up the neural network model；
Second acquisition module, the sample set for obtaining training in advance, the sample set of the training in advance includes multiple instructions
Practice sample, each training sample includes multiple opening values of accelerator pedal and the starting intention value of driver；
Second determining module, the starting for multiple opening values according to the accelerator pedal and the driver is intended to
Value, determines that the output layer changes weight coefficient and the hidden layer to the second of the input layer to the first of the hidden layer
Change weight coefficient；
Second sets up module, for according to the described first conversion weight coefficient, the second conversion weight coefficient, described defeated
Enter layer, the hidden layer and the output layer, set up the neural network model.
In the method and device of control vehicle start provided in an embodiment of the present invention, used acquisition time length can
It is intended to the starting for reflecting driver, therefore, when controlling vehicle start, it is considered to which the starting of driver is intended to so that during starting
The combination of the air throttle and automatic clutch of engine matches with driver to the operation of accelerator pedal, so as to reduce vehicle
The phenomenon for occur play during starting, fluttering, realizes the smooth-going starting of vehicle.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be attached to what is used required in embodiment
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, therefore is not construed as pair
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, can also be according to this
A little accompanying drawings obtain other related accompanying drawings.
Fig. 1 shows the flow chart of the method for the control vehicle start that the embodiment of the present invention is provided；
Fig. 2 is shown in the method for the control vehicle start that the embodiment of the present invention is provided, and controls the flow of vehicle start
Figure；
Fig. 3 shown in the method for the control vehicle start that the embodiment of the present invention provided, throttle opening and clutch
With reference to versus time curve；
Fig. 4 shows the first structural representation for the control vehicle start that the embodiment of the present invention is provided；
Fig. 5 shows second of structural representation of the control vehicle start that the embodiment of the present invention is provided.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
Middle accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only
It is a part of embodiment of the invention, rather than whole embodiments.The present invention being generally described and illustrated herein in the accompanying drawings is real
Applying the component of example can be arranged and be designed with a variety of configurations.Therefore, it is of the invention to what is provided in the accompanying drawings below
The detailed description of embodiment is not intended to limit the scope of claimed invention, but is merely representative of the selected reality of the present invention
Apply example.Based on embodiments of the invention, the institute that those skilled in the art are obtained on the premise of creative work is not made
There is other embodiment, belong to the scope of protection of the invention.
It is considered that in the prior art, vehicle is in starting, the problems such as vehicle play easily occur, flutter, so as to cause nothing
Method realizes the smooth-going starting of vehicle.Based on this, the embodiments of the invention provide a kind of method and device of control vehicle start, under
Face is described by embodiment.
With reference to shown in Fig. 1, the embodiments of the invention provide a kind of method for controlling vehicle start, the method comprising the steps of
S110-S130, it is specific as follows：
S110, before vehicle start, repeatedly obtains the opening value of driver's bend the throttle in predetermined time period；
S120, according to the multiple aperture values pre-set, is determined all more than or equal to each aperture value respectively
The acquisition time length and the ratio of predetermined time period of opening value.
S130, according to the multiple ratios determined, controls vehicle start.
Wherein, when driver's bend the throttle, the aperture of accelerator pedal can gradually increase, and acceleration is stepped in driver
Before pedal, the aperture of accelerator pedal is zero.
In embodiments of the present invention, position sensor is set on the accelerator pedal of vehicle, the position sensor is used for
The opening value of accelerator pedal is gathered, specifically, position sensor collection is magnitude of voltage on accelerator pedal, the magnitude of voltage is used for
The aperture of accelerator pedal is characterized, according to the transformational relation between magnitude of voltage and the opening value of accelerator pedal, each electricity can be obtained
Pressure is worth corresponding opening value, therefore, from position sensor export be accelerator pedal opening value.
When electricity on vehicle, position sensor starts to gather the opening value of accelerator pedal.
The executive agent of the embodiment of the present invention is AMT controller.
In above-mentioned steps S110, before vehicle start, and on vehicle after electricity, in predetermined time period internal controller meeting
Repeatedly obtain the opening value of accelerator pedal that driver steps on, specifically, controller at interval of certain time from position sensor
Obtain the opening value of an accelerator pedal.
Wherein, above-mentioned can be interval 1ms or other numerical value at interval of certain time, above-mentioned interval it is specific
Time numerical value can be configured according to practical application scene, the specific number of the embodiment of the present invention not to above-mentioned interlude
Value is defined.
It is preferred that, above-mentioned predetermined time period is 600ms, and 600ms time is long since driver's bend the throttle
In degree.
In above-mentioned steps S120, according to the multiple aperture values pre-set, determine to be more than or equal to respectively each to open
The acquisition time length and the ratio of predetermined time period of all opening values of value are spent, is specifically included：
According to the multiple aperture values pre-set, determine that all apertures more than or equal to each aperture value take respectively
The number of value；
The ratio of above-mentioned each number and the number of the opening value of acquisition in predetermined time period is calculated respectively；
The ratio of above-mentioned number is defined as to the ratio of above-mentioned acquisition time length and predetermined time period.
In above-mentioned steps S120, the multiple aperture values pre-set can be 10,15 or 20, specific
Number can be configured according to practical application scene, and the embodiment of the present invention is not to the specific of the above-mentioned aperture value pre-set
Number is defined.
General, vehicle is in starting, and the maximum opening value for the accelerator pedal that driver generally steps on is 20%, therefore,
In embodiments of the present invention, 0~20% average division can be obtained multiple aperture values, each aperture value correspondence
One grade, can so determine the corresponding aperture value of each grade.Such as, 0~20% is averagely divided into 10
Grade, then the corresponding value of first grade can be 2%, and the corresponding value of second grade is 4%, Three Estate correspondence
Value be 6%, by that analogy, determine the corresponding aperture value of each grade.
Certainly, it is above-mentioned simply to list a kind of method to set up of aperture value, in embodiments of the present invention, can by 0~
20% average division, can also be one of which embodiment by 0~30% average division, above-mentioned 20%, in actual applications,
Any number more than 20% can also be chosen as the above-mentioned maximum opening value divided.Specifically, being drawn
Timesharing, the maximum opening value when progress of selection is divided can be set according to practical application, and the embodiment of the present invention is not right
It is defined.
In above-mentioned steps S120, corresponding in all opening values more than or equal to each aperture value adopt is determined respectively
Collect time span, can in the following way realize, specifically include：
Still illustrated herein by taking the corresponding aperture value of the above-mentioned grade averagely divided as an example, first grade is corresponding
Aperture value is 2%, and the corresponding aperture value of second grade is 4%, the corresponding aperture value 6% of Three Estate, with this
Analogize, it may be determined that go out the corresponding aperture value of each grade.
In embodiments of the present invention, such as, obtain 10 opening values in step s 110, respectively 1%, 1.2%,
2%th, 2.1%, 2.5%, 3%, 3.6%, 4.5%, 5.2%, 5.6%, the aperture value set according to the example above then can be with
Determine all opening values more than or equal to first aperture value and have 2%, 2.1%, 2.5%, 3%, 3.6%, 4.5%,
5.2%th, 5.6% this eight, have 4.5%, 5.2%, 5.6%, greatly more than or equal to all opening values of second aperture value
In or equal to 2% opening value totally eight, and above-mentioned preset time in obtain opening value number be 10, therefore, it can
8/10=0.8 is defined as being more than or equal to the acquisition time length and preset time of all opening values of first aperture value
The ratio of length.When calculating the collection more than or equal to all opening values of each aperture value respectively in the same way
Between length and predetermined time period ratio, by the above-mentioned means, multiple ratios can be determined.
Certainly, it is above-mentioned simply to list the method that one of which determines above-mentioned ratio, in addition to this it is possible to by as follows
Mode is realized：
Such as, obtain 10 opening values in step s 110, respectively 1%, 1.2%, 2%, 2.1%, 2.5%,
3%th, 3.6%, 4.5%, 5.2%, 5.6%, the aperture value set according to the example above can then be determined and is more than or equal to
All opening values of first aperture value have 2%, 2.1%, 2.5%, 3%, 3.6%, 4.5%, 5.2%, 5.6% this eight
It is individual, have 4.5%, 5.2%, 5.6% more than or equal to all opening values of second aperture value, opening more than or equal to 2%
Angle value totally eight, it is assumed that the acquisition time between each opening value can be then determined more than or equal to first at intervals of 1ms
The acquisition time length of aperture value is 8ms, calculates the ratio of 8ms and predetermined time period.
Therefore, a series of ratio can be obtained in above-mentioned steps S120.
With reference to shown in Fig. 2, in step s 130, according to the multiple ratios determined, vehicle start, including step are controlled
S210-S230, it is specific as follows：
S210, according to above-mentioned ratio and the weight conversion coefficient of the neural network model pre-established, calculates driver's
Start to walk intention value；
S220, according to the starting intention value of driver, determines the starting mode of vehicle；
S230, control vehicle is started to walk according to above-mentioned starting mode.
In embodiments of the present invention, the above-mentioned neural network model pre-established is BP (Back Propagation) nerves
Network model.
Neural network model in the embodiment of the present invention is divided into input layer, hidden layer and the part of output layer three, input layer bag
Containing multiple input nodes, hidden layer includes multiple input nodes, and the output node of output layer is also to be multiple, it is preferred that the present invention
The output node of the output layer of neural network model used in embodiment is two, certainly, the tool of every layer of node number
Body numerical value can be configured according to actual conditions.
In embodiments of the present invention, according to the weight conversion coefficient of above-mentioned a series of ratio and neural network model, lead to
Equation below is crossed, the starting intention value of driver is calculated：
Wherein, in above-mentioned formula, wijRepresent the conversion between the input node i and concealed nodes j of neural network model
Weight coefficient, vjkThe conversion weight coefficient between the concealed nodes and output node of neural network model is represented, what i was represented is defeated
Enter the input node of layer, i=1,2 ..., N, what N was represented is the number of above-mentioned ratio, what j was represented is the node of hidden layer, hide
The number of the node of layer can be set according to actual conditions, in embodiments of the present invention, and that M is represented is for hiding node layer
Number, j=1,2 ..., M, α is the threshold value of hidden layer set in advance, and β is the threshold value of output layer set in advance, piRepresent be
The above-mentioned ratio calculated, siWhat is represented is the activation value for hiding node layer, okTo export the activation value of node layer, bjRepresent
It is the output valve of hidden layer, ykWhat is represented is the output valve of output layer.
In above-mentioned formula, function f is defined as follows：
In embodiments of the present invention, the output valve y of above-mentioned output layerkWhat is represented is then the starting intention value of driver, and
And, in embodiments of the present invention, k value is that the output valve of 1 and 2, i.e. neural network model is [y1,y2]。
Specifically, the above-mentioned starting intention value according to driver, determines the starting mode of vehicle, specifically includes the following two kinds
Situation：
The first situation,
Select the corresponding starting mode of starting intention value with driver；
In the case of in this, the mapping relations between starting intention value and vehicle start pattern are pre-established with, so, when
After the starting intention value for determining driver, then according to the starting intention value, search and anticipated with the starting from above-mentioned mapping relations
The corresponding vehicle start pattern of map values, the starting mode found out as vehicle starting mode.
Second of situation,
Mapping between being intended to according to the starting intention value of driver, and starting intention value set in advance and starting is closed
System, determines that the starting of driver is intended to；
Selection and the starting of driver are intended to the starting mode matched.
In that case, what is pre-established is the mapping relations that starting is intended between starting mode, therefore, is needed first
According to the starting intention value of the driver calculated, to determine that the starting of the corresponding driver of the starting intention value is intended to, tool
Body, the starting of driver is intended to be intended to including slow starting intention, medium starting and suddenly starting is intended to, different [y1,y2] value pair
Different startings are answered to be intended to, such as, can be that the slow starting of [0,1] correspondence is intended to, the medium starting of [1,0] correspondence is intended to, [1,1]
Emergent starting is intended to, [0,0] correspondence is intended to without starting.
In embodiments of the present invention, the mapping relations between starting intention value and starting intention are preset with, according to meter
The starting intention value and the mapping relations calculated, it may be determined that the starting for going out driver is intended to, because the starting calculated is intended to
Value may have decimal digits, therefore, use the mode rounded up to round the starting intention value calculated for 0 or 1.
After the starting for determining driver is intended to, be intended to according to the starting of driver and the starting that pre-establishes be intended to and
Mapping relations between starting mode, determine that the starting with driver is intended to the starting mode matched.
After determining that the starting with driver is intended to the starting mode matched, control vehicle enters according to the starting mode
Row starting.
It is that the corresponding throttle opening of every kind of starting intention is combined the song that changes with time with clutch with reference to shown in Fig. 3
Line, Fig. 3 (a) is intended to corresponding throttle opening and clutch combination versus time curve for slow starting, during Fig. 3 (b) is
It is intended to corresponding throttle opening and clutch versus time curve Deng starting, Fig. 3 (c) is intended to corresponding section for anxious starting
Valve opening and clutch versus time curve.
In addition, in method provided in an embodiment of the present invention, multiple aperture grades can also be pre-set, and often
The corresponding span of individual aperture grade, determines to include in each aperture grade according to the corresponding span of each aperture grade
Opening value；According to the opening value included in each aperture grade, it is determined that the minimum opening value in each aperture grade is corresponding
The second acquisition time length between moment and highest opening value corresponding collection moment is gathered, the second acquisition time is calculated afterwards
The second ratio between length and default acquisition time length, regard second ratio as the neural network model pre-established
Input value, determines that the starting of driver is intended to, and be intended to control vehicle start according to the starting of driver.
Specifically, the neural network model that the embodiment of the present invention is used as follows set up by (1)-step (4)
Arrive：
(1) input layer, hidden layer and output layer of neutral net are set up；
(2) sample set of training in advance is obtained, the sample set of the training in advance includes multiple training samples, each trains sample
This multiple opening value for including accelerator pedal and the starting intention value of driver；
(3) according to multiple opening values of above-mentioned accelerator pedal and the starting intention value of driver, determine output layer to hide
The first of layer changes weight coefficient and hidden layer to the second conversion weight coefficient of input layer；
(4) according to the above-mentioned first conversion weight coefficient, the second conversion weight coefficient, input layer, hidden layer and output layer, build
Found above-mentioned neural network model.
In embodiments of the present invention, the input layer of the neural network model of foundation, hidden layer and output layer, the present invention are implemented
In example, foundation is three-layer neural network model, and afterwards, then the sample set obtained in being tested according to vehicle start is to above-mentioned nerve
Output layer in network model changes weight coefficient to the first of hidden layer, and hidden layer is to the second conversion weight system of output layer
Number is calibrated, and determines that first of the neural network model employed in the embodiment of the present invention changes weight coefficient and second turn
Weight coefficient is changed, detailed process is as follows：
Sample set is obtained by carrying out substantial amounts of vehicle start experiment first, the sample set includes multiple training samples,
Acceleration when starting intention value and driver of each training sample including driver are operated in starting to accelerator pedal is stepped on
The opening value of plate；
The opening value of above-mentioned accelerator pedal is handled using with above-mentioned steps S120 identicals method, obtains a series of
Acquisition time length and starting when obtain accelerator pedal opening value whole acquisition time length between ratio, by this one
Serial ratio is defined as the input value of neutral net, can determine that input value and output valve (drive according to each training sample
Member starting intention value) a corresponding relation, specifically, the input value of neutral net can be designated asThe output valve of neutral net is designated asWherein, what r was represented is r-th of instruction
Practice sample, in embodiments of the present invention, the number of the input node of the input layer of setting is 10, the output node of output layer
Number is 2.
Afterwards, the activation value s for hiding node layer is calculated by equation belowjWith output valve bj；
Wherein, in above-mentioned formula,What is represented is swashing for the hiding node layer that is calculated using r-th of training sample
Value living,What is represented is the output valve of the concealed nodes calculated using r-th of training sample, and what j was represented is hidden layer
Output node, what i was represented is the input node of hidden layer, αrWhat is represented is the threshold value of hidden layer set in advance,Represent
It is conversion weight coefficient of the hidden layer to input layer.
The activation value o of output node layer is calculated by equation belowkWith output valve yk；
Wherein, in above-mentioned formula,What is represented is to export swashing for node layer using what r-th of training sample was calculated
Value living,What is represented is the output valve of the output node layer calculated using r-th of training sample, and what k was represented is output layer
Output node, βrWhat is represented is the threshold value of output layer set in advance,What is represented is conversion right of the output layer to hidden layer
Weight coefficient.
Afterwards, the correction error of output layer and hidden layer is calculated by equation below：
Wherein, in above-mentioned formulaWhat is represented is the correction error of the output layer calculated using r-th of training sample,Represent be the hidden layer calculated using r-th of training sample correction error, f ' expressions be function f derivative.
Next, calculating output layer to the first of hidden layer correcting value and the output for changing weight coefficient by equation below
The correcting value of layer threshold value；
Wherein, in above-mentioned formula,What is represented is the first conversion weight system calculated using r-th of training sample
Several correcting values, λ1What is represented is feedback correcting coefficient of the output layer to hidden layer, Δ βrWhat is represented is the correction of output layer threshold value
Amount.
Hidden layer is calculated to the second conversion weight coefficient and the correcting value of hidden layer threshold value of input layer by equation below；
Wherein, in above-mentioned formula,What is represented is the second conversion weight system calculated using r-th of training sample
Several correcting values, λ2What is represented is feedback correcting coefficient of the hidden layer to input layer, Δ αrWhat is represented is using r-th of training sample
Originally the correcting value of the hidden layer threshold value calculated.
After above-mentioned training, new the first conversion weight coefficient and the second conversion weight coefficient can be calculated, and
Hidden layer threshold value and output layer threshold value,
Wherein, in above-mentioned formula,What is represented is the second new weight conversion coefficient,What is represented is new the
One weight conversion coefficient, αr+1What is represented is the threshold value of new hidden layer, βr+1What is represented is the threshold value of new output layer.
Finally, judge whether the correction error of above-mentioned hidden layer and output layer meets following conditions：
Wherein, in above-mentioned formula, IthWhat is represented is error upper limit threshold set in advance.
If meeting, training process terminates, then willIt is defined as the second weight conversion coefficient, willIt is defined as first
Weight conversion coefficient, αr+1It is defined as the threshold value of hidden layer, βr+1It is defined as the threshold value of output layer, if it is not satisfied, by above-mentioned sample
Sequence number r adds 1, redefines threshold value, the threshold of hidden layer of the first conversion weight coefficient, the second conversion weight coefficient and output layer
Value.
After the above-mentioned conversion of neural network model first weight coefficient, the second conversion weight coefficient and threshold value is determined, obtain
To above-mentioned neural network model.
The method of control vehicle start provided in an embodiment of the present invention, used acquisition time length can reflect driving
The starting of member is intended to, therefore, when controlling vehicle start, it is considered to which the starting of driver is intended to so that the section of engine during starting
The combination of valve and automatic clutch matches with driver to the operation of accelerator pedal, so as to occur when reducing vehicle start
Play, the phenomenon fluttered, realize the smooth-going starting of vehicle.
With reference to shown in Fig. 4, the embodiment of the present invention additionally provides a kind of device of control vehicle start, and the device is used to perform
The method of control vehicle start provided in an embodiment of the present invention, specifically, the device can be the controller of AMT vehicles, the dress
Put including the first acquisition module 410, the first determining module 420 and control module 430；
Above-mentioned first acquisition module 410, for before vehicle start, repeatedly acquisition driver to step in predetermined time period
Under accelerator pedal opening value；
Above-mentioned first determining module 420, for according to the multiple aperture values pre-set, determining to be more than or equal to respectively
The acquisition time length and the ratio of predetermined time period of all opening values of each aperture value；
Above-mentioned control module 430, for according to the multiple ratios determined, controlling vehicle start.
Wherein, as one embodiment, the ratio that above-mentioned first determining module 420 is determined, be by the first determining unit,
What the first computing unit and the second determining unit were realized, specifically include：
Above-mentioned first determining unit, for according to the multiple aperture values pre-set, determining to be more than or equal to often respectively
The number of all opening values of individual aperture value；Above-mentioned first computing unit, for calculating each above-mentioned number respectively with presetting
The ratio of the number of the opening value obtained in time span；Above-mentioned second determining unit, for the ratio of above-mentioned number to be determined
For above-mentioned acquisition time length and the ratio of predetermined time period.
Specifically, with reference to shown in Fig. 5, in embodiments of the present invention, above-mentioned control module 430 is according to the multiple ratios determined
Value, controls vehicle start, can be realized by the determining unit 432 of the second computing unit 431 the 3rd and control unit 433, specifically
Including：
Above-mentioned second computing unit 431, turns for the weight according to above-mentioned ratio and the neural network model pre-established
Coefficient is changed, the starting intention value of driver is calculated；Above-mentioned second determining unit 432, for the starting intention value according to driver,
Determine the starting mode of above-mentioned vehicle；Above-mentioned control unit 433, for controlling above-mentioned vehicle to be started to walk according to above-mentioned starting mode.
In embodiments of the present invention, above-mentioned 3rd determining unit, controls above-mentioned vehicle to be started to walk according to above-mentioned starting mode, has
Body is realized by the following two kinds mode：
The first situation,
Select the corresponding starting mode of starting intention value with above-mentioned driver；
Second of situation,
Mapping between being intended to according to the starting intention value of driver, and starting intention value set in advance and starting is closed
System, determines that the starting of driver is intended to；
Selection and the starting of driver are intended to the starting mode matched.
Wherein, in embodiments of the present invention, it is to set up module, the second acquisition mould by first to set up neural network model
Block, the second determining module and second set up module realization, specifically include：
Above-mentioned first sets up module, input layer, hidden layer and output layer for setting up above-mentioned neural network model；It is above-mentioned
Second acquisition module, the sample set for obtaining training in advance, the sample set of the training in advance includes multiple training samples, each
Training sample includes multiple opening values of accelerator pedal and the starting intention value of driver；Above-mentioned second determining module, for root
According to the multiple opening values and the starting intention value of driver of above-mentioned accelerator pedal, determine above-mentioned output layer to first turn of hidden layer
Weight coefficient and hidden layer are changed to the second conversion weight coefficient of input layer；Above-mentioned second sets up module, for according to above-mentioned
One conversion weight coefficient, above-mentioned second conversion weight coefficient, above-mentioned input layer, above-mentioned hidden layer and output layer, set up above-mentioned god
Through network model.
In the device of control vehicle start provided in an embodiment of the present invention, used acquisition time length, which can reflect, drives
The starting for the person of sailing is intended to, therefore, when controlling vehicle start, it is considered to which the starting of driver is intended to so that engine during starting
The combination of air throttle and automatic clutch matches with driver to the operation of accelerator pedal, so as to go out when reducing vehicle start
Existing play, the phenomenon fluttered, realize the smooth-going starting of vehicle.
The device for the control vehicle start that the embodiment of the present invention is provided can be the specific hardware in equipment or installation
In software or firmware in equipment etc..The technique effect of the device that the embodiment of the present invention is provided, its realization principle and generation and
Preceding method embodiment is identical, to briefly describe, and device embodiment part does not refer to part, refers in preceding method embodiment
Corresponding contents.It is apparent to those skilled in the art that, for convenience and simplicity of description, described above is
The specific work process of system, device and unit, may be referred to the corresponding process in above method embodiment, no longer go to live in the household of one's in-laws on getting married herein
State.
, can be by others side in embodiment provided by the present invention, it should be understood that disclosed apparatus and method
Formula is realized.Device embodiment described above is only schematical, for example, the division of the unit, only one kind are patrolled
Collect function to divide, there can be other dividing mode when actually realizing, in another example, multiple units or component can combine or can
To be integrated into another system, or some features can be ignored, or not perform.It is another, it is shown or discussed each other
Coupling or direct-coupling or communication connection can be the INDIRECT COUPLING or communication link of device or unit by some communication interfaces
Connect, can be electrical, machinery or other forms.
The unit illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit
The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs
's.
In addition, each functional unit in the embodiment that the present invention is provided can be integrated in a processing unit, also may be used
To be that unit is individually physically present, can also two or more units it is integrated in a unit.
If the function is realized using in the form of SFU software functional unit and is used as independent production marketing or in use, can be with
It is stored in a computer read/write memory medium.Understood based on such, technical scheme is substantially in other words
The part contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are to cause a computer equipment (can be individual
People's computer, server, or network equipment etc.) perform all or part of step of each of the invention embodiment methods described.
And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.
It should be noted that：Similar label and letter represents similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined in individual accompanying drawing, then it further need not be defined and explained in subsequent accompanying drawing, in addition, term " the
One ", " second ", " the 3rd " etc. are only used for distinguishing description, and it is not intended that indicating or implying relative importance.
Finally it should be noted that：Embodiment described above, is only the embodiment of the present invention, to illustrate the present invention
Technical scheme, rather than its limitations, protection scope of the present invention is not limited thereto, although with reference to the foregoing embodiments to this hair
It is bright to be described in detail, it will be understood by those within the art that：Any one skilled in the art
The invention discloses technical scope in, it can still modify to the technical scheme described in previous embodiment or can be light
Change is readily conceivable that, or equivalent substitution is carried out to which part technical characteristic；And these modifications, change or replacement, do not make
The essence of appropriate technical solution departs from the spirit and scope of technical scheme of the embodiment of the present invention.The protection in the present invention should all be covered
Within the scope of.Therefore, protection scope of the present invention described should be defined by scope of the claims.
Claims (10)
1. a kind of method for controlling vehicle start, it is characterised in that methods described includes：
Before vehicle start, the opening value of driver's bend the throttle is repeatedly obtained in predetermined time period；
According to the multiple aperture values pre-set, determine to be more than or equal to all opening values of each aperture value respectively
Acquisition time length and the predetermined time period ratio；
According to the multiple ratios determined, the vehicle start is controlled.
2. according to the method described in claim 1, it is characterised in that described according to the multiple aperture values pre-set, difference
It is determined that more than or equal to the acquisition time length and the predetermined time period of all opening values of each aperture value
Ratio, including：
According to the multiple aperture values pre-set, determine to be more than or equal to all opening values of each aperture value respectively
Number；
The ratio of each number and the number of the opening value of acquisition in the predetermined time period is calculated respectively；
The ratio of the number is defined as to the ratio of the acquisition time length and the predetermined time period.
3. according to the method described in claim 1, it is characterised in that multiple ratios that the basis is determined, control the car
Starting, including：
According to the ratio and the weight conversion coefficient of the neural network model pre-established, the starting for calculating driver is intended to
Value；
According to the starting intention value of the driver, the starting mode of the vehicle is determined；
The vehicle is controlled to be started to walk according to the starting mode.
4. method according to claim 3, it is characterised in that the starting intention value according to the driver, it is determined that
The starting mode of the vehicle, including：
Select the corresponding starting mode of starting intention value with the driver；
Or,
Mapping between being intended to according to the starting intention value of the driver, and starting intention value set in advance and starting is closed
System, determines that the starting of the driver is intended to；
Selection and the starting of the driver are intended to the starting mode matched.
5. the method according to claim 3 or 4, it is characterised in that set up the neural network model according to following steps：
Set up the input layer, hidden layer and output layer of the neural network model；
The sample set of training in advance is obtained, the sample set of the training in advance includes multiple training samples, each training sample
This multiple opening value for including accelerator pedal and the starting intention value of driver；
According to multiple opening values of the accelerator pedal and the starting intention value of the driver, determine the output layer to described
The first of hidden layer changes weight coefficient and the hidden layer to the second conversion weight coefficient of the input layer；
According to the described first conversion weight coefficient, the second conversion weight coefficient, the input layer, the hidden layer and described
Output layer, sets up the neural network model.
6. a kind of device of control vehicle start, it is characterised in that described device includes：
First acquisition module, before vehicle start, repeatedly obtains the accelerator pedal that driver steps in predetermined time period
Opening value；
First determining module, for according to the multiple aperture values pre-set, determining to be more than or equal to each described open respectively
Spend the acquisition time length and the ratio of the predetermined time period of all opening values of value；
Control module, for according to the multiple ratios determined, controlling the vehicle start.
7. device according to claim 6, it is characterised in that first determining module includes：
First determining unit, for according to the multiple aperture values pre-set, determining to be more than or equal to each described open respectively
Spend the number of all opening values of value；
First computing unit, for calculating each number and of the opening value obtained in the predetermined time period respectively
Several ratio；
Second determining unit, for the ratio of the number to be defined as into the acquisition time length and the predetermined time period
Ratio.
8. device according to claim 6, it is characterised in that the control module includes：
Second computing unit, for the weight conversion coefficient according to the ratio and the neural network model pre-established, is calculated
The starting intention value of driver；
3rd determining unit, for the starting intention value according to the driver, determines the starting mode of the vehicle；
Control unit, for controlling the vehicle to be started to walk according to the starting mode.
9. device according to claim 8, it is characterised in that the 3rd determining unit specifically for,
Select with the starting of the driver with the corresponding starting mode of drawing；
Or,
Mapping between being intended to according to the starting intention value of the driver, and starting intention value set in advance and starting is closed
System, determines that the starting of the driver is intended to；
Selection and the starting of the driver are intended to the starting mode matched.
10. device according to claim 8 or claim 9, it is characterised in that described device also includes：
First sets up module, input layer, hidden layer and output layer for setting up the neural network model；
Second acquisition module, the sample set for obtaining training in advance, the sample set of the training in advance includes multiple training samples
This, each training sample includes multiple opening values of accelerator pedal and the starting intention value of driver；
Second determining module, for multiple opening values according to the accelerator pedal and the starting intention value of the driver, really
The fixed output layer changes weight coefficient and the hidden layer to the second conversion of the input layer to the first of the hidden layer
Weight coefficient；
Second sets up module, for according to the described first conversion weight coefficient, the second conversion weight coefficient, the input
Layer, the hidden layer and the output layer, set up the neural network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710455533.2A CN107161150B (en) | 2017-06-16 | 2017-06-16 | A kind of method and device controlling vehicle start |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710455533.2A CN107161150B (en) | 2017-06-16 | 2017-06-16 | A kind of method and device controlling vehicle start |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107161150A true CN107161150A (en) | 2017-09-15 |
| CN107161150B CN107161150B (en) | 2019-08-30 |
Family
ID=59818793
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710455533.2A Active CN107161150B (en) | 2017-06-16 | 2017-06-16 | A kind of method and device controlling vehicle start |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107161150B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111059269A (en) * | 2019-11-20 | 2020-04-24 | 中国第一汽车股份有限公司 | Self-adaptive control method for starting of wet-type double-clutch automatic transmission and vehicle |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103410624A (en) * | 2013-07-26 | 2013-11-27 | 联合汽车电子有限公司 | Start assisting system for manual-gear vehicles |
| CN104627153A (en) * | 2013-11-08 | 2015-05-20 | 北汽福田汽车股份有限公司 | Hill starting control method and system of electric vehicle |
| CN105398450A (en) * | 2011-06-02 | 2016-03-16 | 丰田自动车株式会社 | Control device for vehicle |
| US9495874B1 (en) * | 2012-04-13 | 2016-11-15 | Google Inc. | Automated system and method for modeling the behavior of vehicles and other agents |
| US9896030B2 (en) * | 2015-04-30 | 2018-02-20 | Honda Motor Co., Ltd. | System and method for vehicle collision mitigation with vulnerable road user context sensing |
-
2017
- 2017-06-16 CN CN201710455533.2A patent/CN107161150B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105398450A (en) * | 2011-06-02 | 2016-03-16 | 丰田自动车株式会社 | Control device for vehicle |
| US9495874B1 (en) * | 2012-04-13 | 2016-11-15 | Google Inc. | Automated system and method for modeling the behavior of vehicles and other agents |
| CN103410624A (en) * | 2013-07-26 | 2013-11-27 | 联合汽车电子有限公司 | Start assisting system for manual-gear vehicles |
| CN104627153A (en) * | 2013-11-08 | 2015-05-20 | 北汽福田汽车股份有限公司 | Hill starting control method and system of electric vehicle |
| US9896030B2 (en) * | 2015-04-30 | 2018-02-20 | Honda Motor Co., Ltd. | System and method for vehicle collision mitigation with vulnerable road user context sensing |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111059269A (en) * | 2019-11-20 | 2020-04-24 | 中国第一汽车股份有限公司 | Self-adaptive control method for starting of wet-type double-clutch automatic transmission and vehicle |
| CN111059269B (en) * | 2019-11-20 | 2021-07-27 | 中国第一汽车股份有限公司 | Self-adaptive control method for starting of wet-type double-clutch automatic transmission and vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107161150B (en) | 2019-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107203134B (en) | Front vehicle following method based on deep convolutional neural network | |
| CN103324085B (en) | Based on the method for optimally controlling of supervised intensified learning | |
| CN107284442B (en) | A kind of longitudinally controlled method of negotiation of bends for automatic driving vehicle | |
| CN113010967B (en) | Intelligent automobile in-loop simulation test method based on mixed traffic flow model | |
| CN103814201B (en) | The driving-force control apparatus of vehicle | |
| CN107182206A (en) | Speed planning method, device and the computing device of Vehicular automatic driving | |
| CN109492763A (en) | A kind of automatic parking method based on intensified learning network training | |
| CN106364432A (en) | Vehicle control method and device as well as vehicle | |
| JP2000250604A (en) | Cooperation method of optimization for characteristic optimization method | |
| CN110990950B (en) | Multi-target train energy-saving optimization method based on hybrid operation mode | |
| CN111547039B (en) | Hybrid vehicle accelerator control method and system based on deep reinforcement learning | |
| JP2000020103A (en) | Evaluating method for genetic algorithm | |
| CN110427690A (en) | A kind of method and device generating ATO rate curve based on global particle swarm algorithm | |
| CN113217619A (en) | Control method and device for clutch friction torque and electronic equipment | |
| Wang et al. | Design and implementation of an adaptive cruise control system based on supervised actor-critic learning | |
| CN107161150A (en) | A kind of method and device of control vehicle start | |
| CN105527963B (en) | Lateral parking method and system | |
| CN111198501A (en) | Method for determining fuel equivalent factor by RBF neural network | |
| CN110009644A (en) | A kind of method and apparatus of characteristic pattern row pixel segmentation | |
| CN112381376A (en) | Method and device for evaluating driving ability process | |
| JP2009129366A (en) | Sensibility estimation system of vehicle | |
| CN110816531B (en) | Control system and control method for safe distance between unmanned automobile vehicles | |
| CN108128301A (en) | A kind of engine power adjusting method, device and electronic equipment | |
| CN109760681A (en) | A kind of lane-change control method and device | |
| CN113034911B (en) | Vehicle queue control method and system with parameter and structure heterogeneity |
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 |