JP3474051B2 - Vehicle driving direction estimation device, vehicle control device, and shift control device for automatic transmission for vehicle - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a driving direction estimating device for estimating the driving direction of a driver in a vehicle.

[0002]

2. Description of the Related Art Generally, a shift control device for controlling a gear ratio or a gear stage of an automatic transmission, a steering force control device for controlling a steering force of a power steering of a vehicle, a damping force and a spring characteristic of a shock absorber of a vehicle suspension system. In a control device mounted on a vehicle such as a suspension control device for controlling the vehicle, it is possible to control the drive torque, the steering force of the steering, the damping force of the shock absorber, or the spring characteristics by reflecting the driving direction of the driver. desired.

On the other hand, the neural network which has been learned in advance based on the sensor signals detected by the respective sensors of the vehicle, determines whether the driver is acceleration oriented with emphasis on power performance or fuel efficiency with emphasis on fuel consumption. A shift control technique has been proposed in which determination is performed using a network and the shift diagram of an automatic transmission is switched. For example, Japanese Patent Laid-Open No. 6-2
That is the shift control method described in Japanese Patent No. 21420.

[0004]

However, in the conventional shift control method as described above, a neural network to which vehicle speed, throttle opening, acceleration, brake signal, gear position, throttle opening temporal change, etc. are inputted. It is used to determine the driving orientation based on the ratio of the estimated output contents from the neural network a plurality of times. Therefore, the estimated output in a short time cannot obtain sufficient accuracy,
If it is attempted to determine the driving orientation based on the contents of the estimated output for a plurality of times in order to increase the reliability, there is a disadvantage that the delay of the driving orientation result with respect to the driver's operation becomes large and the response cannot be sufficiently obtained.

The present invention has been made in view of the above circumstances, and an object of the present invention is to estimate the driving direction of a vehicle in which the accuracy and response of the driving direction estimation result with respect to the driver's operation are sufficiently obtained. To provide a device.
Another object of the present invention is to provide a control device for a vehicle, including a shift control device for an automatic transmission for a vehicle, which is capable of promptly reflecting and controlling the driving direction of the driver.

[0006]

The inventors of the present invention have made various studies in order to solve the above problems, and as a result, as a result, the output operation amount at the time of starting the vehicle, the maximum change rate of the output operation amount, and the braking operation of the vehicle are performed. The driving operation-related variables such as maximum deceleration of vehicle, coasting time of vehicle, constant driving time of vehicle, etc. are closely related to the driving orientation of the vehicle, and such driving operation-related variables are used as an index in the neural network. It was found that a high-precision driving orientation estimation can be quickly obtained by inputting the input. The present invention was made based on such findings.

That is, the gist of the invention according to claim 1 is a driving direction estimation device for estimating the driving direction of a vehicle, comprising: (a) an output operation amount when the vehicle starts and an output operation amount. Driving operation-related variable calculating means for calculating at least one driving operation-related variable selected from the maximum rate of change, the maximum deceleration during braking operation of the vehicle, the coasting time of the vehicle, and the constant vehicle speed running time; A neural network is provided to which the driving operation related variables calculated by the driving operation related variable calculating unit are input, and the driving operation related variable calculating unit outputs the output operation amount when the vehicle starts, the maximum change rate of the output operation amount, and the braking of the vehicle. Every time one of the maximum deceleration during operation, the coasting time of the vehicle, or the constant vehicle speed traveling time is calculated, the vehicle operation is performed based on the output of the neural network. A driving orientation estimating means for estimating the direction, the output operation portion for operating the output of the (c) vehicle engine
And an output operation amount detecting means for detecting the operation amount of
(D) The driving operation related variable calculation means is configured to output the output operation amount.
Maximum change rate of the output operation amount detected by the detection means
Calculate the maximum rate of change in the output manipulated variable except when opening and closing
That is what you do. The invention according to claim 2
The gist of is a driving direction estimation device for estimating the driving direction of a vehicle, which includes (a) an output operation amount at the time of starting the vehicle, a maximum change rate of the output operation amount, and a maximum during the braking operation of the vehicle. Driving operation related variable calculation means for calculating at least one driving operation related variable selected from deceleration, coasting travel time of vehicle, and constant vehicle speed travel time; and (b) driving calculated by the driving operation related variable calculation means. A neural network to which operation-related variables are input is provided, and the output operation amount when the vehicle is started by the driving operation-related variable calculation means,
Every time one of the maximum change rate of the output operation amount, the maximum deceleration at the time of braking operation of the vehicle, the coasting time of the vehicle, and the constant traveling time of the vehicle is calculated, based on the output of the neural network, A driving direction estimating means for estimating a driving direction , and (c) the driving operation related variable calculating means.
Is during coasting or constant speed travel of the vehicle, or
After the start of maximum deceleration detection during braking operation of the vehicle,
Coasting time or constant vehicle speed or vehicle braking
Maximum deceleration during operation is calculated for each preset time
Alternatively, (d) the driving direction estimation means sets the setting.
The coasting time of the vehicle calculated or determined for each hour
Or constant reduction in vehicle speed or maximum reduction during vehicle braking operation
Each time a speed is input to the neural network
Based on the output value output from the neural network
And estimating the driving direction of the vehicle. Also bill
The gist of the invention according to item 3 is a driving direction estimation device for estimating the driving direction of a vehicle, comprising: (a) an output operation amount at the time of starting the vehicle, a maximum change rate of the output operation amount, Maximum deceleration during braking operation, vehicle coasting travel time, driving operation related variable calculation means for calculating at least one driving operation related variable selected from constant vehicle speed running time,
(B) A neural network is provided to which the driving-operation-related variables calculated by the driving-operation-related variable calculating means are input, and the driving operation-related variable calculating means outputs the output operation amount at the time of starting the vehicle and the maximum change rate of the output operation amount , A driving orientation that estimates the driving orientation of the vehicle based on the output of the neural network every time any one of the maximum deceleration during braking operation of the vehicle, the coasting traveling time of the vehicle, and the constant traveling time of the vehicle is calculated. And (c) the driving operation.
The work-related variable calculation means is based on the neural network.
The longitudinal acceleration generated in the vehicle in order to input repeatedly
It is calculated repeatedly and stored in a predetermined storage location.
(D) In the case of the predetermined shift period of the vehicle,
The longitudinal acceleration generated during the shift period is the neural
Acceleration that prohibits input to the network
It is to further include input prohibition means. In addition, claim 4
The gist of the invention relating to is a driving direction estimation device for estimating the driving direction of a vehicle, comprising: (a) an output operation amount when the vehicle starts, a maximum change rate of the output operation amount, and a braking operation of the vehicle. Driving operation-related variable calculation means for calculating at least one driving operation-related variable selected from the maximum deceleration at time, the coasting travel time of the vehicle, and the constant vehicle speed travel time, and (b) calculated by the driving operation-related variable calculation means. Equipped with a neural network to which the driving operation related variables are input,
The driving operation-related variable calculation means calculates any one of the output operation amount at the time of starting the vehicle, the maximum change rate of the output operation amount, the maximum deceleration during the braking operation of the vehicle, the coasting time of the vehicle, and the constant vehicle speed traveling time. (C) the driving operation related means for estimating the driving direction of the vehicle based on the output of the neural network each time
The variable calculation means repeats to the neural network.
In order to output, the output operation amount, engine speed,
Calculate at least one of the maximum values of the longitudinal acceleration in the specified section
(D) A vehicle turning that determines the turning of the vehicle.
And a driving direction estimating means (f)
It is estimated that the vehicle is in the high speed direction, and the vehicle turning determination means determines
If it is determined that the vehicle is turning,
Maximum value is output to the neural network
Maximum value holding means for holding
It

[0008]

According to the invention of the present claim 1, according to the present invention, the driver's intention estimating means, the output operation amount when the vehicle starts, the maximum rate of change of the output operation amount maximum deceleration during braking operation of the vehicle, the vehicle Based on the output of the neural network to which at least one of coasting running time and constant vehicle speed running time is input,
The driving orientation of the vehicle is estimated. In this way, since the driving orientation is estimated based on the above-mentioned driving operation-related variables that are not conventionally used, it is possible to more accurately estimate the driving orientation, and each time the driving operation-related variable is calculated, the driving orientation Since the estimation result is obtained, it is possible to obtain the driving direction estimation device for a vehicle in which the responsiveness of the driving direction result to the driver's operation is sufficiently obtained. In addition, the driving direction estimation device for the vehicle
Is the operation of the output operation unit that operates the output of the vehicle engine.
The operation control device includes output operation amount detection means for detecting the amount of work.
The work-related variable calculation means uses the output operation amount detection means.
Rapid opening / closing operation based on the maximum change rate of the detected output operation amount
During acceleration operation to calculate the maximum change rate of the output operation amount excluding time
Of the maximum output manipulated variable change rate calculation means.
And during a temporary sudden opening / closing operation (so-called tip-in operation)
The maximum change rate of the output manipulated variable excluding
There is an advantage that the reliability of the orientation estimation result is enhanced. Well
According to the invention of claim 2, the driving direction estimating means is
The maximum amount of output operation amount and output operation amount when the vehicle starts
Rate, maximum deceleration when braking the vehicle, when coasting the vehicle
During this period, at least one of the constant vehicle speed travel times is input.
Based on the output of the Ural network, the driver's finger of the vehicle
The direction is estimated. Thus, it has not been used in the past
The driving orientation is estimated based on the above-mentioned driving-related variables.
Therefore, the driving orientation can be estimated more accurately, and the driving operation
Each time the continuous variable is calculated, the driving-oriented estimation result is obtained.
Therefore, the responsiveness of the driving-oriented result to the driver's operation is
It is possible to obtain a sufficient vehicle driving direction estimation device.
Wear. The driving operation-related variable calculation means is
While coasting or at a constant vehicle speed or when braking the vehicle
After the maximum deceleration detection is started, the vehicle speed is constant
Or vehicle coasting time or maximum reduction when braking the vehicle
Calculate or determine the speed for each preset time,
The driving direction estimating means calculates or sets every set time.
The determined coasting time or constant vehicle speed of the vehicle
Alternatively, the maximum deceleration during braking operation of the vehicle is the neural network
Every time the neural network is input to the network
Direction of the vehicle based on the output value output from the
The vehicle is coasting or
While the vehicle is traveling at a constant speed, the driving-oriented estimation results are set every set time.
Is obtained, so it is suitable for fuel-efficient driving with few driving operations.
However, there is an advantage that the responsiveness is enhanced. Also driving-oriented
Driving orientation based on maximum deceleration during braking closely related to
Since the estimation result of
There is an advantage that reliability can be improved. Also, according to claim 3.
According to the invention, the driving direction estimation means allows
Output operation amount, maximum change rate of output operation amount, vehicle braking operation
Maximum deceleration, coasting time of vehicle, constant vehicle speed
Neural network to which at least one of
The driving direction of the vehicle is estimated based on the output of the vehicle. This
Such as above, which is not used in the past.
More accurate because the driving orientation is estimated based on the continuous variables
Driving orientation can be estimated and driving-related variables are calculated
Each time, a driving-oriented estimation result is obtained, so the driver's
Vehicles with sufficient responsiveness of driving-oriented results to work
Can be obtained. Also, the luck
The rolling operation-related variable calculation means is the neural network.
Acceleration in the vehicle in order to repeatedly input to the
To be repeatedly calculated and stored in a predetermined storage location
If the predetermined shift period of the vehicle is reached,
Input of the above-mentioned longitudinal acceleration that occurs during the gear shifting period is prohibited.
Since it further includes a longitudinal acceleration input prohibition means,
For example, upshift gearshifting and downhill control 4 → 3 downshifting
Shift period in a predetermined shift that is not related to driving orientation
Input of the longitudinal acceleration that occurs during the shifting period,
It is stopped and the information is sent to the neural network.
More reliable driving estimation results
Can be According to the invention of claim 4, the driving finger
Output operation amount and output operation when the vehicle starts by the direction estimation means
Maximum rate of change in quantity, maximum deceleration during vehicle braking operation, vehicle
Based on the output of the neural network to which at least one of coasting running time and constant vehicle speed running time is input,
The driving orientation of the vehicle is estimated. In this way, since the driving orientation is estimated based on the above-mentioned driving operation-related variables that are not conventionally used, it is possible to more accurately estimate the driving orientation, and each time the driving operation-related variable is calculated, the driving orientation Since the estimation result is obtained, it is possible to obtain the driving direction estimation device for a vehicle in which the responsiveness of the driving direction result to the driver's operation is sufficiently obtained. Further, the driving operation related variable calculation means calculates at least one of the maximum values in a predetermined section among the input signal values of the output operation amount, the engine rotation speed, and the longitudinal acceleration in order to repeatedly output to the neural network. In the case where the vehicle turning determination means for determining the turning of the vehicle and the driving orientation estimating means estimate that the vehicle is in the acceleration direction and the vehicle turning determination means determines that the vehicle is turning, the maximum Since it further includes maximum value holding means for holding the output of the value and for holding the output of the held maximum value to the neural network, it is fuel-efficient with respect to the driving operation even during the acceleration-oriented running. The maximum value in the above section is not output to the neural network before there is no difference and before cornering. , The maximum value before the turning is the estimation of the driving manner by its pending value is pending is performed, the reliability of the driver's intention estimation result is increased.

[0009]

Other Embodiments of the Invention Here, it is preferable that the vehicle be driven.
The pointing estimation device is an output that controls the output of the vehicle engine.
Output operation amount detection means for detecting an operation amount of the operation unit, and a vehicle
And a vehicle speed detecting means for detecting the speed of the vehicle.
The continuous variable calculation means is detected by the output operation amount detection means.
Detected by the output operation amount and its vehicle speed detection means.
Start based on the calculated vehicle speed and output operation amount when the vehicle starts
The time output manipulated variable calculating means is included. in this way
The output manipulated variable at the start, which is closely related to driving
Is calculated, the reliability of the driving-oriented estimation result is improved.
There is an advantage.

The output operation amount calculating means at the time of starting is preferably
Is a vehicle stop determination means for determining the state in which the vehicle is stopped.
And that the vehicle speed has reached the preset vehicle speed.
Set vehicle speed reaching determination means to determine the
Output operation amount when the vehicle speed reaches the set vehicle speed
Output operation amount determining means for determining as the output operation amount of the vehicle
Including and

[0011]

Calculation of maximum change rate of output operation amount during acceleration operation
The output means preferably increases the rate of change of the throttle valve opening.
The maximum value of the rate of change in the throttle valve opening is stored and updated.
Throttle valve opening change rate maximum value updating means and access
Tip-in operation in which the pedal is opened and closed quickly within a short time
And a tip-in operation determining means for determining
When the chip-in operation is not judged by the operation judging means
The above-mentioned throttle valve opening change rate maximum value updating means
New throttle valve opening change rate maximum value maximum slot
The maximum throttle valve opening change determined as
And the maximum throttle valve opening change rate determination means
The maximum throttle valve opening change rate is determined by means
Throttle to determine the maximum throttle valve opening up to
Valve opening degree determining means.

The tip-in operation determining means is the maximum slot.
After a certain period of time has passed since the throttle valve opening was calculated,
If the torque valve opening is less than the judgment reference value and the throttle opening
If the conversion rate is zero or negative, or its throttle
Even if the rate of change of opening is positive, the throttle valve opening is judged again.
Rapid opening / closing of the accelerator pedal when the value is less than the standard value
That is, it is determined to be a tip-in operation.

Also, preferably, the driving direction estimation of the vehicle is performed.
In the device, the driving operation-related variable calculation means is
Vehicle speed constant travel time during constant travel or coasting of the vehicle
Alternatively, the coasting time of the vehicle is set every preset time.
Vehicle speed constant travel time calculation means to be calculated or determined by
The driving direction estimation means including coasting travel time calculation means
Is the coasting of the vehicle calculated or determined at each set time.
Running time or constant running speed
Each time it is input to the network, the neural network
The driving direction of the vehicle is estimated based on the output value output from
Set. This will allow the vehicle to coast or travel faster.
During a certain period of driving, a driving-oriented estimation result is obtained every set time.
Therefore, in fuel economy-oriented driving with few driving operations
Also has the advantage that the responsiveness is enhanced.

The constant vehicle speed running time calculating means calculates the vehicle speed
The vehicle speed changes when it is determined that it does not change more than a predetermined width.
Not doing the first thing is that it lasts for a preset time.
1 Running at a constant vehicle speed based on being timed by the timekeeping means
Vehicle speed constant travel determination means for determining line and constant vehicle speed travel
In the state where the vehicle traveling speed is determined to be constant by the line determination means,
2 Based on the elapsed time measured by the timekeeping means,
While determining the constant travel time,
Is a constant vehicle speed for each preset start cycle
Vehicle speed constant traveling time determining means for repeatedly determining
Mu.

The above-mentioned coasting traveling time calculating means is used for the output during traveling.
Determine coasting of the vehicle based on zero force operation
Coasting judgment means and coasting judgment means
The time is measured by the third time measuring means while the traveling is determined.
The coasting time is determined based on the elapsed time
During the coasting, the preset start cycle elapses.
Determine coasting time by repeatedly determining coasting time for each
And means.

Also, preferably, the driving direction estimation of the vehicle is performed.
In the device, the driving operation-related variable calculation means is a vehicle
After the maximum deceleration detection during braking of the
Maximum during braking, where speed is calculated for each preset time
Deceleration calculation means, the driving direction estimation means,
Maximum reduction during vehicle braking calculated or determined at each set time
Every time the speed is input to the neural network, that two
Based on the output value output from the Ural network
Estimate the driving orientation of the vehicle. If you do this,
Driving finger based on maximum deceleration during braking closely related to direction
As the result of estimating the driving direction is obtained, the result of estimating the driving direction
There is an advantage that the reliability of is improved.

The braking maximum deceleration calculating means is preferably
Sequentially records the increasing longitudinal acceleration while the vehicle is braking.
Maximum braking deceleration updating means to be remembered and before and after
When the acceleration is smaller than the preset judgment reference value
Is the value stored by the above maximum braking deceleration updating means.
Determined as the maximum deceleration during braking, and the subsequent longitudinal acceleration
If (negative) is equal to the maximum deceleration during braking, or
During braking Even if it is larger than the maximum deceleration (zero side)
If the maximum deceleration is not more than a specified value, the 4th timer
The above new longitudinal acceleration at every predetermined time measured by the step
Is determined as the maximum deceleration during braking periodically
And deceleration determining means.

[0019]

The driving operation-related variable calculating means includes:
Stores and stores the maximum value of longitudinal acceleration in a predetermined storage location
If a longitudinal acceleration that is larger than the value
A longitudinal acceleration maximum value updating means for updating it is provided,
The longitudinal acceleration input prohibiting means is used for upshifting and downshifting.
Do not relate to driving orientation such as 4 → 3 downshift of slope control.
The longitudinal acceleration input during a predetermined gear change period is
Prohibition of input to the maximum longitudinal acceleration updating means
It is a thing.

The driving operation related variable calculation means is preferably
Repeatedly outputs to the neural network
In order to control the output operation amount, engine speed, and longitudinal acceleration.
At least one of the maximum values of the input signal value within the predetermined interval
In order to determine
A new output manipulated variable that is larger than the stored value
When input, the new output manipulated variable is stored
Output operation amount maximum value updating means to be updated, engine rotation
In order to memorize the maximum value within a predetermined section of speed,
If a new engine speed is input,
To update the new engine speed of
Jin rotation speed section maximum value updating means, predetermined section of longitudinal acceleration
Before and after larger than the stored value to store the maximum value in the interval
If a new acceleration is input, the new longitudinal acceleration
Longitudinal acceleration section maximum value updating means for updating as a stored value
It is equipped with at least one of
The vehicle turning determination means for determining the vehicle turning direction and the driving direction estimation means.
Is estimated to be acceleration-oriented, and the above vehicle turning determination is made.
If the vehicle determines that the vehicle is turning, the maximum
The output of the value is suspended and the maximum retained value is
Maximum value holding means for holding output to network
It further includes and. This way, the acceleration finger
Even when driving in the opposite direction, there is a difference
In front of a corner where there is no
The maximum value within the interval is output to the neural network.
The maximum value before turning will be retained and
Driving direction estimation is performed, so driving direction estimation results
The reliability of is improved.

Further , it is preferable that the accelerator pedal is used for a short time.
When the tip-in operation, which is the sudden opening / closing operation of
Is the output operation amount maximum value updating means, engine rotation
Velocity section maximum value updating means, longitudinal acceleration section maximum value updating means
A check that prevents the maximum value from being updated in at least one of the stages.
A plug-in determination means is provided. If you do this,
Rather than turning direction, it is more suitable for road conditions such as obstacles on the road
Since the related information is removed, the accuracy of driving-oriented estimation is improved.
To be enhanced.

Further, it is preferable to check the acceleration during acceleration.
Accelerator return speed excluding the push-in operation is equal to or higher than a predetermined value, or
Or, the vehicle turns when the deceleration during braking is above a specified value.
Vehicle turning determination means for determining that As a result,
Without installing a detection device such as a steering angle sensor,
It is determined whether the vehicle is traveling in front of the corner or traveling in a corner.

Preferably , the maximum value holding means is used.
Output is greater than the maximum value
If a threshold input signal is input, the maximum value holding
Even if the output is suspended by the stage, update it as the maximum value
The pending update means to output to the neural network
Further included. This will give you a more driving-oriented
Reliability can be increased. More preferably, the maximum value holding hand
Re-depressing the accelerator pedal while holding a step
Is canceled, the hold by the maximum value holding means is canceled.
A hold releasing means is provided for removing.

The driving operation-related variable calculation means is preferably
Repeatedly outputs to the neural network
In order to determine the maximum value of the engine speed within a predetermined section
In order to achieve
The engine speed that is greater than the stored value
If newly input, the new engine speed
Maximum value of engine speed section updated as stored value
It is equipped with a means, the current gear is more than the highest speed
Is one gear lower, and there is steady running where the vehicle speed is constant.
If it continues for more than the specified duration,
Engine detected when traveling in the next lower gear
After correcting the rotation speed to the value of the highest gear speed,
The engine speed section maximum value updating means
Engine rotation speed correction means is further provided. like this
If so, the actual engine speed is lower than that
After being corrected to the value, it is updated as the maximum value input signal in the predetermined section
And is input to the neural network
At a constant speed at a gear one step below the highest speed
Output signal of neural network during high-speed driving
Is suppressed and it is erroneously estimated to be acceleration-oriented.
And are preferably eliminated.

Further, the gist of the invention according to claim 5 is a driving direction estimation device for estimating the driving direction of a vehicle, comprising: (a) an output operation amount when the vehicle starts and an output operation amount. Driving operation-related variable calculating means for calculating at least one driving operation-related variable selected from the maximum rate of change, the maximum deceleration during braking operation of the vehicle, the coasting time of the vehicle, and the constant vehicle speed running time; A driving direction estimation unit that includes a neural network to which the driving operation related variables calculated by the driving operation related variable calculation unit are input, and that estimates the driving direction of the vehicle based on the output of the neural network; and (c) the turning of the vehicle. When the vehicle turn is determined by the vehicle turn determination means for determining (4) and the vehicle turn determination means (d), the neural network in the driving direction estimation means is determined. It prohibits driving manner estimation calculation of click,
And driving direction estimation calculation prohibiting means for holding the output value of the neural network before turning the vehicle.
Such claim 5 by the invention according to lever, in the fuel economy-oriented and not different pre-corner and corner turning with respect to the driving operation, since no output value of the neural network is newly output, the neural It is possible to prevent the output value of the network from being reduced, and there is no risk that it will be determined to be fuel efficiency-oriented even though it is acceleration-oriented.
The reliability of the driving-oriented estimation result is improved.

Also, preferably, the driving-oriented estimation of the above invention
The control content changes based on the driving intention estimated by the means.
A vehicle controller to be added is configured. Like this
Therefore, the responsiveness of the driver-oriented result to the driver's operation is sufficient.
There is provided a vehicle control device obtained as described above.

The vehicle control device is preferably configured to operate in the above-mentioned manner.
Based on the driving direction estimated by the turning direction estimation means.
A shift control device for an automatic transmission for a vehicle in which a speed ratio is controlled,
Based on the driving direction estimated by the driving direction estimation means
Steering that controls the steering force of the vehicle power steering
Estimated by the force control device or the driving direction estimation means
Shock absorber of the vehicle suspension system based on the determined driving direction.
Suspension with controlled damping force or spring characteristics
Control device.

Further , preferably, the automatic transmission for a vehicle described above is
The shift control device is estimated by the driving direction estimation means.
Based on different driving directions, multiple types of shifts stored in advance
A shift diagram switching means for selecting one of the diagrams and its
Based on the shift diagram selected by the shift diagram switching means,
Change of automatic transmission based on throttle valve opening and vehicle speed.
And a shift control means for changing the speed ratio. This allows
A gear that is suitable for driving direction is selected, and sufficient acceleration force or
Fuel efficiency can be obtained.

Further , it is preferable that the climbing control or the downhill control is performed.
Not only the neural network of the driving direction estimation means
Network output signal is above a specified value
Vehicles that have a constant vehicle speed but prohibit the highest gear of a speed machine
Both of the steady running, the output of the neural network
If the force signal is smaller than the specified value,
Maximum speed gear prohibition means for releasing the prohibition is provided. This
If the output of the neural network is
When the value is equal to or more than the predetermined value, sufficient driving force can be obtained.
There is a point.

[0031]

[0032]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an automatic transmission and a shift control device for a vehicle. In FIG. 1, power output from an engine 10 of a vehicle is transmitted to a drive wheel (not shown) via a torque converter 12, an automatic transmission 14, a differential gear device (not shown) and an axle.

The torque converter 12 is connected to a pump impeller 18 connected to the crankshaft 16 and an input shaft 20 of the automatic transmission 14 and is connected to the pump impeller 18 via a fluid.
The turbine impeller 22 to which power is transmitted from the vehicle, the fixed impeller 28 fixed to the position-fixed housing 26 via the one-way clutch 24, the pump impeller 18 and the turbine impeller 22 are directly connected via the damper 30. The lock-up clutch 32 is provided. This lockup clutch 32
Is engagement-controlled by the pressure difference between the release-side oil chamber 33 and the engagement-side oil chamber 35.

The automatic transmission 14 is, for example, a fourth forward speed.
Alternatively, a planetary gear-type multi-stage shift that achieves a fifth gear
It is a machine. The automatic transmission 14 for the fourth forward speed is
Three sets of single pinion type planetary gears arranged on the shaft
Units 34, 36, 38, the input shaft 20, and the planetary gear device.
The output gear 39 that rotates with the ring gear of
Counter shaft that transmits power to the differential gear unit (output
Force axis) 40. Those planetary gear units 34,
Some of the components of 36 and 38 are integrally connected to each other.
Not only the three clutches C₀ , C₁ , C₂ By
Are selectively connected to each other. Also, the above planetary gears
Some of the components of the devices 34, 36, 38 are four blurs.
Key B₀ , B₁ , B₂ , B₃ By the housing 26
In addition to being selectively connected, some of the components
Is three one-way clutch F₀ , F ₁ , F₂ By that
Depending on the direction of rotation, they may be engaged with each other or with the housing 26.
It is designed to be sent.

The clutches C ₀ , C ₁ and C ₂ and the brakes B ₀ , B ₁ , B ₂ and B ₃ are provided with, for example, a multi-plate clutch or one band or two bands with opposite winding directions. It is composed of a band brake or the like, and is operated by a hydraulic actuator (not shown). As shown in FIG. 2, the gear ratio γ (= input shaft 20 of the input shaft 20 is controlled by controlling the operation of each hydraulic actuator by a hydraulic control circuit 44 that operates in accordance with a command from an electronic control unit 42 described later. It is possible to obtain four forward gears and one reverse gear having different rotational speeds / rotational speeds of the counter shaft 40. In FIG. 2, “1st”,
"2nd", "3rd", "O / D (overdrive)" indicate the forward first speed, second speed, third speed, and fourth speed, respectively. The gear ratio gradually decreases from the first gear to the fourth gear. Since the torque converter 12 and the automatic transmission 14 are configured symmetrically with respect to the shaft center, the lower side of the rotation axis of the input shaft 20 and the upper side of the rotation axis of the counter shaft 40 in FIG. It is omitted.

The hydraulic control circuit 44 includes the automatic transmission 1
A hydraulic control circuit for shift control for controlling the fourth gear and a hydraulic control circuit for engagement control for controlling engagement of the lockup clutch 32 are provided. The hydraulic control circuit for gear shifting control is solenoid No. 1 and solenoid N
First solenoid valve 4 which is turned on / off by o.2
6 and a second solenoid valve 48, and a clutch and a brake are selectively activated as shown in FIG. Any of the gear stage to the fourth speed gear stage is established.

Further, the engagement control hydraulic control circuit switches the clutch switching valve (not shown) for switching between the disengagement side position where the lockup clutch 32 is disengaged and the engagement side position where the lockup clutch 32 is engaged. The third solenoid valve 50 for generating a switching signal pressure for turning on and off the valve and the pressure difference ΔP between the engagement side oil chamber 35 and the release side oil chamber 33 are adjusted to control the slip amount of the lockup clutch 32 (not shown). And a linear solenoid valve 54 for generating a slip control signal pressure for operating the slip control valve in accordance with a drive current from the electronic control unit 42.

The electronic control unit 42 includes a CPU 60, R
A so-called microcomputer including an AM 62, a ROM 64, an input / output interface (not shown), etc., in which a throttle sensor 7 for detecting an opening TA of a throttle valve 68 provided in an intake pipe 66 of the engine 10 is provided.
0, an engine rotation speed sensor 72 that detects the rotation speed of the engine 10, an input shaft rotation speed sensor 74 that detects the rotation speed of the input shaft 20 of the automatic transmission 14, and a counter of the automatic transmission 14 to detect the vehicle speed V. An operation position sensor 80 for detecting the rotation speed of the shaft 40, an operation position of the shift lever 78, that is, an operation position sensor 80 for detecting one of the L, S, D, N, R, and P ranges, and an operation of the brake pedal 82 are detected. From the brake switch 84, a signal representing the throttle valve opening TA, a signal representing the engine rotation speed N _E , a signal representing the input shaft rotation speed N _IN , a signal representing the rotation speed N _OUT of the output shaft (counter shaft 40), Shift lever 7
The signal indicating the operation position P _S of No. 8 and the signal S _BK indicating the operation of the brake pedal 82 are respectively supplied. The throttle valve 68 is mechanically connected to an accelerator pedal 58 corresponding to an output operation unit, and the throttle valve opening degree TA is increased with the depression operation thereof.
The manipulated variable (i.e., output manipulated variable) A _ACC is also substantially detected.

The CPU 60 of the electronic control unit 42 pre-sets the R
The input signal is processed according to the program stored in the OM 64, and for example, driving direction estimation control, shift control, lockup clutch control, etc. are executed. Therefore, in this embodiment, the electronic control unit 42 functions as a driving direction estimation device and a shift control device.

In the driving direction estimation control of the electronic control unit 42, a predetermined driving operation-related variable is calculated from the input signal,
The driving orientation is estimated based on the output of the neural network NN to which the driving operation-related variables are input. In the shift control and lock-up clutch control of the electronic control unit 42, a plurality of types of shift diagrams stored in advance in the ROM 64, that is, the acceleration-oriented shift diagram of FIG. 3, and the intermediate-oriented (normal) shift line of FIG. A shift diagram corresponding to the driving direction is selected from the fuel consumption-oriented shift diagrams in FIGS. 5 and 5, and a predetermined gear stage is selected from the selected shift diagram based on the actual vehicle speed V and the throttle valve opening TA. Shift determination or lock-up on / off determination is performed. For example, in the normal shift diagram of FIG. 4, when the point indicating the actual vehicle speed V and the throttle valve opening TA intersects the upshift line or the downshift line, the upshift determination or the downshift determination is performed. Then, the first solenoid valve 46 and the second solenoid valve 48 shown in FIG. 1 are driven so as to establish the gear stage for which the shift determination is performed, or the third solenoid valve is used for engagement control of the lockup clutch 32. The solenoid valve 50 and the linear solenoid valve 54 are driven.

In FIGS. 3, 4, and 5, the solid line indicates a shift-up line, the broken line indicates a shift-down line, and
The one-dot chain line shows the engagement line of the lock-up clutch, and the two-dot chain line shows the release line of the lock-up clutch. In the acceleration-oriented shift diagram of FIG. 3, the shift line is set so that the shift is executed at a higher vehicle speed (high engine rotation speed) than in FIG. Further, in the fuel consumption-oriented shift diagram of FIG. 5, the shift line is set so that the shift is performed at a low engine rotation speed as compared with FIG.

FIG. 6 is a functional block diagram for explaining the control function of the electronic control unit 42. In the figure, the shift control means 90 uses an actual vehicle speed V detected by a vehicle speed sensor 76 and a throttle from a shift diagram selected by a shift diagram switching means 92 from a plurality of types of shift diagrams stored in advance in the ROM 64. Based on the actual throttle valve opening TA detected by the sensor 70, a shift determination to a predetermined gear stage is executed, and solenoid valves 46, 48, 50 for achieving the gear stage determined by the shift determination are provided. On the other hand, a gear shift output is performed to control the switching of the gear stage of the automatic transmission 14.

When the D range is selected, the highest speed gear stage prohibiting means 93 is not only used during uphill control or downhill control, but the output signal NN _OUT of the neural network NN of the driving direction estimation section 94 is set to a predetermined value. Even when the above is the case, that is, when the content of the flag X _{SP ORT} indicating the acceleration direction is set to "1", the highest gear is prohibited in order to obtain a sufficient driving force. However, when the vehicle is traveling at a constant speed and the output signal NN _OUT of the neural network NN is smaller than the predetermined value K, the highest gear stage prohibiting means 93 releases the highest gear stage. .

The shift line diagram switching means 92 includes a ROM 64.
The shift diagram corresponding to the driving direction estimated by the driving direction estimation unit 94 is selected from the plurality of types of shift diagrams shown in FIGS. 3, 4, and 5 stored in advance in FIG. For example, in the shift line diagram switching means 92, when the driving direction estimation unit 94 sets, for example, a flag X _SPORT indicating acceleration direction, the acceleration direction shift line diagram of FIG. 3 is selected, but the fuel consumption direction is set. When the flag X _ECO shown is set, the fuel consumption-oriented shift diagram of FIG. 5 is selected, and when the flag X _NORM indicating intermediate orientation is set, those acceleration-oriented shift diagram and fuel consumption-oriented diagram are selected. The intermediate-oriented shift diagram of FIG. 4 having the intermediate characteristics of the shift diagram of FIG.

The driving direction estimation section 94 is provided with a plurality of types of operation.
Every time one of the operation related variables is calculated, the operation related
A neural network that inputs variables and starts estimation
Network NN, and its neural network NN
The driving direction of the vehicle is estimated based on the output of. For example
As shown in FIG. 7, the driving direction estimation unit 94 is configured to
The detection signals from the sensors 70, 72, 74, 76, 84, etc.
A signal reading means 96 for reading in a relatively short predetermined period,
Is the signal sequentially read by the signal reading means 96?
, Which are closely related to driving maneuvers that reflect driving orientation.
Types of driving-related variables, namely, output control when the vehicle starts.
Work amount (accelerator pedal operation amount), that is, when the vehicle starts
Rotor valve opening TA_STMaximum output operation amount during acceleration operation
Rate of change, ie maximum rate of throttle valve opening
A_CCMAX, The maximum deceleration G when braking the vehicle _NMAX,vehicle
Coasting time T_COAST, Constant vehicle speed T_VCONST,
The section of the signal input from each sensor within the predetermined section
Maximum value, maximum vehicle speed V after starting operation_maxEtc.
Driving operation related variable calculating means for calculating each
The operation means related to the operation means by the processing means 98 and the preprocessing means 98.
Each time a variable is calculated, that driving-related variable is allowed
Neural network NN that performs driving direction estimation calculation by
Which is the output of the neural network NN.
And a driving direction estimation means 100 that outputs a driving direction estimation value.
Provided, output signal NN of neural network NN_OUT
Directly, or its output signal NN_OUTIs getting bigger
Eg flag X_SPORT, X_NORM, X_ECOIndicated by
The signal indicating the driving direction is converted into three levels and then
It is supplied to the speed line diagram switching means 92 and mounted on the vehicle.
And it is necessary to change the control in relation to the driving orientation
Supply to other control devices.

As the other control device, for example, the steering force control device 102 for controlling the steering force of the power steering of the vehicle based on the driving direction estimated by the driving direction estimation unit 94, and the driving direction estimation unit 94. The suspension control device 104 controls the damping force or the spring characteristic of the shock absorber of the vehicle suspension device based on the driving direction estimated by the above. The driving direction estimation means 10
0, the throttle valve opening TA _ST at the start of the vehicle, the maximum change rate A _{CCMA X} of the throttle opening during the acceleration operation, the maximum deceleration G _NMAX during the braking operation of the vehicle, the coasting time T _{COAST of} the vehicle, Since the driving direction of the vehicle is estimated based on the output of the neural network to which the constant vehicle speed travel time T _VCONST is input, the driving direction is estimated based on the above-mentioned driving operation related variables that have not been used conventionally. Therefore, the driving orientation can be estimated more accurately, and the responsiveness of the driving orientation result to the driver's operation can be sufficiently obtained without requiring the estimation a plurality of times.

The pre-processing means 98 shown in FIG.
Output operation amount at the time of the throttle valve opening when the vehicle starts
TA_STOutput operation amount calculation means 98a for starting to calculate
Maximum rate of change in output manipulated variable during high speed operation
Maximum rate of change of the throttle valve opening A _CCMAXAcceleration operation to calculate
Hour output operation amount maximum change rate calculation means 98b, vehicle braking operation
Maximum deceleration G at work_NMAXMaximum deceleration calculation during braking
Outputting means 98c, vehicle coasting time T_COASTTo calculate
Coasting travel time calculating means 98d, constant vehicle speed travel time T
_VCONSTConstant vehicle speed travel time calculating means 98e for calculating
For example, input from each sensor within a predetermined section of about 3 seconds
Input signal section that periodically calculates the maximum value of the force signal
Maximum value calculating means 98f, maximum vehicle speed after starting operation
V_maxMaximum vehicle speed calculation means 98g for calculating
It is prepared.

The input signal section maximum value calculating means 98f
Maximum value of the input signal in the predetermined section calculated in
As for the throttle valve opening TA_maxt, Vehicle speed V_maxt, D
Engine rotation speed N_Emaxt, Longitudinal acceleration NOGBW_maxt  (Reduction
Negative value at high speed) or deceleration G_NMAXt(Absolute value)
Used. Longitudinal acceleration NOGBW_maxt  Or deceleration G
_NMAXtIs, for example, the vehicle speed V (N_OUT) Calculated from the rate of change
To be

The driving direction estimating means 100 shown in FIG. 7 is provided.
Neural network NN is a computer program
By software on the RAM or by connecting electronic components
The hardware consisting of
It can be configured by converting the operation of FIG.
As illustrated in the block of the direction estimation means 100,
Is made. In FIG. 7, the neural network NN
Is r nerve cell elements (neurons) X_i(X₁ ~ X
_r) Input layer composed of), and s neuronal elements Y_j
(Y₁ ~ Y_s) And t nerve cells
Cell element Z_k(Z ₁ ~ Z_t) And an output layer composed of
It is a structured three-layer hierarchical type. And the above input
Transfers the state of neuronal elements from the layer to the output layer
Therefore, the coupling coefficient (weight) W_XijHave the above r gods
Transcellular element X_iAnd s number of nerve cell elements Y_jAnd respectively
Transfer element D to be coupled_XijAnd the coupling coefficient (weight) W_YjkTo
Has s nerve cell elements Y_jAnd t nerve cell elements Z
_kTransfer element D for respectively coupling and_YjkIs provided
It

[0051] The neural network NN is its coupling coefficient (weight) W _Xij, pattern associative system that is made to learn the coupling coefficient (weight) W _Yjk called backpropagation learning algorithm. The learning has been completed in advance by a traveling experiment in which the values of the driving operation-related variables and the driving directions are associated with each other. Therefore, when the vehicle is assembled, the coupling coefficient (weight) W _Xij and the coupling coefficient (weight) are set.
W _Yjk is given a fixed value. In the above learning, driving with fuel efficiency, acceleration, and intermediate (normal) orientation for a plurality of drivers is performed on various roads such as highways, suburban roads, mountain roads, and city roads. The driving direction at that time is used as the teacher signal, and the n signals (input signals) obtained by preprocessing the teacher signal and the sensor signal are input to the neural network NN. The teacher signal is digitized into values from 0 to 1 for driving, for example, 0 for fuel efficiency, 0.5 for intermediate, and 1 for acceleration. The input signal is used after being normalized to a value between -1 and +1 or between 0 and 1.

In the pre-processing means 98 of FIG. 7, the starting output operation amount calculating means 98a includes a throttle valve opening (output operation amount) TA detected by a throttle sensor (output operation amount detecting means) 70 and a vehicle speed sensor ( Vehicle speed detection means) 76
The throttle valve opening degree (output operation amount) TA _ST when the vehicle _starts is calculated from the vehicle speed V detected by. Throttle valve opening (output manipulated variable) when the vehicle starts, which is closely related to driving orientation in order to input to the neural network NN
Since TA _ST is calculated, the reliability of the driving-oriented estimation result is improved. For example, as shown in FIG. 8, the start-time output operation amount calculating means 98a indicates that the vehicle speed is V when the vehicle is stopped for a predetermined time T _VO1 or more.
A vehicle stop determination means 110 for determining the vehicle speed V based on
There a determined speed setting arrival determining means 111 that it has reached the set vehicle speeds V ₁ to a preset throttle when it reaches the set vehicle speed V ₁ of the order of the vehicle speed V, for example, 10 km / h following the stop state of the vehicle valve opening TA the throttle valve at the start of opening (output operation amount) composed of the starting time of the output manipulated variable determining means 112. which determines the TA _ST.

In addition, in the preprocessing means 98 of FIG.
The output operation amount maximum change rate calculation means 98b at the time of high speed operation is
Detected by the rottle sensor (output operation amount detection means) 70
Change in throttle valve opening (output operation amount) TA
The maximum rate of change in output manipulated variable excluding sudden opening / closing
_CCMAXTo calculate. Input to neural network NN
In order to make the
Maximum rate of change A_CCMAXIs calculated, and also temporary opening and closing
Maximum rate of change A during operation (so-called tip-in operation)_{CC MAX}But
Since it is excluded, the reliability of the driving-oriented estimation result is improved.
Be done. Maximum change in output operation amount during acceleration operation
The rate calculation means 98b, for example, as shown in FIG.
Change rate of the throttle valve opening A_CCTAThat is, accelerator depression speed
The maximum value of the throttle valve opening change rate during
Throttle valve opening change rate maximum value updating means 11
4 and the accelerator pedal is opened and closed quickly within a short time.
Chip-in operation determination means 115 for determining a plug-in operation
And the chip-in operation determination means 115
When the throttle operation is not judged, the throttle valve opening change
Throttle valve updated by rate maximum value updating means 114
The maximum opening change rate is the maximum throttle valve opening change rate A
_CCMAXThe maximum throttle valve opening change rate is determined as
Means 116 and a means for determining the maximum throttle valve opening change rate
The maximum throttle valve opening change rate A due to the step 116_CCMAX
Maximum throttle valve opening A until is determined _CCMXTADecided
And a maximum throttle valve opening determination means 117
To be done.

The tip-in operation determining means 115 determines that the throttle valve opening TA after the elapse of a predetermined time (K _SHRT + T _MAXTA ) from the determination of the maximum throttle valve opening A _CCMXTA is less than the reference value K _TACHIP. When the throttle opening change rate A _CCTA is zero or negative, or even when the throttle opening change rate A _CCTA is positive, the throttle valve opening T
When A is smaller than the re-determination reference value K _THRS , it is determined that the acre pedal 58 is in a sudden opening / closing operation, that is, a tip-in operation.

[0055] Further, in the preprocessing means 98 in FIG. 7, the vehicle speed constant running time calculating means 98e for calculating the constant vehicle speed running time T _VCONST, for example, as shown in FIG. 10, the vehicle speed V
It is determined that the vehicle speed V does not change more than a predetermined width ΔV, and the vehicle speed V remains unchanged for a preset time K _VCONAV
A constant vehicle speed traveling determination means 120 for determining constant vehicle speed traveling based on the determination by the first timing means 119 that the vehicle has a constant duration, and a second state in which a constant vehicle speed traveling is determined by the constant vehicle speed traveling determination means 120. Timing means 121
By and determines the timed constant vehicle speed running time T _VCONST based on the elapsed time is, the vehicle speed constant vehicle speed to determine repeatedly constant vehicle speed running time T _VCONST at every elapse of a preset activation cycle K _VCON is in constant running It is composed of the traveling time determining means 122.

Further, in the preprocessing means 98 of FIG. 7, the coasting traveling time calculating means 98d for calculating the coasting traveling time T _COAST of the vehicle has a throttle valve opening TA of zero during traveling as shown in FIG. The coasting traveling determination means 124 for determining the coasting traveling of the vehicle based on the above, and the coasting traveling time based on the elapsed time measured by the third timing means 125 in the state where the coasting traveling is determined by the coasting traveling determination means 124. and determines the T _COAST, made from a preset activation cycle K _Ilon coasting time determining unit 126. repeatedly determining the coasting time T _COAST at every elapse of its coasting running.

Further, in the preprocessing means 98 of FIG. 7, the braking maximum deceleration calculating means 98c for calculating the maximum deceleration G _NMAX during braking of the vehicle is, for example, as shown in FIG.
Longitudinal acceleration increases to the negative side during vehicle braking NOGBW
Braking maximum deceleration updating means 13 for sequentially storing and updating
If 0 and its longitudinal acceleration NOGBW are smaller than a preset reference value K _SPBKG, the value NOGBW stored by the braking maximum deceleration updating means 130 is determined as the braking maximum deceleration MAXBKG, that is, G _NMAX. If the subsequent longitudinal acceleration NOGBW (negative) is equal to the braking maximum deceleration MAXBKG, or if it is greater than the braking maximum deceleration MAXBKG (zero side), then the braking maximum deceleration MAXBKG is greater than the predetermined value KBKGHYS. If it is not separated, the value determined as described above (the value initially determined as a value smaller than the judgment reference value K _SPBKG ) is reduced at the maximum during braking for each predetermined time K _BKCON timed by the fourth time counting means 131. The braking maximum deceleration determining means 132 for periodically determining the speed MAXBKG is included.

Further, in the pre-processing means 98 of FIG. 7, the maximum value of the input signal section for periodically calculating the maximum value of the input signal sequentially input from each sensor in the predetermined section of, for example, about 3 seconds. For example, as shown in FIG. 13, the calculating means 98f stores the throttle valve opening TA input in a predetermined section in a predetermined storage location, and the value stored in the storage location and the newly input throttle valve TA. A throttle valve opening section maximum value updating means 134 for _obtaining a maximum value TA _maxt of the throttle valve opening TA within a predetermined section by sequentially comparing the opening degree TA and a larger value as a stored value, and a predetermined value. Exiled the engine rotational speed N _E which is newly input and the value stored in the storage location stores the engine speed N _E in a predetermined memory location which is inputted in the interval The engine speed interval maximum value updating means 135 for obtaining the maximum value N _Emaxt the engine speed N _E in the predetermined section by updating the larger value compared to a stored value, before and after being inputted within a predetermined interval Acceleration NOGBW is stored in a specified storage location, and the value stored in that storage location and the newly input longitudinal acceleration are also stored.
Maximum value of longitudinal acceleration NOGBW _maxt (maximum deceleration G _{NM AXt} within a given section) of longitudinal acceleration NOGBW within a given section is updated by sequentially comparing NOGBW and the larger value is updated as a stored value. Means 136 and a maximum throttle valve opening TA _maxt within a predetermined section,
Maximum engine speed N _Emaxt , maximum longitudinal acceleration NOGBW
_{The maxt} is repeatedly input to the neural network NN for each predetermined section.

When the vehicle is in a predetermined shift period that is not related to driving orientation, such as an upshift shift or a 4 → 3 downshift in downhill control, the longitudinal acceleration input prohibiting means 137 is generated within that shift period. Input of the longitudinal acceleration NOGBW is prohibited. As a result, during a shift period in a predetermined shift that is not related to driving direction, such as an upshift shift or a 4 → 3 downshift in downhill control, input of the longitudinal acceleration NOGBW that occurs during that shift period is prohibited and the information is Since it is not sent to the neural network, the reliability of the driving-oriented estimation result is further enhanced.

The vehicle turning determination means 138, for example, when the accelerator return speed excluding the tip-in operation is _{equal to} or higher than the predetermined value K _DTAMX during acceleration, or the deceleration G _N during braking is _{equal to} or higher than the predetermined value K _SPBKG. Sometimes the turning of the vehicle is determined. This makes it possible to determine whether the vehicle is traveling in front of a corner or traveling in a corner without providing a detection device such as a steering angle sensor.

The maximum value holding means 139 determines that the driving direction estimation means 100 is in the acceleration direction and the vehicle turning determination means 138 determines that the vehicle is turning. Value updating means 1
34, the engine rotation speed section maximum value updating means 135, the longitudinal acceleration section maximum value updating means 136 suspends the updating of the maximum value, and the suspended maximum value, that is, the maximum value updated in the section before the determination of the vehicle turning (throttle valve Opening T
A _maxt , engine speed N _Emaxt , longitudinal acceleration NOGBW
_maxt ) is input to the neural network NN. As a result, the maximum value in the above section is not updated before the corner and before the corner where there is no difference in driving operation from the fuel consumption direction even when traveling in the acceleration direction and before the corner. Is held and the driving-oriented estimation is performed by the holding value, so that the reliability of the driving-oriented estimation result is enhanced.

The pending update means 140 holds the maximum value holding means 139 when a new input signal larger than the held maximum value is input during the holding by the maximum value holding means 139. Despite this, the throttle valve opening section maximum value updating means 134, the engine rotation speed section maximum value updating means 135, the longitudinal acceleration section maximum value updating means 136 newly input the throttle valve opening degree TA and the engine rotation speed. Speed N _E , longitudinal acceleration NO
Maximum value _{GBW (TA maxt, N Emaxt,} NOGBW ma xt) is updated as. With this configuration, the information related to the driving orientation is input to the neural network NN as much as possible, so that the reliability of the driving orientation is further enhanced. Further, when the accelerator pedal 58 is re-depressed during the holding by the maximum value holding means 139, the holding release means 141 causes the maximum holding means 139 to update the throttle valve opening section maximum value updating means 134 and the engine rotation speed. The hold commanded by the speed section maximum value updating means 135 and the longitudinal acceleration section maximum value updating means 136 is released.

When the tip-in judging means 142 judges a tip-in operation, which is a short-time rapid opening / closing operation of the accelerator pedal 58, the throttle valve opening section maximum value updating means 134 and the engine rotation speed section maximum value are updated. The updating of the maximum value by at least one of the means 135 and the longitudinal acceleration section maximum value updating means 136 is blocked. As a result, the information relating to the road surface state such as obstacles on the road surface is removed rather than the driving direction, so that the estimation accuracy of the driving direction is improved.

FIG. 14 is a flow chart for explaining the main part of the control operation of the electronic control unit 42, that is, the driving direction estimation operation. In step SM1 in FIG. 14 (hereinafter, steps are omitted), initial processing is executed,
Various storage areas or registers provided in the RAM 62, counters and timers for counting or timing, etc. are cleared, the content of the estimation permission flag X _NNCAL is cleared to "0", and the stop flag X _STOP is set. The content is set to "1". Then, the signal reading means 96
In SM2 corresponding to, the input signal from each sensor is read.

Next, in the SM3 corresponding to the preprocessing means (driving operation related variable calculating means) 98, the occurrence timing and the amount of events related to each driving operation, that is, the driving operation related variables are calculated. That is, the throttle valve opening TA _ST at the start of the vehicle, the maximum change rate A _CCMAX of the throttle opening after the start of operation, the maximum deceleration G _NMAX during the braking operation of the vehicle, the coasting time T _{COAST of} the vehicle, and the constant vehicle speed. Travel time T _VCONST , maximum value of section among input signals from each sensor within a predetermined section, for example, throttle opening TA _maxt , vehicle speed V _maxt , engine speed N _Emaxt ,
Plural kinds of driving operation related variables such as longitudinal acceleration NOGBW ( _GNMAXt ) are calculated. It should be noted that the driving orientation is not estimated in the P and R ranges.

Next, in SM4, it is judged whether or not the content of the estimation permission flag X _NNCAL is set to "1". _If the content of the estimated permission flag X _NNCAL is not set to "1", the determination at SM4 is denied and the above SM2 and _subsequent steps are repeatedly executed, but the content of the estimated permission flag X _NNCAL is set to "1". If it is set, the determination in SM4 is affirmative, and in SM5 corresponding to the driving orientation estimation means 100, the driving orientation estimation calculation is executed. The estimation permission flag X _NNCAL is set every time the driving operation-related variable is calculated, or the coasting time T
_{During COAST} measurement, during constant vehicle speed travel time T _VCONST measurement,
Alternatively, when the maximum deceleration G _NMAX during braking is being measured, the content is set to “1” at predetermined time _intervals ,
The SM5 permits the driving orientation estimation calculation of the SM4 for each set of the estimation permission flag X _NNCAL .

In SM5, the driving direction estimating means 1 is used.
As described in _No. 00, the throttle valve opening degree TA _ST when the vehicle starts, the maximum change rate A _CCMAX of the throttle valve opening degree after the start of operation, the maximum deceleration G _NMAX during the braking operation of the vehicle, and the coasting time of the vehicle. T _COAST , constant vehicle speed traveling time T _VCONST , maximum value of section among input signals from each sensor within a predetermined section, that is, throttle opening T
A neural network NN to which a plurality of types of driving operation-related variables such as A _maxt , vehicle speed V _maxt , engine rotation speed N _Emaxt , longitudinal acceleration NOGBW _maxt ( _GNMAXt ) is input is used, and based on the output signal of the neural network NN. _Is converted into a signal indicating three-step driving direction indicated by flags X _SPORT , X _NORM , and X _ECO , for example.
Driving orientation is estimated.

After the driving direction is once estimated as described above, the estimation permission flag X is calculated in SM6.
The content of _NNCAL is cleared to "0", and then the SM
2 or less is repeatedly executed.

FIG. 15 and subsequent flow charts are as follows.
The routine which shows the structural example of the content of the description SM3 is shown.
FIG. 15 corresponds to the output operation amount calculating means 98a at the time of starting.
The throttle valve opening TA when the vehicle starts
_STThe routine which calculates is shown. FIG. 16 shows the addition
Corresponds to the output operation amount maximum change rate calculation means 98b during fast operation
The maximum stroke when the accelerator pedal is depressed is
Rotor opening change rate A _CCMAXShows a routine to calculate
ing. FIG. 17 shows the constant vehicle speed traveling time calculating means 98.
which corresponds to e and has a constant vehicle speed T_VCONST
The routine which calculates is shown. FIG. 18 shows the coasting
Corresponding to the running time calculation means 98d, coasting
Running time T_COASTThe routine which calculates is shown. Figure
Reference numeral 19 corresponds to the braking maximum deceleration calculating means 98c.
Which is the maximum deceleration G during braking operation of the vehicle_NMAX
The routine which calculates is shown. 20 to 23
Corresponds to the input signal section maximum value calculating means 98f.
Input from each sensor within a predetermined section
Maximum value in the section of the signal (throttle opening TA_maxt,
Vehicle speed V_maxt, Engine speed N_Emaxt, Longitudinal acceleration NO
GBW_maxt(G_NMAXt), Etc.
The routine which calculates number is shown.

In the routine of FIG. 15 for calculating the throttle valve opening degree TA _ST when the vehicle starts, it is determined at SA1 whether or not the vehicle is traveling. If the vehicle is stopped, the determination at SA1 is denied, so after "1" is added to the content of the stop time timer T _V0 at SA2, the stop time at SA3 corresponding to the stop determination means 110 in FIG. The content of timer T _V0 is a preset reference value T
_It is determined whether _V01 or higher. This judgment reference value T _V01 is for judging whether or not the vehicle has definitely stopped, and is set to a value of about 0.2 seconds. Above S
When the determination of A3 is negative, this routine is ended, but when the determination of A3 is affirmative, the content of the stop flag X _V0 is set to "1" in SA4. This stop flag X _V0
Indicates that the vehicle has once stopped when the content is "1".

When the vehicle has traveled, the determination at SA1 is affirmative, so at SA5 it is determined whether the content of the stop flag X _V0 is "0". Since the content of the stop flag X _V0 is "1" at the beginning of the start, the above S
Since the determination in A5 is affirmative, it is determined whether or not the vehicle speed V becomes equal to or higher than the preset vehicle speed V ₁ set in SA6 corresponding to the set vehicle speed arrival determination means 111 in FIG.
The set vehicle speed V ₁ is for determining when the vehicle starts, and is set to a value of about 10 km / h, for example.

The vehicle speed V at the start is the set vehicle speed V₁Don't reach
The decision of SA6 is denied when this routine ends.
However, the vehicle speed V at the start is the set vehicle speed V₁Is over
Then, since the determination of SA6 is affirmative, the output at the time of starting of FIG.
At SA7 corresponding to the manipulated variable determining means 112,
Throttle valve opening degree TA when opening the throttle valve when starting
Degree TA_STIs stored as
_V0And stop flag X _V0Is cleared to "0",
And the estimated permission flag X_NNCALIs set to "1"
Be done.

Next, in the routine of FIG. 16 for calculating the maximum throttle opening change rate A _CCMAX when the accelerator pedal is depressed, at SB1, the accelerator pedal 58 shows that the depression speed is constant or is decelerating. It is determined whether or not the content of the _depression deceleration flag X _ACCTA is "1". Initially, the judgment of SB1 is denied, so
In SB2, the positive throttle opening change rate (for example, the throttle opening TA read at a predetermined cycle of about several tens of ms).
Difference) A _CCTA (%) is a preset judgment reference value K
_It is determined whether _ACTAMX has been _exceeded . This judgment reference value K
_ACTAMX is a preset value for removing a gentle depression, and is set to about 6%, for example.

If the determination at SB2 is negative, SB
Proceed to 20 . However, if the determination at SB2 is affirmative, at SB3, the positive throttle opening change rate A
_CCTA is whether below the maximum value A _CCmax stored until then is determined. If the determination in SB3 is denied, that is, if the stepping operation amount A _CCTA is equal to or more than the previous value,
In SB4 corresponding to the throttle valve opening rate of change maximum value updating means 114 in FIG. 9, after the pedal operation amount A _CCTA input in this cycle is updated as the maximum operation amount A _CCMAX , the accelerator pedal deceleration flag X is set in SB5. _After the content of _ACCTA is cleared to "0", this routine is ended.

As described above, when the accelerator depression speed decreases after the maximum value A _CCMAX of the positive throttle opening change rate A _CCTA due to the accelerator _depression is stored, a short accelerator opening / closing operation (depression) is performed at SB6 and below. The maximum throttle opening change rate A _CCMAX and the throttle valve opening A _{CCMXTA at} that time by the continuous acceleration operation that reflects the intention to accelerate are determined and the estimation permission flag X _NNCAL is set. It is supposed to be set.

That is, if the positive throttle opening change rate A _CCTA sequentially read in a predetermined cycle _falls below the maximum value A _CCMAX stored so far, the determination at SB3 is affirmative, so at SB6. It is determined whether or not the content of the accelerator deceleration flag X _ACCTA is "1". Since the determination in SB6 is initially denied, the throttle valve opening TA at this time is stored in _SB7 as the throttle valve opening during accelerator deceleration, that is, the maximum throttle valve opening A _CCMXTA . At SB8, the timer C _SHRT is started and the content of the accelerator deceleration flag X _{ACCT A} is set to "1".

Next, at SB9, the above-mentioned timer C
It is determined whether or not the _SHRT timing operation has stopped. Since the determination at SB9 is initially denied, it is determined at SB10 whether or not the content of the timer C _SHRT is _{equal to} or greater than the preset determination reference value K _SHRT . This judgment reference value K _SHRT is for judging the period immediately after deceleration of the accelerator pedal, and for example, a value of about 0.1 seconds is used.

Initially, immediately after the accelerator pedal deceleration, the determination at SB10 is denied, so at SB11,
It is determined whether the throttle valve opening degree TA is larger than the throttle valve opening degree A _CCMXTA at the time of deceleration of the accelerator stored in SB7. If the determination in SB11 is negative, the throttle valve opening degree TA is in a reduced state, so this routine is ended, but if the determination is affirmative, in SB12, the throttle valve opening degree TA at that time is ended. Is the throttle valve opening A when decelerating the accelerator
_After being updated as _CCMXTA , this routine is terminated and the above steps are repeated.

The content of the timer C _SHRT is the reference value K.
If the judgment of SB10 above is affirmed when the _SHRT or higher is reached,
In SB13, it is determined whether or not the timed content of the timer C _SHRT is larger than the addition value (K _SHRT + T _MAXTA = 0.2 seconds or so) of the preset judgment reference value K _SHRT and the judgment reference value T _MAXTA. It The determination reference value T _MAXTA is a value for determining the timing for determining a so-called tip-in operation of the accelerator pedal 58 (a sudden opening / closing operation of the accelerator pedal, that is, a fluttering operation), and is, for example, 0.
It is set to a value of about 1 second. If the determination at SB13 is negative, this routine is ended, but if the determination is affirmative, after the timer C _SHRT timing operation is stopped at SB14, at SB15, the throttle valve opening T
It is determined whether A is larger than a preset determination reference value K _TACHIP . This judgment reference value K _TACHIP is a value for judging the above-mentioned tip-in operation, for example, 33
A value of about% is used.

Here, the shorter the judgment reference time (K _SHRT + T _MAXTA ) used in the _{SB 13} , the more advantageous it is to improve the driving-oriented responsiveness, but the reliability of the judgment of the chip-in operation is high. Since it will not be obtained, it is set to a value as short as possible within the range where the reliability of the tip-in operation determination can be obtained, so the maximum throttle opening change rate A _CCMAX at the time of accelerator depression can be promptly obtained,
Driving-oriented responsiveness is enhanced.

If the determination at SB15 is positive,
Judgment standard after depressing the accelerator pedal 58 speed
Value (K_SHRT+ T_MAXTA= 0.2 seconds)
Even if the throttle valve opening TA is the reference value K_TACHIP(3
3%) and continuous depression operation, for example, in FIG.
It is a continuous stepping operation as shown in 5, and the tip-in operation
Since it is not a work, the maximum throttle valve opening change rate determination in Fig. 9
Determining means 116 and maximum throttle valve opening determining means 11
In SB16 corresponding to 7, the estimated permission flag X
_NNCALIs set and the positive throttle opening changes
Rate A_CCTAMaximum value A _CCMAXAnd throttle at that time
Valve opening A_CCMXTAIs input to the neural network NN
Predetermined storage location for storing signals EVENT₆, EVENT₇Navel
Will be remembered.

However, if the determination at _{SB15 is negative} , the throttle valve is _opened even after the reference value (K _SHRT + T _MAXTA = about 0.2 seconds) has elapsed since the depression speed of the accelerator pedal 58 was decelerated. Degree TA is the reference value K
Since the operation is less than _TACHIP (about 33%), SB
At 17, it is determined whether the throttle opening change rate A _CCTA is positive. If the determination at SB17 is affirmative, at SB18 it is determined whether the throttle valve opening TA is greater than or equal to the redetermination reference value K _THRS . This re-judgment reference value K _THRS is the judgment reference value K _TACHIP (33%
The value is set to a value smaller than (degree), and for example, a value of about 20% is used.

If the determination at SB18 is positive,
Since it is considered that it is not the tip-in operation, the SB16
Although but is executed, if a negative, for example, it is considered to be a tip-in operation of the accelerator pedal 58 as shown in FIG. 26, the content of the tip-in flag X _ChipIn representing the tip-in operation in SB19 is "1 SB ”and subsequent steps are executed. When the throttle opening change rate A _CCTA is not positive, the SB1
Since the determination of No. 7 is denied, SB20 and subsequent steps are executed after SB19 is executed. Further, in the state in which the timer C _SHRT is stopped after the depression operation, the determination in SB9 is affirmative, and therefore SB20 and subsequent steps are executed.

In this embodiment, the accelerator pedal 58 is depressed.
The judgment reference value (K _SHRT+ T_MAXTA=
Throttle valve opening even after only 0.2 seconds)
TA is the criterion value K_TACHIP(About 33%) or less
Throttle opening change rate A_CCTAIs zero to negative, or
In other words, the throttle opening change rate A_CCTAIs positive
Throttle valve opening TA is the re-determination reference value K_THRSLess than
In this case, the tip-in operation of the accelerator pedal 58 is determined.
Therefore, in this embodiment, SB15, SB17, SB
18, SB19 is the tip-in operation of the accelerator pedal 58
The chip-in operation determination means 115 of FIG.
I am responding.

At SB20, it is determined whether the negative throttle opening change rate D _ECTA is positive, that is, whether the rate of decrease of the throttle opening TA is positive. This S
If the determination in B20 is affirmative, the accelerator pedal is being returned, and therefore, in SB23, the maximum value A _CCMAX of the positive throttle opening change rate A _CCTA and the accelerator _depression deceleration flag X _ACCTA are cleared. After this, this routine is ended and the above steps are repeated.

However, if the determination at SB20 is negative, the accelerator pedal 58 is not being returned, and therefore, at SB21, a short accelerator pedal opening / closing operation, that is, a fluttering (chip-in) operation is indicated. _After the content of the chip-in flag X _CHIPIN is cleared to "0", the accelerator pedal deceleration flag X is set in SB22.
It is determined whether or not the content of _ACCTA is "0". If the determination in SB22 is negative, the routine is immediately terminated, but if the determination is affirmative, the above SB23 is performed.
After this, this routine is ended.

FIG. 17 for calculating the constant traveling time of the vehicle speed
In the routine of _{step S1, it} is determined in SC1 whether the maximum throttle opening change rate A _CCMAX is already stored in the SB _{16 and} is not zero. If the determination in SC1 is affirmative, the maximum throttle opening change rate A _CCMAX has not been stored yet, and _therefore the time _keeping operation of the vehicle speed constant time counting timer C _VCON and the starting timer C _VCON2 is started in SC2. The actual vehicle speed V at that time is stored as the vehicle speed V _CONT when the vehicle starts traveling at a constant vehicle speed, and then this routine is ended. That is, the above-mentioned SC2 is repeatedly executed until the determination of the above-mentioned SC1 is denied, and the vehicle speed constant time counting timer C _VCON and the start-up timer C _VCON2
Resetting and restarting, and resetting the vehicle speed V _CONT at the start of traveling at a constant vehicle speed are repeated.

When the judgment at SC1 is denied, SC3
At, it is determined whether or not the timekeeping operation of the startup timer C _VCON2 is stopped. If the determination at SC3 is affirmative, the above SC2 and subsequent steps are executed again, but if not, SC4, SC5 corresponding to the constant vehicle speed traveling determination means 120 of FIG. 10 for determining the constant vehicle speed traveling state, SC6
Is executed. First, at SC4, the throttle valve opening T
It is determined whether A is larger than a preset reference value K _THRM . This judgment reference value K _THRM is a value for judging whether or not the _vehicle is running under a relatively high load, and is set to a value of about 30%, for example. Then, in SC5,
It is determined whether or not a value obtained by subtracting the determination reference width ΔV preset to the vehicle speed V exceeds the vehicle speed V _CONT at the time when the vehicle starts traveling at a constant vehicle speed. At SC6, the determination reference width ΔV preset to the vehicle speed V is set. It is determined whether or not the added value is lower than the vehicle speed V _CONT at the start of traveling at the constant vehicle speed. The judgment reference width ΔV is a value for judging the fluctuation range of the vehicle speed, and for example, a value of about 1 km / h is used.

If any of the above SC4, SC5, SC6 is affirmed, it means that the vehicle is running under high load or the vehicle speed V fluctuates, and the vehicle speed is not constant. Therefore, first in SC7. , A fixed vehicle speed timekeeping timer C _VCON is set to a judgment reference value K of about 1 second
It is determined whether _VCON or more. If the determination in SC7 is negative, the routine is ended because the vehicle is in a high load traveling state or a vehicle speed varying traveling state. If the determination at SC7 is affirmative, SC
8, the time counting operation of the startup timer C _VCON2 is stopped, and subsequently, in SC9 corresponding to the first time counting means 119 of FIG.
_It is determined whether or not the content of _VCON has become _{equal to} or greater than a preset determination reference value K _VCONAV . This judgment reference value K
_VCONAV is a constant travel duration for determining whether the vehicle travels at a constant speed, and is set to a value of, for example, about 3 to 4 seconds. As described above, in the traveling state in which the vehicle is running under a high load or the vehicle speed V varies, the determination at SC9 is also negative, so that this routine is ended. While the vehicle is traveling under such a high load or the traveling state in which the vehicle speed V fluctuates continues, the determination at SC3 is affirmative in the next cycle and thereafter, and SC2 is repeatedly executed.

However, when a relatively low load vehicle speed constant running state is started, the determinations at SC4, SC5, and SC6 are all denied, so the start-up timer C at SC10.
_It is determined whether or not the content of time _measurement of _VCON2 is still smaller than the preset determination reference value K _VCON . This judgment reference value K
_VCON is a value for determining that the determinations of SC4, SC5, and SC6 are all denied for the activation period, and for example, a value of about 1 second is used.

If the determination at SC10 is positive,
Since the state in which the determinations of SC4, SC5, and SC6 are all denied is a state in which it does not last for more than 1 second, this routine is ended. Therefore, in SC11, the timekeeping operation of the start-up timer C _VCON2 is restarted and the vehicle speed V at that time is set as the vehicle speed V _CONT at the start of the constant vehicle speed running for the determination of the next relatively low load vehicle speed constant running state. Remembered.

Next, at SC9, the content of the vehicle speed constant time counting timer C _VCON is set to a preset judgment reference value K.
_It is determined whether _VCONAV or more. If the determination in SC9 is negative, it means that the elapsed time from the determination of the relatively low load vehicle speed constant traveling state has not _exceeded the determination reference value K _VCONAV set to about 3 to 4 seconds. Therefore, this routine is ended and the above steps are repeated.

However, if the determination at SC9 is affirmative, at SC12 the vehicle speed constant time counting timer C
_It is determined whether or not the content of _VCON exceeds a preset reference value K _VCON2 . The determination reference value K _VCON2 is an upper limit guard value of the vehicle speed constant time counting timer C _VCON , and is set to a value of about 16 seconds, for example. If the determination in SC12 is negative, in SC13, the content of the vehicle speed constant time counting timer C _VCON indicating the vehicle speed constant travel time T _VCONST is stored in a predetermined storage location EVENT ₈ storing the input signal to the neural network NN. but it is the, if affirmative been in SC14, the upper limit guard value K _{VC ON2} is the maximum limit value of the vehicle speed constant running time T _VCONST is stored in the memory location EVENT _8. Then, in SC15, the content of the estimation permission flag X _NNCAL is set to "1", and then this routine is ended.

As described above, when the vehicle speed constant traveling time calculating means 98e is configured by the routine of FIG. 17, SC4 to SC6 have passed the duration K _VCONAV for initially determining the vehicle speed constant traveling. This corresponds to the constant vehicle speed traveling determination means 120 of FIG. 10 that determines the constant vehicle speed traveling under the condition that the time is measured by SC9 corresponding to the first timing means 119. SC11 and SC12 are the initial constant vehicle speed traveling determinations. At a constant vehicle speed T
_A constant vehicle speed traveling time determining means of FIG. 10 for determining _VCONST and also determining a constant vehicle speed traveling time T _VCONST for each predetermined start cycle of each time period K _VCON measured by the SC 10 corresponding to the second time measuring means 121. It corresponds to 122. As a result, the delay in estimation is significantly reduced, as compared with the case where a command for estimation calculation of driving orientation is issued at the time of completion of traveling at a constant vehicle speed.

Even when the constant vehicle speed running is completed, if the time corresponding to the predetermined period has elapsed after the constant vehicle speed running is completed, the determination at SC7 is affirmative and the determination at SC9 following SC8 is made. Is positive, the _advantage is that the vehicle speed constant traveling time T _VCONST is determined and the content of the estimation permission flag X _NNCAL is set to "1". Therefore, even after the vehicle travels at a constant vehicle speed by SC4, it is possible to output the permission for the periodical driving direction estimation calculation only once.

The routine of FIG. 18 for calculating the coasting time T _COAST is, for example, when the idle switch in the throttle sensor 70 is in the ON state and the brake switch 84 is in the ON state.
Is started while the coasting state, which is the off state, is detected and during the timing operation of the coasting traveling time counting timer C _{IL ON} . In SD1 of FIG. 18, for example, it is determined whether the vehicle is coasting based on the fact that the vehicle speed V shows a positive value when the throttle valve opening TA is zero.
If the determination in SD1 is negative, it means that coasting such as acceleration traveling or deceleration traveling has ended, so S
After the start timer C _ILON2 is stopped at D2,
SD5 and below are executed. During acceleration traveling, the coasting traveling time counting timer C _ILON is stopped, the determination at SD5 is denied, and this routine is ended.

If the determination at SD1 is affirmative, S
At D3, it is determined whether or not the content of the startup timer C _ILON2 is smaller than a preset determination reference value K _ILON . The determination reference value K _ILON is a value corresponding to the activation cycle of the driving orientation estimation calculation during coasting, and for example, a value of about 1 second is used. If the determination at SD3 is affirmative, this routine is ended, but if the determination is negative, at SD4, after the start timer C _ILON2 is restarted, at SD5, the throttle sensor 7
The content of the coasting traveling time counting timer C _ILON that is timed while the coasting state in which the idle switch in 0 is on and the brake switch 84 is off is detected is _{equal to} or greater than the preset reference value K _AVEILON. It is judged whether or not it has become. This judgment reference value K _AVEILON is the duration of the coasting state for judging the coasting initially,
For example, a value of about 1.3 seconds is used.

[0098] If the determination in SD5 is negative, but is terminated is the routine because it is a state which can not be determined from the coasting, if affirmative been, the coasting time counting timer C _Ilon in SD6 It is determined whether or not the content is smaller than the preset determination reference value K _ILON2 . The determination reference value K _ILON2 is an upper limit guard value of the coasting traveling time counting timer C _ILON , and is set to a value of about 16 seconds, for example. If the determination in SD6 is affirmative, in SD7, the contents of the coasting traveling time counting timer C _ILON indicating the coasting traveling time T _C0AST are stored in a predetermined storage location EVENT ₉ for storing the input signal to the neural network NN. If the _answer is _{NO, the} upper limit guard value K _ILON2 indicating the maximum limit value of the coasting time T _C0AST is stored in the predetermined storage location EVENT ₉ in _SD8 .

Next, in SD9, the content of the estimation permission flag X _NNCAL is set to "1". And SD
At 10, it is determined whether or not the startup timer C _ILON2 is stopped. If the determination of SD10 is denied, this routine is ended, but if the determination is affirmative, the timing operation of the coasting traveling time counting timer C _ILON is also stopped at _SD11 .

As described above, the coasting traveling time calculating means
In the case where 98d is configured by the routine of FIG.
Indicates SD1 and SD5 are the throttle valve opening while driving.
The state such as TA becomes zero for a predetermined time K_AVEILONLasting
FIG. 11 that determines coasting of the vehicle based on things
SD7 and SD corresponding to the coasting traveling determination means 124 of
8, the coasting traveling determination means 124
SD3 corresponding to the third timing means 125 in the determined state
Coasting time T at each elapse of the starting cycle timed by
_COASTCoasting travel time determining means 126 for repeatedly determining
It corresponds to. As a result, even while the vehicle is coasting,
Judgment reference value K of about 1 second_ILONSpecified start cycle defined by
Coast running time T every period_C0ASTIs estimated as
Permission flag X _NNCALIs set to "1"
In this way, the driving-oriented estimation calculation is performed every predetermined period.
And high responsiveness is obtained.

In the routine of FIG. 19 for calculating the maximum deceleration G _NMAX during the braking operation of the vehicle, in SE1, the vehicle speed V _BK at the start of the braking operation is set to a preset vehicle speed lower limit value K.
_It is determined whether it is greater than _{BKST 1} . Also, this S
When the determination in E1 is affirmative, it is determined in SE2 whether the vehicle speed V _BK at the start of the brake operation is lower than a preset vehicle speed upper limit value K _BKST2 . The vehicle speed lower limit value K _BKST1 and the vehicle speed upper limit value K _BKST2 are lower limit values and upper limit values for setting a vehicle speed range for calculating the maximum deceleration G _NMAX during braking, for example, a value of about 25 km / h and 185 km. A value of about / h is used for each.

If either of the determinations of SE1 and SE2 above is denied, this routine is ended, but if both determinations are affirmed, the pulse interval detected by the vehicle speed sensor 76 at SE3 It is determined whether or not the longitudinal acceleration NOGBW of the vehicle calculated from the change is greater than or equal to zero. If the determination in SE3 is affirmative, it means that deceleration due to braking has not occurred, so this routine is ended.
Since NOGBW is a negative value and the deceleration action due to braking is occurring, SE4 and subsequent steps for calculating the maximum deceleration G _NMAX during braking operation are executed.

In SE4, the longitudinal acceleration NOGBW of the vehicle
Is smaller than the value in the storage location MAXBKG which stores the maximum value of the braking period. This memory location MAXBKG
The value inside is set to zero at the moment when the braking operation is started. Normally, the longitudinal acceleration NOGBW increases toward the negative side, and therefore the determination at SE4 is affirmed and the value in the memory location MAXBKG is updated to a new longitudinal acceleration NOGBW at SE5. As a result, the largest negative value is stored as the value in the storage location MAXBKG. Then, after the start-up timer C _BK is started in SE6, SE7
At, it is determined whether or not the content of the flag X _BKGSM indicating a relatively strong braking operation is "1".

Since the determination of SE7 is initially denied, it is determined at SE8 whether the longitudinal acceleration NOGBW of the vehicle is larger than a preset determination reference value K _SPBKG . The determination reference value K _SPBKG is a negative value for determining a relatively strong brake operation that affects the estimation of driving (acceleration) orientation, and is experimentally obtained in advance. If the braking operation is relatively weak, the determination at SE8 is affirmative.
In the case of a relatively strong braking operation, the determination in SE8 is denied, so the content of the flag X _BKGSM indicating a strong braking operation is set to "1" in SE9. And S
At E10, it is determined whether the value in the storage location MAXBKG is less than or equal to a preset determination reference value K _BKGAVE .
This judgment reference value K _BKGAVE is the above judgment reference value K in order to judge the relatively strong braking necessary for judging the driving orientation.
_It is a value smaller than _SPBKG , that is, a value set on the positive side, and is experimentally obtained in advance. And the above SE1
When the determination of 0 is denied, this routine is terminated, but when the determination is affirmative, the value in the memory location MAXBKG indicating the maximum deceleration G _NMAX during braking is the input signal to the neural network NN in SE11. The estimated permission flag X _NNCAL is set to “1” while being stored in a predetermined storage location EVENT ₁₀ for storing.

_Once the content of the flag X _BKGSM is set to "1" as described above, the determination of SE7 in the next cycle is affirmed, so the longitudinal acceleration NOGBW is equal to or greater than the value in the storage location MAXBKG (decrease). Unless the speed becomes equal to or less than), SE8 and SE11 are not executed. Therefore, while the above steps are repeatedly executed, if the longitudinal acceleration NOGBW becomes equal to or greater than the value in the memory location MAXBKG,
Since the determination in SE4 is denied, it is determined in SE12 whether the longitudinal acceleration NOGBW is the same as the value in the memory location MAXBKG, that is, the maximum deceleration up to that point.

If the longitudinal acceleration NOGBW is the same as the value in the storage location MAXBKG, the determination at SE12 is affirmative, so that at SE13, the content of the activation timer C _BK is _{equal to} or greater than the preset value K _BKCON. It is judged whether or not it has become. This set value K _BKCON is a range in which the longitudinal acceleration NOGBW does not become smaller than the judgment reference value K _BKGAVE which is smaller than the judgment reference value K _SPBKG once it exceeds the judgment reference value K _SPBKG . Maximum value MAXBKG
More preset hysteresis value K _BKGHYS (positive value)
Within the period, it is a cycle in which the driving direction estimation permission is repeatedly issued in order to enhance the driving direction responsiveness, for example, 0.
A value of about 2 seconds is used.

When the determination at SE13 is negative,
Since the cycle has not yet been reached, it is determined at SE14 whether the brake has been turned off. If the determination in SE14 is negative, braking is in progress, so the above steps are repeated after the end of this routine. While the steps are repeated, if the determination at SE13 is affirmative, the start timer C _BK is started at SE15 and the steps from SE10 onward are executed. That is, the memory location MAXB
If the value of the KG is equal to or less than a preset determination reference value K _BKGAVE the storage location value in MAXBKG braking time maximum deceleration G
_It is stored as _NMAX and the estimated permission flag X
The content of _NNCAL is set to "1".

If the longitudinal acceleration NOGBW is not the same as the value in the memory location MAXBKG, that is, if the longitudinal acceleration NOGBW becomes small, the determination at SE12 is denied, so S
At E16, whether or not the value obtained by subtracting the preset hysteresis value K _BKGHYS (positive value) from the longitudinal acceleration NOGBW is less than or equal to the maximum deceleration value MAXBKG stored up to that point, that is, the longitudinal acceleration NOGBW is the maximum It is determined whether or not the hysteresis value K _{BKGH YS} has become smaller than the value MAXBKG. When the determination in SE16 is denied, the longitudinal acceleration NOGBW is still not in the state of becoming smaller than the maximum value MAXBKG, and therefore it is determined in SE18 whether or not the timer C _BK is stopped. If the determination in SE18 is negative, the timer C _BK is stopped in _{SE19 and the} flag X _BKGSM indicating a strong braking operation is _set.
After the content of is reset to "0", the SE10 and below are executed to store the maximum deceleration MAXBKG, and
The estimation permission flag X _NNCAL is set to "1".

However, if the determination at SE18 is affirmative, this routine is ended. Then, while the above steps are repeatedly executed, if the determination at SE16 is affirmative, at SE17 it is determined whether or not the timer C _BK is stopped. If the determination in SE17 is negative, the steps from SE13 onward are executed and the startup timer C
Every time the arrival of the activation cycle K _BKCON is decided by _BK ,
The steps from SE10 onward are executed to store the maximum deceleration MAXBKG and the estimation permission flag X _NNCAL is set to "1". However, if the determination at SE17 is affirmative, this routine is ended.

The point shown in FIG. 27 is that the maximum deceleration MAXBKG is stored by the operation of FIG.
_It shows the time when _NNCAL is set to "1". According to the present embodiment, when the magnitude of the longitudinal acceleration NOGBW once exceeds the judgment reference value K _SPBKG , it is within a range that does not become smaller than the judgment reference K _BKGAVE set to a value smaller than the judgment reference value K _SPBKG. While the longitudinal acceleration NOGBW of is within the preset hysteresis value K _BKGHYS (positive value) from the maximum value MAXBKG of the previous deceleration, the maximum deceleration is reduced for each start cycle K _BKCON set to about 0.2 seconds. Since the speed MAXBKG is stored and the estimation permission flag X _NNCAL is set to “1” to permit the estimation of the driving orientation, there is an advantage that the responsiveness for obtaining the estimation result of the driving orientation can be suitably obtained.

The above-mentioned FIG. 1 corresponding to the braking maximum deceleration calculating means 98c for calculating the maximum deceleration G _NMAX during braking of the vehicle
In the routine of FIG. 9, SE5, which sequentially stores and updates the longitudinal acceleration NOGBW that increases to the negative side during braking of the vehicle, corresponds to the braking maximum deceleration updating means 130 of FIG.
When the longitudinal acceleration NOGBW is smaller than the preset judgment reference value K _SPBKG 11, the value NOGBW stored by the braking maximum deceleration updating means 130 is determined as the braking maximum deceleration MAXBKG, that is, G _NMAX. The subsequent longitudinal acceleration NOGBW (negative) is the maximum deceleration MAXBKG during braking.
Or when the braking maximum deceleration MAXBKG is greater than the braking maximum deceleration MAXBKG (on the zero side) and the braking maximum deceleration MAXBKG is not more than a predetermined value K _BKGHYS , the fourth timing means 131.
The maximum deceleration that is periodically determined as the braking maximum deceleration MAXBKG by the value determined as described above (the value initially determined as a value smaller than the judgment reference value K _SPBKG ) for each predetermined time K _BKCON timed by It corresponds to the determination means 132.

20 to 23 corresponding to the input signal section maximum value calculating means 98f, the section maximum value of the input signals from the respective sensors within the section, that is, the throttle opening, is set for each predetermined section of, for example, about 3 seconds. Degree TA _maxt ,
_Vehicle speed V _maxt , engine speed N _Emaxt , longitudinal acceleration NO
A plurality of types of driving operation-related variables such as GBW _maxt ( _GNMAXt ) are calculated. First, in SF1 of FIG. 20, the throttle valve opening TA and the engine rotation speed N _E are read and stored in predetermined storage locations INPVAL ₁ and INPVAL ₂ , respectively. Next, in SF2, it is determined whether or not the actual shift speed SHIFT1 of the automatic transmission 14 is equal to or lower than the SHIFT requested by the shift determination.

If the determination at SF2 is affirmative, it means that there is no upshift or no shift.
In 3, it is determined whether the actual shift speed SHIFT1 at that time is the same as the SHIFT requested by the shift determination. If the determination in SF3 is affirmative, it means that there is no gear shift. Therefore, _after the content of the upshift flag X _PTUP is cleared to "0" in SF4, in _SF5 , the timing operation is started for a predetermined period from the time of the upshift determination. It is determined whether the timer C _GMCAN is stopped.
If the determination at _SF5 is affirmative, it means that it is not immediately after the downshift and the timer _CGMCAN is determined at SE7.
Is stopped and the content of the upshift flag X _PTUP is cleared to "0", the longitudinal acceleration NOGBW (absolute value) indicating the maximum deceleration G _NMAXt in the predetermined section is calculated in SF8 and the predetermined value is calculated. Each is stored in the memory location INPVAL ₃ and SF13 and subsequent steps are executed.

However, if the determination at _SF5 is _{negative, the} timer C _GMCAN, which operates the timekeeping only during the upshift or the downshift 4 → 3 downshift, is not stopped, that is, immediately after the upshift. Therefore, in SF6, it is determined in SF6 whether or not the content measured by the timer C _GMCAN is less than or equal to a preset determination reference value K _GMCAN . This judgment reference value K _GMCAN is a value set to be larger than the shift period in order to judge the shift period, and for example, a value of about 1.5 seconds is used. This SF
If the result of the determination in No. 6 is negative, it means that the gear change period is not in progress, so SF7 and SF8 are executed and the longitudinal acceleration NO
GBW is stored, but if the determination in SF6 is affirmative, SF8 is not executed so as not to store the longitudinal acceleration NOGBW that is generated in connection with gear shifting regardless of the driver's orientation because the gear shifting period is in progress. Then, SF13 and subsequent steps are executed.

When the determination in SF2 is negative, the downshift determination is being performed and the downhill control is 4 → 3.
SF9 so that the longitudinal acceleration NOGBW when shifting is not remembered
At, it is determined whether or not the brake is being operated. If the determination in SF9 is denied, the descending slope control 4 → 3
Since the gear shift state is not generated, SF7 and below are executed. However, if the determination in SF9 is affirmative, it is determined in _SF10 whether the timer C _GMCAN is stopped. If the determination in _SF10 is negative, the upshift flag X _PTUP is determined in SF11.
It is determined whether or not the content of "1" is "1", but if the result is affirmative, the timer C _{GM CAN} is started in _{SF12 and} the content of the upshift flag X _PTUP is "1".
Is set to. In the downshift during braking, that is, in the downshift of downhill control, the determination at SF11 is denied, so the timer _CGMCAN is started at _{SF12 and} the content of the upshift flag X _PTUP is "1".
However, in the upshift, the determination in SF11 is affirmed and SF6 and subsequent steps are executed. Further, in the upshift request state, the determination in SF3 is denied, so that the above SF10 and thereafter are executed.

As a result, the longitudinal acceleration NOGBW is not stored in the downshift shift during the brake operation performed by the upshift shift and the downhill control, because it occurs regardless of the driving direction. That is, in the present embodiment, SF3, SF6, and SF9 prohibit the input in order to prevent the longitudinal acceleration NOGBW from being stored during the upshift shift and the downshift shift during the braking operation performed in the downhill control. This corresponds to the longitudinal acceleration input prohibiting means 137 of.

When the step for storing the longitudinal acceleration NOGBW is executed as described above, it is determined in SF13 whether or not the shift lever is operated to the N range. If the determination in SF13 is positive,
In SF17 of FIG. 21, the storage locations INPVAL ₁ , IN
Throttle valve opening TA stored in PVAL ₂ and INPVAL ₃
The engine speed N _E , the longitudinal acceleration NOGBW, the accelerator return speed D _ECTA , and the actual gear shift SHIFT1 are stored in predetermined memory locations EVENT ₁ , EVENT ₂ , EVENT ₃ , EVENT ₄ , and EVENT ₅ , respectively, and the vehicle starts. The flag X _PTST is cleared to "0" and the section timer C _MAX3 is started.

However, if the determination in SF13 is negative, the shift lever is set to D, which is used to drive the vehicle forward.
Since the vehicle is being operated to either the 2 range or the L range, it is determined at _SF14 in FIG. 21 whether or not the content of the start flag X _PTST for starting is "1". If the determination in SF14 is affirmative, it is determined in SF15 whether or not the vehicle speed V is equal to or higher than the preset determination reference value K _START . The determination reference value K _START is a value for determining whether or not the vehicle is starting, and for example, a value of about 10 km / h is used. If the determination in SF15 is negative, the routine is terminated because the vehicle is in a low speed state, but if the determination is affirmative, it means that the vehicle speed exceeds 10 km / h at the start, and therefore the estimation permission flag X _NNCAL of _SF16 is set in _SF16 . After the content is set to "1",
By executing SF17, each input signal is initially stored.

However, the judgment of SF14 is denied.
In the case of SF18, since it is not in the start state,
Change in maximum throttle opening when the accelerator pedal is depressed
Rate A _CCMAXIs already stored by the SB16 and at zero,
It is determined whether or not there is. The decision of this SF18 is denied
If the accelerator pedal is pressed, the accelerator pedal 58 is not depressed.
Therefore, the section timer C in SF19_MAX3Stopped
If it is judged whether it is in the middle and this judgment is affirmed,
Section timer C in SF20_MAX3Was started
It This section timer C_MAX3Finds the maximum value of the input signal
It is for timing the section.

Next, in SF20, the section timer C
_After starting _MAX3 , in SF21, the accelerator return speed D is determined by the accelerator return speed extraction routine.
_ECTA is required. However, in the state where the accelerator pedal 58 is depressed, the determination in SF18 is affirmative, so the SF20 is directly executed. If the determination in _SF19 is negative because the section timer C _MAX3 is in the timekeeping operation, SF21 is directly executed.

In the subsequent SF22, the hold flag X for holding the operation for obtaining the maximum value of the input signal in the predetermined section in the transient operation state where the accelerator pedal 58 is returned at the predetermined speed or the deceleration is more than the predetermined speed. It is determined whether or not the content of _MODF is "1". Since the determination in SF22 is initially denied, it is determined in SF23 whether the content of the flag X _SPORT indicating acceleration (sports) orientation is "1". If the determination in SF23 is negative, the driving is on the intermediate directing side and the transitional operation state is not performed, so SF30 and subsequent steps in FIG. 22 are executed, but if affirmative, the driving is on the acceleration directing side. Therefore, SF24 and SF2 for determining whether or not the transient operation state is set.
5, SF26 is executed.

That is, at _SF24, it is judged if the accelerator return speed D _ECTA is smaller than a preset judgment reference value K _DTAMX . This judgment reference value K _DTAMX is
This is a value for determining whether or not the accelerator is returned promptly, and for example, a value of about 13% is used. If the determination in SF24 is affirmative, in SF25, is the value in the storage location MAXBKG, that is, the maximum value (extreme value) MAXBKG of the longitudinal acceleration during brake operation, is less than or equal to the preset determination reference value K _SPBKG ? It is determined whether or not. This judgment reference value K _SPBKG is used in SE8 and is a value for judging that the longitudinal acceleration is relatively large. If the determination in SF25 is negative, the accelerator pedal is returned relatively gently, and the longitudinal acceleration is relatively small.
30 Although the following is executed, if an affirmative been, since the longitudinal acceleration even return speed is relatively gentle is relatively large state, the pending flag in service function 27 X _MODF
Is set to "1".

If the determination in SF24 is negative, it is determined in SF26 _whether the content of the chip-in flag X _CHIPIN is set to "1".
If the determination in SF26 is affirmative, it means that the accelerator return speed has become relatively fast due to the tip-in operation of the accelerator pedal 58, so SF30 in FIG.
Although the following is executed, if it is affirmative, since it is when the speed accelerator return in the absence of tip-in operation becomes relatively quickly, the flag X _MODF in SF27
Is set to "1".

When the content of the suspension flag X _MODF is set to "1" as described above, the determination at SF22 in the next cycle is affirmed, so that at SF28, A _CCTA corresponding to the accelerator depression speed is previously set. Set judgment reference value K
_It is determined whether it is larger than _ACCTAS . This judgment reference value K _ACCTAS is for judging the re-depressing operation of the accelerator pedal 58, and a value of about 3% is used, for example. If the determination in SF28 is affirmative, SF29
In, after the contents of the value N _MODF indicating the number of times to hold the said pending flag X _MODF is cleared to "0", SF 30
The following is executed, but if the determination in SF28 is negative, SF30 and below are directly executed. That is, in the state where the hold flag X _MODF is set to "1",
Every time the accelerator pedal 58 is depressed again, the contents of the holding flag X _MODF and the value N _MODF indicating the number of _holdings are cleared to “0” and the holding is released.

As described above, when the accelerator return speed D _ECTA when the accelerator pedal 58 is not rapidly opened (chip-in) is not _less than the predetermined value K _DTAMX ( _SF24 , _S24).
F26) and accelerator return speed D _ECTA is a predetermined value K
_{Although smaller} than _DTAMX, the deceleration during braking (maximum value of longitudinal acceleration NOGBW MAXBKG) is less than or equal to the specified value K _SPBKG , and when it is considered to be before the corner (SF25), the maximum value for the last 3 seconds. Hold flag X to hold
_{Since the MODF} is set, it is possible to preferably suppress the reduction of the driving-oriented estimated value due to the maximum value of the input signal within the predetermined section that temporarily decreases. Acceleration (sports) orientation There is almost no difference from the fuel consumption orientation as far as driving operation is seen before or during a corner, and information for corner determination (wheel speed, steering angle, lateral acceleration, yaw rate, etc.) This is because there is a possibility that fuel efficiency may be estimated in this control that cannot obtain In this embodiment, the above-mentioned SF24, SF25,
SF26 is, corresponds to the vehicle turning decision unit 138 determines 13 the turning of the vehicle, the SF27 sets the suspension flag X _MODF when turning of the vehicle is determined, and pending flag X _MODF is set Sometimes memory location EVEN
SF48 that blocks the storage of the section maximum value in T _i corresponds to the maximum value holding means 139 in FIG. 13 that holds the output of the section maximum value of the input signal to the neural network NN during the vehicle turning determination.

Further, in this embodiment, the holding flag X is set.
_{When MODF} is set, the accelerator pedal 58 is judged to have been operated again (SF28), by its pending flag X _{MO DF} is cleared (SF29) that the maximum value of the previous 3 seconds interval The output of is suspended. Since the larger the maximum value of the input signal when the accelerator pedal 58 is re-operated is, the more the acceleration direction is meant, the reliability of the driving direction estimated value is enhanced by inputting such information to the neural network NN. In the present embodiment, the SF 28 and SF 29 are the accelerator pedal 5
It corresponds to the hold canceling means 141 of FIG. 13 which cancels the output of the maximum value in the section (3 seconds) during the re-operation of 8.

In SF30 of FIG. 22, it is determined whether or not the braking operation is being performed. If the determination in SF30 is denied, SF32 and below are executed after the maximum deceleration value MAXBKG during braking in SF31 is set to zero, but if affirmative, SF32 and below are executed directly. To be done. SF32 to SF39 are, for example, memory locations
For the throttle valve opening TA, engine speed N _E , and longitudinal acceleration NOGBW stored in INPVAL ₁ , INPVAL ₂ , and INPVAL ₃ , respectively, a predetermined number of times is determined in order to determine the maximum value within a predetermined section (3 minutes). (K _MAXNUM -1)
That is, this embodiment is a loop-like routine that is repeated only three times.

In the loop routine of SF32 to SF39, the number of repetitions i is "1" in SF32.
When "1" is set, "1" is added to the number of repetitions i in SF38, and the number of repetitions i is determined to be a predetermined number (K _MAXNUM -1) in _SF39 , that is, K.
_{Since MAXNUM} is set to 4, SF33 to SF39 are repeatedly executed until it is judged as "3". First,
In _SF33 , it is determined whether or not the content of the chip-in flag X _CHIPIN is "1". If the determination is affirmative, it means that the accelerator pedal 58 has been rapidly opened and closed within a short time. In such a case, execution of SF34 is avoided because the maximum value is not stored.

When the determination at SF33 is negative,
The accelerator pedal 58 is opened and closed within a short time.
Since it is not in the state, in SF34, the memory location IN
PVAL _iThe input signal value stored in
A memory place for storing MAXVAL_iSignal stored in
It is determined whether it is less than or equal to the value. With this,
Read in INPVAL₁, INPVAL₂, INPVAL₃To each
Stored throttle valve opening TA and engine speed N
_E, Longitudinal acceleration NOGBW is stored in MAXVAL₁, MAXVAL₂,
MAXVAL₃Is greater than the value previously stored in
Is judged. If this SF34 judgment is affirmed
In SF35, the input signal value at that time is
Applicable storage location MAXVAL_iBe stored in. And where
By repeating the routine within the fixed interval,
Each memory location MAXVAL₁, MAXVAL₂, MAXVAL₃In the designated area
Maximum value in the period, that is, throttle valve opening TA_maxt, En
Jin rotation speed N_Emaxt, The maximum value of longitudinal acceleration NOGBW (maximum
Large deceleration G_NMAXt) Is stored. In this embodiment, the above
SF34 and SF35 indicate the throttle valve opening of FIG.
Section maximum value updating means 134, engine rotation speed section maximum
Value updating means 136, longitudinal acceleration section maximum value updating means 13
It corresponds to 6 respectively.

In the loop routine of SF32 to SF39, SF36 is executed when the number of repetitions i is 2, that is, when the engine rotation speed N _E becomes maximum.
This is for storing the shift speed SHIFT1 in the storage location MAXVAL ₅ in SF37.

Following the above loop-shaped routine, SF40
Then, section timer C_MAX3Setting area where the contents of
Interval K_MAX3Is less than or equal to. This setting area
Interval K _MAX3Is a driving-oriented neural network NN
In order to perform the estimation with sufficient responsiveness, the input signal
It is a section for repeatedly obtaining the maximum value, for example, 3
It is set to a value in seconds. The judgment of SF40 above is affirmative
If it does, the set section has not yet expired.
The routine is terminated. However, the above SF40 format
If the determination is denied, SF41 and subsequent steps are executed.

At SF41, it is determined whether or not the content of the flag X _SPORT indicating acceleration (sports) orientation is "1". If the determination in SF41 is negative, it means that the vehicle is in the normal steering mode and the normal steering shift map is selected. Therefore, in SF42, it is determined whether or not the brake operation is being performed. If this SF42 judgment is affirmed,
In SF43, it is determined whether the longitudinal acceleration NOGBW is lower than a preset determination reference value K _AVEBKG (negative value). If this SF43 judgment is affirmed,
Since a relatively strong braking operation has been performed, SF46 and below, which will be described later, are executed, but if the result is negative, the braking operation is relatively weak, and therefore the storage location for storing the maximum value in the section in SF45. MAXVAL _{1 to} MAXVAL ₅
After the content of is cleared, the section timer C _MAX3 is restarted in _SF60 . If the determination in SF42 is negative, it is determined in SF44 whether or not the akuru return operation is being performed. If the determination in SF44 is negative, the vehicle is in a non-braking operation, and therefore, the processes in and after SF46 described below are executed, but if the determination is affirmative, it means that the accelerator return operation has been performed. ,
Storage location MA for storing the maximum value in the section in SF45
After the contents of XVAL _{1 to} MAXVAL ₅ are cleared, SF60
At, the section timer C _MAX3 is restarted. That is, when the passage of the set section has expired, when a relatively weak braking operation with intermediate orientation is performed, or when an accelerator return operation is performed, the neural network N
The pointing estimation operation by N is not calculated.

If the determination at SF41 is positive,
Since the acceleration-oriented state is reached when the elapse of the set section has expired, in SF46, "1" is added to the content of the value N _{MO DF} indicating the number of times of holding. And a memory location
SF47 to S47 so that the maximum value of the input signal stored in MAXVAL _i is stored in the storage location EVENT _i for storing the signal to be input to the neural network NN.
The loop routine of F56 is executed. This SF47
In the loop routine from SF56 to SF56, if "1" is set as the number of repetitions i in SF47, SF55
Is added to the number of repetitions i at, SF5
In 6 repetition number i is SF48 to SF56 until it is determined to have reached the predetermined number K _{MA XNUM} is repeatedly executed.

[0134] First of all, hold the flag X _MODF in SF48
It is determined whether or not the content of is "1". When this determination is denied, it is determined in _SF49 whether or not the accelerator pedal depression speed (throttle opening change rate) A _CCTA is larger than a preset determination reference value K _ACCTA . This judgment reference value K _CCTA is a value smaller than K _ACCTAS for judging even a relatively gentle depression of the accelerator pedal 58, and for example, a value of about 1.3% is used. If the determination in SF49 is positive is, in the case where there is re-depression of the accelerator pedal 58, as in the case where the content of the suspension flag X _MODF in the SF48 is determined to be "1", it SF50 to allow it to be stored in EVENT _i by rewriting the interval maximum value of the input signal by the time the current input signal is greater than the stored value stored in EVENT _i is executed up.

However, if the determination in SF49 is negative, it is considered that the operation amount of the accelerator pedal 58 does not change so much and is still in the same state as the cornering, so that the execution of SF50 is avoided. And SF
51 and below are executed.

In the above-mentioned SF50, it is judged whether or not the value MAXVAL _i in the maximum value storage location is equal to or more than the neural input storage location EVENT _i . If this determination is affirmed, the content of MAXVAL _i in SF53 is determined. Is stored in EVENT _i , and the contents of MAXVAL _i are cleared in SF54. This allows storage locations MAXVAL ₁ through MAXVA
Throttle valve opening TA, engine speed N _E , longitudinal acceleration _{NO GBW} , accelerator return speed DECTA stored in L ₄
Is stored in the memory locations EVENT _{1 to} EVENT ₄ and is output to the neural network NN only when is larger than the value stored in the memory locations EVENT _{1 to} EVENT ₄ . In addition,
When it is determined in SF51 that the number of repetitions i is 2, in SF52, the shift stage SHIFT1 stored in MAXVAL ₅ is synchronized with the storage of the maximum value of the engine rotation speed N _E.
It is stored in EVENT ₅ .

Then, in SF57, the number of holding times N
_It is determined whether or not _MODF is smaller than a preset determination reference value K _{MO DF} . This judgment reference value K _MODF is the number of times of holding for holding an input signal that is _likely to be erroneously estimated for driving orientation, and is set to, for example, "1". If the determination in SF57 is affirmative, it means that the number of holding times is still insufficient. Therefore, _after the content of the estimation permission flag X _NNCAL is set to "1" in SF59, the section timer C _MAX3 is restarted in _SF60. To be

In the loop routine, SF4 is used.
If it is determined that the content of the hold flag X _MOF is “1” in 8, that is, if the determination in SF25 is affirmative or the determination in SF26 is negative, the determination in SF48 is affirmative. Since SF50 is executed by the above, only when the maximum value of the input signal in the predetermined section is larger than the value stored in EVENT _i stored so far,
EVENT value in _i is updated, in the present embodiment is not updated and if smaller than the value of the maximum value is EVENT in _i stored so far in the predetermined section of the input signal, the SF50 is pending 13 This corresponds to the middle updating unit 140.

Further, in this embodiment, the accelerator pedal 58
Since the maximum section value of the input signal is updated in SF35 only when the rapid opening / closing (chip-in) operation is not performed (SF33), the estimation of the driving orientation based on the input signal during the chip-in operation is prevented. , The accuracy of driving orientation estimation is improved. In this embodiment, SF33 corresponds to the chip-in determination means 142 of FIG.

FIG. 24 shows a routine corresponding to the maximum vehicle speed calculating means 98g. In SG1 of FIG. 24, it is determined whether or not the vehicle speed V input in a predetermined sampling period is equal to or higher than the maximum vehicle speed V _max (initially “0”) stored in a predetermined storage location. If the determination in SG1 is negative, the routine is terminated, but if the determination is affirmative, the new vehicle speed V is stored in SG2 as the maximum vehicle speed V _max . As a result, the maximum value V _max of the vehicle speed after the start of traveling is determined.

FIG. 28 shows the shift line diagram switching means 9 of FIG.
The shift line diagram switching routine corresponding to 2 is shown. Figure 2
In SH8, it is determined in SH1 whether or not the content of the acceleration directing flag X _SPORT is "1". If the determination in SH1 is affirmative, in SH2, for example, as shown in FIG.
Although the acceleration-oriented shift diagram shown in (1) is selected, if it is denied, it is determined in SH3 whether or not the content of the intermediate orientation flag X _NORM is "1". If the determination in SH3 is affirmative, in SH4, for example, the intermediate directional shift line diagram is selected, but if not, in SH5, the content of the fuel consumption directional flag X _ECO is "1". It is determined whether there is any. If the determination in SH5 is affirmative, for example, the fuel economy-oriented shift diagram of FIG. 5 is selected in SH6, but if not, this routine is ended.

FIG. 29 shows a maximum gear stage prohibiting routine corresponding to the maximum gear stage inhibiting means 93 of FIG. In FIG. 29, the shift lever 78 is D in SI1.
It is determined whether or not the range is operated. This SI
When the determination of 1 is denied, it is determined in SI1 whether or not the shift lever 78 is operated to the engine braking range that is one step lower than the D range. This D
The engine braking range that is one step below the range is 3 if the automatic transmission 14 has a fourth gear.
If the automatic transmission 14 is provided with a fifth gear, the range is four.

When the determination of SI2 is affirmative, it becomes possible to shift to the highest gear when the shift lever 78 is operated from the engine brake range which is one step lower than the D range to the D range. In addition, the highest gear is permitted in SI9. However, if the determination of SI2 is negative, if the shift lever 78 is the one in which the automatic transmission 14 is provided with the fifth speed gear stage, the third, second and L ranges, and the automatic transmission 14 is the fourth speed gear. When the gear is provided with gears, the L range is in the 2nd state, so the routine is terminated without permitting the highest gear.

When the determination of SI1 is positive,
In SI3, it is determined whether or not the uphill control is being performed, and when the determination of SI3 is denied, it is determined in SI4 whether or not the downhill control is being performed, and when the determination of SI4 is denied. Is acceleration-oriented in SI5, that is, the output signal NN of the neural network NN.
It is determined whether or not the content of the flag X _SPORT that indicates the acceleration direction set when _OUT is large is "1".
The uphill control and the downhill control are performed by a well-known routine according to whether the actual vehicle acceleration force is larger or smaller than the reference acceleration force on the flat road surface obtained from the actual throttle valve opening TA and the vehicle speed V. To be judged.

When the vehicle is in the uphill control or the downhill control to obtain the driving force or engine braking force of the vehicle,
Since the above judgment of SI3 or SI4 is affirmed, SI
At 6, the highest gear is prohibited. Further, when the output signal NN _OUT of the neural network NN is large and the content of the flag X _SPORT indicating the acceleration direction is “1”, the determination at SI5 is affirmative, so that the highest speed gear stage at SI6 is set. prohibited. This prohibition of the highest speed gear is, for example, when the automatic transmission 14 has four forward gears, the 3 → 4 shift-up line included in the shift diagram used for the shift control is removed, and the automatic transmission 14 is prevented. In the case of 5 forward gears, 4 included in the shift diagram used for shift control
→ 5 This is done by removing the shift-up line.

However, if the determinations of SI3, SI4, and SI5 are all denied, the determination of SI7 is made to determine whether the vehicle is traveling normally (constant vehicle speed), and the output signal NN of the neural network NN. _The highest speed gear of SI6 can be obtained by permitting the highest gear in SI9, provided that the determinations of SI8 for determining whether _OUT is, for example, a predetermined value K or less corresponding to the intermediate orientation are positive. Ban prohibition is lifted.

Next, another embodiment of the present invention will be described. In the following description, the same parts as those in the above-described embodiment will be designated by the same reference numerals and the description thereof will be omitted.

FIG. 30 is a functional block diagram showing another example of the control function of the electronic control unit 42 and corresponds to FIG. 30, when the vehicle turning determination unit 138 determines that the vehicle is turning, the driving direction estimation calculation prohibiting unit 143 prohibits the operation of the neural network NN in the driving direction estimation unit 100 in the driving direction estimation unit 94. , Holds the output signal NN _OUT ^-1 of the neural network NN before turning the vehicle and outputs it. As a result, FIG. 13 and FIGS.
As in the third embodiment, the output signal NN _OUT of the neural network NN is prevented from decreasing during turning of the vehicle,
It is avoided that it is determined to be fuel efficiency oriented even though it is acceleration oriented. In this embodiment, the vehicle turning determination means 138 and the maximum value holding means 139 shown in FIG. 13 may not be provided.

FIG. 31 is a flow chart for explaining the control operation of the electronic control unit 42 in the embodiment shown in FIG. 30, and SM7 and SM8 are added to the steps shown in FIG. In FIG. 31, when the determination of SM4 is affirmative, the SM7 corresponding to the vehicle turning determination means 138 of FIG. 30 determines whether or not the vehicle is turning prior to the estimation of the driving direction of SM5. To be executed. This SM7
Is the vehicle turning determination means 138 of FIG. 13 or SF2 of FIG.
4, SF25 and SF26. This SM
If the determination of No. 7 is denied, the estimation of the driving orientation is executed in SM5, but if the determination of SM7 is affirmed, the neural network in SM8 corresponding to the driving orientation estimation calculation inhibiting means 143 of FIG. NN's driving direction estimation calculation is prohibited and estimated output before turning NN
_After holding _OUT ^-1 and outputting it, SM6 and subsequent steps are executed.

FIG. 32 is a functional block diagram showing another example of the control function of the electronic control unit 42 and corresponds to FIG. In FIG. 32, the engine speed correction means 145 indicates that when the current gear is one gear lower than the highest gear and the steady running at a constant vehicle speed continues for a predetermined duration K _VCONAV or more. Is one gear lower than the highest gear (third gear: gear ratio i ₃ = 1.
0) engine speed N detected during traveling in 0)
The _E after correction to the value N _E × i _{O / D} of the fourth speed running, is updated in the engine rotational speed interval maximum value updating means 135.
This i _{O / D} is one gear below the highest gear (3rd gear).
High speed gear: Gear ratio (1 overdrive gear)
A smaller value, for example 0.8). As a result, the actual engine rotational speed N _E is corrected to a smaller value N _E × i _{O / D} , and then updated as the predetermined section maximum value input signal N _Emaxt , and is input to the neural network NN. , The output signal NN of the neural network NN during high-speed running at the highest speed (fourth speed) at a constant speed of, for example, about 120 km / h.
The size of _OUT is suppressed, and erroneous estimation that acceleration is directed is preferably eliminated.

FIG. 33 is a flow chart for explaining the control operation of the electronic control unit 42 in the embodiment of FIG. 32, and shows another example of step SF1 of FIG. 33, first, in SF1-1, the throttle valve opening degree TA is read and stored in a predetermined storage location INPVAL ₁ as in step SF1 of FIG. Next, in SF1-2, it is determined whether or not the vehicle speed constant timer C _{VCON, which} is timed when the vehicle speed is constant, is stopped. If the determination in SF1-2 is affirmative, it means that the vehicle speed is not constant. Therefore, in SF1-6, the engine rotation speed N _E is read and a predetermined memory is stored to determine the maximum value within the predetermined section. Store in place INPVAL ₂ .

If the determination in SF1-2 is negative, the vehicle speed is constant, and therefore, in SF1-3, the content measured by the constant vehicle speed timer C _VCON exceeds the preset duration K _VCONAV . Whether or not it is determined. This continuation time K _VCONAV is a value for determining steady running that is not acceleration-oriented even at high speed, and is, for example, 3 to 4
Values on the order of seconds are used. If the determination in SF1-3 is denied, the above steps SF1-6 and below are executed, but if the determination is affirmative, in SF1-4, the output signal of the neural network NN is being controlled during the uphill control, the downhill control. NN
It is determined whether or not the maximum high speed gear speed inhibition control is being performed due to the reason that _OUT is greater than the predetermined value. This SF1
-4 is denied, the above SF1-6 and below are executed, but if affirmative, the current gear stage of the automatic transmission 14 is higher than its highest gear stage in SF1-5. It is determined whether or not the gear is one gear lower.

If the determination at SF1-5 is negative, the above steps SF1-6 and below are executed, but if the determination is positive, at SF1-7, the actual engine speed N is reached.
_E is the value of traveling at the highest speed (4th speed) N _E × i _{O / D}
After being corrected to, it is stored in a predetermined storage location INPVAL ₂ . That is, the above SF1-2 to SF1-5 and S
F1-7 corresponds to the engine speed correction means 145.

Although the embodiment of the present invention has been described above with reference to the drawings, the present invention can be applied to other modes.

For example, in the above-mentioned embodiment, the neural network NN of the driving direction estimation means 100 is
Throttle valve opening degree TA _ST at start, maximum throttle valve opening change rate A _CCMAX at accelerator _depression , maximum deceleration at braking
MAXBKG, coasting time _{T COAS} T, constant vehicle speed running time T
_{Although VCONST} was input, even if any one of them or any one or more of them is input, it is possible to make a reliable estimation.

Further, in the above-described embodiment, the driving orientation is estimated from the output of the neural network NN into three stages of acceleration orientation, intermediate orientation, and fuel consumption orientation, but it is estimated in two stages of acceleration orientation and fuel consumption orientation. May be done. Furthermore, the acceleration orientation and the fuel consumption orientation may be continuously estimated. In such a case, the orientation between the acceleration orientation and the fuel efficiency orientation is estimated by the complementary function of the neural network NN.

Further, a signal value indicating the road condition of a highway, a suburban road, a mountain road, a city road, etc. is input as an index signal to the neural network NN by a predetermined road condition detecting means or a manual input means. May be. In the above embodiment, the neural network N
Since the signal input to N as an index is also influenced by the road conditions, the accuracy of estimating the driving orientation can be further improved by the above.

Further, in the above-mentioned embodiment, the predetermined time T
A value obtained by adding the weighting of the time difference up to the estimated time to the index signal value to the neural network NN calculated in ₂ and adding it may be input to the neural network NN as a new index signal. In this way, since the estimation can be performed in consideration of the past history, the influence of the disturbance is less likely to occur, and the driving-oriented estimation accuracy can be further improved.

Further, the vehicle turning determination means 1 of the above-mentioned embodiment.
38 may determine that the vehicle is turning because the steering angle detected by the steering angle sensor exceeds a predetermined value.

Also, the driving direction estimation means 1 of the above-mentioned embodiment
The neural network NN of No. 00 had a three-layer structure including an input layer, an intermediate layer, and an output layer, but may have a hierarchical structure of four layers or more, or mutual connection in which each neural cell element is connected to each other. It does not matter if it is a mold.

Further, in the above-described embodiment, the operation amount of the accelerator pedal 58 and the throttle valve opening TA are obtained using the signal from the throttle sensor 70, but the operation amount A _cc of the accelerator pedal 58 is detected. By independently providing the accelerator pedal sensor, the accelerator pedal 58
The operation amount of may be detected directly.

Further, in the above-described embodiment, the throttle valve opening TA and the maximum throttle valve opening change rate A _CCTA are used, but the vehicle not equipped with the throttle valve 68 such as a diesel engine mounted vehicle. Then, in place of the throttle valve opening TA and the maximum throttle valve opening change rate A _CCTA , the accelerator pedal operation amount and the accelerator pedal depression speed can be used.

Further, the automatic transmission 14 of the above-mentioned embodiment was a planetary gear type multi-stage transmission known as so-called A / T, but, for example, a belt type transmission described in Japanese Patent Laid-Open No. 2-271149. It may be a gear transmission.

The above description is merely one embodiment of the present invention, and the present invention can be modified in various ways without departing from the spirit of the invention.

[Brief description of drawings]

FIG. 1 is a diagram illustrating a shift control device for a vehicle automatic transmission having a driving direction estimation function according to an embodiment of the present invention.

FIG. 2 is a chart showing a combination of a gear stage and an operating state of a solenoid valve or a friction engagement device for establishing the gear stage in the automatic transmission of FIG.

FIG. 3 is a shift diagram used in the shift control device of FIG. 1, and is a shift diagram selected when it is estimated that the driving is acceleration (sports) oriented.

FIG. 4 is a shift diagram used in the shift control device of FIG. 1, and is a shift diagram selected when it is estimated that the driving is in the intermediate (normal) direction.

5 is a shift diagram used in the shift control device of FIG. 1, and is a shift diagram selected when it is estimated that the operation is directed to fuel economy (economy).

FIG. 6 is a functional block diagram illustrating a main part of a control function of the electronic control device of FIG.

7 is a functional block diagram for explaining in detail the function of the driving orientation estimation unit in FIG.

FIG. 8 is a diagram illustrating in detail the starting output control variable calculation means of FIG. 7;

FIG. 9 is a diagram illustrating in detail the output operation amount maximum change rate calculation means during acceleration operation in FIG. 7;

FIG. 10 is a diagram illustrating in detail the vehicle speed constant traveling time calculation / calculation means of FIG. 7;

FIG. 11 is a diagram illustrating in detail the coasting travel time calculation / calculation means of FIG. 7.

FIG. 12 is a diagram illustrating in detail the braking maximum deceleration calculation means of FIG. 7.

FIG. 13 is a diagram illustrating in detail the input signal section maximum value calculating means of FIG. 7.

FIG. 14 is a flowchart illustrating a driving orientation estimation control routine, which is a main part of control operation of the electronic control unit of FIG. 1.

FIG. 15 constitutes a part of the pre-processing routine of FIG.
6 is a flowchart illustrating a start throttle valve opening calculation routine.

16 is a part of the pre-processing routine of FIG. 14,
It is a flow chart explaining a maximum throttle valve opening change rate calculation routine when the accelerator pedal is depressed.

FIG. 17 constitutes a part of the pre-processing routine of FIG.
It is a flow chart explaining a constant vehicle speed running time calculation routine.

FIG. 18 constitutes a part of the pre-processing routine of FIG.
It is a flow chart explaining a coasting running time calculation routine.

FIG. 19 is a part of the pre-processing routine of FIG.
It is a flow chart explaining the maximum deceleration calculation routine at the time of braking.

FIG. 20 is a part of the pre-processing routine of FIG.
It is a figure explaining the flowchart explaining the predetermined | prescribed area maximum value calculation routine with FIG. 21, FIG. 22, and FIG.

FIG. 21 constitutes a part of the pre-processing routine of FIG.
It is a figure explaining the flowchart explaining the predetermined | prescribed area maximum value calculation routine with FIG. 20, FIG. 22, and FIG.

FIG. 22 constitutes a part of the pre-processing routine of FIG.
It is a figure explaining the flowchart explaining the predetermined area maximum value calculation routine with FIG. 20, FIG. 21, and FIG.

FIG. 23 is a part of the pre-processing routine of FIG.
It is a figure explaining the flowchart explaining the predetermined | prescribed area maximum value calculation routine with FIG. 20, FIG. 21, and FIG.

FIG. 24 is a part of the pre-processing routine of FIG.
It is a figure explaining the flowchart explaining the maximum vehicle speed calculation routine.

FIG. 25 is a time chart for explaining a depressed state of the accelerator pedal without a tip-in operation in FIG. 16.

FIG. 26 is a time chart for explaining the tip-in operation of the accelerator pedal in FIG.

FIG. 27 is a time chart illustrating an updated state of longitudinal acceleration of the vehicle in FIG. 19.

28 is a flowchart illustrating a shift diagram switching routine, which is a main part of control operation of the electronic control unit of FIG. 1. FIG.

FIG. 29 is a flowchart illustrating a maximum speed gear stage inhibition control routine, which is a main part of control operation of the electronic control unit of FIG. 1.

30 is a functional block diagram showing a control function of an electronic control device according to another embodiment of the present invention, which is a diagram corresponding to FIG. 6. FIG.

31 is a flowchart illustrating a driving orientation estimation control routine showing a main part of control operation of the electronic control unit of the embodiment of FIG. 30, and is a diagram corresponding to FIG. 14. FIG.

32 is a functional block diagram showing a control function of the electronic control device in another example of the present invention, and is a diagram corresponding to FIG. 13. FIG.

33 is a diagram showing a part of a predetermined section maximum value calculation routine showing the main part of the control operation of the electronic control unit of the embodiment of FIG. 32, and showing another example of step SF1 of FIG. 20. .

[Explanation of symbols]

14: Automatic transmission 58: Accelerator pedal (output operation part) 70: Throttle sensor (output operation amount detection means) 90: Shift control means 92: Shift diagram switching means 94: Driving orientation estimation unit 96: Signal reading means 98: Preprocessing means (driving operation related variable calculation means) 100: Driving orientation estimation means NN: Neural network

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁷ Identification code FI F16H 59:24 F16H 59:24 59:42 59:42 59:48 59:48 (72) Inventor Yasuya Nakamura Toyota City, Aichi Prefecture Toyota Town No. 1 Toyota Motor Corporation (72) Inventor Hideo Tomomatsu No. 1 Toyota Town, Toyota City, Aichi Prefecture Toyota Motor Corporation No. 1 (72) Inventor Shoichi Sayo 1 Town Toyota City, Aichi Prefecture Toyota Motor Vehicle Incorporated (72) Inventor Mitsuharu Oshima, 41, Nagachote, Nagakute-cho, Aichi-gun, Aichi Prefecture 1 of Toyota Central Research Institute Co., Ltd. No. 1 Toyota Central Research Institute Co., Ltd. (72) Inventor Masataka Osawa No. 41, Nagakute-cho, Nagakute-cho, Aichi-gun, Aichi Prefecture 1 No. 1 Toyota Central Research Institute Co., Ltd. (72) Inventor Kisaburo Kawa Kawai, Aichi Prefecture, Nagakute-cho, Aichi, Japan 1 41, Yokomichi, Yokosuka Central Research Institute Co., Ltd. (56) Kaihei 4-341657 (JP, A) JP-A-1-238748 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) F16H 59/00-63/48 B60K 41/06

Claims (5)

(57) [Claims] 1. A driving direction estimating device for estimating a driving direction of a vehicle, comprising: an output operation amount at the time of starting the vehicle; a maximum change rate of the output operation amount; a maximum deceleration during a braking operation of the vehicle; Driving operation related variable calculating means for calculating at least one driving operation related variable selected from coasting running time and constant vehicle speed running time, and neural to which the driving operation related variables calculated by the driving operation related variable calculating means are input. Equipped with a network,
Any one of the output operation amount at the time of starting the vehicle, the maximum change rate of the output operation amount, the maximum deceleration during the braking operation of the vehicle, the coasting time of the vehicle, and the constant vehicle speed traveling time is calculated by the driving operation-related variable calculation means. Every time the operation is performed, the operation amount of the driving direction estimating means for estimating the driving direction of the vehicle based on the output of the neural network and the operation amount of the output operation unit for operating the output of the engine of the vehicle are calculated.
And a driving operation-related variable calculation means for detecting the output operation amount.
The maximum change rate of the output manipulated variable detected by the
The maximum change rate of the output operation amount except when opening and closing is calculated.
A driving direction estimation device for a vehicle, characterized in that 2. A driving finger for estimating the driving direction of a vehicle.
A device for estimating the direction, the output operation amount when the vehicle starts, the maximum change rate of the output operation amount, the vehicle
Maximum deceleration during braking operation of both, vehicle coasting time, vehicle
At least one driving operation selected from constant speed travel time
Operation-related variable calculation means for calculating operation-related variables, and driving operation calculated by the operation-related variable calculation means
Equipped with a neural network into which related variables are input,
By the driving operation-related variable calculation means, the output when the vehicle starts
Force operation amount, maximum rate of change in output operation amount, during vehicle braking operation
Maximum deceleration, vehicle coasting time, constant vehicle speed
Each time any one of the
Driving that estimates the driving direction of the vehicle based on the work output
Including the pointing estimation means, the driving operation-related variable calculation means, during the coasting traveling of the vehicle or traveling at a constant vehicle speed, or after the start of the maximum deceleration detection during the braking operation of the vehicle, the coasting traveling time or the vehicle speed of the vehicle. The constant running time or the maximum deceleration at the time of braking operation of the vehicle is calculated or determined for each preset time, and the driving direction estimating means calculates the coasting time of the vehicle calculated or determined for each preset time. or on the basis of the output value maximum deceleration during braking operation of the vehicle speed constant running time or vehicle is output from the neural network for each input to the neural network, that is to estimate the driver's intention of vehicle Characteristic vehicle driving direction estimation device. 3. A driving finger for estimating the driving direction of a vehicle
A device for estimating the direction, the output operation amount when the vehicle starts, the maximum change rate of the output operation amount, the vehicle
Maximum deceleration during braking operation of both, vehicle coasting time, vehicle
At least one driving operation selected from constant speed travel time
Operation-related variable calculation means for calculating operation-related variables, and driving operation calculated by the operation-related variable calculation means
Equipped with a neural network into which related variables are input,
By the driving operation-related variable calculation means, the output when the vehicle starts
Force operation amount, maximum rate of change in output operation amount, during vehicle braking operation
Maximum deceleration, vehicle coasting time, constant vehicle speed
Each time any one of the
Driving that estimates the driving direction of the vehicle based on the work output
The driving operation-related variable calculation means repeatedly calculates the longitudinal acceleration generated in the vehicle to repeatedly input the neural network, and stores the acceleration in a predetermined storage location. If the vehicle is in a predetermined gear shift period, it further includes a longitudinal acceleration input inhibiting means for inhibiting the longitudinal acceleration generated in the gear shifting period from being input to the neural network. Dress
Place 4. A driving finger for estimating the driving direction of a vehicle.
A device for estimating the direction, the output operation amount when the vehicle starts, the maximum change rate of the output operation amount, the vehicle
Maximum deceleration during braking operation of both, vehicle coasting time, vehicle
At least one driving operation selected from constant speed travel time
Operation-related variable calculation means for calculating operation-related variables, and driving operation calculated by the operation-related variable calculation means
Equipped with a neural network into which related variables are input,
By the driving operation-related variable calculation means, the output when the vehicle starts
Force operation amount, maximum rate of change in output operation amount, during vehicle braking operation
Maximum deceleration, vehicle coasting time, constant vehicle speed
Each time any one of the
Driving that estimates the driving direction of the vehicle based on the work output
The driving operation-related variable calculating means includes at least one of a directivity estimating means, and calculates at least one of the maximum values of the output operation amount, the engine rotation speed, and the longitudinal acceleration in a predetermined section in order to repeatedly output to the neural network. If the vehicle turning determination means for determining the turning of the vehicle and the driving orientation estimation means estimate that the vehicle is in the acceleration direction and the vehicle turning determination means determines that the vehicle is turning, the predetermined A driving direction of the vehicle , further comprising maximum value holding means for holding the output of the maximum value in the section to the neural network.
Estimator. 5. A driving direction estimation device for estimating a driving direction of a vehicle, comprising: an output operation amount at the time of starting the vehicle; a maximum change rate of the output operation amount; a maximum deceleration during a braking operation of the vehicle; Driving operation related variable calculating means for calculating at least one driving operation related variable selected from coasting running time and constant vehicle speed running time, and neural to which the driving operation related variables calculated by the driving operation related variable calculating means are input. Equipped with a network,
Driving direction estimation means for estimating the driving direction of the vehicle based on the output of the neural network, vehicle turning determination means for determining the turning of the vehicle, and when the vehicle turning determination means determines the turning of the vehicle, And a driving direction estimation calculation prohibiting unit for prohibiting the driving direction estimation calculation of the neural network in the driving direction estimation unit to hold the output value of the neural network before turning the vehicle. apparatus.