JP3156567B2 - 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 shift control device for an automatic transmission for a vehicle, and more particularly to a shift control device for an automatic transmission having a manual mode and an automatic mode.

[0002]

2. Description of the Related Art Conventionally, as a vehicle transmission, an automatic transmission in which a shift operation is automated has been frequently used. This type of automatic transmission mainly employs a torque converter in place of a clutch in the case of a small vehicle, while transmission of a driving torque is performed in a large vehicle such as a bus or a truck. Due to the large amount, it is difficult for the torque converter to sufficiently transmit the driving torque.

[0003] Therefore, in a mechanical transmission similar to a manual transmission, an automatic transmission having a structure in which an actuator for automatically connecting and disconnecting a clutch is provided to thereby release the clutch pedal has been developed for large vehicles. Have been. Normally, in such a shift control device of a mechanical automatic transmission, a target shift speed is set according to a vehicle speed and an accelerator opening. However, in many cases, the driver's intention is not sufficiently reflected in the target gear. Therefore, in this case, there is a possibility that a gearshift contrary to the driver's will is performed.

[0004] In view of the above, fuzzy inference is performed based on information such as road surface inclination, vehicle speed, accelerator opening, and brake operation status, and the shift characteristics of a shift map are switched in accordance with the obtained fuzzy rules. Thus, a shift control device configured to shift the speed by reflecting the driver's intention is disclosed in Japanese Patent Application Laid-Open Nos. 1-255748 and 2-33738.

[0005]

As described above, in the shift control device for an automatic transmission disclosed in the above-mentioned publications and the like, a shift reflecting the intention of the driver is performed. It is not the basis that the shift is performed according to the driver's preference. That is, normally, in the case of a manual transmission, the timing of performing a shift operation is slightly different for each driver, but in the automatic transmission as described above, the shift timing is set to a predetermined timing regardless of the driver's preference. That is, some drivers feel uncomfortable.

Therefore, in recent years, an automatic transmission has been developed which is capable of manual shifting according to the driver's preference while mainly performing automatic shifting. Further, there has been developed an automatic transmission in which the preference of a driver is learned and stored based on the contents of a manual shift, and the stored value can be applied at the time of an automatic shift. By the way, in order to favorably reflect the driver's preference in the shift, many input parameters are required and learning of both the upshift and the downshift is required. Therefore, the calculation at the time of learning may be enormous and complicated, which is not preferable.

The present invention has been made in view of the above-mentioned circumstances, and an object of the present invention is to facilitate both upshifting and downshifting according to the driver's preference without making a complicated structure. An object of the present invention is to provide a shift control device for a controllable automatic transmission for a vehicle.

[0008]

In order to achieve the above-mentioned object, a first aspect of the present invention is directed to a manual mode in which a gear is switched by an operation of a driver and an automatic operation according to a driving state of a vehicle. In a shift control device for a vehicle automatic transmission having an automatic mode for automatically switching gears, a shift preference of a driver is reflected by a neural network using a shift state in at least one of the manual mode and the auto mode as a parameter. Learning means for learning a shift pattern; target gear setting means for setting a target gear based on an output from the learning means and a driving state of the vehicle in the auto mode; and the automatic transmission based on the target gear. Gear shift control means for performing gear shift control , wherein the learning means comprises a
Depending on the frequency of use of the manual mode,
Speed mode of one of the
It is characterized by selecting a state and using it as a parameter .

[0009] Therefore, in the manual mode by the driver,
The shift state in at least one of the manual mode and the automatic mode is used as a parameter in accordance with the frequency of use, and a shift pattern learning that reflects the driver's shift preference is performed by a neural network. The target shift speed is set based on the operation state of. Then, the shift control of the automatic transmission is favorably performed based on the target shift speed.

[0010]

Further, in the invention according to claim 2 , further comprising an engine speed detecting means for detecting an engine speed of the engine, wherein the learning means is provided with the manual mode .
When the use frequency is larger than a predetermined value, the shift state in the manual mode is set as a parameter, and the shift pattern learning is performed using at least the engine speed at the time of a shift operation by a driver as a parameter.

[0012] Therefore, in the manual mode by the driver ,
If the frequency of use is larger than the predetermined value, the shift state in the manual mode is set as a parameter, and at least the engine speed at the time of the shift operation by the driver is set as a parameter, so that the shift pattern learning is favorably performed.

In a preferred embodiment, the learning means may use a shift state in the automatic mode as a parameter when the frequency of use in the manual mode is smaller than a predetermined value. In this case, if the use frequency in the manual mode is smaller than a predetermined value, the shift state in the auto mode is used as a parameter, and the shift pattern learning is performed well.

At this time , the vehicle further includes an accelerator opening detecting means for detecting an accelerator opening of the engine, and the learning means detects the accelerator opening at least when the gear is switched as a shift state in the auto mode. the shift pattern learning may it line in parameter. this
In this case , at least the accelerator opening at the time of shifting gears is set as a parameter as a shift state in the auto mode,
Shift pattern learning is performed well.

[0015] In a further preferred aspect, the learning means may perform the shift pattern learning by including current shift pattern information in a parameter. In this case, the current shift pattern information is used as a reference parameter for the shift pattern learning, and the shift pattern learning is performed well. In the invention according to claim 3 , the vehicle further comprises a flat road detecting means for detecting whether the vehicle is traveling on a flat road,
The learning means performs the shift pattern learning when the vehicle is traveling on the flat road.

Therefore, the shift pattern learning is favorably performed only when the vehicle is traveling on a flat road where the driver's preference tends to appear. In a preferred embodiment , the target shift speed setting unit has a plurality of types of shift maps having different shift patterns, and selects an optimal map from the plurality of types of shift maps in accordance with an output from the learning unit. The target shift speed may be set based on the optimal map and the driving state of the vehicle.

In this case , an optimum map is selected from a plurality of types of shift maps in accordance with the output from the learning means, and the target shift speed is set satisfactorily based on the optimum map and the driving state of the vehicle. Further, at this time , the vehicle further includes flat road detecting means for detecting whether or not the vehicle is traveling on a flat road, and the target gear position setting means is configured to detect a shift speed selected previously when the vehicle is not traveling on the flat road. It is set the target gear position based on the map and the driving state of the vehicle
Good .

In this case, when the vehicle is not traveling on a flat road where the driver's preference is likely to appear, the target gear is set well based on the previously selected shift map and the driving state of the vehicle. Further, in the invention according to claim 4, the learning means performs a shift pattern learning that reflects a driver's shift preference at the time of shift-up by the neural network using a shift state in at least one of the manual mode and the auto mode as a parameter. It is characterized by performing.

Therefore, the shift pattern learning that reflects the driver's shift preference at the time of upshifting is performed satisfactorily by the neural network using the shift state in at least one of the manual mode and the automatic mode as a parameter .

Further, in the invention of claim 5 , the engine is connected to the engine.
Manual that switches gears by driver's operation
Automatically shifts gears according to the vehicle mode and vehicle operating conditions.
Shifting of automatic transmission for vehicles with automatic mode to change
In the control device, the manual mode and the automatic mode
The speed change state in at least one of the
Driver's gear shift preference by neural network
Learning means for performing a shift pattern learning reflecting a favorable condition;
The output from the learning means and the operation of the vehicle in the auto mode
Target gear position setting procedure that sets the target gear position based on the rotation state
And a shift control of the automatic transmission based on the target shift speed.
And shift control means for performing control, and vehicle speed detecting means for detecting a vehicle speed, e Bei a brake operating state detecting means for detecting an operating state of the brake, said learning means, wherein a brake operation state is ON and the vehicle speed When the amount of decrease in the vehicle speed is equal to or more than the predetermined amount, the driver performs at least the shift-down operation until the amount of decrease in the vehicle speed reaches the predetermined amount.
The shift pattern learning is performed using the frequency of use of the manual mode as a parameter, and the mode switching frequency of the driver from the auto mode to the manual mode at least until the amount of decrease in the vehicle speed reaches the predetermined amount. The shift pattern learning is performed as a parameter, and the shift pattern learning that reflects the driver's shift preference at the time of downshifting is performed.

Therefore, when the brake operation state is ON and the amount of decrease in vehicle speed is equal to or more than the predetermined amount, the shift pattern learning that reflects the driver's shift preference at the time of downshifting is performed well. In addition, at least during the shift to the downshift side until the amount of decrease in vehicle speed reaches the predetermined amount.
The frequency of use of the manual mode by the inversion is used as a parameter, and the shift pattern learning is performed well. Further, the frequency of mode switching from the auto mode to the manual mode by the driver until at least the amount of decrease in the vehicle speed reaches the predetermined amount is used as a parameter, and the shift pattern learning is performed well.

[0022] Also, as a preferred embodiment, the shift control means may include a fuzzy control means for correcting the target shift speed by using fuzzy inference based on the vehicle information.

In this case, fuzzy inference is performed on the basis of the vehicle information, and the target shift speed is suitably corrected in accordance with the result of the fuzzy inference, whereby good shift control is performed.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A shift control device for an automatic transmission for a vehicle according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic block diagram showing the main configuration, and FIG. 2 is a schematic configuration diagram showing the entire configuration. First, the overall structure of the transmission control device of the present invention will be described with reference to FIG.

As shown in FIG. 2, the shift control device of the present invention is a diesel engine (hereinafter referred to as engine).
This is a system for automatically changing the rotational driving force of 11 using a gear type transmission (hereinafter referred to as a transmission mechanism) 17 having a clutch 15. Here, the speed change mechanism 17 has seven forward speeds in addition to the reverse speed, and can perform not only automatic shifting but also manual shifting.

The engine 11 is provided with a fuel injection pump (hereinafter, referred to as an injection pump) 21 having a pump input shaft 19 that rotates at half the rotation speed of the engine output shaft 13. An electromagnetic actuator 25 is connected to the control rack 23. Further, a rack position detection sensor 123 for detecting the position of the control rack 23 is also provided. The pump input shaft 19 is provided with an engine rotation sensor 27 for detecting a rotation speed signal of the output shaft 13 of the engine 11.

An exhaust pipe 12a for guiding exhaust gas extends from the engine 11 through an exhaust manifold 12, and an exhaust brake device 119, which is one of engine auxiliary brakes, is interposed in the exhaust pipe 12a. .
The clutch 15 is provided with an air cylinder 33 as a clutch actuator. This clutch 15
Is connected by bringing the clutch plate 31 into pressure contact with the flywheel 29 by well-known holding means (not shown). That is, when the air cylinder 33 shifts from the non-operation state to the operation state, the holding means operates in the release direction, whereby the clutch 15 changes from the connected state to the disconnected state.

The clutch 15 is provided with a clutch stroke sensor 35 for detecting disconnection and connection information based on a clutch stroke amount. Note that a clutch touch sensor 37 may be used in place of the clutch stroke sensor 35. Transmission mechanism 1
7 is provided with a clutch rotation sensor 41 for detecting the number of rotations of the input shaft 39.

An air passage 43 is connected to the air cylinder 33. The air cylinder 33 is connected via a check valve 45 to a pair of air tanks 47 as an air source.
49. In the middle of the air passage 43, there are solenoid valves X1 and X2 that are duty-controlled to supply working air and function as opening / closing means, and solenoid valves Y1 and Y that are duty-controlled to open the air cylinder 33 to the atmosphere.
2, and a three-way solenoid valve W is provided upstream of the solenoid valves X1 and X2.

As shown in FIG.
X2 are connected in parallel with each other and are normally closed. The solenoid valves Y1 and Y2 are also connected in parallel to each other, and are normally open. When the air cylinder 33 is turned on, the solenoid valve W
The air passage is controlled so as to connect the air passage to the air passage when the air cylinder 33 is turned off.

Here, the solenoid valves X1, X2 and the solenoid valves Y1, Y2 are used alternately, so that the solenoid valves X1, X2, Y1, Y2 can be used for a long time. Further, when one of the solenoid valves X1 and X2 fails, or when one of the solenoid valves Y1 and Y2 fails, the other solenoid valve is used. The reliability of the entire device is maintained.

Incidentally, of the pair of air tanks 47, 49, the air tank 49 is an emergency tank. If the air in the main air tank 47 runs out for some reason, the solenoid valve 55 is opened to release the air from the emergency air tank 49. Supply. For this reason, air sensors 57, 59 that output an ON signal when the internal air pressure falls below a specified value are attached to each of the air tanks 47, 49.

The air tank 47 has an air passage 43.
Is connected to an air passage that branches into two systems on the downstream side, and the leading end of the air passage is connected to a braking system (air overhydraulic type) through a pair of solenoid valves MVQ111 and 111. Pair of air masters 109, 10
9 is connected. These air masters 109, 10
9 includes a strong brake pedal force sensor (BPS sensor) 106
Is installed. The BPS sensor 106 is configured as a diaphragm open / close switch that is turned on when receiving strong brake depression force information corresponding to air pressure when a strong braking force exceeding a set value is required.

The change lever 61 is a select lever of the speed change mechanism 17 and can move in a select direction and a direction orthogonal to the select direction as shown in FIG. From the selected position in the shift direction parallel to the select direction. In the select pattern and shift pattern in each of these directions, in the select direction, an N (neutral) range, an R (reverse) range, and a D (drive) range corresponding to an automatic shift mode (auto mode) are set. In the shift direction, the change lever 6 is moved from the D (drive) range in the orthogonal direction.
An I-type shift pattern having an UP (shift up) position and a DOWN (shift down) position across an M (manual) range corresponding to a manual shift mode (manual mode) is set. ing.

In such a select pattern and a shift pattern, the change lever 61 located in the N range, the R range and the D range is held at the position and stopped even if the driver's hand is released after the operation to the position. On the other hand, if the shift operation is performed to the UP position or the DOWN position after the M range is selected,
After the operation, when the driver's hand is released, the driver automatically returns to the M range and is held at that position (indicated by HOLD in FIG. 3). The detection of each range and position of the change lever 61 is performed by the gear position selection switch 63, whereby the gear shift unit 65 is operated, and the gear in the transmission mechanism 17 is switched according to the select range and the shift position.

The gear shift unit 65 has a plurality of solenoid valves (only one is shown in FIG. 2) which is operated by an operation signal from a control unit 71 as a shift control means.
73 and a pair of power cylinders (not shown) for operating a select fork and a shift fork (both not shown) in the transmission mechanism 17. This power cylinder is connected to the air tank 47,
It operates when high-pressure operating air is supplied from 49. That is, each power cylinder is operated by the operation signal given to the solenoid valve 73, and the meshing state of the gear type transmission mechanism 17 is changed in the order of select and shift.

Further, the gear shift unit 65 is provided with a gear position switch 75 as a gear position sensor for detecting each gear position, and a gear position signal from the gear position switch 75 is output to the control unit 71. A vehicle speed sensor 79 for detecting a vehicle speed signal is attached to the output shaft 77 of the transmission mechanism 17, and an accelerator pedal 81 for detecting the amount of depression (accelerator opening VA) as engine load information. Degree sensor 8
5 are provided. This accelerator opening sensor 85
The resistance change according to the amount of depression of the accelerator pedal 81 is detected as a voltage value (VA), which is converted into a digital signal by the A / D converter 83 and output. FIG. 4 shows a map showing the relationship between the accelerator opening and the voltage value (VA), and the accelerator opening VA is set based on this map.

A brake sensor 87 for outputting a high-level brake signal when the brake pedal 69 is depressed is attached to the brake pedal 69. Is mounted. The starter 89 is provided with a starter relay 91, and the starter relay 91 is connected to the control unit 71.

Reference numerals 120 and 121 in the figure denote exhaust brake devices 11 which are engine auxiliary brakes, respectively.
9. An engine brake auxiliary device, that is, an exhaust brake on / off switch (engine auxiliary brake operation detecting means) for switching a compression-release type engine auxiliary brake device (not shown) between an operation standby state and a non-operation state, and an engine brake auxiliary device These are on / off switches, which are arranged near the driver's seat.

In FIG. 2, reference numeral 93 denotes an engine control unit provided separately from the control unit 71. The engine control unit 71 supplies information from each sensor to the electronic governor 25 in the injection pump 21 and the control unit 71.
The drive control of the engine 11 is performed in accordance with the accelerator opening information VA and the like received from the ECU. That is, in the electronic governor 25 which has received the command signal from the engine control unit 93, the control rack 23 operates to increase or decrease the fuel, thereby controlling the increase or decrease in the rotation speed of the output shaft 13 of the engine 11.

The control unit 71 comprises a microcomputer (hereinafter referred to as a CPU) 95, a memory 97, and an interface 99 as an input / output signal processing circuit. The input port (input interface) 101 of the interface 99 includes the above-mentioned gear position selection switch 63, brake sensor 87, accelerator sensor 85, engine rotation sensor 27, clutch rotation sensor 41, gear position switch 75, vehicle speed sensor 79, clutch. Stroke sensor 35, clutch touch sensor 37
(Used when outputting the connection / disconnection information of the clutch 15 instead of the clutch stroke 35), the air sensors 57, 5
9. BRS sensor 10 that outputs strong brake pedal force information
6. Exhaust brake on / off switch 120, engine brake auxiliary device on / off switch 121, rack position detection sensor 123, and slope start switch 103 described later.
Are connected, and detection information is input to the control unit 71 from each of these sensors.

The slope start switch 103 is for operating a system (hereinafter, referred to as AUS) for preventing the vehicle from moving backward when the vehicle starts on an uphill. This AUS is
Air master 1 for a plurality of wheel brakes 107
The supply of air to the valves 09 and 109 is controlled by a pair of solenoid valves MVQ
This is a system in which a vehicle is started while being controlled via 111,111.

On the other hand, the output port (output interface) 113 is provided with the above-described engine control unit 93, starter relay 91, and solenoid valves X1, X2, Y.
1, Y2, W and the solenoid valves 55, 73, 111 are respectively connected. Reference numeral 115 in the figure denotes an air warning lamp which is turned on in response to a detection signal from the air sensors 57, 59 when the air pressure in the air tanks 47, 49 has not reached the set value. Reference numeral 117 denotes wear of the clutch 15. Reference numeral 116 denotes a stop lamp switch that is turned on when the brake pedal 69 is depressed, the clutch warning lamp being turned on in response to a detection signal when the amount exceeds a specified value.

The memory 97 includes a read-only ROM in which various flowcharts are written as programs and data, and a writable RAM. In the ROM, in addition to the control program, the duty ratios of the solenoid valves X1, X2, Y1, and Y2 corresponding to the accelerator opening information VA are stored in advance as a map.
Read out appropriate values from this map as appropriate.

The above-mentioned shift position selection switch 63 outputs a select signal and a shift signal as shift signals. The shift position corresponding to a paired combination of the two signals is stored in advance in the ROM as a data map. ing.
Accordingly, when the control unit 71 receives the select signal and the shift signal, the CPU 95 calculates an output signal from this map, and further provides this output to each of the solenoid valves 73 of the gear shift unit 65 to set the target gear corresponding to the shift signal. Set the gear. The gear position signal from the gear position switch 75 is output when the shift is completed, and it is determined whether all the gear position signals corresponding to the select signal and the shift signal have been output. That is, the gear position signal is used to generate a signal indicating whether the meshing is normal.

When a target shift speed in the D range exists in the ROM, a shift map for determining an optimum shift speed based on the values of the vehicle speed V, the accelerator opening VA, and the engine speed Ne is also provided. It is remembered. Table 1 shows a shift map as a selection map according to the on / off status of the foot brake signal and the exhaust brake signal from the brake sensor 87 and the exhaust brake on / off switch 120.
As shown in FIG. 3, three types, # 1, # 2, and # 3, are set. The three types are set in order to improve the running feeling on a downhill. Normally, when no braking is performed, the selection map # 1 is applied.

[0047]

[Table 1]

In general, the required driving torque of a bus or truck changes depending on the loading situation or driving situation, and the ROM stores the loading situation (for example, a vehicle load degree αVL described later), the driving situation, and the like. A shift map that is switched automatically according to the situation and the driver's preference, which will be described later, or based on the driver's intention is also stored. For example, there are three types of shift maps according to the loading situation and the driving situation, such as a fuel economy-oriented economy mode, a normal mode, and an acceleration-oriented power mode.
It is provided for each shift map of # 1, # 2 and # 3.
That is, in the ROM, for example, the vehicle speed V and the accelerator opening V
Speaking of the relationship with A, a total of nine types of shift maps are stored. FIGS. 5 to 13 show these nine types of shift maps. FIGS. 5 to 7 show the shift maps of # 1, # 2, and # 3 in the economy mode, and FIGS. Are # 1, # 2, # in normal mode
11 to 13 show shift maps of # 1, # 2, and # 3 in the power mode. That is, the CPU 95 appropriately selects a shift map according to the situation from the shift map group. In these shift maps, shift-up shift characteristics are indicated by solid lines, and shift-down shift characteristics are indicated by broken lines.

By the way, the basic operation of the above-mentioned automatic transmission is well-known, and detailed description thereof is omitted here.
This automatic transmission operates, for example, in the same manner as that disclosed in Japanese Utility Model Laid-Open No. 2-49663. Next, control contents in the control unit 71 as a main part of the present invention will be described.

As shown in FIG. 1, the control unit 71 of this automatic transmission selects a suitable shift map from the above-mentioned shift map group, and outputs the vehicle speed information V from the vehicle speed sensor 79 and the vehicle speed information V from the accelerator opening sensor 89. A target gear position setting means 3 for setting a target gear position from the selected map based on the accelerator opening information VA, and an optimum gear position determining means 1 for correcting the target gear position by reflecting the driver's intention are provided. Have been.

The target shift speed setting means 3 is provided with a storage means 3A for storing a shift map and a learning expression selecting means 3B. The storage means 3A is provided in addition to the shift map groups shown in FIGS. In addition, a downshift restriction map (see FIG. 33) for restricting 5-4 downshift and 4-3 downshift as described later is stored for each mode of the selection map # 1. Therefore, the target gear position setting means 3 is configured as a map-type storage means.

The learning expression selecting means 3B of the target gear position setting means 3 has a function of performing learning according to the driver's driving preference and selecting any one of the above-described modes. . The driver's driving preference refers to, for example, a driving method of whether the driver desires an early shift-up with an emphasis on fuel efficiency or a driving with an emphasis on acceleration. Therefore, the learning expression selecting means 3B learns the driving operation of the driver by using the driving operation information by the driver as input information, and optimally selects from the plurality of shift maps in the storage means 3A according to the driver's preference and personality. Select the shift map.

More specifically, the learning expression selecting means 3B is provided with the function shown in FIG.
30 is provided with a plurality of neural networks as shown in FIG. 30. When various kinds of driving operation information are input to these neural networks, arithmetic processing including learning is performed in the neural networks. An output signal (judgment output) optimal for the user's preference is output. Then, a shift map corresponding to the output signal from each neural network is selected, and based on the shift map, a target gear position corresponding to the vehicle speed V and the accelerator opening VA is set.

Further, the optimum gear position determining means 1 determines the driver's intention and the running state of the vehicle using fuzzy logic, and can correct the target gear position set by the target gear position setting means 3 by a fuzzy type optimum gear position determining means. It is constituted as a gear position determining means. In the fuzzy type optimum gear position determining means 1, a fuzzy rule described later is applied using vehicle information having vehicle load information as a parameter. As the vehicle load information, the vehicle is accelerated on a straight flat road in an empty state. The difference between the acceleration α0 in the case and the actual acceleration α when the vehicle is actually accelerated (= vehicle load αVL) is used as a parameter.

For this reason, as shown in FIG. 1, the control unit 71 is provided with a vehicle load degree calculating means 2 for calculating the vehicle load degree αVL. The vehicle load degree calculating means 2 is provided with an engine torque calculating means 4, a driving force calculating means 5, an air resistance calculating means 6, an acceleration calculating means 7 equivalent to a straight flat road vehicle, and a subtracting means 8. The engine torque calculation means 4 calculates the engine torque Te from the position information SRC of the control rack 23 supplied from the rack position detection sensor 123 and the engine speed information Ne. As in the case of No. 3, it is configured as a storage means of a map format.

Here, the position information SRC of the control rack 23 is used to calculate the engine torque Te. The position information SRC, which is the actual rack position and voltage value, is used.
Has a relationship as shown in FIG. In using the position information SRC, a filtering process for removing noise is performed. Specifically, a known low-pass filter (Butterworth-type primary digital filter) as shown in FIG. 15 is used as the filter. In this low-pass filter, processing coefficients a and b
Are represented by the following equations.

A = ωc / (ωc + 2) b = (ωc−2) / (ωc + 2) where ωc is a cutoff frequency, for example, 0.0
628 rad / sec. FIG. 16 shows a flowchart of the low-pass filter processing routine of FIG. 15, and the filter processing will be briefly described with reference to FIG.

In step F10, an initial value 0 is set for the processing variables buffer1 and buffer2. And step F
At 12, the processing coefficients a and b are calculated based on the above equation, and at step F14, the processing variable buffer2 is temporarily changed to the processing variable buffer1.
And In step F16, a processing variable buffer1 is set from the following equation based on the processing variable buffer2 obtained as described above.

Buffer1 = buffer2 · (−b) + (input value) In step F18, the final output value, that is, SRC is obtained from the following equation. (Output value) = SRC = (buffer1 + buffer2) · a In this manner, the position information SRC, which is a voltage value, is filtered, and these steps F14, F16, and F1 are performed.
8 is true if the result of the determination in step F20 is true (Ye
The process is continued as long as the input value exists in s).

The same filtering process is performed on the engine speed information Ne. The engine torque Te is determined based on a preset map (not shown) from the position information SRC and the engine speed information Ne filtered as described above. Since it differs depending on the use of the brake, in a situation where the exhaust brake or the compression release type engine auxiliary brake is used, a map (not shown) separately set according to the situation is used.

The values obtained from the respective maps are subjected to a first-order lag processing. The low-pass filter and the processing routine shown in FIGS. 15 and 16 are applied to the first-order lag processing. At this time, the processing coefficients a and b are determined based on the exhaust brake (EXB) and the compression-release type engine auxiliary brake (P
It is set as shown in Table 2 according to the usage situation of T).

[0062]

[Table 2]

Here, T indicates a calculation cycle, and τA, τB,
τC is a time constant preset for each use condition of the exhaust brake (EXB) and the compression release type engine auxiliary brake (PT). However, in this case, the exhaust brake (EXB) and the compression release type engine auxiliary brake (PT)
At the point in time when the usage status is switched, the processing variables buffer1 and b in step F10 of the processing routine shown in FIG.
The initial value of buffer2 is calculated by the following equation using the processing coefficients a and b.

Buffer1 = y / a-buffer2 buffer2 = (y / ax) / (1-b) where x is an input value immediately before switching and y is an output value immediately before switching.

Further, the driving force calculating means 5 generates the engine torque information Te obtained by the engine torque calculating means 4.
The driving force F of the vehicle is calculated based on the following formula. The driving force F is calculated by the following equation, for example. F = (Te · it · if · η) / R where it is the gear ratio of the shift speed, if is the final reduction gear ratio (differential gear ratio), η is the power transmission efficiency, and R is the tire radius.

The air resistance calculating means 6 calculates the air resistance Rl as the running resistance of the vehicle from the actual vehicle speed information V, and is calculated by the following equation. Rl = λ · A · V ² where λ is the air resistance coefficient, A is the front projected area of the vehicle, V
Is the actual vehicle speed. Next, the acceleration calculating means 7 (hereinafter, referred to as acceleration calculating means) 7 equivalent to a straight flat road vehicle will be described. The acceleration calculating means 7 includes the driving force information F calculated by the driving force calculating means 5 and the air resistance. Based on the air resistance coefficient information Rl calculated by the calculation means 6, the acceleration α0 when the vehicle accelerates on a straight flat road in an empty state, that is, the acceleration equivalent to a straight flat road empty vehicle is calculated. The acceleration α0 corresponding to the straight flat road vehicle is calculated by the following equation using the driving force F of the vehicle.

Α0 = g · {F− (μW0 + R1)} / (W0 + Wr) where g is the gravitational acceleration, μ is the road surface friction coefficient, W0 is the weight of the empty vehicle, and Wr is the weight of the rotating part. Then, in the subtraction means 8, the acceleration information α 0 corresponding to the straight flat road vehicle calculated by the acceleration calculation means 7 and the actual acceleration information α from the vehicle speed sensor 79.
And the vehicle load degree information αVL is calculated by the following equation. The vehicle load information αVL corresponds to the weight of the vehicle and the gradient resistance of the vehicle.

ΑVL = α0−α That is, if αVL> 0, the vehicle load is heavy, and if αVL <0, the vehicle load is light. Then, it is determined from the magnitude of the value of αVL how heavy (or light) the vehicle load is. 17 to 19 are examples in which the vehicle load αVL is simulated and calculated. The horizontal axis represents the total vehicle weight gvw (gross vehicle weight), and the vertical axis represents the vehicle load αVL. FIG.
Is a flat road, FIG. 18 is a 10% gradient uphill road, and FIG.
This is a calculation example of the vehicle load αVL on a downhill road with a 0% gradient, which is calculated by keeping the conditions of the vehicle constant except for the vehicle weight.

When the vehicle load degree αVL is calculated by the vehicle load degree calculation means 2 in this way, the fuzzy type optimum gear position determination means 1 adds the accelerator opening information VA, in addition to the vehicle load degree αVL. Each of the accelerator opening change information ΔVA, the vehicle speed information V, the brake information, and the current gear position information is fetched, and the target gear position set by the target gear position setting means 3 is corrected. As the brake information, in addition to the operation information of the foot brake input from the brake sensor 87, the operation information of the exhaust brake, the compression-release type engine auxiliary brake, and the like is also input. Is corrected in consideration of the above information.

That is, the optimum gear position determining means 1 corrects the target gear position by fuzzy inference of each information from a predetermined fuzzy rule. 20 to 22 show membership functions of the information αVL, VA, and ΔVA that are input to the optimal gear position determining means 1 as parameters of the driving information. FIG. 20 shows the membership function of the vehicle load degree αVL. Here, α1, α2, α3, α4,
α5 and α6 are preset, and these values take different values for each of the seven forward speeds.
That is, each shift speed has a unique membership function for the vehicle load αVL.

FIG. 23 shows a membership function for correcting the gear position reflecting the driver's intention. Note that the membership function of the vehicle speed information V is omitted. In the optimum gear position determining means 1, the target gear position is corrected by a fuzzy inference method using the min-max composite center of gravity method. FIG. 24 shows the min-max composite centroid method.
A brief description will be given step by step using (a) to (c).

1) First, the degree of conformity of the membership function to the input value of the fuzzy rule antecedent is obtained for all the fuzzy rules. For example, assuming that the outputs from the sensors at a certain time t are a1 (t), a2 (t),..., As shown in FIGS. 24 (a) and (b), each fuzzy rule R1, R2,. In step (1), membership grades (degrees of conformity) A1 and A2 for a1 (t) and a2 (t) are obtained from each membership function.

2) Next, the minimum value m of the conformity for each rule
Find in. That is, the fuzzy rule R shown in FIG.
In the antecedent part 1, the fitness A1 for a1 (t) is smaller than the fitness A2 for a2 (t), and therefore the parameters of the fuzzy rule are a1 (t), a2
In the case of only two of (t), A1 is the minimum value min. Similarly, as shown in FIG. 24B, in the fuzzy rule R2, the degree of conformity A2 to a2 (t) is the minimum value mi.
n.

3) The membership function of the consequent part is defined by the minimum value min. Then, of the conformity of the membership function of the consequent part, a part having the minimum value min or more obtained in the above 2) is cut, and the membership function of the consequent part is defined. This is also shown in FIGS. 24 (a) and (b).
This is performed for each fuzzy rule R1, R2,....

4) A graph obtained by each fuzzy rule is superimposed to obtain a max figure. Here, FIG.
As shown in (c), the maximum value of the membership function is obtained by superimposing the graphs (graphics) obtained in 3) above. 5) The position of the center of gravity of the figure in 4) is obtained. Then, by obtaining the graphic centroid C1 of the membership function shown in FIG. 24 (c), each fuzzy rule R1, R2,
.. Can be weighted in accordance with the degree of matching. Then, in the membership function of the consequent part, the intention of the driver is determined from the fitness corresponding to the output C1.

The membership functions shown in FIGS. 24A to 24C are for explaining the min-max composite centroid method, and the member functions of the present embodiment shown in FIGS. It does not always match the ship function. By the way, as shown in FIG. 25, fifteen rules R1 to R15 are set as fuzzy rules. Of these, in each of the rules R1 to R10, the target shift speed determined from the shift maps shown in FIGS. 5 to 13 (that is, the shift speed set by the normal target shift speed setting means 3)
Even if the current gear is different from the current gear, the optimal gear determining means 1 estimates that the driver has little intention to shift and outputs a correction signal to maintain the current gear.

The fuzzy rules R1 to R10 will be described below. In the first fuzzy rule R1, the vehicle load degree αVL> 0, the absolute value is medium, the accelerator opening VA is large, and When the target shift speed is larger than the current shift speed and an upshift is instructed based on the shift map of the target shift speed setting means 3 (for example, the above-described # 1 selection map), the optimum shift speed determining means is selected. In 1,
It is estimated that the vehicle is traveling on an uphill or traveling in a loaded state. In this case, even if the upshift is instructed by the target gear position setting means 3, the optimal gear position determining means 1 estimates that the driver does not intend to upshift, and shifts the gear speed to the current state. The gear shift command signal from the target gear position setting means 3 is corrected so as to be maintained. As a result, upshifting of the shift speed is suppressed, and the current shift speed (optimum shift speed) is maintained.

Further, in the second fuzzy rule R2, the vehicle load degree αVL> 0, the absolute value thereof is medium, the accelerator opening VA is medium, and the shift map (for example, ## When an upshift is instructed based on 1), the first fuzzy rule R
As in the case of 1, the optimum gear position determining means 1 estimates that the vehicle is traveling on an uphill road or is running in a loaded state. In this case also, the optimal gear position determining means 1 gives the driver an intention to shift up. It is presumed that there is no gear speed, and the gear speed is maintained in the current state as the optimum speed.

As described above, in the first and second fuzzy rules R1 and R2, unnecessary upshifts are suppressed and the drivability is improved when the vehicle is traveling on an uphill or in a loaded state. In the third fuzzy rule R3, when the vehicle load αVL <0, the absolute value is small and the accelerator opening VA
Is small and the upshift is instructed based on the shift map (for example, # 1) of the target shift speed setting means 3, the optimum shift speed determining means 1 estimates that the vehicle is traveling on a downhill road. I do. In this case as well, it is estimated that the driver does not intend to shift up, and the shift command signal from the target shift speed setting means 3 is corrected to maintain the gear shift speed in the current state.

In the fourth fuzzy rule R4, the vehicle load degree αVL <0, the absolute value of which is medium, and the accelerator opening degree V
When A is small and an upshift is instructed based on the shift map (for example, # 1) of the target gear position setting means 3, as in the case of the above-mentioned third fuzzy rule R3, the optimum gear position determining means 1 It is assumed that the vehicle is traveling on a downhill. Then, similarly to the above, the gear shift command signal from the target gear position setting means 3 is corrected, and the gear speed is maintained at the current state.

As described above, in the third and fourth fuzzy rules R3 and R4, unnecessary shift-up during traveling on a downhill road is suppressed, whereby an effective engine brake can be obtained. become. Therefore,
The safety of the vehicle is improved and the drivability is improved. Next, the fifth fuzzy rule R5 will be described. In the fifth fuzzy rule R5, the accelerator opening VA is small, the vehicle speed V is low, and the shift map (for example, , # 1), when an upshift is instructed, the optimum gear position determining means 1 estimates that the vehicle is traveling on a congested road. In this case, the optimal gear position determining means 1 estimates that the driver does not intend to upshift, and corrects the gear shift command signal from the target gear position setting means 3 so as to maintain the gear gear position in the current state. As a result, upshifting of the shift speed is suppressed, and the current shift speed is maintained as the optimal shift speed. Therefore, unnecessary upshifts on congested roads are suppressed, and drivability is improved.

In the sixth fuzzy rule R6, when the vehicle load αVL> 0, its absolute value is small and the accelerator opening change Δ
If VA <0, the absolute value is large, and the target gear position setting means 3
When an upshift is instructed based on the shift map (for example, # 1), the value 1 is set in the corner flag fc.
And the optimum gear position determining means 1 estimates that the vehicle has decelerated just before the curve. Also in this case, it is estimated that the driver does not intend to upshift, and the gear shift speed is corrected so as to be maintained in the current state. Therefore, upshifting of the shift speed is prohibited, and the current shift speed is maintained as the optimal shift speed. As described above, according to the sixth fuzzy rule R6, it is possible to prevent an upshift when decelerating just before a curve, to obtain an effective engine brake, and to improve drivability.

In the seventh fuzzy rule R7, the vehicle load degree αVL <0, the absolute value of which is large, the accelerator pedal opening VA is small, and either the exhaust brake or the compression-release type engine auxiliary brake is used. When the upshift is instructed based on the shift map of the target gear position setting means 3 (for example, the selection map of # 3) with the auxiliary brake on, the vehicle is turned on by the optimal gear position determining means 1 It is estimated that the vehicle is traveling on a rapidly descending slope road where the vehicle speed further increases despite the above. Also in this case, it is estimated that the driver does not intend to upshift, and the gear shift speed is corrected so as to be maintained in the current state. Therefore, upshifting of the shift speed is prohibited, and the current shift speed is maintained as the optimal shift speed. That is, according to the seventh fuzzy rule R7, even if the target shift speed is a shift speed higher than the current shift speed, the shift speed is one shift speed lower than the target shift speed, ie, The shift command signal is corrected so as to maintain the current shift stage as the optimal shift stage.
Therefore, it is possible to continuously and satisfactorily obtain the engine brake at the current gear position.

In the eighth fuzzy rule R8, when the accelerator opening change ΔVA> 0, the absolute value is small and the vehicle speed V is low, and the target gear is smaller than the current gear, ie, When a downshift is instructed based on the shift map of the target gear position setting means 3 (for example, the selection map of # 1), the optimum gear position determining means 1 performs the re-acceleration after the vehicle is once decelerated. It is estimated that there is. And
In this case, it is estimated that the driver does not intend to downshift, and correction is made so as to maintain the current gear position. Therefore, downshifting of the gear position is suppressed, and the current gear position is maintained as the optimal gear position. This also mainly takes into account running in a traffic jam, and is based on the fact that in such a traffic jam, the vehicle often decelerates immediately even after performing a shift.

In the ninth fuzzy rule R9, when the accelerator opening change ΔVA> 0, the absolute value is medium, the vehicle speed V is low, and a downshift is instructed based on a shift map (for example, # 1). If it is, it is assumed that the vehicle is re-accelerated after decelerating the vehicle once in a traffic jam or the like, and the gear position is maintained at the current gear position. That is, in the eighth and ninth fuzzy rules R8 and R9, unnecessary downshifting can be suppressed during acceleration in a traffic jam or the like, and the drivability is also improved. In addition, there is an advantage that fuel economy can be improved by suppressing downshifting.

In the tenth fuzzy rule R10, the operation of auxiliary brakes such as the exhaust brake and the compression release type engine auxiliary brake is on, the foot brake is off, and the downshift is performed based on the shift map (for example, # 3). When instructed, it is assumed that the vehicle is to be decelerated lightly, and the shift command signal from the target shift speed setting means 3 is corrected so as to maintain the gear shift speed in the current state. As a result, downshifting of the shift speed is suppressed,
The gear is kept at the current gear (optimum gear). Therefore, it is possible to suppress a large deceleration that is contrary to the driver's intention,
Shift shock associated with downshifting can be prevented, and drivability is improved.

The eleventh fuzzy rule R11 to the fourteenth fuzzy rule R14 will be described. In these fuzzy rules, the shift speed set based on the shift maps shown in FIGS. Even if they match, the optimum gear position determining means 1 estimates that the driver has a will to change gears and outputs a correction signal so as to shift to a gear position one speed lower than the current gear position. .

That is, in the eleventh fuzzy rule R11,
When the vehicle load degree αVL> 0, the absolute value is large, the accelerator opening VA is large, and there is no shift instruction by the target shift speed setting means 3, that is, a shift set based on a shift map (for example, # 1). When the gear and the current gear are the same, the optimum gear deciding means 1 estimates that the vehicle is traveling on a steep hill or traveling in a loaded state. Then, in this case, even if the target gear position setting means 3 does not set an instruction to change the gear position,
The optimal gear position determining means 1 estimates that the driver intends to downshift, and sets the gear gear position to be one gear lower than the current gear position.
The shift command signal is corrected so as to shift to the lower gear. As a result, the shift speed is shifted down to the optimum shift speed, so that a larger driving torque can be obtained, and therefore, the acceleration performance in traveling on a steep uphill road or traveling in a loaded state is improved. .

In the twelfth fuzzy rule R12,
When the vehicle load degree αVL <0, the absolute value is large, the auxiliary brake is on, and there is no gear shift instruction based on the shift map (for example, # 3) of the target gear position setting means 3, the optimum gear position determining means 1 It is estimated that the vehicle is traveling on a steep downhill road. By the way, in this case, since there is no shift instruction to the upshift side even though the vehicle is on a steep downhill road, the auxiliary brake is more intense than in the case of the above-mentioned seventh fuzzy rule R7, which is also assumed to be traveling on a steep downhill road. It seems that it is working well. However, even in such a case, the optimum gear position determining means 1 estimates that the driver intends to downshift, and sets the gear speed position one step lower than the current gear position which is the target gear position. The gear change command signal is corrected so that the gear is shifted to the gear position. This makes it possible to reliably obtain a larger engine brake.

In the thirteenth fuzzy rule R13,
The vehicle load degree αVL <0, the absolute value of which is large, the auxiliary brake is on, the foot brake is on, and there is no shift instruction based on the shift map of the target shift speed setting means 3 (for example, the selection map of # 2). In this case, the optimal gear position determining means 1 estimates that the vehicle is traveling on a steep downhill road as in the case of the twelfth fuzzy rule R12. However,
In this case, the driver operates the foot brake in addition to the case of the fuzzy rule R12, and it is estimated that the vehicle is traveling on a steeper downhill. Therefore, the optimum gear position determining means 1 again estimates that the driver intends to downshift, and instructs the gear position to shift to the gear position one step lower than the current gear position. Correct the signal. This makes it possible to reliably obtain a larger engine brake.

As described above, in the twelfth and thirteenth fuzzy rules R12 and R13, the shift speed is shifted down to the optimum shift speed, so that a large engine brake can be obtained and the load on the foot brake can be reduced. As a result, a more safe driving state can be achieved. In the fourteenth fuzzy rule R14, the absolute value is small when the vehicle load αVL is greater than 0, and the accelerator opening VA is medium, the absolute value is large when the accelerator opening change ΔVA is greater than 0, and the vehicle speed V is moderate. ,
In addition, the shift map (for example, #
If there is no gear change instruction based on 1), the optimum gear position determining means 1 estimates that the vehicle is about to perform overtaking acceleration, and shifts the gear gear position one gear lower than the current gear position. The shift command signal is corrected so as to shift to the first gear. As a result, the shift to the lower gear is performed, and smooth acceleration can be performed.

In the fifteenth fuzzy rule R15, the vehicle load degree αVL> 0, the absolute value of which is larger, the target gear is larger than the current gear, and the shift map of the target gear setting means 3 (for example, When an upshift is instructed based on # 1), the optimal gear position determining means 1 determines whether the vehicle is traveling on a steep uphill road or in a loaded state, regardless of the accelerator opening VA or the like. It is estimated that the vehicle is running. In this case, even if an instruction to shift up is given by the target shift speed setting means 3, the optimal shift speed determining means 1 estimates that the driver does not intend to shift up and sets the gear shift speed to the current state. The shift command signal from the target shift speed setting means 3 is corrected so as to hold the speed change command signal. As a result, the shift-up of the gear is suppressed, and the gear is maintained at the current gear, which is also the optimum gear.

In the shift control device for an automatic transmission for a vehicle according to one embodiment of the present invention, the target shift speed is corrected as described above to determine the optimum shift speed. Is determined by the control unit 71 in accordance with the flowchart of the main routine as shown in FIG. The automatic shift, that is, the procedure for determining the optimal shift position in the D range will be described in detail below with reference to FIG.

First, in step S10, the engine torque Te is calculated by the engine torque calculating means 4 as described above. Then, in step S12, the driving force F of the vehicle is calculated by the driving force calculating means 5, and in the next step S14, the air resistance Rl of the vehicle is calculated by the air resistance calculating means 6. In step S16, the linear flat road empty vehicle equivalent acceleration calculating means 7 calculates the linear flat road empty vehicle equivalent acceleration α0 from the driving force information F calculated in step S12 and the air resistance information Rl calculated in step S14.
Is calculated.

Then, in Step S18, Step S
The vehicle load degree information αVL is calculated based on the acceleration α0 equivalent to the straight flat road empty vehicle calculated in step 16 and the actual acceleration information α from the vehicle speed sensor 79. In the next step S19, the learning type selection means 3B of the target gear position setting means 3 performs the learning control of the speed change according to the driver's preference,
An optimal shift map is selected from the storage unit 3A. Hereinafter, the learning control of the shift according to the driver's preference will be described.
This will be described with reference to the flowchart of FIG.

First, in step S70 in FIG. 27, data necessary for learning the shift according to the driver's preference is calculated. The learning data includes the following.
Here, since the effect of the learning is large, the learning of the upshift is targeted for the 2-3 upshift and the 3-4 upshift. It is intended for. (1) M range use frequency FMU: The use frequency FMU is the use frequency of the M (manual) range at the time of upshifting, and is the number of times the M range has been used during the last 10 2-3 upshifts and 3-4 upshifts. Ask for. (2) 2-3 accelerator opening average AVA (23): 2-3 accelerator opening average AVA (23) is an average value of accelerator opening VA at the time of shifting from the second gear to the third gear in the D range. And the average of the past five flat roads is obtained from the following equation.

AVA (23) = (VA (23) 0 + VA (23) 1 + ·
.. + VA (23) 4) / 5 (3) 2-3 standard deviation SVA (23): 2-3 standard deviation SV
A (23) is the standard deviation (3σ) of the accelerator opening VA at the time of shifting from the second gear to the third gear in the D range, and the standard deviation of the past five flat roads is also obtained from the following equation. Ask.

SVA (23) = 3 · ((1/5) · Σ (VA (2
3) i-AVA (23)) ² ) ^1/2 , i = 0 to 4 (4) 3-4 accelerator opening average AVA (34): 3-4 accelerator opening average AVA (34) is in D range The average value of the accelerator opening VA at the time of shifting from the third gear to the fourth gear is obtained from the following equation.

AVA (34) = (VA (34) 0 + VA (34) 1 + ·
.. + VA (34) 4) / 5 (5) 3-4 standard deviation SVA (34): 3-4 standard deviation SV
A (34) is a standard deviation (3σ) of the accelerator opening VA at the time of shifting from the third gear to the fourth gear in the D range. Similarly, the standard deviation of the past five flat roads is obtained from the following equation. Ask.

SVA (34) = 3 · ((1/5) · Σ (VA (3
4) i-AVA (34)) ² ) ^1/2 , i = 0 to 4 (6) 2-3 engine rotation average ANe (23): 2-3 engine rotation average ANe (23) is a manual in the M range. This is the average value of the engine speed at the time of shifting from the second gear to the third gear by the operation, and the average value of the past five flat roads is obtained from the following equation. Specifically, it is obtained from the average value of the engine rotation speed Ne immediately before the start of the shift.

ANe (23) = (Ne (23) 0 + Ne (23) 1+...
・ + Ne (23) 4) / 5 (7) 3-4 engine rotation average ANe (34): The 3-4 engine rotation average ANe (34) is changed from the third gear to the fourth gear by manual operation in the M range. This is the average value of the engine speed at the time of shifting, and the average value of the past five flat roads is obtained from the following equation. Similarly to the above, it is obtained from the average value of the engine rotation speed Ne immediately before the start of the shift.

ANe (34) = (Ne (34) 0 + Ne (34) 1+...
・ + Ne (34) 4) / 5 (8) Hold range (HOLD range) frequency of use F
H: HOLD range use frequency FH, that is, the frequency of switching to the M range side is such that the vehicle speed V increases from a predetermined value V2 (for example, 40 km / h) to a predetermined value V1 (for example, 1
0 km / h) is the ratio of the total time ts to the HOLD range use time tH, and the average value of the past five times is obtained from the following equation.

FHi = tHi / tsi, i = 0 to 4 FH = (FH0 + FH1 +... + FH4) / 5 (9) Frequency of use of M range at shift down FMD: Frequency of M range at shift down FMD is foot brake Is the shift-down frequency by manual operation while the vehicle speed V is changed from the predetermined value V2 (for example, 40 km / h) to the predetermined value V1 (for example, 10 km / h) while the switch is turned on. Obtain from the formula.

FMDi = NMDi / NDi, i = 0 to 4 FMD = (FMD0 + FMD1 +... + FMD4) / 5 where NMDi is the number of downshifts by manual operation in the above period, and NDi is the total shift including manual operation. This is the number of downs. Then, in the next step S72, it is determined whether or not the vehicle is traveling on a flat road for the above reason. Here, it is determined whether or not the vehicle load degree αVL is within a predetermined range (near 0 value) set for each gear. As described above, by determining whether the vehicle is traveling on a flat road, it is possible to perform learning only on a flat road where the driver's preference greatly appears.

If the decision result in the step S72 is false (No),
That is, when it is determined that the vehicle load αVL is outside the predetermined range and the vehicle is not traveling on a flat road, the process proceeds to step S82 without performing the upshift learning. on the other hand,
When the result of the determination in step S72 is true (Yes), that is, when it is determined that the vehicle load αVL is within the predetermined range and the vehicle is traveling on a flat road, the process proceeds to step S74.

In step S74, M at the time of upshift is set.
It is determined whether or not the frequency of use FMU of the range is small. Specifically, it is determined whether or not the usage frequency FMU is equal to or less than a predetermined threshold. If the result of the determination is true, it can be determined that the frequency of use of the M range is low during upshifting, and in this case, the process proceeds to step S76. In step S76, learning of upshifting preference is performed based on upshifting by the D range operation, and an output necessary for selecting a shift map, that is, a judgment output is set. That is, here
Since the use frequency of the M range is low, the learning based on the automatic shift is performed without learning the preference based on the use of the M range. More specifically, learning based on the back propagation learning method and calculation of the judgment output are performed by the first neural network shown in FIG. Hereinafter, processing in the first neural network will be described.

In the input layer of the first neural network shown in FIG. 28, the previously selected shift map information
-3 average accelerator opening AVA (23), 2-3 standard deviation SV
A (23), 3-4 accelerator opening average AVA (34) and 3-4
Each input value I1, I2, I3, I4, I of the standard deviation SVA (34)
5 is entered. Here, the previously selected shift map information is the output of the previous determination, and has the character as a reference value for the current learning. The input values of I1 to I5 are normalized by the following equation to obtain Ii '.
Are input to the three intermediate layers.

Ii '= (Ii-Iimin) / (Iimax-Iimi
n), i = 1 to 5 Here, Iimin is the minimum value of Ii, that is, the minimum value of each input value of I1 to I5, and Iimax is the maximum value of Ii, that is, I1
To I5 are the maximum values of the input values. Then, upon input to the three intermediate layers, different coupling coefficients Wij (i = 1 to 5, j = 1 to 3) are added to the respective intermediate layer input values Ii '. That is, each of the input values Ii ′ input from the five input layers to the three intermediate layers is set to 1 by an experiment or the like in advance.
Five coupling coefficients Wij (W11, W12, W13, W2)
1, W22... W53) are integrated.

From the three intermediate layers, output values O1, O2 and O3 calculated by the following sigmoid function are output. Oj = 1 / (1 + exp (-Xj)), j = 1 to 3, where Xj = W1j.I1 '+ W2j.I2' +... + W5j.
I5'-θj, j = 1 to 3 Here, θj (j = 1 to 3) is a predetermined threshold value set in advance by an experiment or the like.

The output values O1, O2, and O3 are also integrated with a predetermined coupling coefficient Wj4 (j = 1 to 3) and input to the output layer. From the output layer, a judgment calculated by the following equation is made. Output O4 is output. O4 = (1 / (1 + exp (-Y4))). (Omax-Om
in) + Omin where Y4 = W14.O1 + W24.O2 + W34.O3-.theta.4 where Omax is the maximum value of the judgment output O4 and Omin is the minimum value of the judgment output O4. Θ4 is a predetermined threshold value set in advance by an experiment or the like.

The above Iimin, Iimax, Omin, Om
Detailed descriptions of ax, Wij, Wj4, θj, and θ4 are omitted here. However, among the input values Ii, the input value I1 and the judgment output value O4 corresponding to the previously selected shift map information take values between the values 1 to 3 as described later. The value I1min and the minimum value Omin of the output value O4 become the value 1, while the maximum value I1max of the input value I1 and the maximum value Omax of the output value O4 become the value 3
Becomes

As described above, when the frequency of using the M range at the time of the shift up by the driver is small, learning is performed based on the shift up by the D range operation, not the M range operation, and the shift map is selected. output,
That is, the judgment output O4 is set satisfactorily and stably. On the other hand, if the determination result of step S74 is false and it is determined that the use frequency of the M range at the time of upshifting is high, it can be determined that the driver frequently uses the M range and is not satisfied with only the D range, Next, the process proceeds to step S78.

In step S78, step S74 is executed.
In the above, based on the fact that the use frequency of the M range is increased, the preference of the upshift is learned based on the M range operation, that is, the upshift by the manual operation, and the output for selecting the shift map, ie, the judgment output is set. . More specifically, learning and calculation of the judgment output are performed by a second neural network as shown in FIG. 29 provided separately from the first neural network. Less than,
Processing in the second neural network will be described.

The input layer of the second neural network shown in FIG.
The input values I′1, I′2, I′3 of the −3 engine rotation average ANe (23) and the 3-4 engine rotation average ANe (34) are input. Here, similarly to the above, the shift map information selected last time is the previous determination output, and has the character as the reference value for the current learning. Then, the input values of I1 to I5 are normalized by the following equation to be I'i ', and are input to three intermediate layers.

Then, the output values O'1,
O'2 and O'3 are obtained, and the judgment output O'4 is finally derived. The arithmetic processing in the second neural network is the same as the arithmetic processing in the first neural network. Here, detailed description is omitted. In this way, if the frequency of using the M range at the time of upshifting is high, learning is performed based on manual upshifting by operating the M range, and the shift map according to the driver's preference is determined. The output O'4 will be set appropriately.
In other words, the output for selecting which of the shift maps of the economy mode, the standard mode, and the power mode is set according to the operating condition of the M range by the driver.

When the judgment outputs O4 and O'4 relating to the upshift are set by the processing of the first neural network or the second neural network as described above, the process proceeds to step S80. In step S80,
A shift map is selected according to the judgment outputs O4 and O'4. That is, any one of the shift maps of the economy mode, the standard mode, and the power mode is selected here.

More specifically, the shift map is selected based on a graph showing the relationship between the judgment outputs O4 and O'4 and the selection map as shown in FIG. For example, the judgment output O4, O'4 value from the first neural network or the second neural network is 1 or more and less than 1.5 (1 ≦ O
When 4, O'4 <1.5, the economy mode shift map is selected, and the judgment output O4, O'4 values are 1.5 or more and less than 2.5 (1.5? O4, O'4 < In the case of 2.5), the shift map in the standard mode is selected, and the decision output O
4, O'4 value is 2.5 or more and less than 3 (2.5 ≦ O4, O'4 <
In the case of 3), the power mode shift map is selected.

Here, the selection map of the economy mode is a shift map focusing on fuel consumption as described above. In the standard mode having the same shift timing as the normal D range, the shift timing at each shift speed is lower than that of the standard mode. Side is set. The power mode selection map is a shift map with an emphasis on acceleration, and the upshift timing of each gear is set on the high vehicle speed side with respect to the standard mode. In other words, when the judgment outputs O4 and O'4 change due to the learning, the shift timing in the D range (automatic shift) is appropriately switched to a shift pattern according to the driver's preference.

In the next step S82, it is determined whether or not the vehicle speed V has changed from a predetermined value V2 (for example, 40 km / h) to a predetermined value V1 (for example, 10 km / h) with the foot brake on. This determination is for determining whether to learn the downshift preference. If the decision result in the step S82 is false, the vehicle speed V is kept at the predetermined value V2 while the foot brake is on.
(For example, 40 km / h) to a predetermined value V1 (for example, 10 km / h).
If it has not changed to h), it can be determined that it is not necessary to learn the downshift preference, and in this case,
Exit this routine without doing anything.

On the other hand, if the decision result in the step S82 is true,
The vehicle speed V is increased from a predetermined value V2 (for example, 40 km / h) to a predetermined value V1 (for example, 10 km / h) while the foot brake is on.
If it has been changed to a state, it can be determined that the learning of the downshift preference is to be performed.
Proceed to 4. In step S84, the shift down preference is determined. That is, the learning of the downshift preference is performed according to the driver's use condition of the M range, and the judgment output necessary for selecting the shift map is set. Specifically, learning and calculation of the judgment output are performed by the third neural network shown in FIG. Hereinafter, processing in the third neural network will be described.

The input layer of the third neural network shown in FIG.
The input values I "1, I" 2, I "3 of the OLD range use frequency FH and the M range use frequency FMD are input, and the shift map information selected last time is the reference value of the current learning. And these I "1,
The input values of I "2 and I" 3 are normalized by the following equation to obtain I "i '
Are input to the three intermediate layers.

Then, the output values O "1, O" 2,
O "3 is obtained, and the decision output O" 4 is finally derived. The arithmetic processing in the third neural network is the same as the arithmetic processing in the first and second neural networks. Here, the description is omitted. Here, among the input values I "i, the input value I" 1 and the judgment output value O "4 corresponding to the previously selected shift map information are values between values 0 and 1 as described later. Therefore, the minimum value I "1min of the input value I" 1 and the output value O "
The minimum value O "min of 4 has a value of 0 , while the maximum value I" 1max of the input value I "1 and the maximum value O" max of the output value O "4 have a value of 1 .

In this manner, the learning related to the downshift is performed based on the use condition of the M range at the time of the downshift, and the judgment output O "4 for selecting the shift map according to the driver's preference is set. That is, here, 5-4 shift down and 4-3
The output is set as to whether or not to select the downshift restriction map that restricts downshifting. Here, the shift down regulation map is provided for each of the economy mode, the standard mode, and the power mode selection map # 1 as described above, and is a shift map different from the maps of FIGS. 5, 8, and 11 described above. is there.

FIG. 33 shows a typical downshift map in the economy mode. As shown in FIG. 33, in the downshift restriction map, the accelerator opening V
When A is small, 5-4 shift down and 4-3
The shift timing of the downshift has shifted to the lower vehicle speed side. That is, the timing of the 5-4 shift down and the 4-3 shift down is the same as the timing of the 3-2 shift down. Therefore, in this shift-down regulation map, the fifth gear or the fourth gear is held unless the vehicle speed V becomes extremely low, and the 5-4 shift down and the 4-3 shift down are not performed.

Then, in step S86, it is determined whether or not downshifting is possible in accordance with the determination output O "4. The normal economy mode, standard mode, and power mode selection maps # 1 and shift maps provided for each mode are provided. One of the down regulation maps is selected.
As shown in FIG. 2, from the graph showing the relationship between the judgment output O "4 and the downshifting availability, the downshifting availability according to the determination output O" 4 is selected, and the downshifting restriction map is appropriately selected based on the result. Is done. That is, if the value of the judgment output O "4 is equal to or more than 0 and less than 0.5 (0≤O" 4 <0.5), it is determined that there is no downshift, and the downshift restriction map is selected. 4 value is 0.5 or more and 1 or less (0.5
If .ltoreq.O "4.ltoreq.1), the shift down is possible, and the selection map # 1 for the normal economy mode, the standard mode, and the power mode is selected.

For example, when the driver frequently performs a manual downshift in the M mode during deceleration, the normal economy mode, standard mode, and power mode selection map # 1 is selected.
At this time, in the D range, 5-
A 4,4-3 downshift is performed, whereby the engine brake works properly. On the other hand, during deceleration,
In the case where the driver often operates the foot brake while holding the HOLD range, that is, when the driver does not like downshift shocks, the downshift restriction map is selected. In other words, the fifth gear and the fourth gear are satisfactorily held irrespective of the decrease in the vehicle speed V, so that discomfort such as a shift shock is suitably prevented.

As described above, when the selection of the shift map according to the driver's preference is completed by taking the learning into account,
Next, the process proceeds to step S20 in FIG. This step S2
From 0 onward, a routine for setting the optimal gear position mainly based on the driving state of the vehicle is performed. In step S20, it is determined whether or not the corner flag fc has a value of 1 and the vehicle has approached a corner. As described above, the corner flag fc is set to a value of 1 based on the sixth fuzzy rule R6, and is set in step S32, which will be described later, for inferring the driver's intention in this routine. If the decision result in the step S20 is true (Yes), the corner flag f
If c is equal to 1, that is, if it is determined that the vehicle is approaching a corner and decelerating just before the corner, then step S26 is executed.
Proceed to.

At step S26, post-corner processing is performed. This corner post-processing means that the corner flag fc is set to the value 0.
Here, the corner flag fc is reset when the following condition is satisfied. fc reset condition: vehicle load αVL is smaller than predetermined value TH2 (αVL <TH2), accelerator opening VA is equal to or larger than predetermined value TH3 (VA ≧ TH3), or target gear matches current gear (current = target) Note that the predetermined value TH3 is 75 times the actual accelerator opening.
% Open state.

Normally, immediately after the corner flag fc is set to the value 1, the condition for resetting the corner flag fc is not satisfied. In this case, the corner flag fc is held at the value 1 without being reset. Then, the process proceeds to step S28. In step S28, the final target stage is set to the current stage. As a result, the above step S2
If the determination at 0 indicates that the corner flag fc has a value of 1 and the vehicle is decelerating just before the corner,
Even if the target gear is larger than the current gear, that is, when an upshift is instructed, the gear is maintained at the current gear regardless of the target gear without performing the driver intention inference to be described later. (The sixth fuzzy rule R6).

On the other hand, if the decision result in the step S20 is false (N
In o), when the corner flag fc is set to the value 0 by executing step S26 and it is determined that the vehicle is traveling straight, the process proceeds to step S22. Step S
At 22, a target gear suitable for the running state is obtained from the shift map selected by the target gear setting means 3 based on the vehicle speed V and the accelerator opening VA.

Then, in step S24, step S
10 Based on the fuzzy rules R1 to R15, based on the fuzzy rules R1 to R15, based on the information from the sensors and the target shift speed, in addition to the vehicle load degree αVL obtained in step S18, the driver's driving intention is determined. presume. In step S30, the shift command to the target shift speed is corrected for each of the fuzzy rules R1 to R15 based on the driver's intention estimated in step S24 (see FIG. 25). Specifically, in this step S30, a subroutine for target gear position correction shown in FIGS. 34 and 35 is executed. Less than,
The procedure for correcting the target shift speed will be described with reference to FIGS.

In step S32 in FIG. 34, it is determined whether or not a predetermined time t1 (for example, 3 seconds) has elapsed since the previous shift was completed. If the determination result is false and it is determined that the predetermined time t1 (for example, 3 seconds) has not yet elapsed after the shift is completed, the shift is continuously performed so that the shift shock is not caused. Step S4
Proceed to 2 to hold the final target stage at the current stage. On the other hand, if the result of the determination is true and it is determined that the predetermined time t1 (for example, 3 seconds) has elapsed since the end of the shift, the process proceeds to step S34.

In step S34, a predetermined time t2 (for example, after the on / off switching of the auxiliary brake such as the exhaust brake or the compression-release type engine auxiliary brake) is performed.
3s) is determined. If the determination result is false and it is determined that the predetermined time t2 (for example, 3 seconds) has not yet elapsed after the on / off switching of the auxiliary brake, the process also proceeds to step S42, and the final target stage is held at the current stage. That is, as described above, the engine torque Te is calculated based on the map of the on / off state of the auxiliary brake such as the exhaust brake and the compression release type engine auxiliary brake, but immediately after this on / off switching,
There is a delay in the operation of the exhaust brake and the compression-release type engine auxiliary brake, and an unmatch occurs between the calculated engine torque Te and the actual engine torque, thereby causing unnecessary shifting contrary to the driver's intention. It is preventing it.

The result of determination in step S34 is true, and a predetermined time t2 (for example, 3 seconds) after the on / off switching of the auxiliary brake
If it is determined that has elapsed, then step S36
Proceed to. In step S36, the output value of the fuzzy inference,
That is, it is determined whether or not the membership grade (degree of conformity) of each membership function described above is greater than a predetermined value TH0 (for example, value 0) and equal to or less than a predetermined value TH1 (for example, value 0.5). That is, it is determined whether or not the driver's intention is such that the gear position is desired to be maintained at the current gear position.

If the decision result in the step S36 is true, the output value is larger than the predetermined value TH0 (for example, the value 0) and equal to or less than the predetermined value TH1 (for example, the value 0.5) (TH1 ≧ output value> TH)
When it is determined as 0), the process proceeds to step S42, and the final target stage is held at the current stage. On the other hand, the determination result of step S36 is false, and the output value is a predetermined value TH0 (for example, value 0).
Alternatively, if it is determined that the value is larger than the predetermined value TH1 (for example, the value 0.5), the process proceeds to step S38.

In step S38, the output value of the fuzzy inference, that is, the membership grade (degree of conformity) is set to a predetermined value TH.
It is determined whether it is greater than 1 (for example, a value of 0.5).
In other words, it is determined whether or not the driver's intention is to shift down the shift speed by one speed. If the decision result in the step S38 is true, and the output value is a predetermined value TH1 (for example, the value 0.
5) If it is determined that the value is greater than (output value> TH1), the process proceeds to step S40.

In step S40, it is determined whether or not the vehicle speed V is equal to or higher than the overrun vehicle speed. If the determination result is true, the process proceeds to step S42, and the final target stage is held at the current stage. On the other hand, if the determination result is false, the process proceeds to step S44. In step S44, step S38
As a result, the output value becomes the predetermined value TH1 (for example, the value 0.
5) Based on the determination that the target value is greater than (output value> TH1), the final target speed is shifted down to a speed one speed lower than the current speed.

If the result of the determination in the previous step S38 is false and it is determined that the output value is not larger than the predetermined value TH1 (for example, a value of 0.5), the process proceeds to step S46.
Such a situation is a case where the output value is equal to a predetermined value TH0 (for example, value 0). In this case, it is not necessary to apply the above fuzzy rules, and step S4
In 6, the final target gear is set to the target gear determined in step S22 in FIG.

As described above, the optimum gear position reflecting the driver's intention with respect to the target gear position is set based on the fuzzy rules R1 to R15. That is, even if the target gear position different from the current traveling gear position is set by the target gear position setting means 3 and a shift down or shift up command signal is input to the optimum gear position determination means 1, When it is estimated that the optimum gear position reflecting the intention corresponds to the current traveling gear position, the correction for prohibiting the shift command to the target gear position is performed by the optimum gear position determining means 1. Accordingly, in this case, the vehicle travels while maintaining the current gear position, and the most appropriate gear position according to the load condition of the vehicle and road conditions (uphill, downhill, curve presence, congested road, etc.). It is possible to run with.

Further, even if the target gear set by the target gear setting means 3 matches the current traveling gear and the gearshift command signal is not input to the optimum gear deciding means 1, the driver's If the optimum gear position reflecting the intention is different from the current traveling gear position (here, the target gear position), the optimum gear position determining means 1 sets a correction signal for executing a gear shift operation. As a result, the speed change mechanism 17 performs the speed change operation, and the vehicle can run at the most appropriate speed according to the load state of the vehicle and the road condition.

If the target shift speed and the optimum shift speed match, the target shift speed naturally becomes the optimum shift speed, and a shift operation to this target shift speed is performed. FIG.
In step S50, it is determined whether or not a jump shift-up command (for example, 4th → 6th) has been output. That is, in step S46, the final target gear is set as the target gear based on the shift map. However, it is determined whether the target gear at this time is not a gear shift command for each gear but a jump shift-up command.

If the decision result in the step S50 is false, that is, it is determined that the jump up command has not been output, the routine is terminated and the process returns to the main routine of FIG. On the other hand, when the result of the determination in step S50 is true and it is determined that the jump up command has been output, the process proceeds to step S52.

In step S52, it is determined whether or not a predetermined time t3 (for example, 3 seconds) has elapsed since the output of the jump shift-up command, and whether a later-described jump shift flag fj has a value of 1. If the determination result is false, that is, from the time immediately after the output of the jump shift-up command to the time until a predetermined time t3 (for example, 3 seconds) elapses, the next step S
At 54, it is assumed that the final target stage is shifted up by only one stage from the current stage irrespective of the jump up instruction. As a result, the jump up can be performed for a predetermined time t3.
(For example, for 3 seconds), the shift feeling is effectively prevented from being deteriorated. Then, in step S56, a value 1 is set to the jump shift flag fj in order to store that the final target stage has been shifted up by one stage from the current stage.

Then, the routine is executed next time, and as determined in step S52, a predetermined time t3 (for example, 3 seconds) has elapsed since the jump shift-up command was output, and the jump shift flag fj has a value of 1. Is determined, the jump shift flag fj is reset to a value of 0 in step S58. As described above in detail, in the shift control device for an automatic transmission for a vehicle according to the present invention, learning is performed based on a back propagation learning method using a neural network. In addition to selecting the optimum shift map according to the driver's preference from a plurality of shift maps taking into account the preference, the target shift speed determined by this shift map is further corrected by the desired fuzzy rule using the vehicle load degree αVL as a parameter Therefore, the optimum gear position can be suitably determined by reflecting not only the driver's preference but also the driver's intention. Accordingly, in the shift control device for the automatic transmission for a vehicle, the shift control device is an automatic transmission,
Shift control very close to upshifting and downshifting of a manual transmission mainly composed of manual operation can be performed, and drivability can be greatly improved.

Further, a neural network for performing learning control of the driver's preference, a first neural network for performing learning based on the D range operation according to the frequency of use FMU of the M range, and a learning based on the M range. The learning control at the time of upshift is performed separately from the second neural network, and the learning control at the time of downshift is individually performed by another third neural network, so that the neural network is complicated. It is possible to easily learn the driver's preference without any problem, and to always set the optimum gear position satisfactorily and smoothly perform the shift control of the automatic transmission without feeling uncomfortable.

[0146]

As described above, according to the shift control device for an automatic transmission for a vehicle according to the first aspect, the vehicle is in a manual mode in which the shift speed is switched by an operation of a driver by a driver's operation. In a shift control device for a vehicle automatic transmission having an automatic mode in which a shift stage is automatically switched according to a shift state in at least one of a manual mode and an auto mode, a driver reflects a shift preference of a driver through a neural network. Learning means for learning the selected shift pattern, target gear setting means for setting a target gear based on the output from the learning means and the operating state of the vehicle in the auto mode, and gear shift control of the automatic transmission based on the target gear. Speed change control means for performing the operation of the manual mode by the driver.
Depending on the frequency, manual mode or auto mode
Since the selected parameters and to so that one or the other shifting state of, based on the frequency of use of the manual mode,
The shift pattern learning that reflects the driver's shift preference can be satisfactorily performed, and the target shift speed can be suitably set based on the output from the learning and the driving state of the vehicle in the auto mode. Thus, the shift control of the automatic transmission can be made similar to the shift operation of the manual transmission.

[0147]

According to the shift control device for an automatic transmission for a vehicle of the second aspect , the learning means uses the manual mode .
When the frequency of use is greater than the predetermined value, the shift state in the manual mode is used as a parameter. Therefore, when the frequency of use in the manual mode is greater than the predetermined value, good shift pattern learning can be performed using the shift state in the manual mode as a parameter.

At this time , there is further provided an engine speed detecting means for detecting the engine speed of the engine, and the learning means sets the shift state in the manual mode to at least the engine speed at the time of the shift operation by the driver as a parameter. Therefore, good shift pattern learning can be performed by using at least the engine speed at the time of the shift operation by the driver as a shift state in the manual mode as a parameter.

In a preferred embodiment, the learning means comprises:
When the use frequency of the manual mode is smaller than the specified value,
The shift state in the auto mode may be used as a parameter, whereby when the frequency of use in the manual mode is smaller than a predetermined value, good shift pattern learning can be performed using the shift state in the auto mode as a parameter.

At this time , there is further provided an accelerator opening detecting means for detecting the accelerator opening of the engine, and the learning means sets the gear shift state in the auto mode as a parameter by using at least the accelerator opening at the time of gear shift as a parameter. learning may it line, thereby, can perform a good shift pattern learning accelerator opening at least shift stage switched as the shift state in the automatic mode as a parameter.

In a further preferred embodiment, the learning means may perform the shift pattern learning by including the current shift pattern information as a parameter, whereby the current shift pattern information is used as a reference parameter for the shift pattern learning. Thus, good shift pattern learning can be performed. According to the shift control device for an automatic transmission for a vehicle of the third aspect, the vehicle further includes a flat road detecting unit that detects whether or not the vehicle is traveling on a flat road, and the learning unit includes: Since the shift pattern learning is performed while the vehicle is traveling, the shift pattern learning can be favorably performed when the vehicle is traveling on a flat road where the driver's preference easily appears.

In a preferred embodiment , the target shift speed setting means has a plurality of types of shift maps having different shift patterns, and selects an optimum map from the plurality of types of shift maps in accordance with an output from the learning means. setting the target gear based on the operating condition of the optimum map and vehicle
Well, this makes it possible to select the optimum map from a plurality of types of shift maps according to the output from the learning means,
The target shift speed can be satisfactorily set based on the optimal map and the driving state of the vehicle.

At this time , the vehicle further comprises a flat road detecting means for detecting whether or not the vehicle is traveling on a flat road, and the target gear position setting means is selected when the vehicle is not traveling on a flat road. The target shift speed may be set based on the shift map and the driving state of the vehicle , so that when the vehicle is not traveling on a flat road where the driver's preference is likely to appear, the previously selected shift map and the vehicle The target shift speed can be satisfactorily set based on the operating state.

Further, according to the shift control device for an automatic transmission for a vehicle according to the fourth aspect , the learning means uses the shift state in at least one of the manual mode and the automatic mode as a parameter by the neural network when the driver shifts up. Since the shift pattern learning reflecting the shift preference is performed, the shift pattern learning reflecting the driver's shift preference at the time of upshifting can be favorably performed by the neural network using the shift state in at least one of the manual mode and the auto mode as a parameter. it can.

According to the shift control device for an automatic transmission for a vehicle of the fifth aspect , the vehicle is connected to the engine and operated by the driver.
Manual mode in which the gears are switched by driving and vehicle operation
Auto mode that automatically switches gears according to conditions
In a shift control device for a vehicle automatic transmission having
At least one of the new mode and the auto mode
Neural network using the speed change state
Shift pattern learning that reflects the driver's shifting preferences
Learning means for performing the
To set the target gear position based on the driving conditions of the vehicle in
Target gear position setting means and an automatic transmission based on the target gear position.
A shift control unit that performs a shift control; a vehicle speed detection unit that detects a vehicle speed; and a brake operation state detection unit that detects an operation state of a brake. The learning unit is configured such that the brake operation state is on and the vehicle speed decreases. When the amount is equal to or more than a predetermined amount, the shift pattern learning that reflects the driver's shift preference during downshifting is performed. Therefore, when the brake operation state is ON and the amount of decrease in vehicle speed is equal to or more than the predetermined amount, the driver's shift is performed. Good shift pattern learning that reflects shift preferences at the time of a down can be performed.

[0157] At this time, the learning means is provided on the downshift side at least until the amount of decrease in vehicle speed reaches the predetermined amount.
Since the shift pattern learning is performed using the frequency of use of the manual mode by the driver in the shift to <br/> as a parameter,
Put until reduction of at least the vehicle speed reaches a predetermined amount
Good shift pattern learning can be performed using the frequency of use of the manual mode by the driver in the shift to the downshift side as a parameter.

Further , the learning means performs the shift pattern learning using at least the mode switching frequency from the auto mode to the manual mode by the driver until the vehicle speed decreases by a predetermined amount, so that at least the vehicle speed decreases. Good shift pattern learning can be performed by using the mode switching frequency of the driver from the auto mode to the manual mode until the amount becomes the predetermined amount as a parameter.

In a preferred embodiment , the shift control means may include fuzzy control means for correcting a target shift speed using fuzzy inference based on vehicle information .
In other words , fuzzy inference is performed based on the vehicle information, and the target shift speed can be suitably corrected according to the result of the fuzzy inference, thereby achieving good shift control.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a main configuration of a shift control device for an automatic transmission for a vehicle according to an embodiment of the present invention.

FIG. 2 is a configuration diagram showing an overall configuration of a shift control device for an automatic transmission for a vehicle as one embodiment of the present invention.

FIG. 3 is a diagram showing a select pattern of a change lever.

FIG. 4 is a graph showing a relationship between an accelerator opening and VA.

FIG. 5 is a graph showing a shift map # 1 in the economy mode.

FIG. 6 is a graph showing a shift map # 2 in the economy mode.

FIG. 7 is a graph showing a shift map # 3 in the economy mode.

FIG. 8 is a graph showing a shift map # 1 in a normal mode.

FIG. 9 is a graph showing a shift map # 2 in a normal mode.

FIG. 10 is a graph showing a shift map # 3 in a normal mode.

FIG. 11 is a graph showing a shift map # 1 in a power mode.

FIG. 12 is a graph showing a shift map # 2 in the power mode.

FIG. 13 is a graph showing a shift map # 3 in the power mode.

FIG. 14 is a graph showing a relationship between a rack position and an SRC.

FIG. 15 is a diagram showing a configuration of a low-pass filter.

FIG. 16 is a flowchart illustrating a filter processing procedure of a low-pass filter.

FIG. 17 is a diagram showing characteristics of a vehicle load degree as a shift control parameter on a flat road.

FIG. 18 is a diagram showing characteristics of a vehicle load degree as a shift control parameter on a 10% uphill road.

FIG. 19 is a diagram showing a characteristic of a vehicle load degree as a shift control parameter on a 10% downhill road.

FIG. 20 is a diagram illustrating a membership function of a vehicle load degree.

FIG. 21 is a diagram showing a membership function of an accelerator opening.

FIG. 22 is a diagram showing a membership function of an accelerator opening change.

FIG. 23 is a diagram showing a membership function of driver intention.

FIG. 24 is a diagram for describing fuzzy inference.

FIG. 25 is a diagram showing a fuzzy rule.

FIG. 26 is a flowchart illustrating a main routine of a shift control.

FIG. 27 is a flowchart illustrating a learning control routine in FIG. 26;

FIG. 28 is a diagram showing a first neural network.

FIG. 29 is a diagram showing a second neural network.

FIG. 30 is a diagram showing a third neural network.

FIG. 31 is a graph showing a relationship between judgment outputs O4 and O'4 and a selection map.

FIG. 32 is a graph showing a relationship between a judgment output O ″ 4 and downshiftability.

FIG. 33 is a graph showing a downshift restriction map corresponding to shift map # 1 in the economy mode.

FIG. 34 is a part of a flowchart showing a subroutine for target gear position correction in FIG. 26;

FIG. 35 is the remaining part of the flowchart showing the subroutine for correcting the target gear position following FIG. 34;

[Explanation of symbols]

 DESCRIPTION OF REFERENCE NUMERALS 1 optimal shift speed determining means (shift control means) 2 vehicle load degree calculating means 3 target shift speed setting means 3A storage means 3B learning formula selecting means (learning means) 4 engine torque calculating means 5 driving force calculating means 6 air resistance calculating means 7 Acceleration calculation means equivalent to a straight road road vehicle 8 Subtraction means 11 Diesel engine (engine) 17 Gear transmission (transmission mechanism) 27 Engine rotation sensor (engine rotation speed detection means) 61 Change lever 63 Shift stage selection switch 65 Gear shift unit 71 Control Unit 79 Vehicle speed sensor (vehicle speed detecting means) 81 Accelerator pedal 85 Accelerator opening sensor (accelerator opening detecting means) 87 Brake sensor (brake operating state detecting means)

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ identification code FI F16H 59:42 59:44 59:54 59:66 (58) Investigated field (Int.Cl. ⁷ , DB name) F16H 61 / 02 F16H 61/10 F16H 61/18 F16H 59:02 F16H 59:18 F16H 59:42 F16H 59:44 F16H 59:54 F16H 59:66

Claims (5)

(57) [Claims] 1. A shift control of an automatic transmission for a vehicle, which is connected to an engine and has a manual mode in which a shift speed is switched by a driver's operation and an auto mode in which a shift speed is automatically switched according to a driving state of the vehicle. In the apparatus, learning means for learning a shift pattern reflecting a shift preference of a driver by a neural network using a shift state in at least one of the manual mode and the auto mode as a parameter, and an output from the learning means and the auto mode. of the target gear position setting means for setting a target shift speed based on the driving state of the vehicle, and a shift control means for shifting control of the automatic transmission on the basis of the target gear, said learning means, by the driver Manual mode
Depending on the frequency of use, the manual mode and the auto
Select one of the gears in one of the
Shift control apparatus for a vehicular automatic transmission, characterized in that the over data. 2. An engine speed detecting means for detecting an engine speed of the engine, wherein the learning means sets a shift state in the manual mode as a parameter when a frequency of use of the manual mode is larger than a predetermined value. , and performs the shift pattern learning the engine speed during gear shift operation of at least the driver parameters, the shift control system for an automatic transmission for a vehicle according to claim 1, wherein. 3. The vehicle according to claim 1, further comprising a flat road detecting unit configured to detect whether the vehicle is traveling on a flat road, wherein the learning unit performs the shift pattern learning when the vehicle is traveling on the flat road. The shift control device for an automatic transmission for a vehicle according to claim 1 or 2, wherein 4. The shift means according to claim 1, wherein the learning means performs a shift pattern learning that reflects a shift preference of the driver at the time of shift-up by the neural network using a shift state in at least one of the manual mode and the automatic mode as a parameter. The shift control device for an automatic transmission for a vehicle according to any one of claims 1 to 3 , wherein 5. An engine connected to an engine and operated by a driver.
Mode for changing gears and vehicle operating conditions
Automatic mode that automatically switches gears according to
In the shift control device for an automatic transmission for a vehicle, the manual mode and the automatic mode are at least used.
Neural with the gearshift state on the other hand as a parameter
A gearshift that reflects the driver's shifting preferences through a network
Learning means for performing turn learning, an output from the learning means and a vehicle in the auto mode
Target gear setting that sets the target gear based on the operating conditions of the
Control means for controlling a shift of the automatic transmission based on the target shift speed.
And Cormorant shift control unit, a vehicle speed detecting means for detecting a vehicle speed, e Bei a brake operating state detecting means for detecting an operating state of the brake, said learning means, wherein a brake operation state is ON and decreases the vehicle speed when the amount is more than a predetermined amount, shea in until reduction of at least the vehicle speed becomes the predetermined amount
The shift pattern learning is performed using the frequency of use of the manual mode by the driver during the shift to the downshift side as a parameter, and the auto mode by the driver at least until the amount of decrease in the vehicle speed reaches the predetermined amount. said manual mode switch frequency at which to mode performs the shift pattern learning as a parameter, the driver of the shift transmission control car dual automatic transmission that you wherein performing shift pattern learning that reflects the shifting preferences during down from apparatus.