JPH0581791B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a control device for an automatic transmission, and more specifically to a control device for an automatic transmission, and more specifically, by applying fuzzy control theory, The present invention relates to an automatic transmission control device that enables control similar to the judgment and operation of an expert driver.

(Prior Art) Traditionally, manual transmissions have been used for vehicle transmissions, in which the driver considers the surrounding circumstances and decides when to shift depending on the driving condition, and then operates the clutch pedal and shift lever. I was shifting gears. However, since such manual transmission is cumbersome, automatic transmissions have been developed and are now installed in the majority of passenger cars sold. In the case of a control device for such an automatic transmission, a shift valve is provided in the hydraulic circuit, and a throttle pressure proportional to the throttle opening is applied to one end of the valve, and a governor pressure proportional to the vehicle speed is applied to the other end. The system automatically switches gears by supplying/cutting off hydraulic pressure to the gear clutch depending on the pressure ratio between the two. In addition, with the subsequent shift to electronic control, a control device was configured with a microcomputer, which searched the shift map stored in its memory based on the throttle opening and vehicle speed to detect the shift point, and excited/excited the solenoid valve. The gear is switched by de-energizing and driving the shift valve.

Therefore, in conventional automatic transmission control devices, the timing of the shift, which was determined and operated by the driver himself in the case of manual transmissions, is determined uniquely from the throttle opening and vehicle speed, which inevitably causes problems. It was undeniable that a natural shift would occur. For example, when climbing a hill, if the driver returns the throttle opening to the cruise opening as when driving on flat ground, depending on the speed of the vehicle, the driver may shift up, resulting in insufficient driving force and having to step on the accelerator pedal again. This caused an inconvenience in that the driver had to downshift and shift up repeatedly, giving the driver a feeling of being busy. Such inconveniences also occur when towing a camping car or the like, when the weight of the vehicle increases due to loading, or when driving at high altitudes where engine charging efficiency deteriorates.

If we consider why the driver depresses the accelerator pedal to open the throttle valve, it is because he expects the vehicle to follow the driver's request for acceleration by opening the throttle valve. None other than that. In other words, the above-mentioned inconvenience occurs because the control device issues a shift command even though the margin driving force is reduced and the controllability of the vehicle is not sufficiently ensured. Therefore, in order to do so, it is necessary to accurately grasp the driving force and running resistance in the control device, and to confirm that the driving force exceeds the running resistance and that there is a margin of driving force before shifting up. This is due to the fact that it is not done that way.

From this point of view, a technique described in Japanese Patent Application Laid-Open No. 60-143133 has recently been proposed, in which the torque required by the driver is determined from the amount of accelerator pedal depression, and the separately calculated hill-climbing resistance is subtracted. The required acceleration is calculated using Furthermore, a shift diagram corresponding to the detected hill climbing resistance is selected from among a plurality of best fuel economy shift diagrams, and the throttle opening is adjusted to achieve the required acceleration based on the constant acceleration traveling locus data on the shift diagram. The system is configured to control the throttle opening and then search a shift diagram based on the changed throttle opening and vehicle speed to make a shift decision and maintain the acceleration before the change.

(Problem to be Solved by the Invention) However, in the above-mentioned conventional technology, although a shift decision is made taking into account the torque requested by the driver, the shift decision is made based on a preset shift diagram. This method is different from the previous one mentioned above in that it can only respond to a set situation, and in any case, the timing of the shift is determined uniquely from the throttle opening and vehicle speed. It was difficult to escape the same criticism as technology.

If this is a manual transmission vehicle, the driver would be aware that the vehicle is climbing a slope and would avoid inadvertent upshifts. In other words, in a vehicle with a manual transmission, the driver grasps the driving condition of the vehicle including the surrounding conditions, recognizes the driving force that the vehicle is outputting, and also foresees and masters the increase or decrease in the driving force when shifting. The timing of the shift should have been determined by selecting various empirical rules. That is, the above-mentioned disadvantages are due to the fact that in conventional control, human judgment and actions are ignored and are not reflected in control. In other words, in conventional automatic shift control technology, the timing of shifting is basically determined mechanically from the throttle opening and vehicle speed, and the timing of shifting is not determined based on multiple variables of the vehicle driving condition. Therefore, the occurrence of the above-mentioned inconvenience was unavoidable. Further, the above-mentioned circumstances are similar not only to stepped transmissions but also to continuously variable transmissions. That is, the continuously variable transmission is no different from the case of a stepped transmission in that the speed ratio is changed depending on the running state of the vehicle. Therefore, in this specification, the term "shifting" is used to mean both a change in the shift position in a stepped transmission and a change in the gear ratio (speed ratio) in a continuously variable transmission.

Therefore, an object of the present invention is to eliminate the above-mentioned drawbacks of the prior art, and to apply fuzzy control theory to incorporate the shift operation, which was judged and operated by the driver in a manual transmission vehicle, into automatic shift control. ,
Therefore, it is an object of the present invention to provide a control device for an automatic transmission that enables gear change decisions similar to those made by humans.

Furthermore, in such a control device, the operability of the vehicle is predicted based on the running resistance and the driving force after shifting, and this helps in determining the shifting, and the average value of the driving force over a certain period of time is used to calculate the running resistance. It is an object of the present invention to provide a control device for an automatic transmission, which can be configured to use the present invention, thereby calculating running resistance more accurately and making a more accurate shift judgment.

A further object of the present invention is to provide a control device for an automatic transmission that realizes the above-mentioned control not only for a stepped transmission but also for a continuously variable transmission.

(Means for Solving the Problems) In order to achieve the above object, the automatic transmission control device according to the present invention controls engine speed, throttle opening, amount of change in throttle opening, and vehicle speed, as shown in FIG. , a driving state detecting means 1 for detecting the driving state of the engine or vehicle including at least the amount of change in vehicle speed and the current gear ratio; calculating the torque output by the engine from the engine rotational speed, and a predetermined period of time for the calculated value; Running resistance calculation means 2 calculates the driving force of the vehicle by calculating the average value of A driving state change prediction means 3 for predicting a change in the margin driving force through the calculated running resistance, and a fuzzy product that quantifies the driving state to be detected and the driving state change to be predicted by a membership function. a setting means 4 for presetting a plurality of yellow rules; a gear ratio determining means 5 for determining a gear ratio by performing fuzzy inference from the detected operating state, the predicted operating state change, and the fuzzy production rule; The apparatus is configured to include a driving means 6 for driving the transmission mechanism according to the output of the speed ratio determining means.

Note that in the above, the term "shift" is used to include both a change in the shift position in a stepped transmission and a change in the gear ratio (speed ratio) in a continuously variable transmission.

(Function) In a vehicle with a manual transmission, it is possible to incorporate the shift operation judged and operated by the driver into shift control through fuzzy reasoning, making it possible to make shift decisions similar to human decision-making. Since the running resistance is accurately determined from the average value and changes in the driving state including the margin driving force are predicted, frequent changes in the gear ratio can be avoided.

(Embodiments) Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

FIG. 2 is a schematic diagram showing the entire automatic transmission control device according to the present invention. Referring to the figure, reference numeral 10 indicates the main body of the internal combustion engine. An intake passage 12 is connected to the engine body 10, and an air cleaner 14 is attached to the tip side of the intake passage 12. Therefore, the intake air introduced from the air cleaner 14 is
The flow rate is regulated through a throttle valve 16 that operates in conjunction with an accelerator pedal (not shown) on the floor of the driver's seat of the vehicle and reaches the engine body. A fuel injection valve (not shown) is provided at an appropriate position near the combustion chamber (not shown) of the intake passage 12 to supply fuel,
Intake air is mixed with fuel, enters a combustion chamber, is compressed by a piston (not shown), and is then ignited via a spark plug (not shown) to explode and drive the piston. The piston driving force is converted into rotational motion and extracted from the engine output shaft 18.

Transmission 2 is located after the engine body 10.
0 is connected, and the engine output shaft 18 is connected there to a torque converter 22 and its pump impeller 22a. Torque converter 22
The turbine runner 22b is connected to a main shaft (mission input shaft) 24. A countershaft (mission output shaft) 26 is juxtaposed to the main shaft 24, and one shaft is placed between the two shafts.
Speed gear G1, 2nd speed gear G2, 3rd speed gear G3 and 4
A speed gear G4 and a reverse gear GR are provided, and multi-plate hydraulic clutches CL1, CL2, CL3, CL4 are provided correspondingly to each gear (the reverse gear clutch is omitted for simplicity of illustration). . Further, a one-way clutch 28 is attached to the first gear G1. An oil passage 30 connecting a hydraulic power source (not shown) and a tank (not shown) is connected to these hydraulic clutches, and two shift valves 32 and 34, A and B, are inserted in the middle of the oil passage 30. The shift valve changes its position depending on the energized/de-energized state of two electromagnetic solenoids 36 and 38, and controls the supply/discharge of pressure oil to the clutch group. The torque converter 22 is equipped with a lock-up mechanism 40, which directly connects the turbine runner 22b and the engine output shaft 18 in response to a command from a control unit, which will be described later. The countershaft 26 is connected to a rear axle 44 via a differential device 42, and rear wheels 46 are attached to both ends of the countershaft 26. Furthermore, such engine body 1
0, transmission 20, and differential device 42 are mounted on a chassis (not shown), and a frame (not shown) is mounted on the chassis to form a vehicle.

A throttle sensor 50 consisting of a potentiometer or the like is provided near the throttle valve 16 in the intake passage 12 to detect its opening, and a distributor (not shown) near the engine body 10 is provided. A crank angle sensor 52 made of an electromagnetic pickup or the like is provided on the rotating portion of the piston, and detects the crank angle position of the piston and outputs a signal at every predetermined crank angle. Further, a brake switch 54 for detecting depression of the brake pedal is provided near a brake pedal (not shown) installed on the floor of the driver's seat of the vehicle, and a reed switch or the like is installed at an appropriate position on the transmission 20. A vehicle speed sensor 56 is provided to detect the traveling speed of the vehicle. These sensors 50, 52, 54, 56
The output is sent to the speed change control unit 60.
Further, outputs from a range selector switch 62 for detecting the selected position of the range selector and a shift position switch 64 for detecting the shift position (gear stage) are also sent to the control unit.

FIG. 3 is a block diagram showing the details of the speed change control unit 60. As shown in the figure, the output of the throttle sensor 50 is input to the control unit 60, and then first input to the level conversion circuit 68 and then adjusted as appropriate. amplified to the level of microcomputer 70
is input. The microcomputer 70 is
Input port 70a, A/D conversion circuit 70b,
It includes a CPU 70c, a ROM 70d, a RAM 70e, an output port 70f, and a group of registers and counters (both not shown), and the output of the level conversion circuit 68 is connected to its A/D conversion circuit 70b.
It is input into the RAM70 and converted to a digital value.
It is temporarily stored in e. Similarly, the output of the crank angle sensor 52, etc. is also waveform-shaped by a waveform shaping circuit 72 within the control unit, and then sent to the input port 70a.
is input into the microcomputer via
It is temporarily stored in RAM70e. The CPU 70c determines a speed change command value as described later based on these measured values and various calculated values calculated from them, and sends it from the output port 70f to the first output circuit 74 and/or the second output circuit 76, and solenoid 3
6 and 38 are energized/de-energized to switch gears or hold the current gear. In addition, when changing the gear stage, for example, both solenoids are de-energized (off).
When this happens, the 4th gear is engaged. However, this gear shifting operation via an electromagnetic solenoid is well known and is not a feature of the present application, so a detailed explanation will be omitted. do.

Next, the operation of the present control device will be explained with reference to the flow chart shown in FIG. 4 and subsequent figures.

Here, before going into specific explanation, the characteristics of the present control device will be briefly explained.The characteristics of the control device according to the present invention are as follows: The characteristics of the control device according to the present invention apply fuzzy control theory to determine the shift point in a manner similar to human decision making. The point is that it is structured in such a way that it is determined. That is, the feature of the control device according to the present invention lies not in the configuration of the device itself but in the operation of the control device, that is, the control method. Since the fuzzy control theory itself has recently been applied in various fields, its detailed explanation will be omitted, but simply put, it vaguely grasps the state recognition of the controlled object, and based on that state recognition. Control rules that determine control values (referred to as "production rules")
itself is expressed in language in the form of ``if...then...'', and in the production rules, the criteria for judging the situation or the contents of the operation are treated as ambiguous quantities, and are quantified by membership functions. It is something that exists. In other words, although it uses ambiguous information that humans perform, flexible and highly adaptable control actions are modeled using fuzzy theory, and control values are calculated using fuzzy inference. Since it is easy to express the knowledge possessed by humans, it is easy to adapt to so-called expert systems that incorporate the knowledge and judgment of experts into a computer system. This control device is based on such a theory.

Therefore, when designing the control system for automatic transmissions, we need to create fuzzy production rules necessary for introducing fuzzy control theory when designing the control system for automatic transmissions. The control value is determined, and specifically, it is performed as follows.

(1) Creation of pull-action rules As described below, create an appropriate number of rules expressed in language, such as ``When the engine speed reaches extremely high speeds, shift up to 1st gear to protect the engine.'' When creating these rules, we analyze the judgments and operations of expert drivers in manual transmission vehicles, and then select the rules of thumb that are derived from the analysis.

(2) Determination of parameters and membership functions At the same time, determine the parameters from which the state of the controlled object will be recognized, select the parameters (variables) to be used for each of the above-mentioned pull-down rules, and further determine the parameters. The evaluation criteria are determined by determining the membership function of (the state expressed by such a membership function is called a fuzzy label). As this parameter, this control device uses a physical quantity consisting of an actual value detected through a sensor and a calculated value (including estimated value and predicted value) obtained by differentiating the actual value.
Specifically, engine speed, throttle opening, vehicle speed,
Throttle change amount, acceleration, etc. are used as parameters, and as shown in Fig. 25, on the coordinate system, the parameters are given an appropriate waveform (the membership function) along the horizontal axis (hereinafter referred to as the "domain"), and the vertical axis is Values from "0" to "1.0" (referred to as "membership values (grades)") are assigned to the axis.

The above is the preparatory work during vehicle design. In addition, in the preparation stage, the selection of a sensor for detecting the determined parameters, the storage of control rules, etc. in the memory of the microcomputer of the control unit, or the storage of instructions for calculation procedures, etc. are performed.

(3) Control during actual running During running, the CPU 70c in the microcomputer detects (calculates) parameters.
Then, after referring to the control rules and performing fuzzy reasoning to select one of the control rules and determining the control result, for example, 1st gear up, based on the control rules, predetermined electromagnetic solenoids 36 and 38 are energized/de-energized. This will engage the first gear. In addition, in this fuzzy inference, membership values are calculated for parameters related to each control rule,
The minimum value is taken as the evaluation value of that control rule, and the control rule with the maximum evaluation value is selected from all the control rules. Such mini-max operations themselves are often used in fuzzy inference.

Next, the operation of this control device will be explained with reference to the flow chart of FIG. Note that this program is activated at an appropriate timing of, for example, 10 ms to 40 ms.

FIG. 4 is a flow chart showing the main routine of speed change control. First, in S10, the values detected by the sensor group when the program is started this time are read and temporarily stored in the RAM. Detected values include engine speed Ne (rpm) (calculated by integrating the output of the crank angle sensor 52 mentioned above for a predetermined time), vehicle speed V (Km/h), and throttle opening θTH.
(degrees), current shift position (current gear) signal
So bar (calculated from the ratio of transmission input shaft rotation speed to output shaft rotation speed, or engine speed, throttle opening, vehicle speed, etc.), elapsed time after shift
tSFT(s) (This is determined by measuring the time with a timer counter of the microcomputer, not the sensor output. Specifically, when a shift command is issued in the microcomputer, the bit of the flag register is turned on as appropriate. ) and the on/off signal BKe-ON/OFF of the brake switch 54
A range position signal P RANGE is also used.

Subsequently, in S12, it is confirmed that the range selector is in the D range, and then in S14, it is determined whether or not a gear shift operation is currently being performed. This determination is performed by referring to the shift command flag described above.
If it is confirmed in S14 that the shift is not in progress, the process proceeds to S16 and a shift command value is determined. This will be discussed later. Incidentally, if the results in S12 and S14 are negative or affirmative, this program is immediately terminated.

FIG. 5 is a flow chart showing a subroutine for determining a shift command value. To explain according to the figure, first, in S100, the vehicle speed V and the throttle opening θTH are read out from the sensor output values detected when the program was started last time, and the acceleration α (Km/
h/s) (vehicle speed deviation) and throttle change amount ΔθTH (degrees/S). That is, as shown in FIG. 6, when the program starts this time (time n)
The deviation (first-order differential value divided by unit time n-(n-1)) between the value and the value at the previous program startup (time n-1) is calculated. In the actual calculation, the acceleration is calculated as "Km/h/0.1s" and the throttle change amount is calculated as "degrees/0.1s".

Next, in S102, the output desired by the driver is estimated from the throttle opening θTH at the current time n, and the ratio between it and the force actually output by the vehicle (hereinafter referred to as "PS ratio") is calculated. . Furthermore, this
Horsepower (PS), driving force (Kgf), etc. are used as units for the PS ratio and calculation parameters described below, but in addition, torque (Kgf・m), acceleration (Km/
h/s) may also be used.

7 to 9 are subroutine flow charts showing the calculation of this PS ratio. To explain according to the figure, first, in S200, the throttle opening θTHn at the current time is calculated from the table stored in the ROM 70d. The value is searched to determine the degree of horsepower utilization (hereinafter referred to as "PS%") desired by the driver. Fig. 8 is an explanatory diagram showing this table value, and as shown in the figure, the output characteristic proportional to the throttle opening θTH shown on the horizontal axis has been determined through experiments and stored in advance, and from this characteristic diagram For example, if the throttle is opened to WOT, the driver wants the maximum horsepower that the engine can generate at that point, and if the throttle opening is θTH - α, the driver wants to use the horsepower that is α% of the engine's maximum horsepower. You can understand what you want.

Next, in S202, the map in ROM70d is searched based on the throttle opening θTHn and engine speed Ne at the current time, and the horsepower actually output by the vehicle is determined.
Calculate PSD. FIG. 9 is an explanatory diagram showing this output map stored in the ROM. Needless to say, this must also be determined in advance through experiments.

Next, in S204, add the maximum horsepower (the maximum horsepower that the vehicle can output) to the PS% calculated in S200.
The actual horsepower PSD obtained in the previous step is divided by the product to obtain the PS ratio described above. In other words, PS ratio = actual horsepower retrieved from the map / horsepower desired by the driver, and from now on it is possible to understand the ratio of the horsepower the vehicle is actually outputting to the horsepower desired by the driver. I can do it. Therefore, the PS ratio is “1”
If it is close to or larger than , the horsepower desired by the driver is sufficiently satisfied, and in other words, the driver is considered to be highly motivated to shift up and reduce the horsepower. If the PS ratio is less than "1", it can be determined that the horsepower desired by the driver is not being obtained, and that the driver has low motivation to shift up. Therefore, this PS ratio can be used as an index for upshifting.

Returning to FIG. 5 again, in S104,
The horsepower change expected by the driver is calculated from the throttle change amount ΔθTH, and the ratio of this to the horsepower change actually output by the vehicle (hereinafter referred to as "expected PS ratio") is calculated from the throttle change amount ΔθTH.
(referred to as “EPSRTO”). As mentioned below,
This expected PS ratio determines the motivation for downshifting.

Fig. 10 is a subroutine flow chart showing the calculation procedure for this expected PS ratio. To explain it according to the figure, first, in S300, it is determined whether or not the throttle change amount ΔθTH is a negative value. If so, it means that the throttle valve has been returned, so proceed to S302 and set the expected PS ratio to zero.
In other words, as will be explained later, this expected PS ratio determines whether or not to downshift, so when the throttle opening is decreasing, the driver's acceleration request (intention to downshift) is not seen. be.

If it is confirmed in S300 that the throttle valve has not returned to its original position, the process moves to S304, and the throttle opening θTHn-1 at the previous detection (time n-1) is
The map stored in the ROM is searched based on the amount of throttle change ΔθTH that occurred between the previous detection and the current detection, and the amount of horsepower change expected by the driver (hereinafter referred to as "expected PS change amount DEPS") is determined. Calculate. FIG. 11 is an explanatory diagram for explaining such a map, and it goes without saying that this map is also obtained through experiments and stored in advance.

Next, in S306, the actual horsepower change amount (hereinafter referred to as "actual PS change amount DLTPSD") is calculated as follows.

Actual PS change amount = Actual horsepower retrieved from the map (at time n) - Actual horsepower retrieved from the map (at time n-1) The actual horsepower retrieved from this map is calculated from the output map shown in Figure 9 when the throttle is opened. Therefore, in the above formula, the difference between the value retrieved from θTHn and Ne at time n and the value retrieved from θTHn and Ne at time n-1 is calculated using the above formula. By determining this, it is possible to determine the actual horsepower change per unit time between times n-1 and n. Then,
In S308, the table in Fig. 12 (stored in the ROM) is searched from the actual horsepower change amount obtained in the previous step and the constant GARD (set as appropriate), and the correction coefficient kPS is calculated.
seek.

Next, in step 310, the expected PS ratio
Find EPSRTO as follows.

Expected PS ratio = (kps x expected horsepower change) / (actual horsepower change + GARD) In the above formula, kps and GARD are provided for calculation convenience, and the horsepower change is zero in the low rotation range. This is used to eliminate such inconveniences.

As mentioned above, this expected PS ratio indicates the ratio of the change in horsepower expected by the driver to the change in horsepower actually output by the vehicle, and the driver's motivation to downshift is determined from this value. I can do it. That is, expected PS ratio<1...motivation for downshifting is low expected PS ratio≧1...motivation for downshifting is determined to be high. In other words, if it is greater than 1, the driver's expectations are greater and the vehicle cannot meet them, so it is necessary to downshift to increase the driving force, and if it is less than 1, the driver's expectations cannot be met. This is because there is no need to downshift. In addition, PS, which is the shift-up judgment index mentioned above,
Rather than using the ratio to judge downshifts, we introduced a new concept called expected PS ratio and used it as an indicator for downshifting.The PS ratio is calculated from the throttle opening, whereas the expected PS ratio is calculated from the amount of throttle change. This is because it is done. That is, the amount of throttle change is considered to be more appropriate for estimating the motivation for downshifting, which is intended to increase output.

Returning to FIG. 5 again, in S106, the engine speed after shifting for all shift positions (gears) that can be increased or decreased from the current shift position (hereinafter referred to as "post-shift rotation speed")
seek.

FIG. 13 shows the calculation procedure, and will be explained according to the same figure. First, in S400, the value of a counter SFT1 that sequentially indicates shift positions that can be changed is initialized (initial value "1"). In other words, this post-shift rotation speed is not for a specific gear, but for all gears other than the current shift position So bar, specifically for the 4th forward gear, so it is calculated separately for the remaining 3rd gear from the remaining gear. Therefore, in order to use this counter to display the gear stage being calculated, the count value is initialized in this step SFT1=
1 (that is, the shift destination is set to 1st speed without hesitation).

Next, in S402, the first speed (counter value S
FT1) and the current shift position So bar,
Calculate the maximum number of gears that can be downshifted, CHMIN. As shown in the calculation example in FIG. 14, this is the number of gears that can be lowered by two gears if the vehicle is currently in third gear, for example.

Next, in S404, it is determined whether or not the current gear is in first gear, and if it is not in first gear, the process proceeds to S406 to calculate the post-shift rotation speed in first gear assuming that the gear has been shifted to first gear. This is calculated as follows: Post-shift rotation speed = 1st gear total reduction ratio GR/current gear total reduction ratio GR [rpm]. Incidentally, such total reduction ratio is stored in advance in the ROM as data for each gear stage.

Next, in S408, 1st speed (counter value)
The number of gears is calculated by calculating the difference between the current gear and the current gear.
The post-shift rotation speed calculated in S410 is stored in the column of the gear in RAM. In this case, as shown in FIG. 14, the value of the gear on the down side is stored as CnDNE, and that of the gear on the up side is stored as CnUNE (n: gear. Therefore, in this case n=
1).

Next, in S412, the counter value S FT1 is
4", that is, whether the fourth gear has been reached. 1st
In the case of start-up, the rotation speed after shifting is calculated from 1st gear, so it will naturally not reach the desired speed, so the counter value is incremented in S414, and the same procedure is followed for 2nd gear and higher unless it matches the current gear. Calculate the final step after confirming that 4th gear has been reached.
In S416, the difference between the fourth gear and the current gear is calculated to determine the maximum number of gears CHMAX that can be increased, and the process ends.

Returning to the flow chart in Figure 5 again,
In S108, the ratio between the horsepower change expected by the driver and the expected horsepower change of the actual vehicle after downshifting (hereinafter referred to as "post-shift expected PS ratio CnDPSR") is calculated. In other words, in this control device, the downshift is performed by estimating the horsepower change expected by the driver from the throttle operation performed by the driver, and comparing it with the horsepower change actually output by the vehicle. Whether or not to downshift is determined based on whether or not the expected change has been realized, and this comparison corresponds to the above-mentioned expected PS ratio.
As a result, if it is determined that it is necessary to downshift, the expected post-shift PS ratio that will be calculated from now on will be used as an index to determine which gear (shift position) to shift down to. This expected post-shift PS ratio indicates which gear should be lowered to in order to achieve the change in horsepower that the driver expects.

Next, regarding shift up, we estimate the horsepower expected by the driver from the current throttle opening, and compare it with the horsepower output by the actual vehicle (PS ratio mentioned above) to determine shift up. This is confirmed through the concept of control toughness, which is established as a coefficient that indicates the appropriateness of the vehicle's response to throttle changes, in order to avoid excessively reducing surplus horsepower and losing vehicle operability due to unreasonable upshifts. be. This control toughness will be described later. In this control device, these various indicators are included in the parameters, and the degree of satisfaction of the fuzzy production rule is determined through fuzzy inference to determine the control command value.

The expected PS ratio after shift will be explained with reference to FIG.

First, in S500, as with the expected PS ratio mentioned above, after confirming that the throttle valve is not in the closing direction and that the driver is not in a state where there is no intention to downshift, in S502 Initialize the value of the counter that displays the number of gears STEP obtained in S408 of the flow chart (initial value "-1"). This initial value means the case where it is assumed that the gear is down by one speed.

Next, in S504, it is determined whether the initial value, ie, the first speed, exceeds the maximum number of downshiftable gears CHMIN similarly determined in S402 of the previous flow chart. For example, if the current gear is 1st gear and it is not possible to lower the gear by one gear, the calculation will end immediately, and if it is possible to lower the gear without exceeding the gear, proceed to S506 and calculate the post-shift rotation speed and the current speed. Figure 9 shows the throttle opening of
Calculate the PS map and calculate the horsepower CPS that the vehicle will output if it is assumed that the vehicle is down by one gear. In this case, the post-shift rotational speed uses the one-speed down value C1DNE among the down-side values in the data stored in S410 of the flow chart of FIG. 13 above.

Next, in S508, calculate the current horsepower PSD from the expected horsepower CPS (calculated using the flow chart in Figure 7)
Calculate the horsepower increment CDELTA due to the shift by subtracting
CnDPSR (n: the number of downs) is calculated as follows.

Expected PS ratio after shift = expected PS change amount / (horsepower increment due to shift + GARD) Here, the expected PS change amount uses the change amount DEPS calculated in FIG. 10. Also, GARD is a constant to prevent division by zero.

Next, in S512, the value of the gear stage number counter is decremented, and the above operation is repeated until it is determined in S504 that the maximum value that can be lowered has been reached. Note that when it is determined in S500 that the valve is closed, the expected post-shift PS ratio is set to zero in S514 and the process ends.

Returning to the flow chart in Figure 5 again,
In S110, the control toughness described above is calculated. FIG. 16 is a flow chart showing this calculation subroutine.

Here, before going into a specific explanation of the flow chart, a brief explanation of control toughness will be given with reference to FIG. 17. This is a term coined by the inventors, A coefficient that represents the appropriateness of a vehicle's response to changes in As described above, this concept was devised with one purpose of eliminating the busy feeling caused by frequently repeated shifts when climbing a hill or towing a camper car. In other words, the above-mentioned inconvenience arises from not securing sufficient margin driving force, which is obtained by subtracting running resistance, which is an extrinsic load of the vehicle, from the driving force. This becomes noticeable during upshifts when the amount decreases. This point will be explained with reference to FIG. 17. Assuming that the vehicle is currently running at the engine speed Ne0, the amount equivalent to the surplus horsepower shown as the difference from the full-open driving force is shown as shown in the figure. In this case, running resistance increases when running uphill compared to when running on a flat road because gradient resistance is added. When the throttle opening is returned to the cruise opening in this state, the shift point is uniquely determined from the vehicle speed and the throttle opening in conventional control devices, so the shift up is automatically performed. Since the engine speed drops to Ne1 and the full-open driving force (after shifting) value also drops, the equivalent horsepower after the shift also decreases as shown in the diagram, and as a result, downshifting is performed again. . In other words, in this case, the vehicle is in a state in which it is unable to respond appropriately to the driver's request due to the large running resistance corresponding to the surplus horsepower after the shift, and such a state should be taken into consideration when making a shift decision. If this is possible, it should be possible to avoid meaningless shift-ups.
Therefore, in this control system, the appropriateness of the vehicle's response is expressed by the concept of control toughness, which is determined by the current running resistance against the driving force after the shift, and the decision to shift up is made taking this concept into account. It was decided to. More precisely, as mentioned above, when determining shift up, the driver's expected horsepower and actual horsepower are compared based on the PS ratio to determine when to shift up, and at the same time, the control toughness is used to determine the vehicle's operability when shifting up. The final decision will be made as to whether or not it should be uploaded. The calculation of this control toughness will be explained below.

First, in S600, the current torque TE is calculated as follows.

Current torque = (716.2 x actual horsepower) / engine speed
[Kgf・m] As is well known, 716.2 is a constant for horsepower-torque conversion.

Next, in S602, the torque ratio map is searched to calculate the torque ratio TR. That is, in the automatic transmission, the transmission input torque is amplified via the torque converter 22, so the degree of amplification is calculated to correct the torque. FIG. 18 is an explanatory diagram showing this torque ratio map (stored in ROM), where the horizontal axis shows the speed ratio and the vertical axis shows the corresponding torque ratio. The speed ratio is the rotation ratio between the main shaft 24 and the countershaft 26 of the transmission, and these are specifically substituted by the engine rotational speed and vehicle speed. Calculated torque ratio
TR is then multiplied by the torque TE calculated in S600 in S604 to obtain a corrected torque T0.

Subsequently, in S606, the corrected torque values calculated in this manner are averaged by going back an appropriate period. That is,
Since there is a slight time delay before throttle changes are reflected in the engine output, it is possible to calculate the engine output more accurately by understanding and averaging the engine output based on the impulse of a given engine. It is. 1st
Figure 9 is an explanatory diagram showing this averaging work, in which the torques are summed up from the current time n (current control cycle) to a predetermined cycle section n-M, and then divided by the total number of cycles to calculate the average. Calculate the value.

Next, in S608, brake switch 54
After confirming from the detection signal that the brake is not depressed, the brake timer is decremented in S610. This is the same as applying a load or running resistance to the vehicle when the brakes are in operation, and since control toughness is calculated from the ratio of driving force and running resistance, running resistance is This is because it is difficult to ensure the accuracy of the calculation. Therefore, when it is determined that the brake is in operation,
Control toughness R1/Q1 is 1.0 in S612
As a result, the rules are configured so that the shift-up command is not issued in fuzzy inference. In this case, R1 means the current running resistance, and Q1 means the full-throttle driving force at that gear assuming that the gear has been shifted (in addition, since the running resistance does not change before and after the shift, R1 means the (You can also call it running resistance). In addition, this flow chart is designed to avoid calculating control toughness not only during braking, but also for a certain period of time after the braking is completed and the brake is returned, in order to further improve the accuracy of the calculation. There is. To this end, if it is determined in S608 that the brake pedal has been depressed, a brake timer (built in the microcomputer) is started in S614, and each time the completion of the brake operation is confirmed in S608, a count value is set in S610. decrements and
If it is detected in S608 that the brake has been operated again during that time, the count value is reset in S614.

If it is confirmed in S616 that the brake timer value has reached zero, then in S618 the vehicle speed V is set to a predetermined lower limit value VMINCT, for example 2 km/h.
Determine whether or not it exceeds. This is because, in the case of such a low vehicle speed, a gear shifting operation is not required in any case, and in this case, the control toughness is set to 1.0 in S620 and the program is terminated.

If S618 determines that the vehicle speed is above the predetermined value,
Next, in S622, the throttle change amount ΔθTH is set to the predetermined valve opening speed ΔθTH−OPEN as shown in FIG.
Determine whether it exceeds or not, and if it does not exceed
Similarly, in S624, the predetermined valve closing speed ΔθTH−
Determine whether it exceeds CLOSE. In other words, when the throttle suddenly changes, it indicates a sudden transient state, but in a sudden transient state, especially when accelerating rapidly, the fuel increased by opening the throttle as described above is distributed to each cylinder through the intake manifold, and the engine Since there is a predetermined time delay before the output increases, the calculation of running resistance R0 is stopped at such sudden throttle change, and the calculation for the following predetermined time is also stopped. Specifically, S622 or
When a sudden change in the throttle is detected in S624
Move to S626 and reset the throttle timer/
At the same time as the start, the timer value is decremented in S628 each time it is detected in S624 that the throttle rapid change operation has ended.

Subsequently, after it is confirmed in S630 that the timer value has reached zero, the current running resistance R0 is calculated as follows in S632.

Running resistance R0 = [(average torque TRQ x transmission efficiency η x total reduction ratio GR of current stage) / (tire effective radius r)] - [(1 + equivalent mass coefficient) x (vehicle body mass M x acceleration α)] [Kgf ]...(1) In addition, the transmission efficiency η, total reduction ratio GR, tire effective radius r, equivalent mass coefficient, and vehicle body mass M (ideal value) are obtained in advance and stored in the ROM, and the torque TRQ The value calculated in S606 is used, and the value calculated in S100 of the flow chart of FIG. 5 is used for acceleration α.

Here, to explain why running resistance is calculated as in the above formula, the power performance of a vehicle is determined from the equation of motion as follows: Driving force F - Running resistance R = (1 + equivalent mass coefficient)
× (vehicle weight Wr/gravitational acceleration G) × acceleration α
[Kgf]...(2) Here, F = (Torque TRQ x Gear ratio GR x Efficiency η) / Tire effective radius r [Kgf] R = (Rolling resistance μ0 + Gradient sin θ) x Vehicle weight Wr +
Air resistance (μA×V ² ) [Kgf] In the above equation, what changes depending on the driving condition is the vehicle weight, which changes depending on the number of passengers and the amount of cargo loaded.
The slope sinθ varies depending on Wr and the driving road surface,
All of these are included in running resistance R.
Therefore, by modifying the above formula (2), running resistance R = driving force F - [(1 + equivalent mass coefficient)
x vehicle body mass M x acceleration α] [Kgf]. Equation (1) is based on this.

Next, after confirming that the acceleration α is not a negative value in S633, the acceleration guarantee rate map (h
MAP) to calculate the acceleration guarantee rate,
In S636, the aforementioned running resistance R0 is corrected to calculate the corrected resistance R1 as described below. In addition, Fig. 16 Flow・
In the chart, when it is determined that the throttle is suddenly changing, the value of running resistance R0 is the previously calculated value R0n−
1 is used (S638). Further, if the acceleration is in the negative direction, no correction is made (S633).

Corrected running resistance R1 = R0 + (acceleration guarantee rate h x (1
+ equivalent mass coefficient) x vehicle mass M x acceleration α)
×SIGN (R0) [Kgf] Note that SIGN (R0) means that the sign (R0) is the same as the above-mentioned R0.

To explain this acceleration correction, Fig. 21 shows an acceleration guarantee rate map, in which the horizontal axis represents the acceleration α, and for example, the guarantee rate (correction coefficient) shown on the vertical axis increases as the acceleration increases. Therefore, it is set so that it decreases. 22 on this point
To explain with reference to the figure, assume that when the vehicle speed V is in the state shown in the figure, a shift-up decision is made at time tn. Now, let's assume that the control toughness in the shift up judgment is given by R0/Q1. In this case, R0 lacks the driving force necessary to maintain the acceleration state, so
The index of control toughness is not appropriate as an index because it represents the surplus horsepower when it is sufficient to maintain the current vehicle speed. Conversely, the entire driving force necessary to maintain the acceleration state is contained in R0.
If R1 is used in addition to R0, and control toughness is considered by R1/Q1, and considering that a reduction in driving force will always occur due to a decrease in gear ratio or engine speed due to a shift up, R1
> Q1, and there was almost no upshifting, which also did not match our feelings. Naturally, people expect that acceleration will be impaired by upshifting, and it is necessary to express that person's expectations in some way and make corrections. Therefore, with this configuration, even during acceleration, as long as the acceleration before the shift can be maintained, the shift-up can be performed effectively to ensure smooth driving, and the acceleration can suddenly change after the shift-up, causing the driver to feel uncomfortable. There are no inconveniences that I can remember.

Subsequently, in S640, the value of the gear stage number counter is initialized, and in S644 and thereafter, the post-shift full-open driving force Q1 is calculated for each possible gear stage until it is determined that the shift-up upper limit number has been reached in S642. To explain, first, in S644, assuming that the counter value STEP=1, that is, the first gear has been shifted up, the maximum horsepower CPSMAX at that gear is searched. This is calculated by searching the output map shown in Figure 9 from the post-shift rotational speed C1UNE calculated in the flow chart of Figure 13 and the fully open throttle opening value.

Next, in S646, horsepower-driving force conversion is performed to calculate full-open driving force Q1 as follows.

Full-open driving force Q1 = (716.2 x full-open horsepower after shift
CPSMAX × Total reduction ratio after shift GR × Gear transmission efficiency after shift η) / (Revolutions after shift
CnUNE × Tire effective radius) [Kgf] Next, in S648, calculate the control toughness C1UCT when moving up to 1st gear by dividing running resistance R1 by full-throttle driving force Q1, then increment the counter value in S650, and in S642 The control toughness C2UCT and C3UCT of 2nd gear up and 3rd gear up are calculated until it is determined that the upper limit value has been reached. As mentioned above, control toughness indicates how much of the running resistance accounts for the maximum driving force obtained when shifting by n speeds, that is, it indicates the surplus horsepower after shifting. Therefore, in this sense, it can also be regarded as a coefficient indicating how appropriately the vehicle can react to the driver's intention to shift gears as indicated by throttle changes.

FIG. 23 is an explanatory diagram showing a case where such control toughness is defined by a membership function. In other words, when R1/Q1 is close to 1 or greater than 1, there is no extra driving force, and therefore, if you shift up, you will lack horsepower, so the evaluation value (grade)
μ also becomes lower. On the other hand, when the value is negative, Mα is large, which means that the vehicle is in a downhill state, and similarly, the controllability of the vehicle is low, so the evaluation value is also low. Therefore, in the case of the example, even if the gear is shifted up by a predetermined range of about 0.2 to 0.5, there is still some margin in the driving force. As will be described later, in this control device, a shift command value is determined by evaluating the suitability of a shift rule, for example, if the control toughness is good, shift up by one gear, etc., through fuzzy inference regarding the control toughness. The configurations of FIG. 16, FIG. 4, and FIG. 5 correspond to claim 1.

Returning to FIG. 5 again, after the control toughness is calculated in S110, the shift position is determined in accordance with the fuzzy production rule in S112.

FIG. 24 is a flow chart showing the main routine of this rule search. Before entering into the explanation of this figure, the rules used in this control device will be briefly explained with reference to FIG. 25. It should be noted that, as described above, these rules and parameters used or their fuzzy labels are set at the time of designing the control system of the vehicle. In this embodiment, 20 rules are used as shown in the figure.

Rule 1 Parameters used...Engine speed Ne [rpm. The same applies hereafter] Conclusion... 1st gear up rule Implications... ``When the engine speed reaches extremely high speeds, 1st gear is increased to protect the engine.'' This is a rule for engine protection, and in the red zone where the engine speed exceeds 6000 rpm. This means that if the engine enters the engine or there is a risk of it entering, shift up and lower the rotation speed to protect it. In this rule, "1st gear up" means to shift up by one gear, for example, if you are currently in 2nd gear, shift up to 3rd gear, but does not mean to shift up to 1st gear.

Rule 2 Parameters used...Current shift position So bar Vehicle speed V [Km/h. Same hereafter] Throttle opening θTH [WOT/8 degrees. same as below. Furthermore, WOT=84
] Conclusion... Implications of the downshift rule for 1st gear... "If the vehicle is fully closed and the vehicle is at an extremely low speed, if the current shift position is 4th gear, shift down to 1st gear." From this rule to Rule 4 This rule defines the initial shift operation that commands a downshift to 1st gear when the throttle is fully closed and the vehicle speed is extremely low. When this rule is set to 4th gear, Rule 3 is set to 3rd gear. When in speed and Rule 4
is scheduled to be in second gear.

The evaluation of such rules using fuzzy reasoning will be explained in detail with reference to Fig. 24, but here I will briefly explain that if the current shift position is
Assuming that the vehicle speed is 10 km/h and the throttle opening is 1/8, the grades for each fuzzy label in Rule 2 are: Current shift position = 0 (score is 0 because it does not intersect with the waveform), Vehicle speed = 0.95 , throttle opening = 0.95. In this case, three fuzzy labels are involved, and each score is different, but since the minimum evaluation value satisfies all related matters at least within that range, the minimum evaluation value, in the case of the example is the evaluation value of the shift position 0
is the evaluation value of rule 2. The 20 rules are evaluated in sequence, and the rule with the highest evaluation value is selected as the one with the highest degree of satisfaction, and the shift command value is determined based on that rule. As for the actual example, when evaluating rule 3, the grade is: current shift position = 0, vehicle speed = 0.95,
Throttle opening = 0.95, and the evaluation value of rule 2 is also 0. Similarly, regarding Rule 4, the current shift position = 0.95, vehicle speed = 0.95, throttle opening = 0.95, and 0.95 is the evaluation value.
Therefore, if we ignore the existence of other rules,
According to Rule 4, the vehicle will shift from second gear to first gear. In this case, it goes without saying that Rule 4 was selected among similar Rules 2 to 4.
This is because the current driving state is closest to the downshift from second gear to first gear as scheduled by Rule 4. In this embodiment, the maximum value of the membership function is made different depending on the rules. That is, rule 1 has a maximum value of 1.0, rules 2 to 6 have a maximum value of 0.95, and rules 7 and onwards have a maximum value of 0.9. The reason for this will be explained later.

Continuing the explanation of the rules below, Rule 5 Parameters used... Current shift position So bar Vehicle speed V Throttle opening θTH Conclusion... Implications of the downshift rule to 2nd gear... ``If the vehicle is fully closed and at low vehicle speed, If the current shift position is 4th gear, shift down to 2nd gear.'' This is a rule similar to rules 2 to 4, and even if the vehicle speed is not that low, if it is still slow, shift down to 2nd gear. It has been decided that there will be a shift.
Note that Rule 6 has the same effect except that the current shift position is scheduled to be the third gear.

Rule 7 Parameters used...Engine speed Ne Acceleration α [Km/h/0.1s. The same applies hereafter] Throttle change amount ΔθTH [degrees 0.1s. The same applies hereafter] Control toughness R1/Q1 PS ratio conclusion... Implications of the 1st gear up rule... "Shifting up at a constant throttle during acceleration is done if the PS ratio approaches 1 and the control toughness is good." This rule applies during acceleration. In other words, if the engine is accelerating, the engine speed should be relatively high, the acceleration should be increasing, and the throttle should be open (not returned).As mentioned above, Shift up is determined from the PS ratio and control toughness, so if these are in a satisfactory state, it is possible to shift up by one gear even during acceleration.

Rule 8 Parameters used...Current shift position So bar Expected PS ratio Conclusion...Do not shift Implications of the rule..."If the throttle suddenly returns to full closure, hold the shift" This means that 4th gear This means that when the expected PS ratio (a measure of downshift motivation) is small while driving, the gear will not be shifted.

Rule 9 Parameters used... Acceleration α Throttle change ΔθTH Control toughness R1/Q1 Engine speed Ne Conclusion... Implications of the 1st gear up rule... "The shift up during slow acceleration is
If the engine speed is not low and the control toughness is good, do it.'' If the acceleration is gradual, the acceleration α cannot be used as an indicator, so the shift up should be determined based on the requirement that the engine speed is relatively high. I will do it. Since it is an upshift, it naturally requires good control toughness. The reason why the PS ratio is not judged is because it is considered appropriate that the PS ratio can be used as an index only when the vehicle acceleration is above a certain level.

Rule 10 Parameters used...Elapsed time after shift [s] Throttle change amount ΔθTH Conclusion...Do not shift Implications of the rule..."If the throttle does not move immediately after a shift change, the gear will not shift." If the throttle valve is not depressed greatly, it is assumed that the driver has no intention of shifting, and a dead zone of a predetermined time, for example, about 1.6 to 2.5 seconds, is provided.

Rule 11 Parameter used... Expected PS ratio throttle change ΔθT Conclusion... Implications of the rule without shifting... "If the expected PS ratio is small even when the throttle is depressed (when the car follows the throttle movement) does not change gears. Regarding downshifts, we aim to motivate downshifts based on the expected PS ratio, and also
The destination stage is determined from the PS ratio, but the expected
A small PS ratio means that the change in horsepower of the actual vehicle is greater than the change in horsepower expected by the driver, so there is no need to increase horsepower by downgrading.
Therefore, there is no need to change gears.

Rule 12 Parameters used: Control toughness, number of revolutions after shifting [rpm]. Same hereafter] PS ratio throttle change amount ΔθTH Conclusion... Implications of the 3rd gear up rule... ``When the throttle returns and cruise is intended, 3rd gear is increased in consideration of both control toughness and fuel efficiency.'' The throttle is on the return side. In some cases, Cruise's intentions can be read. Also, if it is expected that the rotational speed will decrease by shifting, it is a good idea from the viewpoint of fuel efficiency.
Follow. If the PS ratio, which is the ratio between the actual horsepower and the horsepower desired by the driver, is close to 1 or greater than 1, it can be seen that the motivation to shift up is high, so the control toughness after shifting is satisfactory. Upload. Note that Rules 13 and 14 also intend to increase speeds from 2nd gear to 1st gear for the same purpose.

Rules 15-17 Parameters used...Expected PS ratio Expected PS ratio after shift (1st to 3rd gear down value) Revolution speed after shifting (1st to 3rd gear down value) Conclusion...3rd gear (2nd gear, 1st gear) Implications of the down rule... ``If the car does not follow the throttle movement even when the throttle is depressed, it will shift down to 3rd gear (2nd gear, 1st gear) so that the expected PS ratio after shifting is 1.

Rules 15 to 17 are kickdown rules. It is determined that a downshift is necessary because the expected PS ratio, which is the ratio between the horsepower change expected by the driver and the actual vehicle's horsepower change, is large. Therefore, for 1st to 3rd gear down, the change in horsepower of the actual vehicle relative to the change in horsepower expected by the driver after shifting (expected PS after shift)
ratio).

Rule 18 Parameter used...Vehicle speed only conclusion...Implications of the shift hold rule..."When the vehicle speed is extremely low or the vehicle is stationary, wait at the current shift (1st gear)" This means that there is no rule that is adopted when the vehicle is stopped. This rule prevents other rules from being adopted with lower grade values.

Rules 19, 20 Parameters used: Control toughness (when 1st gear up) Conclusion: Implications of the shift hold rule: ``If you can predict that there will be no control toughness as a result of 1st gear up, don't shift.'' This is to compensate for the shift-up rule, and the shift-up rule states that the shift-up is performed when the control toughness is good, so even if the control toughness is not good, if the satisfaction level of other rules is low, the shift-up rule will be changed as a result. This set of rules was created because one of the purposes of the present application, which is to avoid busy shifts, would not be achieved if the system was adopted.

Next, rule search will be explained with reference to the flow chart of FIG. In the figure, first, in S700, the grade value of the membership function is calculated. This is done according to the subroutine shown in FIG. To explain with reference to the figure, first, in S800, data is set for each physical quantity (parameter) No., in S802, the address code No. of the address register is initialized (initial value = 1), and in S804, the CN number is set. Reads the membership value (grade) (DAT) of a value.

The above will be explained with reference to FIGS. 27 to 29.
Data is stored in the ROM as shown in FIG. The data is set for each parameter such as vehicle speed, and the corresponding membership function is defined and stored in a table format with the relevant physical quantity attached to the domain (horizontal axis).
Each one has a physical quantity number and address (code).
No) will be added. The membership function of this physical quantity (parameter) has been explained with reference to the rules shown in FIG. Note that if different membership functions (waveforms) are defined for one physical quantity, special addresses are given. or,
FIG. 28 shows a calculation table prepared in the RAM, and is set to write actually measured or calculated values for each physical quantity. Figure 29 shows
Apply the data in Figure 28 to Figure 27 and code
This is a calculation table that writes the results of calculating the membership value (grade) for each No.
Provided in RAM.

Therefore, in the flow chart of Figure 26,
S800 means writing the measured or calculated data into the calculation table in Figure 28, and S802
sets the value of the address register specifying the address code in Figure 27 to an initial value of 1 (indicates the first column).
S804 calculates (reads) the grade value for each address (code number) by applying the measured value to the membership function table in Figure 27 using the operation table in Figure 28, that is, the first column. The actual measured value of the vehicle speed, for example 120
This refers to the work of applying values such as Km/h and reading grade values such as 0.0. The read data is then converted to the grade value μ of the code in S806.
(CN), and then the code number is incremented in S808, and the process is repeated until it is confirmed in S810 that the grade values have been read for all codes.

Returning to FIG. 24 again, a search matrix is then created in S702. FIG. 30 is a flow chart showing the creation subroutine. That is, the rule group shown in Fig. 25 is actually stored in the ROM in a matrix form as shown in Fig. 31, but by searching it and applying the grade value obtained earlier, the rules shown in Fig. 32 are stored. The purpose of this subroutine is to write to the arithmetic matrix stored in the RAM shown in FIG. This will be explained below.

First, in S900, the total number of rules N is read.
In this example, there are 20 pieces. Next, in S902, the value n of a counter that counts rule numbers is initialized (n=1, meaning rule 1), and in S904, the value 1 of a counter that counts label numbers is similarly initialized (1=1). (means the first label of rule 1).
For example, this label corresponds to the current shift position, vehicle speed, and throttle opening according to Rule 2.
They will be numbered label 1, label 2, label 3, respectively. Next, in S906, the total number of labels QL is read. According to rule 2, there are 3 pieces.
Subsequently, the code number of the corresponding rule is read from the rule matrix shown in FIG. 31 in S910 via S908. According to Rule 2, the code corresponds to the shift position, vehicle speed, and throttle opening.
No. (shown in the table in Figure 27) will be read. Next, in S912, read the previously calculated grade value corresponding to the code No.
In S914, it is written to the calculation matrix shown in FIG. 32, and in S916, the label number is incremented.

When it is confirmed in S908 that all the labels of the rule have been searched, S918
Proceed to step S920 to update the rule number and perform the same operation for the next rule, and confirm that all rules have been completed in S920.

Returning again to the main routine in FIG. 24, the output is determined in the final step S704, but this is done based on the subroutine shown in FIG. 33. This subroutine calculates the fitness of each rule and the label that determines its fitness based on the membership value obtained earlier.
This section shows the work to find No. and the so-called mini-max operation in which the control command value is determined by selecting the rule with the highest degree of compliance.

First, a rule number counter is initialized in S1000, and the conclusion of the first rule is read in S1002. FIG. 34 shows a rule map stored in the ROM, and the conclusion is read by referring to this map. For example, the first rule is 1st gear up (+1).

Next, in S1004 and 1006, it is determined whether the conclusion is executable (for example, if the current shift position is 3rd gear, it is possible to shift up to 1st gear), and it is determined whether or not the upshift and downshift are possible, and then in S1008, a comparison is made. Initialize the starting membership value for (initial value =
1.0), read the total number of labels of the first rule in S1010, initialize the label No counter in S1012,
After going through S1014, the starting value 1.0 is compared with the grade value previously obtained for the first label in S1016, and if the grade value is smaller, it is replaced with the starting value in S1018, and then in S1020, that value is used as the grade for the label. value, increment the label number in S1022 and repeat the same operation, and when it is determined in S1014 that the search for all labels for the rule has been completed, proceed to S1024 and set the minimum value searched as the representative value of the rule. Then, in S1026, the process proceeds to search for the next rule. Note that if the results in S1004 and 1006 are negative, the rule representative value is set to 0 (S1028).

After initializing the rule number counter in S1030, the second starting value for comparison is initialized in S1032 (initial value = 0), and then in S1034, the representative value (minimum value) and the above-mentioned value are initialized from the first rule. Compare it with the second starting value, and if the representative value is larger, proceed to S1036 and replace it with the starting value, and then in S1038 set that rule as the rule with the largest matching value;
In step S1040, the rule is incremented and all rules are searched in the same way. When it is confirmed in S1042 that the search for all rules has been completed, S1044
Let the maximum value among them be the final selection rule compliance value.

Next, in S1046, the selected value is compared with the appropriately set reference value μTH, and if it exceeds it, in S1048, the output shift position SA is determined from the current shift position So bar according to the conclusion of the rule, and it is If not, the rule selected in S1044 is discarded, and the previous control value S An-1 is set in S1050.
Use as is. In other words, the reason for setting this reference value is that in mini-max calculations, rules are selected relatively, so a rule that cannot be said to be suitable for a given driving condition may have lower scores than other rules. This is to avoid the possibility that it may be adopted. FIG. 35 is an explanatory diagram showing a calculation table used in the output determination routine. As described above, in this embodiment, the maximum membership value is different depending on the rule, but such a configuration is also useful for avoiding selection of an inappropriate rule. In other words, by assigning the maximum values in descending order of importance, it is possible to increase the possibility that a rule with a high degree of importance will be selected in the intended driving state, and as a result, the selection of an inappropriate rule can be avoided. can be prevented.

Finally, returning to FIG. 4 again, in accordance with the determined control command value, the electromagnetic solenoids 36, 3
8 is energized/de-energized to drive or hold the transmission. At the same time, the shift command flag is turned on in the microcomputer.

As described above, in this embodiment, not only the actual measured values such as throttle opening or vehicle speed, but also the output amount of the actual vehicle in relation to the driver's expected amount are quantitatively measured and set as parameters, and based on these parameters. The system is configured to set multiple control laws that are derived by analyzing the judgments and operations seen in a manual transmission vehicle by an expert driver, and to select the optimal control value by evaluating the control laws through fuzzy reasoning. This technology captures and instantaneously processes the vehicle driving state in multiple variables, including the surrounding conditions, making it possible to perform automatic gear shift control similar to the judgment and operation of a skilled driver with a manual transmission. In other words, control using the fuzzy method enables more appropriate control similar to manual gear shifting operations by humans, and as seen in the above-mentioned prior art,
There are no inconveniences such as being restricted by setting data or having the timing of the gear change temporarily determined from the throttle opening and vehicle speed. Furthermore, by further increasing the number of disclosed rules, it is possible to realize shift control that is compatible with emission countermeasures, and furthermore, it is possible to respond more flexibly to shift control characteristics desired by users. In this sense, the present invention is completely different from the prior art in purpose, structure, and effect.

Furthermore, it predicts the vehicle's operability based on the running resistance and the driving force after deformation, and helps in gear shifting decisions.
Since the running resistance is calculated using the average value of the engine output during a predetermined control cycle, the running resistance can be calculated accurately and a gear shift judgment can be made accurately.

FIG. 36 and subsequent figures show a second embodiment of the present invention, and show an example in which the shift control using the above-mentioned fuzzy inference is applied to a continuously variable transmission (CVT).

To explain the following, FIG. 36 is an overall schematic diagram illustrating the continuously variable transmission focusing on the hydraulic circuit. In the same figure, a continuously variable transmission 100 includes a constant displacement hydraulic pump 102 having a transmission input shaft 24 driven by an engine main body 10 via an engine output shaft 18.
and a variable displacement hydraulic motor 106 having a drive shaft 104 and disposed on the same axis as the hydraulic pump.
are connected to each other to form a hydraulic closed circuit 108. In the closed circuit, the hydraulic pump 10
The discharge port of the hydraulic motor 106 and the inlet of the hydraulic motor 106 are connected to each other by a high pressure oil passage 108h.
02 are mutually connected by a low pressure oil passage 1081. In the hydraulic closed circuit, a short-circuit path 110 is provided between the discharge port and the suction port of the hydraulic pump 102, that is, between the high-pressure and low-pressure oil paths 108h and 1081, and a short-circuit path 110 is provided in the middle of the short-circuit path 110. provided. Moreover, the hydraulic pump 10
In addition to 2, a replenishment pump 114 is provided,
The replenishment pump is driven by a mission input shaft 24, and its discharge port is connected to check valves 116, 118, 12.
0 through the high pressure and low pressure oil passages 108h, 10
81, and supplies hydraulic oil pumped up from the oil tank 122 to the hydraulic closed circuit. In addition, code 12
4 indicates a relief valve.

The transmission output shaft 26 connected to the rear wheel 46 is arranged parallel to the drive shaft 104 of the hydraulic motor 106, and a forward/reverse switching device 130 is provided therebetween. This forward/reverse switching device includes first and second drive gears 132 and 134 fixed to the drive shaft 104 with an interval in the axial direction, and an output shaft 2.
a first driven gear 136 that is rotatably supported by 6 and meshes with the first drive gear; and an intermediate gear 13
A second driven gear 1 is connected to the second driving gear via 8 and is rotatably supported by the output shaft 26.
40 and a driven clutch gear 1 fixed to the output shaft 26 between the first and second driven gears 138 and 140.
42, and a clutch member 144 selectively connecting the driven clutch gear and both driven gears 136, 140. First and second driven gears 13
Driving clutch gears 136a and 140a are provided at the ends of the driven clutch gears 142 of No. 6 and 140, and the clutch member 144 is located at a position where the driving clutch gears 136a and the driven clutch gears 142 are connected. , and a position connecting the driven clutch gear 142 and the drive clutch gear 140a. In such a forward/reverse switching device 130, as shown in FIG.
When the output shaft 26 is connected to the driven clutch gear 142, the output shaft 26 is rotated in a direction opposite to the rotational direction of the drive shaft 104, and the rear wheel 46 can rotate in the forward direction. Conversely, when the driven clutch gear 142 and the drive clutch gear 140a are connected, the output shaft 2
6 is rotated in the same direction as the drive shaft 104, and the rear wheel 4
6 is rotatable in the reverse direction.

The clutch valve 112 is driven by a servo cylinder 150, the forward/reverse switching device 130 is operated by a range drive mechanism 152 consisting of a hydraulic cylinder, etc., and the displacement of the hydraulic motor 106 is controlled by a hydraulic cylinder 154. It will be done. To explain them individually below, the servo cylinder 150 includes a cylinder 156, a piston 162 which defines the inside of the cylinder into a head chamber 158 and a rod chamber 160, a piston rod 164 integrated with the piston, and a piston inside the rod chamber. 162 toward the head chamber. The tip of the piston rod 164 is connected to the clutch valve 11 via a link 168, and when the piston 162 moves to the maximum right by a spring 166, the clutch valve 1
12 is fully open, and in this state, the hydraulic oil discharged from the hydraulic pump 102 flows through the short circuit path 110, the hydraulic motor 160 is not driven, and therefore, power is transmitted to the rear wheels 46 via the output shaft 26. Never. On the other hand, when the piston 162 moves to the left against the spring force of the spring 166, the opening degree of the clutch valve 112 becomes smaller, the continuously variable transmission 100 becomes a half-clutch state, and the piston 162 is further driven to reach the maximum clutch position. When the clutch valve 112 is moved to the left, the clutch valve 112 closes and power is completely transmitted to the rear wheels 46.

Here, referring to FIG. 37, the hydraulic motor 106
To explain this, the motor is, for example, a variable displacement axial piston motor, and as shown in the figure, a cylinder block 172 connected to the drive shaft 104 has a plurality of cylinders arranged annularly around the rotation axis of the shaft. The piston 174 is slid together,
A swash plate 176 that defines the reciprocating stroke of the pistons is arranged at a position facing these pistons with a variable inclination angle θTRU. In the piston group,
Cylinder chamber 1 corresponding to the piston in the expansion stroke
78 is communicated with the high pressure oil passage 108h, and the cylinder chamber 180 corresponding to the piston in the contraction stroke is communicated with the oil passage 1081 on the low pressure side. Therefore, the high-pressure oil discharged from the hydraulic pump 102 is sucked into the cylinder chamber 178, and the low-pressure oil discharged from the cylinder chamber 180 is returned to the hydraulic pump 102. During this period, the piston 174 in the expansion stroke receives air from the swash plate 176. Cylinder block 1 due to reaction torque
72 and the drive shaft 104 are rotationally driven.

Since the capacity of the hydraulic motor is determined by the stroke of the piston 174, by moving the inclination angle θTRU of the swash plate 176 to the left from the maximum position shown by the solid line to the minimum position shown by the chain line, the speed ratio The ratio (G/R) can be changed steplessly from the minimum to the maximum. Here, the speed ratio e is expressed by the following equation.

Speed ratio e=Output rotation speed/Input rotation speed=Pump capacity/Motor capacity Therefore, at one end of the swash plate 176, there is a swing link 18.
One end of 2 is connected via a pin 184,
The other end of this link 182 is connected to the aforementioned hydraulic cylinder 154 via a second pin 186.

The hydraulic cylinder 154 is slidably connected to the cylinder 190 and is connected to the head chamber 1 inside the cylinder.
92 and a piston 196 defined in a rod chamber 194.
and a piston rod 1 integrated into the piston.
It consists of 98. One end of the swing link 182 is connected to the tip of the piston rod 198 via the pin 186, and when the piston 196 moves to the maximum right, the tilt angle θTRU of the swash plate 104 becomes maximum, and the capacity of the hydraulic motor 106 becomes maximum. Therefore, the speed ratio e becomes the minimum. Conversely, when the piston 196 moves to the left as much as possible, the inclination angle of the swash plate becomes the minimum as shown by the chain line, the motor capacity becomes the minimum, and the speed ratio e becomes the maximum.

Now, returning to Figure 36 again, we will explain the pilot valves that control these hydraulic cylinders.
The aforementioned servo cylinder 150 is provided with an electromagnetic pilot valve 200. The pilot valve 20
0 indicates oil passages 202 and 20 communicating with the head chamber 158 and rod chamber 160 of the servo cylinder 150, respectively.
4 and a supply oil passage 206 that communicates with the discharge port of the replenishment pump 114 and a return oil passage 208 that communicates with the oil tank 122, and includes a sleeve 210 and a spool 212 that is relatively movable within the sleeve. . Further, a piston rod 164 of a servo cylinder 150 is connected to the sleeve 210 via a link 214, and the action of the servo cylinder 150 is fed back to the pilot valve 20.

Thus, the spool 212 is movable between three positions on the left and right including the neutral position, and when it moves to the left and reaches the right position, the head chamber 158 of the servo cylinder 150 is moved.
The discharge oil pressure of the replenishment pump 114 is introduced to the valve 114, and the rod chamber 160 is opened to the oil tank 122, thereby the piston 162 moves to the left and the clutch valve 112 closes as described above. Furthermore, when the spool 212 moves to the right, the discharge hydraulic pressure is guided to the rod chamber 160 of the servo cylinder 150 and the head chamber 158 is released to the oil tank, so the piston 162 moves to the right and the clutch valve 112
opens the valve. Since the cylinder 150 is a servo cylinder, as the piston rod 164 moves due to the movement of the piston 162, the sleeve 210 also moves and stops at an appropriate position. Through such a mechanism, by moving the piston 162 according to the amount of movement of the spool 212, the opening degree of the clutch valve 112, that is, the opening degree of the short circuit path 110 can be adjusted as desired.

Next, the second pilot valve 220 that controls the hydraulic cylinder 154 will be explained. This pilot valve 220 is connected to the head chamber 1 of the hydraulic cylinder.
Oil passage 222 and rod chamber 194 that can communicate with 92
an oil passage 224 that communicates with
4 inserted between the return oil passage 226 connected to 2
Consists of a port restriction switching valve (solenoid valve). This pilot valve 220 includes a sleeve 228 and a spool 230 that is movable within the sleeve, and the spool is movable between a neutral position and three left and right positions, including an intermediate position where the degree of throttling changes continuously. be. That is, when the spool 230 moves to the left, high pressure hydraulic oil is introduced into the head chamber 192 of the hydraulic cylinder 154 and discharged from the rod chamber 194.
The piston 196 and piston rod 198 move to the left, and as described above, the angle of inclination of the swash plate 176 becomes smaller, the hydraulic motor capacity becomes smaller, and the speed ratio e becomes larger. Conversely, when the spool 230 moves to the right,
Hydraulic oil is discharged from the head chamber 192 and guided to the rod chamber 194, causing the piston 196 and piston rod 198 to similarly move to the right, the swash plate inclination angle increases, the hydraulic motor capacity increases, and the speed ratio e
becomes smaller. Thus, the pressure distribution between the head chamber 192 and the rod chamber 194 in the hydraulic cylinder 154 is determined by the degree of restriction of the pilot valve 22 (i.e., the position of the spool), and the piston 196 and piston rod 198 are The hydraulic motor 10 operates at a speed dependent on the pressure difference.
6 can be changed, and the speed ratio e, that is, the gear ratio (G/R) can be changed.

The pilot valves 200, 220 are connected to the speed change control unit 60, and the speed change control unit energizes/deenergizes the solenoids of the pilot valves to move the spools 212, 230 to arbitrary positions. That is, the speed change control unit receives the outputs of the throttle sensor 50, crank angle sensor 52, brake switch 54, and vehicle speed sensor 56 as in the first embodiment, and calculates the control value from these input values as described below. and pilot valve 200
The transmission/cutoff of power is controlled via the pilot valve 220, and the gear ratio is arbitrarily controlled via the pilot valve 220. Furthermore, the speed change control unit also controls the operation of a range drive mechanism 152 equipped with a hydraulic cylinder, and changes F (forward) and N ( Based on the neutral) and R (reverse) signals, a forward or reverse gear train is established in the forward/reverse switching device 130 described above. It goes without saying that such range position information is also input to the speed change control unit via the range selector switch 62. Although the main focus of the present invention is on continuously variable transmission control applying fuzzy reasoning and not on the continuously variable transmission mechanism itself, the spools 212 and 230 in the pilot valves 200 and 220 are controlled by hydraulic pressure (not shown). Through spring force or duty control of a solenoid valve, it is driven by a predetermined amount in either the left or right direction, and for example, in the case of the pilot valve 220, a desired throttle is applied to set the speed ratio of the continuously variable transmission mechanism to a desired value. Needless to say, it is constructed in such a way that it can be controlled.

Next, the operation of the speed change control unit in the second embodiment will be explained with reference to FIG. 38 and subsequent figures. still,
The description of the second embodiment will focus on the points that are essentially different from the first embodiment. For example, the details of the speed change control unit and the like are similar to those shown in FIG. 3 of the first embodiment, and therefore will be omitted. It goes without saying that the same applies to preparatory work such as creating rules before the actual race.

FIG. 38 is a main flow chart schematically showing the speed change control of the second embodiment. Referring to the figure, first, measured parameter values are read and stored in S2010. The current speed change ratio (gear ratio G/R) is calculated from the input/output rotation speed ratio of the transmission, etc., as in the first embodiment. Next, in S2012, it is determined whether or not the forward range is in the forward range (the above-mentioned F range). If determined, this program will be terminated immediately.

Figure 39 is S2014 of Figure 38 flow chart
2 is a flow chart showing a routine for determining a shift command value. In the figure, in S2100, S2102, and S2104, the acceleration α, throttle change amount ΔθTH, PS ratio, and expected PS ratio are calculated in the same manner as in the first embodiment, and then the process proceeds to S2106 to calculate the post-shift rotation speed.

Figure 40 is a flow chart showing the calculation method. First, in S2200, the gear ratio change coefficient γ
Initialize, specifically, initialize the counter value that counts it.

To summarize the control of this embodiment, since this embodiment uses a continuously variable transmission,
The concept of gear ratio is used instead of the shift position in the first embodiment, and the target gear ratio is sequentially inferred. Therefore, the object of inference is not the gear ratio itself, but a coefficient that changes the gear ratio (the gear ratio change coefficient γ described above). That is, target gear ratio (G/R) = current gear ratio (G/R)
Calculate by ×γ. As shown in Fig. 41, the gear ratio (G/R) is expressed as a total reduction ratio, and the definition range is shown in the figure as a closed interval from 3.0 (highest side) to 11.0 (lowest side). It is assumed that this fuzzy set covers the following. In the diagram, the fujii label HI means "largely move the gear ratio to the High side", and MH means "the gear ratio is set to medium High side".
"Move the gear ratio to the Low side" for ML, "Slightly lower gear ratio" for LO.
Increase the gear ratio to the Low side.

Further, as shown in FIG. 42, the gear ratio change coefficient γ is also defined by seven fuzzy sets for values ranging from 0.27 to 3.74. In addition, in the fuzzy label shown in the figure, BU is "largely on the High side" and MU is "
"Moderately on the High side", SU "Slightly on the High side",
HD "holds the current gear ratio", SD "holds the current gear ratio"
On the Low side, MD is on the Medium Low side, BD is on the Low side.
Means "largely on the Low side".

Therefore, the initialization in S2200 means setting the value of the counter indicating the gear ratio change coefficient γ to the highest value of "0.27".

Subsequently, in S2202, the post-shift rotation speed is calculated. The rotation speed after shifting is calculated at several points, specifically at the 4th point.
In Figure 2, the values of the gear ratio change coefficient γ when the grade value is 1.0 are 0.27, 0.42, 0.65, 1.0, 1.56,
Multiply the current gear ratio by 7 types, 2.42 and 3.74, and calculate each value obtained.

Next, the post-shift rotation speed calculated in S2204 is stored in the corresponding address, CFLNE (here, FL means HI, MH, ML, and LO described above). In addition, here, seven gear ratio change coefficients γ are given, but only four FLs indicating the gear ratio are given, which is why, for example, when the vehicle is currently on the lowest side, it cannot go any lower than that. This is because there are restrictions. This also applies when you are on the highest side.

Next, in S2206, the gear ratio change coefficient counter value γ is updated by multiplying by 1.55, and is calculated individually until it is confirmed in S2208 that it has reached the lowest value of 3.74.

Returning to FIG. 39 again, the expected post-shift PS ratio is then calculated in S2108. FIG. 43 is a subroutine flow chart showing the calculation procedure. First, in S2300, it is confirmed that the throttle valve is not in the closing direction, and then in S2302, the gear ratio change coefficient counter value is initialized. Here, the initial value was set to ``1.0'' because the expected value after the shift
This is because it is sufficient to calculate the PS ratio only for the Low side (down side). Then, in S2304, the
S2306 to 2312 until it is determined that the Low value has been reached.
Each value obtained by multiplying the gear ratio by the coefficient is calculated individually. Note that when the throttle valve is in the closing direction, the expected PS ratio after shifting is set to zero, as in the first embodiment.

Returning to FIG. 39 again, control toughness is then calculated in S2110.

Figure 44 is a flow chart showing the calculation procedure. After calculating the running resistance R1 after acceleration correction in S2400 to S2436, the gear ratio gear coefficient counter value is initialized in S2438, and the value is set to the highest side in S2440. S2442~ until it is determined that the
In S2448, the target gear ratio is calculated by dividing by the difference between the coefficients, 1.55, and for each value, the driving force Q1 at full opening is calculated and stored.

Returning to Figure 39 again, the final step
In S2112, a target gear ratio is determined through fuzzy reasoning.

FIG. 45 is a flow chart showing the subroutine.
In S2500, the grade values of membership functions are calculated for all rules, and the values of the conclusion part are waveform-synthesized and summed. Next, in S2502, the center of gravity is calculated, and in S2504, the gear ratio change coefficient γ is determined from the values on the domain that intersects with the center of gravity.
In S2506, a target gear ratio (G/R) is calculated.

FIG. 46 shows 17 rules used in the second embodiment. The difference in the first embodiment from the rule shown in FIG. 25 is that the shift position is indicated by the gear ratio (G/R) instead of the shift position, and the conclusion is indicated by the gear ratio change coefficient γ described above. This is true, and the implications of the rules differ accordingly, but the rest of the structure is the same. For example, the control toughness is also the same as the first embodiment in that the conclusion part of the rule searches for a value corresponding to the gear ratio scheduled from among pre-calculated values to determine the degree of rule satisfaction.

Therefore, in the fuzzy reasoning of the first embodiment, a mini-max calculation was performed, the rule that obtained the maximum grade value was selected, and the shift command value was determined according to the conclusion determined by that rule. In the example, after finding the mini value, synthesize the obtained waveform,
Find the center of gravity and determine the shift command value. To explain this point with reference to Fig. 47, the current engine speed Ne is 1500 rpm, the current gear ratio (G/R)
is 4.6, vehicle speed V is 10km/h, throttle opening θTH
is 1/8 WOT (degrees). When calculating the grade value for Rule 1, if a line is drawn upward from the position of 1500 rpm in the defined area, it will not intersect with the waveform, so the grade value will be 0. Regarding rule 2, although the vehicle speed V and the throttle opening θTH intersect with the waveforms, the current gear ratio (G/R) does not intersect, so the grade value is 0 as a mini value. The same applies to rule 3. Therefore, in the case of Rule 4, the current gear ratio (G/R) intersects with the waveform, and the same is true for vehicle speed, etc., so the waveform shown in the conclusion is obtained (in this case, the maximum membership (The waveform will remain as it is because the waveforms will intersect at .) As shown at the end of the figure, when the results obtained from the above four rules are mapped onto the same domain, an isosceles triangle shown by diagonal lines is obtained, its centroid G is found, and then the domain is mapped again. The value at the intersection of the perpendicular lines, in this case γ = 0.65, is the conclusion. Although only the cases of rules 1 to 4 are shown in the illustrated example, the procedure up to rule 17 is carried out, and the obtained waveforms are synthesized to find the center of gravity. Note that the rules are created assuming different driving conditions, so even if fuzzy inference is performed on all the rules, only about two or three will remain.

Therefore, based on the results obtained, the 38th
Output processing is performed in S2016 of the main flow chart, and the electromagnetic pilot valve 220 described above is
to change the capacity of the hydraulic motor 106,
Control the gear ratio to a desired value.

Since the present embodiment is configured as described above, when controlling the speed change of the continuously variable transmission, the speed change judgment and operation performed by an expert driver in a manual transmission vehicle can be performed by controlling the speed ratio through fuzzy reasoning. This can also be incorporated into unauthorized disguise control, providing smooth shifting characteristics without a busy feeling, and also making it possible to accurately meet emission countermeasures and individual user requests.

In this embodiment, only the gear ratio was controlled, but it is also possible to control the speed of change of the gear ratio.Furthermore, although a swash plate type continuously variable transmission mechanism was explained as an example, a pulley type continuously variable transmission mechanism was explained. Needless to say, it may be a mechanism, and furthermore, it can be applied to traction control.

(Effect of the invention) According to claim 1, it is possible to incorporate the gear shifting operation that was judged and operated by the driver in a vehicle with a manual transmission into the gear shifting control through fuzzy reasoning, which is similar to human decision making. This makes it possible to judge gear changes. In other words, by understanding the driving state of the vehicle including surrounding conditions in multiple variables and processing it through fuzzy reasoning, it is possible to make judgments and actions similar to those made by expert drivers in manual transmission vehicles. It can be reproduced during control. Furthermore, since the running resistance is accurately determined from the average value over a predetermined period of time and changes in the margin driving force are predicted, frequent changes in the gear ratio can be prevented, which is different from the conventional technology. Since there is no need to mechanically determine the shift point based on the throttle opening degree and vehicle speed based on a preset shift diagram, it is possible to realize shift control that immediately responds to constantly changing driving conditions.
It is also possible to realize a shift control that is compatible with emission countermeasures or a shift control that can individually respond to the shift characteristics desired by each user.

[Brief explanation of the drawing]

FIG. 1 is a diagram corresponding to the claims of the present invention, FIG. 2 is a schematic diagram showing the overall configuration of an automatic transmission control device according to the present invention, FIG. 3 is a block diagram showing the configuration of a control unit therein, and FIG. Fig. 4 is a flow chart of the main routine showing the operation of the unit, Fig. 5 is a flow chart showing the subroutine for determining the shift command value, and Fig. 6 is a flow chart showing the calculation of acceleration and throttle change amount in the subroutine. Explanatory diagram, Fig. 7 is a flow chart showing the PS ratio calculation subroutine in the flow chart of Fig. 5, and Fig. 8 is a flow chart showing the PS ratio calculation subroutine in the flow chart of Fig. 5.
An explanatory diagram showing the calculation of PS%, Figure 9 is also shown in Figure 7.
Figure 10 is an explanatory diagram showing the calculation of generated horsepower in the flow chart shown in Figure 5. Figure 10 is a flow chart showing the subroutine for calculating the expected PS ratio in the flow chart shown in Figure 5.
Chart, Fig. 11 is an explanatory diagram showing the calculation of the expected PS change amount, Fig. 12 is an explanatory diagram showing the calculation of the correction coefficient used in the flow chart of Fig. 10, and Fig. 13 is an explanatory diagram showing the calculation of the expected PS change amount. is a flow chart showing a subroutine for calculating the rotation speed after shifting in the flow chart of Fig. 5, Fig. 14 is an explanatory diagram showing an example of the calculation, and Fig. 15 is a subroutine for calculating the rotation speed after shifting in the flow chart of Fig. 5. A flow chart showing the expected PS ratio calculation subroutine, Fig. 16 is a flow chart showing the control toughness calculation subroutine in the flow chart of Fig. 5, and Fig. 17 is a driving force diagram explaining the premise of control toughness. , FIG. 18 is an explanatory diagram showing the torque ratio used in the 16th flow chart, FIG. 19 is an explanatory diagram similarly showing the average torque calculated in the flow chart of FIG. 16, and FIG. FIG. 21 is an explanatory diagram showing the torque calculation method, FIG. 21 is an explanatory diagram showing the acceleration correction, FIG. 22 is an explanatory diagram showing the premise, FIG. 23 is an explanatory diagram showing the membership function of control toughness, and FIG. 24 is an explanatory diagram showing the membership function of control toughness. A flow chart showing the main routine for retrieving fuzzy production rules, Fig. 25 is an explanatory diagram showing fuzzy production rules, and Fig. 26 is a flow chart showing the membership value calculation subroutine of the flow chart in Fig. 24. , FIG. 27 is an explanatory diagram showing the ROM storage table used in the calculation, FIGS. 28 and 29 are explanatory diagrams similarly showing the calculation table used in the calculation, and FIG. 30 is an explanatory diagram showing the calculation table used in the calculation. Fig. 31 is an explanatory diagram showing the rule matrix stored in the ROM used in the calculation, Fig. 32 is an explanatory diagram showing a similar calculation map, Fig. 33 is a flow chart showing the output determination subroutine of the flow chart in FIG. 24, and FIGS. 34 and 35 are used therein.
An explanatory diagram showing the tables stored in the ROM and RAM, FIG. 36 is a schematic diagram showing the entire control device for a continuously variable transmission according to the second embodiment of the present invention, centering on the hydraulic circuit, and FIG. 37 is its diagram. 38 is a flow chart showing the main routine of the transmission control unit, and FIG. 39 is a flow chart showing the transmission command value determination routine therein. Figure 40 is a flow chart showing the post-shift rotation speed calculation subroutine in the flow chart of Figure 39, and Figure 41 is the flow chart of the second
FIG. 42 is an explanatory diagram showing the membership function of the gear ratio change coefficient used in the example. FIG. 43 is an explanatory diagram showing the membership function of the gear ratio change coefficient. FIG. FIG. 44 is a flow chart showing the calculation routine; FIG. 44 is a flow chart showing the control toughness calculation subroutine; FIG. 45 is a flow chart showing the fuzzy inference routine.
FIG. 46 is an explanatory diagram showing a group of rules used in the second embodiment, and FIG. 47 is an explanatory diagram showing a fuzzy inference method. 10... Internal combustion engine body, 16... Throttle valve, 18... Engine output shaft, 20... Transmission, 22... Torque converter, 24... Main shaft, 26... Counter shaft, 30
... Oil passage, 32, 34 ... Shift valve, 36,
38... Electromagnetic solenoid, 42... Differential device, 46... Rear wheel, 50... Throttle sensor, 52... Crank angle sensor, 54... Brake switch, 56... Vehicle speed sensor, 60...
Gear change control unit, 62... Range selector switch, 64... Shift position switch, 80
...Micro computer, 100...Continuously variable transmission, 102...Quantitative discharge type hydraulic pump, 104
... Drive shaft, 106 ... Variable displacement hydraulic motor,
108...Hydraulic closed circuit, 110...Short circuit, 11
2...Clutch valve, 114...Replenishment pump, 11
6,118,120... Check valve, 122... Oil tank, 124... Relief valve, 130... Forward/forward switching device, 132... First drive gear, 134...
Second driving gear, 136...First driven gear, 138
...Intermediate gear, 140...Second driven gear, 142
... Driven clutch gear, 144 ... Clutch member, 150 ... Servo cylinder, 152 ... Range drive mechanism, 154 ... Hydraulic cylinder, 156 ...
...Cylinder, 158...Head chamber, 160...Rod chamber, 162...Piston, 164...Piston rod, 166...Spring, 168...Link,
170... Connection member, 172... Cylinder block, 174... Piston, 176... Oblique plate, 17
8,180...Cylinder chamber, 182...Swing link, 184,186...Pin, 190...Cylinder, 192...Head chamber, 194...Rod chamber,
196... Piston, 198... Piston rod, 200... Pilot valve, 202, 204...
...oil path, 206...supply path, 208...return oil path, 210...sleeve, 212...spool,
214...link, 220...pilot valve, 2
22, 224... Oil path, 226... Return oil path, 2
28...sleeve, 230...spool.

Claims (1)

[Scope of Claims] 1 a. Driving state detection means for detecting the driving state of the engine or vehicle, including at least the engine rotation speed, throttle opening, amount of change in throttle opening, vehicle speed, amount of change in vehicle speed, and current gear ratio. , b Running resistance calculating means for calculating the torque output by the engine from the engine rotation speed, and calculating the driving force of the vehicle by calculating the average value of the calculated value over a predetermined period, and calculating the running resistance, c. Driving state change prediction means for predicting, using the calculated running resistance, a change in the margin driving force that would occur if the speed is changed to the speed change ratio for all the speed change ratios that can be changed from the speed change ratio; d. a setting means for presetting a plurality of fuzzy production rules in which the desired operating state and the expected operating state change are quantified by a membership function, e.・An automatic transmission characterized by comprising: a transmission ratio determining means for determining a transmission ratio by performing fuzzy inference from a production rule; and f) a driving means for driving a transmission mechanism according to the output of the transmission ratio determining means. Machine control device.