JP3485945B2 - Throttle control device for internal combustion engine - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a throttle control device for an internal combustion engine, and more particularly to a predetermined reference position such as a throttle sensor for detecting a throttle opening or a pedal sensor for detecting an operation amount of an accelerator pedal. For example, the present invention relates to a throttle control device for an internal combustion engine that learns detection values at the fully closed position and the fully open position and uses the learned values for various controls.

[0002]

2. Description of the Related Art As a conventional throttle control device for an internal combustion engine of this type, there is, for example, one disclosed in Japanese Patent Publication No. 63-15467.

This throttle control device has a function of learning the fully closed position of the throttle valve of the internal combustion engine,
The minimum value of the detected voltage of the throttle sensor is learned as the fully closed position, and every time the detected voltage falls below the learned value, the learned value is updated to the detected voltage at that time. Then, the fully closed position learned in this way is used, for example, in comparison with the detection voltage of the throttle sensor to determine whether or not the throttle valve is currently in the fully closed position, and based on the determination result. Ignition timing control during idle operation and fuel cut during deceleration are executed.

[0004]

By the way, the output voltage of the above-mentioned throttle sensor may temporarily fluctuate under the influence of noise such as EMI (electromagnetic waves) or abrasion powder of a resistor. is there. Since the conventional throttle control device does not take measures against such noise and the like, the learning value is updated and erroneous learning is performed even when the detection voltage is lower than the learning value due to the influence of noise or the like. Will occur. As a result, the fully closed state of the throttle valve may not be accurately determined, and emission and drivability may be deteriorated due to inappropriate ignition timing control or fuel cut.

Therefore, an object of the present invention is to provide a throttle control device for an internal combustion engine which can prevent erroneous learning due to noise or the like and can always obtain an accurate learning result.

[0006]

As shown in FIG. 1, a throttle control device for an internal combustion engine according to the present invention has a plurality of operating position detecting means M for individually detecting the operating positions of detection targets.
1 and the detection value of each of the operating position detecting means M1,
The preset reference position within the working range of the detection target
A plurality of reference position learning means M2 for their respective learning, the reference position is learned in each reference position learning means M2 compared with one another, the comparison result to determine the presence or absence of erroneous learning on the basis of the erroneous learning determination Means M3 and learning result invalidating means M4 for invalidating the learning result of the reference position by the reference position learning means M2 when it is determined by the incorrect learning determining means M3 that the learning is incorrect. .

[0007]

In the present invention, the operating position of the object to be detected, such as the opening of the throttle valve and the operation amount of the accelerator pedal, is detected by the plurality of operating position detecting means M1 and is fully closed based on the respective detected values. The reference position learning means M2 learns a preset reference position such as a position or a fully open position. When noise or the like occurs in this and can only either the reference position at the impact is erroneously learned. And
The reference positions are compared with each other by the erroneous learning determining means M3, and, for example, when the difference between the reference positions is large, it is determined as erroneous learning, and the learning result of the reference position by the reference position learning means M2 is learned. The result is invalidated by the invalidation means M4. Therefore, erroneous learning due to noise or the like is prevented in advance.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment embodying a throttle control device for an internal combustion engine according to the present invention will be described below.

FIG. 2 is an overall configuration diagram showing a throttle control device for an internal combustion engine which is an embodiment of the present invention, and FIG. 3 is a detailed diagram showing a pedal sensor of the throttle control device for an internal combustion engine which is an embodiment of the present invention. FIG. 4 is an explanatory diagram showing an output characteristic of a pedal sensor of a throttle control device for an internal combustion engine which is an embodiment of the present invention.

The internal combustion engine 1 is mounted on a vehicle (not shown), and as shown in the figure, a throttle valve 4 which is opened and closed by a DC motor 3 is provided in an intake passage 2 thereof. The intake air amount supplied to the internal combustion engine 1 is adjusted according to the opening of the throttle valve 4, and the throttle opening V
Th is detected by the throttle sensor 5. In addition, the first pedal sensor 7 is provided on the accelerator pedal 6 at the driver's seat of the vehicle.
And a second pedal sensor 8 are provided, and both pedal sensors 7 and 8 detect the operation amount of the accelerator pedal 6, respectively.

Electronic control unit 1 of throttle control device
1 includes a CPU 12, a ROM 13, a RAM 14, a backup RAM 15, an interface 16 and a DC motor driver 17. CPU12 in ROM13
Various programs relating to throttle control and the like executed by the CPU 12 are stored, and the processing data of the CPU 12 and the like are temporarily stored in the RAM 14. Further, the backup RAM 15 stores learning values and the like obtained by the learning process executed by the CPU 12, and the stored contents are retained even when an ignition switch (not shown) is turned off.

The interface 16, the throttle sensor 5, the first pedal sensor 7 and the second pedal sensor 8 are connected, and the detection signals from these sensors are input to the CPU 12 via the interface 16. The DC motor 3 is connected to the DC motor driver 17, and the drive signal output from the DC motor driver 17 drives and controls the DC motor 3.

Next, the structures of the first pedal sensor 7 and the second pedal sensor 8 will be described in detail. As shown in FIG. 3, these pedal sensors 7 and 8 include accelerator pedal 6
Is provided at a position of a support shaft 21 that rotatably supports. Both pedal sensors 7 and 8 utilize the principle of variable resistance, the first pedal sensor 7 is composed of a first resistor 7a and a first slider 7b, and the second pedal sensor 8 is It is composed of a second resistor 8a and a second slider 8b.

The first resistor 7a and the second resistor 8a
Is formed in a double arc shape centering on the support shaft 21 on a base (not shown).
5V is applied to one end of a and the other end is grounded.
Further, the first slider 7b and the second slider 8b are
The base end is provided on an arm (not shown) fixed to the support shaft 21, and the tips of the sliders 7b and 8b are in contact with the corresponding resistors 7a and 8a. The arm together with the support shaft 21 rotates counterclockwise (throttle opening side) in FIG. 3 when the accelerator pedal 6 is depressed, and clockwise (throttle close side) by a return spring (not shown) when the depression operation is stopped. When the sliders 7b and 8b move, the tips of the two sliders 7b and 8b slide on the resistors 7a and 8a. A fully closed stopper 22 is provided at a position close to the sliders 7b and 8b, and the fully closed stopper 22 restricts the rotation of the arm toward the throttle closed side. It is controlled to a small opening degree, and this position is hereinafter referred to as a fully closed position. Therefore, as shown in FIG.
The potentials of b and 8b increase / decrease in proportion to the pedal operation amount, and the potentials are output from the pedal sensors 7 and 8 as detection voltages VAP1 and VAP2.

Next, the throttle control process executed by the CPU 12 of the throttle control device for an internal combustion engine configured as described above will be described.

FIG. 5 is a flowchart showing an accelerator operation amount calculation routine executed by the CPU of the throttle control device for an internal combustion engine which is an embodiment of the present invention, and FIG. 6 is a throttle control for an internal combustion engine which is an embodiment of the present invention. FIG. 7 is an explanatory view showing a map for calculating the throttle opening command value θcmd from the accelerator operation amount AP of the device, and FIG. 7 is a diagram showing a throttle from the throttle opening command value θcmd of the throttle control device for an internal combustion engine which is an embodiment of the present invention. It is explanatory drawing which shows the map for calculating command voltage Vcmd.

First, the outline of the throttle control will be described according to the routine of FIG. This accelerator operation amount calculation routine is executed every 10 msec, and the CPU 12 executes step S1.
This time the detected voltage VAP1i, V of both pedal sensors 7,8
Detection voltage VAP1i-1, VAP from AP2i at the time of the previous processing
2i-1 is subtracted to obtain the change amounts ΔVAP1, ΔVA
Find P2. Then, in step S2, the change amount ΔVAP
1, the absolute value of the difference between ΔVAP2 and the preset value Δ
KV is compared, and the absolute value of the difference between the variations ΔVAP1 and ΔVAP2 is equal to or less than the set value ΔKV (| ΔVAP1-ΔVAP2
| ≦ ΔKV), the current detection voltage VAP1i, V
AP2i is considered to be reliable because it is not affected by noise or the like, and the process proceeds to step S3. And step S
3 is the detected voltage VAP1i of the first pedal sensor 7 of this time.
Then, a final learning value VAPO, which will be described later, is subtracted and corrected, and then a preset constant KAP1 is multiplied to obtain an accelerator operation amount AP, and this routine is ended.

Further, in the step S2, the change amount ΔVAP
1. When the absolute value of the difference between ΔVAP2 exceeds the set value ΔKV (| ΔVAP1−ΔVAP2 |> ΔKV), it is considered that the detected voltages VAP1i and VAP2i at this time are unreliable due to the influence of noise, etc. Go to S4. Then, in step S4, the fully closed position learning value VA is calculated from the detected voltage VAP1i-1 of the first pedal sensor 7 at the previous time.
After correcting by subtracting PO, the constant KAP1 is multiplied to obtain the accelerator operation amount AP, and this routine is ended.

That is, the detection voltage VAP1i of the first pedal sensor 7 is used to calculate the actual accelerator operation amount AP, and the detection voltage VAP2i of the second pedal sensor 8 is
It is used to determine whether the detection voltage VAP1i of the first pedal sensor 7 is reliable. CPU 12
From the accelerator operation amount AP calculated in this way, FIG.
The throttle opening command value θcmd is calculated according to the map of FIG. 7 and the throttle command voltage Vcmd is calculated from the throttle opening command value θcmd according to the map of FIG. Then, the DC motor driver 17 drives the DC motor 3 so that the throttle opening Vth detected by the throttle sensor 5 becomes equal to the throttle command voltage Vcmd, and the opening of the throttle valve 4 is adjusted.

FIG. 8 is a flow chart showing an accelerator operation determination routine executed by the CPU of the throttle control device for an internal combustion engine which is an embodiment of the present invention, and FIG. 9 is a throttle control device for an internal combustion engine which is an embodiment of the present invention. 10 is a flowchart showing a fully closed position learning routine executed by the CPU of FIG. 10, FIG. 10 is a flowchart showing a learning value comparison / update routine executed by the CPU of the throttle control device for an internal combustion engine which is an embodiment of the present invention, and FIG. 3 is a time chart showing a learning state of a fully closed position of a pedal by a throttle control device for an internal combustion engine which is an embodiment of the invention.

Next, the learning processing of the pedal fully closed position will be described according to the routines of FIGS. The accelerator operation determination routine of FIG. 8 is executed every 10 msec, and the CPU 1
In step S11, the detected voltage VAP1 of the first pedal sensor 7 is compared with a set value KVAPO set as a voltage when the accelerator pedal 6 is near the fully closed position in advance in step S11. The detection voltage VAP1 is less than or equal to the set value KVAPO (VAP
If the condition (1 ≦ KVAPO) continues for 500 msec or more, it is determined that the accelerator pedal 6 is fully closed, and the pedal fully closed flag XAPO is set (= 1) in step S12.
In other cases (VAP1> KVAPO), it is determined that the accelerator pedal is being operated, and step S13 is performed.
Then, the pedal fully closed flag XAPO is reset (= 0), and then the process proceeds to step S14.

Then, in step S14, the pedal fully closed flag XAPO is switched from set to reset (1
→ 0), that is, it is the moment when the accelerator pedal 6 is depressed. When a negative determination is made in step S14, the pedal fully closed flag XAP is set in step S15.
When the pedal fully closed flag XAPO is reset (= 0), the accelerator pedal 6 is judged.
Is considered not to be fully closed, and this routine is once terminated. Further, in step S15, the pedal fully closed flag XAPO
Is set (= 1), it is considered that the accelerator pedal 6 is in the fully closed state, and the fully closed position of the accelerator pedal 6 is learned in the processing from step S16.

First, the CPU 12 determines the state of the fully closed learning selection flag XAPOS in step S16. here,
The fully closed learning selection flag XAPOS is set in step S described later.
Each time the routine is executed in step 20 and step S22, the setting (= 1) and the reset (= 0) are alternately switched. Therefore, the CPU 12 alternately repeats the case of shifting from step S16 to step S17 and the case of shifting from step S16 to step S18, and in step S17, the learning processing of the fully closed position for the first pedal sensor 7 is performed. In step S18, learning processing of the fully closed position of the second pedal sensor 8 is executed.

The above-mentioned step S17 or step S
When the fully closed position learning process is called in 18, the CPU 12
Executes the routine of FIG. It should be noted that this routine also displays learning processing for the first pedal sensor 7 and the second pedal sensor 8 in common, and the portion marked with * in the flowchart is the first processing in the processing of step S17.
“1” is applied to execute the learning process of the pedal sensor 7, and in the process of step S18, “2” is applied to execute the learning process of the second pedal sensor 8. Therefore, it is assumed that the learning process for the first pedal sensor 7 is executed.

At step S21, the CPU 12 determines whether or not the pedal fully closed flag XAPO has been switched from reset to set (0 → 1), that is, when the accelerator pedal 6 is fully closed, the learning process at step S17 is started. Or not. When an affirmative decision is made in step S21, the learning value VAPO for the first pedal sensor 7 stored in the RAM 14 during the previous processing in step S22.
A preset offset value KDVAP is added to 1i-1 to obtain a tentative learning value RVAPO1. Next, in step S23, the detection voltage VAP1 of the first pedal sensor 7 and the temporary learning value RVAPO1 are compared, and the detection voltage VAP1 is equal to or lower than the temporary learning value RVAPO1 (VAP1 ≦ RVAPO1).
If the state of continues for 32 msec or more, step S2
In step 4, the temporary learning value RVAPO1 is updated to the detection voltage VAP1, and this routine is once terminated.

When executing the subsequent routine, the CPU 1
2 shifts directly from step S21 to step S23 and repeats the processing of step S23 and step S24,
Until the accelerator pedal 6 is operated again from the fully closed position, the minimum value of the detection voltage VAP1 of the first pedal sensor 7 is learned as the temporary learning value RVAPO1 and stored in the RAM 14. Here, in step S22, the temporary learning value RVO
P is set to a value larger than the previous learning value VAPO1i-1 because consideration is given so that learning can be performed even when the minimum value of the detection voltage VAP1 fluctuates to the open side. That is, the previous learning value VAPO1i-1 is used as it is as the temporary learning value R.
With VAPO1, learning is not performed unless the minimum value of the detection voltage VAP1 fluctuates to the open side. However, if the temporary learning value RVAPO1 is set to a large value in advance as described above, the minimum value of the detection voltage VAP1 fluctuates to the open side. Even if the temporary learning value RVA
Learning is possible as long as it is PO1 or less.

As described above, the learning process for the first pedal sensor 7 in step S17 is executed, and although the description is omitted, the learning process for the second pedal sensor 8 in step S18 is performed in exactly the same procedure. Done. Therefore, as shown in FIG. 11, every time the accelerator pedal 6 is fully closed, learning processing of the fully closed positions of both pedal sensors 7 and 8 is alternately performed, and the temporary learning value thereof is acquired as necessary. RVAPO1 and RVAPO2 (in the figure, shown as learned values VAPO1 and VAPO2 after confirmation, which will be described later) are updated.

On the other hand, when the affirmative determination is made in step S14, it is determined that the moment the accelerator pedal 6 is depressed, and in step S19, the fully closed learning selection flag XAPO is selected.
Determine the state of S. When the fully closed learning selection flag XAPOS is reset (= 0), step S20
Then, set this fully closed learning selection flag XAPOS (= 1)
Then, this routine is once terminated. If the fully closed learning selection flag XAPOS is set (= 1), the learning value comparison / update process is executed in step S21, and then the fully closed learning selection flag XAPO is executed in step S22.
S is reset (= 0) and this routine is once ended.

That is, as shown in FIG. 11, when the learning process for the first pedal sensor 7 in step S17 is completed and the learning process for the second pedal sensor 8 in step S18 is completed. , Step S2
At 1, the learning value comparison / update process is called. Then, the CPU 12 at this time executes the routine of FIG. 10, and first, in step S31, the temporary learning values RVAPO1, RVAP.
O2 is the current learning value VAPO1i, VAPO2i
And the learning value VAP determined in step S32.
From O1i, VAPO2i to the previous learning value VAPO1i-1, V
APO2i-1 is subtracted to change the learning value ΔVAPO1, Δ
Ask for VAPO2.

Next, at step S33, the learning value change amount Δ
The absolute value of the difference between VAPO1 and ΔVAPO2 is compared with a preset setting value ΔKVO, and the learning value change amount ΔVAPO is compared.
1, when the absolute value of the difference between ΔVAPO2 is less than or equal to the set value ΔKVO (| ΔVAPO1−ΔVAPO2 | ≦ ΔKVO), it is considered that neither learning value VAPO1i or VAPO2i is reliable because it is not affected by noise. The process moves to S34. Then, the last learned value V stored in the backup RAM 15 in step S34.
APOi-1 is the current learning value change amount ΔVAPO1, ΔVA
Add the average value of PO2 to obtain the final learning value VAPO
ask for i.

In step S33, the absolute value of the difference between the learning value change amounts ΔVAPO1 and ΔVAPO2 is the set value ΔK.
It exceeds VO (| ΔVAPO1-ΔVAPO2 |
> ΔKVO), one of the learning values VAPO1i,
Since the VAPO2i is affected by noise or the like and is considered unreliable due to erroneous learning, the process proceeds to step S35. Then, in step S35, the last learned value VAP of the last time.
Let Oi-1 be the final learning value VAPOi for this time.

After that, the CPU 12 guards the upper limit of the final learning value VAPOi to the upper limit Vmax in steps S36 and S37, and further steps S38 and S.
At 39, the lower limit of the final learning value VAPOi is guarded to the lower limit value Vmin. Further, in step S40, the final learning value V of this time
Let APOi be the last learned value VAPOi-1 and the current learned values VAPO1i and VAPO2i be the last learned value VA.
These are stored in the RAM 14 as PO1i-1 and VAPO2i-1, respectively, and this routine ends.

Therefore, as shown by points a and b in FIG. 11, every time the fully closed learning selection flag XAPOS is set (= 1), the learning value VAPO1i of the both pedal sensors 7 and 8 learned immediately before is set. , VAPO2i, the learning value change amounts ΔVAPO1, ΔVAPO2 are calculated, and it is determined whether or not the final learning value VAPO should be updated based on the difference between the learning value change amounts ΔVAPO1, ΔVAPO2. For example, at the point a in FIG. 11 in which noise or the like has not occurred in the immediately preceding learning process, the learning value change amount ΔVA
Since the difference between PO1 and ΔVAPO2 is small, the final learning value VAPO is updated, and noise b or the like occurs at the point b in the learning process of the second pedal sensor 8 immediately before.
In this case, since the difference between the learning value change amounts ΔVAPO1 and ΔVAPO2 becomes large due to erroneous learning of the learning value VAPO2i, updating of the final learning value VAPO is prohibited, that is, the learning values VAPO1i and VAPO2i learned immediately before are invalidated.

That is, since the learning processing of the fully closed positions of both pedal sensors 7 and 8 is performed at different timings alternately, when noise or the like occurs, only one of the detection voltages VAP1 and ΔVAP2 has an influence. Received
The learning values VAPO1i and VAPO2i on that side are erroneously learned. In this case, both learning value changes ΔVAP
Since a large difference occurs between O1 and ΔVAPO2, it is possible to determine the presence or absence of erroneous learning based on the difference between the learning value change amounts ΔVAPO1 and ΔVAPO2 as described above.

As described above, when the learning values VAPO1i, VAPO2i are erroneously learned due to noise or the like, the learning values VAPO1i, VAPO2i are invalidated and the update of the final learning value VAPO is prohibited. The erroneous learning due to is prevented in advance, and the fully closed position of the accelerator pedal 6 can always be learned accurately.

On the other hand, as described above, both pedal sensors 7,
8 forms each of the resistors 7a and 8a on a common base, and provides the sliders 7b and 8b on a common arm to make most of the components common, for example, both resistors 7a and 8a.
Variation in characteristics of a, resistors 7a and 8a, and slider 7
The difference in the output characteristics due to the mounting error of b and 8b is suppressed to the minimum. However, even if such a measure is taken, as shown in FIG. 4, since there is still an error ΔV in the outputs of both pedal sensors 7 and 8, there is a difference due to noise or the like in both learning values VAPO1i and VAPO2i. In addition to this, there is a difference caused by the difference in the output characteristics of the pedal sensors 7 and 8, and if the learning values VAPO1i and VAPO2i are directly compared, it may not be possible to accurately determine the presence or absence of erroneous learning. Here, in this embodiment, since the learning value change amounts ΔVAPO1, ΔVAPO2, which represent how much the learning values VAPO1i, VAPO2i have changed from the previous processing, are compared to determine whether or not there is erroneous learning, both pedals are determined. Whether or not there is erroneous learning can be accurately determined without being affected by the difference in the output characteristics of the sensors 7 and 8.

As described above, in the present embodiment, the first pedal sensor 7 and the second pedal sensor 8 function as the operating position detecting means M1, and the steps S21 to S24 are executed as the reference position learning means M2. CP of time
CPU12 when U12 performs the processing of step S32 to step S33 as an erroneous learning determination means M3
However, the CPU 12 at the time of executing the processing of step S34 functions as the learning result invalidating means M4.

As described above, the throttle control device for the internal combustion engine of this embodiment has the first pedal sensor 7 and the second pedal sensor 8 for individually detecting the operation amount of the accelerator pedal 6.
And the detection voltages VAP1, V of the pedal sensors 7, 8
Based on AP2, the fully closed position of the accelerator pedal 6 is alternately learned at different timings, and the learning value change amounts ΔVAPO1, ΔVAPO2 calculated from the learning values VAPO1i, VAPO2i are compared with each other,
When the difference between the two is large, those learning values VAPO1
The CPU 12 for invalidating i, VAPO2i and prohibiting update of the final learning value VAPO is provided.

Therefore, since the fully closed positions of both pedal sensors 7 and 8 are learned at different timings, when noise occurs, one of the learned values VAP is learned.
O1i and VAPO2i are erroneously learned, and the difference between the learning value change amounts ΔVAPO1 and ΔVAPO2 becomes large. At this time, the learning values VAPO1i and VAPO2i are invalidated, and updating of the final learning value VAPO is prohibited. Therefore, false learning due to noise or the like is prevented, and the fully closed position of the accelerator pedal 6 is always accurately learned. Therefore, the throttle control can be performed more accurately.

In the above embodiment, the throttle control device for learning the fully closed position of the accelerator pedal 6 is embodied. However, the present invention is not limited to this and is not limited to this. The learning content and the learning target are not limited as long as there is a risk of erroneous learning due to the influence. Therefore, for example, it may be embodied as a throttle control device that learns the fully open position instead of the fully closed position, or learns the fully closed position or the fully open position of the throttle valve 4.

Further, in the above embodiment, when the learning values VAPO1i, VAPO2i of any one of the pedal sensors 7, 8 are not erroneously learned, those learning values VAPO1i, VAPO1i, VAPO2i.
Although the final learning value VAPOi is calculated based on the average of PO2i and the final learning value VAPOi is used for the correction of the accelerator operation amount AP, the learning values VAPO1i and VAPO2i of both pedal sensors 7 and 8 are not necessarily used.
Need not be reflected in.

Therefore, for example, one of the learning values VAPO1i and VAPO2i is the learning value change amount ΔVAPO.
It may be used only to determine whether or not there is erroneous learning by comparing 1 and ΔVAPO2, and only the other learning values VAPO1i and VAPO2i may be used to correct the accelerator operation amount AP.

Further, in the above embodiment, the learning value change amount ΔV.
The presence / absence of erroneous learning is determined based on the difference between APO1 and ΔVAPO2, but the present invention is not limited to this. For example, the learning value change amount ΔVAPO
1, learning value VAPO1i, VAPO instead of ΔVAPO2
2i may be directly compared, or the presence or absence of erroneous learning may be determined based on a ratio or the like instead of the difference between the two.

[0044]

As is evident from the foregoing description, according to the throttle control apparatus for an internal combustion engine of the present invention, will learn the reference position of the detection target, the learning result when it is determined that compared to erroneously learns the learning result to each other Since it is invalidated, erroneous learning due to noise or the like can be prevented in advance, and an accurate learning result can always be obtained. Consequently, various controls using the learning result can be performed more accurately.

[Brief description of drawings]

FIG. 1 is a claim correspondence diagram conceptually showing the contents of an embodiment of the present invention.

FIG. 2 is an overall configuration diagram showing a throttle control device for an internal combustion engine which is an embodiment of the present invention.

FIG. 3 is a detailed view showing a pedal sensor of a throttle control device for an internal combustion engine which is an embodiment of the present invention.

FIG. 4 is an explanatory diagram showing an output characteristic of a pedal sensor of a throttle control device for an internal combustion engine which is an embodiment of the present invention.

FIG. 5 is a flowchart showing an accelerator operation amount calculation routine executed by a CPU of a throttle control device for an internal combustion engine according to an embodiment of the present invention.

FIG. 6 is an explanatory diagram showing a map for calculating a throttle opening command value θcmd from an accelerator operation amount AP of a throttle control device for an internal combustion engine which is an embodiment of the present invention.

FIG. 7 is an explanatory diagram showing a map for calculating a throttle command voltage Vcmd from a throttle opening command value θcmd of a throttle control device for an internal combustion engine which is an embodiment of the present invention.

FIG. 8 is a flowchart showing an accelerator operation determination routine executed by a CPU of a throttle control device for an internal combustion engine according to an embodiment of the present invention.

FIG. 9 is a flowchart showing a fully closed position learning routine executed by a CPU of a throttle control device for an internal combustion engine according to an embodiment of the present invention.

FIG. 10 is a flowchart showing a learning value comparison / update routine executed by a CPU of a throttle control device for an internal combustion engine according to an embodiment of the present invention.

FIG. 11 is a time chart showing a learning state of a pedal fully closed position by a throttle control device for an internal combustion engine according to an embodiment of the present invention.

[Explanation of symbols]

M1 operating position detection means M2 reference position learning means M3 False learning judgment means M4 Learning result invalidation means 6 accelerator pedal 7 First pedal sensor 8 Second pedal sensor 12 CPU

Continued front page       (56) Reference JP-A-3-122513 (JP, A)                 JP 63-247610 (JP, A)                 JP-A-4-41944 (JP, A)                 JP-A-63-192934 (JP, A)                 JP 63-27812 (JP, A)                 JP-A-2-42157 (JP, A)

Claims (2)

(57) [Claims] 1. A plurality of operating position detecting means for individually detecting an operating position of a detection target, and preset values within the operating range of the detection target based on detection values of the respective operating position detecting means detected at different timings. A plurality of reference position learning means for respectively learning the learned reference positions at different timings, and the different tie positions learned by the reference position learning means.
Erroneous learning determining means for comparing the reference positions learned by mingling with each other and determining the presence or absence of erroneous learning of the reference position based on the comparison result, and when erroneous learning is determined by the erroneous learning determining means. A throttle control device for an internal combustion engine, comprising: learning result invalidating means for invalidating the learning result of the reference position by the reference position learning means. 2. The erroneous learning determination means compares the amounts of change of the reference positions learned by the reference position learning means with each other, and determines whether or not the reference positions are erroneously learned based on the comparison result. The throttle control device for an internal combustion engine according to claim 1 , wherein: