JP2000104813A - Vehicular transmission - Google Patents

Abstract

PROBLEM TO BE SOLVED: To instruct the gear shift to the direction in which a fuel consumption is improved in response to a vehicular operation state and an engine load state precisely and ensure the versatility of a control system. SOLUTION: This is provided with a means (S1-S4, S6-S8) for judging a vehicular running state from the time series information of a car speed by a neural network NN1 at the semi-automatic mode time and a means (S9) which can find out an average engine fuel injection amount from the time series information of the engine fuel injection amount, at the running state time in which the fuel consumption is preceded from the judging result and an average engine rotation speed from the time series information of an engine rotation speed and an average engine torque from these values by the neural network NN2, and judge whether the fuel consumption is improved when the speed of a present step is changed by the neural network NN3 from at least the average engine rotation speed, average engine torque and the detection information of a gear position and instruct the gear shift to the direction in which the fuel consumption is improved from the judging result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a transmission for a vehicle.

[0002]

2. Description of the Related Art In an automatic transmission for a vehicle, when a shift lever in a driver's cab is set to a D (drive) range, a target shift speed corresponding to a driving state (vehicle speed or the like) of the vehicle is obtained from a shift map. The shift operation of the vehicle is controlled so that the actual shift speed matches the target shift speed. The shift lever has a D range signal, 1 according to its position.
A speed range signal, a second speed range signal, an R (reverse) range signal, an N (neutral) range signal, and a P (parking) range signal are selectively output.

[0003] Among them, there is one in which a shift instruction for urging the shift lever range to be switched is notified by an indicator or a buzzer or the like in the driver's cab (actually open 5-108).
No. 62). In this case, during the D range, when shifting to downhill traveling and the brake fluid temperature exceeds a predetermined value or when shifting to steep ascent traveling, a shift instruction prompting the shift lever range switching is output, while the second speed range is output. During the transition, when the vehicle shifts to a steep ascending speed or when the vehicle speed becomes equal to or more than a predetermined value (30 km / h), a shift instruction for urging the shift lever to switch the range is output. In addition, during the first speed range, when the vehicle speed becomes equal to or higher than a predetermined value (10 km / h) instead of traveling downhill, a shift instruction prompting switching of the first speed range is output.

[0004]

In such a conventional apparatus, when traveling on a congested road in the second speed range or the first speed range, and when the vehicle speed is equal to or higher than a predetermined value, a shift instruction for prompting the shift lever to switch the range is provided. Is output, there is a possibility that the driver receives this instruction and switches the shift lever to the D range. Therefore, when the vehicle is accelerated or decelerated in the D range on a congested road, there is a possibility that unnecessary speed change operations (shift up and down) that are contrary to the driver's intention are repeated. Therefore, it is necessary not only to cope with the state of climbing up and down, but also to issue an appropriate shift instruction according to the traveling state of the vehicle including traffic congestion.

The applicant sets a downshift instruction area and an upshift instruction area in relation to fuel efficiency, and the driving state of the vehicle enters these instruction areas only at a high speed of 80 km / h or more. When a predetermined time has elapsed in that state, an indicator that displays a shift-up instruction or a shift-down instruction on an indicator has been developed.However, the shift-down instruction area and the shift-down instruction area differ depending on the engine model and vehicle type. Also, it is necessary to set the shift instruction characteristic for the control device according to the type, which is a factor that greatly increases the cost.

An object of the present invention is to provide means for solving such a problem.

[0007]

According to a first aspect of the present invention, there is provided a gear shift actuator for driving a gear shift mechanism of a transmission, a clutch actuator for engaging / disengaging a clutch, a means for detecting a gear position of the transmission, and an artificial gear shift. Means for generating a shift request in accordance with the operation, and means for controlling a gear shift actuator and a clutch actuator so as to match the gear position of the transmission with the requested gear; Means for determining the driving situation of the vehicle in which fuel efficiency is prioritized based on the detection information; and, based on the result of the determination, the driving state and the gear position detection information in the driving situation where fuel efficiency of the vehicle is prioritized. To determine whether changing the gear position improves fuel economy Comprising the stage, and means for instructing a gear shift operation in the direction in which the fuel consumption is improved when it is determined that the fuel economy is improved, the.

In the second invention, a gear shift actuator for driving a gear shift mechanism of a transmission,
A clutch actuator for intermittently operating the clutch;
Means for detecting the gear position of the transmission, means for generating a shift request according to an artificial shift operation, and means for controlling a gear shift actuator and a clutch actuator so that the required gear and the gear position of the transmission match. Means for detecting a vehicle speed at predetermined time intervals, means for detecting an engine fuel injection amount at predetermined time intervals, means for detecting an engine rotation speed at predetermined time intervals, and time-series information of vehicle speed. Means for determining the traveling state of a vehicle in which fuel efficiency is prioritized by the neural network NN1, and the average engine fuel injection amount obtained from the time sequence information of the engine fuel injection amount in the driving state in which fuel efficiency is prioritized based on the determination result And the average engine speed obtained from the time series information of the engine speed Means for calculating an average engine torque by means of a neural network NN2; means for determining whether fuel efficiency is improved by shifting the current gear by means of a neural network NN3, based on at least the average engine speed, the average engine torque and the detection information of the gear position. Means for instructing a shift operation in a direction in which fuel efficiency is improved when it is determined that fuel efficiency is improved.

In a third aspect based on the second aspect, the vehicle speed detection information VS is converted into standard data VS = 2 / [1 + expｅｘ- (VS) using constants αVD and TT1 as input data to the neural network NN1. −αVD)｝
/ TT1] −1.

[0010]

According to the first aspect of the present invention, the driving condition of the vehicle is determined from the detection information of the driving condition. In the case of the traveling pattern, it is determined from the detection information of the driving state and the detection information of the gear position whether the current gear position improves the fuel efficiency if the current gear position is improved. An operation is instructed. When the driver performs a shift operation in accordance with this instruction, the fuel economy can be greatly increased. On the other hand, in a driving situation where fuel efficiency is not prioritized (a traffic congestion driving pattern in which acceleration and deceleration of the vehicle are repeated, a low-speed driving pattern in which the vehicle speed is substantially constant), a shift instruction is not output, so that driving is not disturbed by these. The user can easily drive on a congested road or the like without performing unnecessary shift operations at a desired shift speed.

According to the second aspect of the present invention, the running condition of the vehicle is determined by the neural network NN1 from the time-series information of the vehicle speed, and from the result of the determination, the running condition in which the mileage is prioritized (the high-speed running pattern in which the vehicle speed is substantially constant; In the case of the middle speed running pattern), the average engine torque is obtained by the neural network NN2 from the average engine fuel injection amount obtained from the time series information of the engine fuel injection amount and the average engine rotation speed obtained from the time-series information of the engine rotation speed. Is required. Then, it is determined by the neural network NN3 from the average engine speed, the average engine torque, and the detected information of the gear position whether or not the current stage is shifted to improve the fuel efficiency. If the fuel efficiency is improved, the fuel efficiency is improved. Is instructed. When the driver performs a shift operation in accordance with this instruction, the fuel economy can be greatly increased. On the other hand, in a driving situation where fuel efficiency is not prioritized (a traffic congestion driving pattern in which acceleration and deceleration of the vehicle are repeated, a low-speed driving pattern in which the vehicle speed is substantially constant), a shift instruction is not output, so that driving is not disturbed by these. The user can easily drive on a congested road or the like without performing unnecessary shift operations at a desired shift speed. Neural networks NN1 to NN
With respect to 3, the learning (adjustment of the weighting factor) is required in advance, but by changing the learning content, it is possible to apply to various engine models and vehicle models at low cost without replacing the control device. .

In the third invention, VD = 2 / [1 + exp
{− (VS−αVD)} / TT1] −1 is a constant αV
The rate of change (gain) of the specific area with respect to the vehicle speed information VS can be increased by taking D and TT1. Therefore, these differences clearly appear between the driving situation in which the fuel efficiency of the vehicle is prioritized and the driving situation in which the fuel efficiency is not prioritized, and the driving situation of the vehicle is accurately determined from the firing pattern of the neural network. be able to.

[0013]

FIG. 1 is a system diagram showing an embodiment of the present invention. 1 is a diesel engine, 2 is a mechanical clutch, 3 is a gear type transmission, and an output shaft of the transmission 3 is connected to a rear axle via a propeller shaft (not shown). Governor device 1A for controlling injection quantity to engine fuel injection pump
(Electronically controlled governor) is a clutch booster 2A that performs connection and disconnection operations on the clutch 2, and a transmission 3 is a gear shift unit 3A that drives the gear shift mechanism.
Are respectively provided. 27 denotes a supply / discharge valve of the clutch booster 2A, and 31 denotes an air tank.

In addition to a normal brake as a vehicle braking device, an exhaust brake and a third brake (hereinafter referred to as an engine extra brake) are provided as auxiliary brakes (not shown) for increasing negative work of the engine (not shown). Also, a retarder 18 (electromagnetic) is attached to the propeller shaft.
Is provided.

When the engine brake is operated when the operation switch 32 (exhaust brake switch) is turned on, the exhaust brake closes the shutter of the engine exhaust passage from the fully open position to a predetermined throttle opening. When the exhaust brake is actuated when the operation switch 9 (engine extra brake switch) is turned on, the engine extra brake releases the in-cylinder pressure to the exhaust passage through the third valve of the engine. When the operation switch 8 (retarder switch) of the retarder 18 is turned on, the magnet inside the retarder 18 is energized to generate an eddy current in the rotor on the propeller shaft side. 15 is a control valve of an engine extra brake, 16 is a retarder control unit that controls energization of a magnet,
Reference numeral 30 denotes an exhaust brake control valve.

As means for detecting driving information of the vehicle necessary for shifting control of the vehicle, an engine rotation sensor 29 for detecting an engine rotation speed and an accelerator opening sensor for detecting an amount of depression of an accelerator pedal 7 (accelerator opening). 28, a clutch stroke sensor 22 for detecting a stroke position of the clutch 2, a gear position sensor for detecting a shift position of the transmission 3 (built in the gear shift unit 3A), and a vehicle speed sensor for detecting a rotation speed of an output shaft thereof. 21, a main shaft rotation sensor 23 for detecting the rotation speed of the main shaft, and a range shaft rotation sensor 17 for detecting the rotation speed of the range shaft. From the governor device 1A, detection information of the fuel injection amount is input to an engine control unit 12 described later.

In order to switch the clutch 2 between automatic control and manual control, clutch switches 24 and 25 for detecting an initial position (clutch is in a connected state) and an operating position (clutch is in a disengaged state) of a clutch pedal are provided. A shift lever unit 4 (shifter) is provided in the operator's cab as a shift operation means of the transmission 2. When the shift lever 4A is shifted according to a predetermined pattern, shift instruction information (shift up signal, shift down signal) corresponding to the position is provided. Signal, reverse signal, neural signal, and hold signal).

The cab has an operation switch 10 (mode changeover switch) for selecting an automatic transmission (full automatic) mode and a semi-automatic transmission (semi-automatic) mode, an operation switch 32 for operating the exhaust brake, and an engine. An operation switch 14 for operating an extra brake, an operation switch 8 for operating a retarder 18, and a brake pedal switch 26 for detecting depression of a brake pedal (not shown)
And an alarm buzzer 13A that is activated immediately before automatic shifting of the vehicle
And a direction indicator switch 6 for the vehicle.

The transmission (T / M) control unit 11 and the engine control unit 12 control the shift operation of the vehicle based on these output signals, and these are connected by serial communication. When the control units 11 and 12 are in the semi-automatic mode, the shift lever unit 4
Each of the actuators (clutch booster 2A, gear shift unit, etc.) performs a gear shift to a target position in accordance with the operation state of the vehicle so as to perform a gear shift to a requested position according to shift instruction information from the vehicle. 3A, governor device 1A).

That is, after the clutch 2 is disengaged, if the transmission 3 is not in neutral, the gear is disengaged. Then, when the vehicle enters the synchronization region while controlling the engine rotational speed as necessary, the transmission 3 is engaged. When the clutch pedal is depressed, the clutch pedal switches 24 and 25 are turned on and off alternately, so that automatic control of the clutch is switched to manual control, and manual operation of the clutch according to the pedal depression amount becomes possible.

Reference numeral 13 denotes a display unit for displaying the gear position of the transmission 3 in the operator's cab, and a shift-up instruction lamp 13B and a shift-down instruction lamp 13C for urging the driver to perform a shift operation are added to the display panel. .

FIG. 2 shows a block diagram relating to the shift control. The T / M control unit 11 has a memory (not shown) for storing a shift map. The shift map is a map in which a shift point (shift up point, shift down point) is set for each gear using the accelerator opening and the engine speed as parameters.
The / M control unit 11 obtains a target shift speed from a shift map based on each detection information from the engine rotation sensor 29, the accelerator opening sensor 28, and the gear position sensor in the full automatic mode, and sets the actual gear position to the target gear position. In the semi-automatic mode, based on the shift designation information of the shift lever unit 4 and the detection information of the gear position sensor 33, the engine control unit 12 and the engine control unit 12 take a gear position corresponding to the shift designation information so as to match the gear position. Controls a series of shifting operations.

The T / M control unit 11 operates the shift-up designating lamp 13B and the shift-down designating lamp 13C of the display unit 13 in the semi-automatic mode as a shift instruction for urging the driver to perform a shift operation (shift lever operation). Control means (shift instruction control means 11a) is provided. Shift instruction control means 1
Although not shown, 1a reads out detection information from the vehicle speed sensor 21 every predetermined time TC, and reads out injection amount information from the governor device 1A every predetermined time TC.
Means for reading detection information from the engine rotation speed sensor 29 at predetermined time intervals TC, means for determining the running state of the vehicle by the neural network NN1 from the time-series information of the vehicle speed, and a running state in which fuel efficiency is prioritized based on the determination result. Means for calculating the average engine fuel injection amount from the time series information of the engine fuel injection amount, the average engine rotational speed from the time-series information of the engine rotational speed, and the average engine torque from these values by the neural network NN2. From the processing information (average fuel injection amount, average engine speed, average engine torque) and the information detected by the gear position sensor 33, the neural network NN
3 means for determining whether or not the fuel economy is improved by shifting the current gear position, and means for outputting to the display unit 13 a shift instruction signal in the direction of improving the fuel efficiency based on the determination result.

Also, the detection information VS of the vehicle speed sensor 21 is input as standard data VD = 2 / [1 + exp {-(VS-αVD)} as input data to the neural network NN1.
/ TT1] -1. Here, the standard data VD = 2 / [1 + exp ｛− (VS−α
VD)｝ / TT1] -1 is the neural network N
Represents a function that characterizes the input data to N1,
The vehicle speed detection information V
The gain (change rate) of the specific region with respect to S is increased.

As the neural network NN1,
A hierarchical type for performing unsupervised learning as shown in FIG. 3 is used.
In the input layer, a predetermined number of standard data VD1 to VD5
(Vehicle speed data pattern), each neuron adds weighting coefficients W11 to W11 to data VD1 to VD5, respectively.
W14, (hereinafter abbreviated), W51 to W54,
The sum i is transmitted to each neuron in the output layer. In the output layer, a function y = f using the sum i as a parameter
(I) is compared with the internal threshold value k, and when i ≧ k, 1
(Fire), and outputs 0 (no fire) when i <k.

Then, the running condition of the vehicle is determined from the firing patterns of the outputs x1 to x4, and learning (adjustment of the weighting factor) is performed in advance, as shown in (a) to (d) of FIG. Outputs x1 to x4 are 1 for each of four basic traveling patterns (substantially constant high-speed traveling pattern, approximately constant medium-speed traveling pattern, low-speed traveling pattern with small acceleration / deceleration, and traffic congestion traveling pattern in which acceleration / deceleration is repeated). Only one neuron fires.
Note that the correspondence between the firing pattern of the neural network NN1 and the basic running pattern determined as shown in FIG.
Is shown in Table 1. ○ represents the firing state of the neuron, and × represents the non-firing state of the neuron.

As the neural network NN2,
As shown in FIG. 6, an input layer receiving a set of input data (average engine speed NEm, average fuel injection amount Qm), an intermediate layer (hidden layer) for inputting values output from this layer, and finally output An output layer that outputs a value (average engine torque TQm) is used. Then, a teacher signal (output based on experimental data) considered to be ideal is given to the input data, the output value from the neural network NN2 is compared with the teacher signal, and the weight coefficient W11 is set so that the error is minimized. ~ W14 (hereinafter abbreviated), the neural network NN2 is constructed, and the average engine speed NEm and the average fuel injection amount Qm
, The calculation is performed in each layer and the average engine torque is output.

As the neural network NN3,
Hierarchical type that performs supervised learning is used like NN2,
As shown in FIG. 7, it is composed of an input layer, an intermediate layer, and an output layer, and it is determined from the firing patterns of the outputs x1 to x3 whether to shift down, shift up, or hold (maintain the current stage). By performing supervised learning based on experimental data in advance, as shown in FIG.
For (shift down, shift up, hold), each of the outputs x1 to x3 is such that only one neuron fires. ○ indicates the firing state of the neuron, ×
Represents the non-firing state of the neuron.

FIG. 9 illustrates the learning contents of the neural network NN3. In FIG. 9, the horizontal axis represents the engine rotation speed, the vertical axis represents the engine torque, and the equal fuel consumption line, the equal horsepower line, and the full load torque of the engine represent the running load. It is also displayed. On such a map, weight coefficients W11 to W1 are set in order to minimize an error from a teacher signal (experimental data).
By adjusting (3) (hereinafter abbreviated), three types of gear information DG are obtained from the average engine torque TQm, the average engine rotation speed NEm, the average fuel injection amount Qm, and the gear position. That is, by performing supervised learning in advance, the neural network NN3 is constructed so as to obtain the result shown in FIG.

For example, under a certain traveling load, the 6th speed Lo
When the vehicle speed is substantially constant at the U1 point of the w stage, the U2 point on the iso-horsepower line is the same when maintaining the constant speed traveling state at the 6th High stage, and the U1 point and the U2 point When the fuel efficiency is compared at point U2, the upshift instruction is output as the gear position information DG, and the vehicle speed is substantially constant at point D1 of the 6th High position under a certain traveling load. In the traveling state, when maintaining the constant-speed traveling state at the 6th low speed, the point D2 on the iso-horsepower line is the same. Comparing the fuel consumption rates at the points D1 and D2, D
Since the two points are better, a downshift designation is output as the gear position information DG.

FIG. 10 is a flow chart for explaining the control contents relating to the shift instruction. In step 1, it is determined whether or not the vehicle is in the semi-automatic mode, and only when the determination is yes (no processing is performed if no). Proceed to 2. In step 2, the vehicle speed sensor 21
, The fuel consumption amount information Q from the governor device 1A, and the detection information NE from the engine rotation sensor, and in step 3, the vehicle speed information VS is converted into the standard data VD =
2 / [1 + exp {-(VS-αVD)} / TT1]-
The data VD is converted to a predetermined number (in this case, five) together with the fuel injection amount information Q and the engine speed information NE.
Save until you reach.

In step 4, it is determined whether the number of stored data VD has reached a predetermined number. When the determination is yes, the process proceeds to step 5, and when the determination is no, the process returns to step 2. In step 5, the average value Qm is obtained from a predetermined number (five in this case) of fuel injection amount information Q, and the average value NE is obtained from a predetermined number (five in this case) of engine rotation information NE.
m is obtained. In step 6, a predetermined number of standard data VD is input to the neural network NN1, and its firing pattern PD is obtained. In steps 7 and 8, the running condition of the vehicle is determined from the firing pattern PD of the neural network NN1, and the shift instruction process in step 9 is executed when the running condition gives priority to fuel efficiency. That is, when PD = high-speed running pattern in which the vehicle speed is substantially constant or PD = medium-speed running pattern in which the vehicle speed is substantially constant, the process proceeds to step 9.

Here, if the determination in both steps 7 and 8 is no (PD = low speed running pattern with little acceleration / deceleration and PD = congestion running pattern in which acceleration / deceleration is repeated), step 9 is passed. One control is ended. This is because when the accelerator opening is fluctuating, it is considered that the shift instruction process is difficult, and the fuel economy improvement margin is also small.

FIG. 11 is a subroutine for explaining the shift instruction process performed in step 9. In step 10, the average fuel consumption injection amount Qm and the average engine speed NEm are input to the neural network NN 2 to determine the average engine torque TQm. . In step 11, these pieces of processing information (the average engine speed NEm and the average engine torque T
Qm and the average fuel consumption injection amount Qm) and the detection information of the gear position sensor are input to the neural network NN3 to obtain gear position information DG.

The gear position information DG is basically obtained from three data of the average engine speed NEm, the average engine torque TQm, and the current position information, and does not require the average fuel consumption amount Qm. The average fuel consumption injection amount Qm is also set to the neural network N so that a gear change instruction suitable for the actual situation (actual fuel consumption characteristics) can be performed even for disturbances such as changes over time.
Input to N3.

In step 12, it is determined whether the gear position information DG is upshifted. If the determination is yes, in steps 13 to 17, the timer TD
The GU counts the elapse of the predetermined time T _- DGU (TD
GU ≧ T ₋ DGU) and the display unit 1
The third upshift instruction lamp 13B is operated (lit). If the determination in step 12 is no, step 14
It is determined whether the gear position information DG = shift down.
If this determination is yes, step 15 to step 1
At 9, the timer TDGD counts the elapse of the predetermined time T _- DGD (TDGD ≧ T _- DGD),
Shift down instruction lamp 1 of display unit 13
Activate (turn on) 3C. The determination in step 14 is no
In step 20, the timer TDGU = 0 and the timer TDGD = 0 are reset in step 20.

With such a configuration, the running condition of the vehicle is determined by the neural network NN1 based on the time-series information of the vehicle speed, and based on the determination result, the running condition in which fuel efficiency is prioritized (high-speed running at a substantially constant vehicle speed) is determined. In the case of the pattern, also the middle speed running pattern), the average fuel injection amount Qm is obtained from the time sequence information of the fuel injection amount Q, the average engine rotation speed NEm is obtained from the time series information of the engine rotation speed NE,
From these values, the average engine torque TQm is obtained by the neural network NN2.

The processing information (the average fuel injection amount Qm,
Average engine speed NEm, average engine torque TQ
m) and the gear position detection information, a determination is made by the neural network NN3 as to whether or not the current stage is shifted to improve fuel economy. A shift operation (shift down, shift up) in the direction of improving fuel economy is performed based on the determination result. ) Is a shift down instruction lamp 13 of the display unit 13.
This is instructed via C and the upshift instruction lamp 13B. Therefore, when the driver performs the shift operation in accordance with the instruction, the fuel efficiency can be greatly increased.

On the other hand, in a driving situation where fuel efficiency is not prioritized (a traffic congestion driving pattern in which acceleration and deceleration of the vehicle are repeated, a low-speed driving pattern in which the vehicle speed is substantially constant), the shift-down instruction lamp 13 of the display unit 13 is provided.
Neither C nor the upshift instruction lamp 13B operates, that is, the shift instruction is not output. Therefore, the driver does not needlessly perform the unnecessary shift operation at the desired shift speed without burdening the driver, without burdening them. It can be easily driven without.

As preprocessing for characterizing the input data to the neural network NN1, the vehicle speed information VS is converted into the standard data VD = 2 / [1 + exp {-(VS-αVD)} / T.
T1] -1, the difference between the four types of traveling patterns clearly appears, and the traveling state of the vehicle can be accurately determined based on the firing pattern of the neural network NN1. Neural networks NN1 to NN
In the case of No. 3, learning (adjustment of weighting factors) is required in advance, but by changing the learning content, the control device (T / M control unit 11) does not need to be replaced, so that various types of vehicles and engine models are used. The effect that it can be applied at low cost is obtained.

The shift map stored in the memory of the T / M control unit 11 may be one in which shift points (shift up points, shift down points) are set using the accelerator opening and the vehicle speed as parameters. In that case, the detection information of the gear position sensor 33 becomes unnecessary in the control processing for obtaining the target shift speed from the shift map.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing an embodiment of the present invention.

FIG. 2 is a block diagram of a control system according to a shift instruction.

FIG. 3 is an explanatory diagram of the neural network NN1.

FIG. 4 is an explanatory diagram relating to a traveling pattern of the vehicle.

FIG. 5 is a correspondence table between the ignition pattern of the NN1 and the traveling pattern of the vehicle.

FIG. 6 is an explanatory diagram of the neural network NN2.

FIG. 7 is an explanatory diagram of the neural network NN3.

FIG. 8 is a correspondence table between the ignition pattern of the NN3 and the traveling pattern of the vehicle.

FIG. 9 is an explanatory diagram showing learning contents of NN3.

FIG. 10 is a flowchart illustrating control contents according to a shift instruction.

FIG. 11 is a flowchart for explaining control contents according to a shift instruction.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1A Governor device of fuel injection pump 2A Clutch booster 3A Gear shift unit 4 Shift lever device 10 Mode changeover switch 11 Transmission control unit 12 Engine control unit 13 Display unit 13B Shift up instruction lamp 13C Shift down instruction lamp 17 Range shaft rotation sensor 21 Vehicle speed sensor 22 Clutch stroke sensor 23 Main shaft rotation sensor 26 Brake pedal switch 28 Accelerator opening sensor 29 Engine rotation sensor 33 Gear position sensor

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) F16H 59:74

Claims (3)

[Claims] 1. A gear shift actuator for driving a gear shift mechanism of a transmission, a clutch actuator for engaging and disengaging a clutch, a means for detecting a gear position of the transmission, and a means for generating a shift request according to an artificial shift operation. And a means for controlling a gear shift actuator and a clutch actuator so as to make the required gear and the gear position of the transmission coincide with each other. Means for determining the traveling state of the vehicle, and when the current gear position is shifted from the detection information of the driving state and the detection information of the gear position in the traveling state in which the fuel efficiency of the vehicle is prioritized based on the determination result, the fuel efficiency is improved. Means to determine whether fuel efficiency is improved Sometimes fuel transmission of a vehicle, characterized in that it comprises means for instructing the shift operation to the well becomes direction. 2. A gear shift actuator for driving a gear shift mechanism of a transmission, a clutch actuator for engaging and disengaging a clutch, a means for detecting a gear position of the transmission, and a means for generating a shift request according to an artificial shift operation. Means for controlling a gear shift actuator and a clutch actuator so as to make the required gear and the gear position of the transmission coincide with each other, a means for detecting a vehicle speed at predetermined time intervals, and a method for setting an engine fuel injection amount to a predetermined value. Means for detecting at every time, means for detecting the engine rotational speed at every predetermined time, means for determining the running situation of the vehicle whose fuel efficiency is prioritized by the neural network NN1 from the time series information of the vehicle speed, and For driving situations where fuel efficiency is prioritized Means for obtaining an average engine torque by a neural network NN2 from the average engine fuel injection amount obtained from the time series information of the engine fuel injection amount and the average engine rotation speed obtained from the time series information of the engine rotation speed. Means for determining whether or not the current stage is shifted to improve fuel efficiency by the neural network NN3 based on the average engine speed, the average engine torque, and the detected information of the gear position, and a direction in which the fuel efficiency is improved when it is determined that the fuel efficiency is improved. Means for instructing a shift operation to the vehicle. 3. Using the constants αVD and TT1 as input data to the neural network NN1, the vehicle speed detection information VS is converted into standard data VS = 2 / [1 + expｅｘ− (V
The transmission according to claim 2, further comprising a unit configured to convert the value into (S−αVD)｝ / TT1] −1.