BACKGROUND OF THE INVENTION
Field of the Invention
[0001] The present invention relates to a torque estimation system that estimates a torque
generated in an engine and a vehicle.
Description of the Background Art
[0002] In a vehicle with an automatic or semi-automatic transmission, a torque generated
in an engine is estimated with the use of a torque map. Then, an output of the engine
after speed change is adjusted based on the estimated torque. This allows for smooth
speed change, resulting in improved drivability of the vehicle.
[0003] The foregoing torque map is created based on the torque of the engine measured in
a standard driving environment. Therefore, the torque estimated based on the torque
map could be greatly different from a torque actually generated in the engine depending
on the driving environment of the vehicle.
[0004] In an engine output torque monitor device described in
JP 2007-291856 A, for example, an output torque of the engine is estimated to be zero when a difference
between the number of input rotations and the number of output rotations of a torque
converter is not more than a predetermined value, so that errors in the torque map
are corrected.
SUMMARY OF THE INVENTION
[0005] An object of the present invention is to provide a torque estimation system arranged
to estimate a torque of an engine at high accuracy and allow for reduction of a vehicle
in cost and size and a vehicle including the torque estimation system.
- (1) According to an aspect of the present invention, a torque estimation system arranged
to estimate a torque generated in an engine of a vehicle includes a storage arranged
to store a first equation for estimating the torque generated in the engine based
on a rotational speed of the engine and a second equation for calculating an inertial
torque of a crank of the engine based on the rotational speed of the engine, and an
arithmetic processor arranged to calculate an estimated torque of the engine based
on the first equation, wherein the arithmetic processor calculates the inertial torque
of the crank when the torque generated in the engine is not transmitted to a drive
wheel based on the second equation, and corrects the first equation such that the
estimated torque calculated based on the first equation approaches the inertial torque
calculated based on the second equation.
In the torque estimation system, the estimated torque of the engine is calculated
from the rotational speed of the engine by the arithmetic processor based on the first
equation. The arithmetic processor calculates the inertial torque of the crank from
the rotational speed of the engine based on the second equation. Then, the arithmetic
processor corrects the first equation such that the estimated torque calculated based
on the first equation approaches the inertial torque calculated based on the second
equation.
Here, the inertial torque of the crank is calculated when the torque of the engine
is not transmitted to the drive wheel in the torque estimation system. In this case,
it can be considered that the inertial toque is equivalent to the torque generated
in the engine. Accordingly, the first equation is corrected such that the estimated
torque of the engine approaches the inertial torque of the crank as described above,
thereby allowing a value of the estimated torque calculated based on the first equation
to come close to the torque actually generated in the engine.
Moreover, since the inertial torque of the crank is calculated from the rotational
speed of the engine based on the second equation, the inertial torque of the crank
can be accurately calculated even when the rotational speed of the engine changes
in the torque estimation system. Thus, transient characteristics of the inertial torque
when the rotational speed of the engine changes can be reflected in correction of
the first equation. This allows transient characteristics of the estimated torque
calculated based on the first equation to come close to transient characteristics
of the torque actually generated in the engine.
Furthermore, the estimated torque of the engine and the inertial torque of the crank
are calculated based on the first equation and the second equation that are previously
stored in the storage in the torque estimation system. Accordingly, the estimated
torque and the inertial torque can be calculated without providing such a device as
a torque converter.
As a result, the torque of the engine can be estimated at high accuracy while the
vehicle can be reduced in cost and size.
- (2) The first equation may include a time constant of a first-order delay element
of rising of the torque generated in the engine (the first order delay element represents
the rise of the torque generated in the engine), and the arithmetic processor may
correct a time constant of the first-order delay element in the first equation such
that a difference between the estimated torque calculated based on the first equation
and the inertial torque of the crank calculated based on the second equation is reduced.
In this case, the transient characteristics of the estimated torque calculated based
on the first equation can be brought sufficiently close to the transient characteristics
of the torque actually generated in the engine.
- (3) The first equation may include a dead time of rising of the torque generated in
the engine, the arithmetic processor may correct the dead time in the first equation
such that a difference between the estimated torque calculated based on the first
equation and the inertial torque of the crank calculated based on the second equation
is reduced.
In this case, a rising timing of the torque generated in the engine can be estimated
at high accuracy based on the first equation.
- (4) The first equation may include a first-order delay element and a dead time, the
arithmetic processor may correct a time constant of the first-order delay element
and the dead time in the first equation such that a difference between the estimated
torque calculated based on the first equation and the inertial torque of the crank
calculated based on the second equation is reduced, and the dead time is corrected
before the time constant is corrected.
In this case, the time constant in the first equation can be corrected while a difference
between the dead time of rising of the estimated torque (i.e., the dead time in the
first equation) and the dead time of rising of the torque actually generated in the
engine is sufficiently decreased, thereby allowing the first equation to be corrected
at high accuracy.
- (5) The vehicle may include a transmission, the torque estimation system may further
include a controller arranged to increase the rotational speed of the engine in the
case of down-shifting of the transmission, and the arithmetic processor may correct
the first equation in the case of the down-shifting of the transmission.
In this case, the transient characteristics of the torque generated in the engine
when the rotational speed of the engine is increased can be sufficiently reflected
in the first equation. This sufficiently improves the accuracy of the estimated torque
calculated based on the first equation.
- (6) The vehicle may include a transmission, the torque estimation system may further
include a controller arranged to decrease the rotational speed of the engine in the
case of up-shifting of the transmission, and the arithmetic processor may correct
the first equation in the case of the up-shifting of the transmission.
In this case, the transient characteristics of the torque generated in the engine
when the rotational speed of the engine is decreased can be sufficiently reflected
in the first equation. This sufficiently improves the accuracy of the estimated torque
calculated based on the first equation.
- (7) The vehicle may include a transmitting mechanism arranged to transmit the torque
generated in the engine to the drive wheel, and the torque estimation system may further
include a controller arranged to control, when the vehicle is stopped, the transmitting
mechanism to interrupt transmission of the torque from the engine to the drive wheel
and to increase or decrease the rotational speed of the engine.
In this case, since the transmission of the torque from the engine to the drive wheel
is interrupted by the controller, the rotational speed of the engine can be changed
in a wide range while the vehicle is stopped. Accordingly, the first equation can
be corrected in a wide range of engine rotational speeds. This further improves the
accuracy of the estimated torque calculated based on the first equation.
- (8) The vehicle may include a throttle valve arranged to adjust an amount of air supplied
to the engine and a plurality of rotation members arranged to transmit the torque
of the crank to the drive wheel, the storage may store a torque map previously created
based on a relationship among the rotational speed of the engine, an opening of the
throttle valve and the torque generated in the engine, the first equation may be Tcal=Tmap·e-T2·s/(1+T1·s), the second equation may be Tr=J×(dω/dt), the Tcal may be the estimated
torque, the Tmap may be a torque obtained from the torque map based on the rotational
speed of the engine and the throttle opening, the T1 may be a time constant of a first-order
delay element of rising of the torque generated in the engine, the T2 may be a dead
time of the rising of the torque generated in the engine, and the s may be a Laplacean
in the first equation, and the Tr may be the inertial torque of the crank, the J may
be moments of inertia of the crank and the rotational member, which is rotated by
the torque transmitted from the crank, of the plurality of rotational members, and
the (dω/dt) may be an angular velocity of the crank in the second equation.
In the torque estimation system, the torque map created based on the relationship
among the rotational speed of the engine, the opening of the throttle valve and the
torque generated in the engine is previously stored in the storage. Then, the arithmetic
processor calculates the estimated torque using the first equation of the first-order
delay element with the torque Tmap calculated based on the torque map as the gain.
In this case, the estimated torque can be calculated based on the torque map previously
stored in the storage, thus facilitating calculation of the estimated torque.
In addition, the arithmetic processor calculates the inertial torque of the crank
based on the moments of inertia of the crank and the rotation member that is rotated
by the torque transmitted from the crank.
Here, the moment of inertia of the rotation member that is rotated by the torque of
the engine while the torque of the engine is not transmitted to the drive wheel can
be easily obtained by experiments, simulations and so on. Accordingly, the moment
of inertia of the rotation member that is easily obtained is used as an element of
the second equation, so that the inertial torque of the crank can be easily calculated
at high accuracy in the torque estimation system.
- (9) According to another aspect of the present invention, a vehicle includes a drive
wheel, an engine, a transmitting mechanism arranged to transmit a torque generated
by the engine to the drive wheel, and the torque estimation system according to any
of the claims 1 to 8.
[0006] In the vehicle, the torque generated by the engine is transmitted to the drive wheel
through the transmitting mechanism. This causes the vehicle to drive.
[0007] In addition, the torque estimation system according to the foregoing embodiment is
provided in the vehicle. Thus, the estimated torque of the engine is calculated from
the rotational speed of the engine by the arithmetic processor based on the first
equation. Moreover, the inertial torque of the crank is calculated from the rotational
speed of the engine by the arithmetic processor based on the second equation. Then,
the first equation is corrected such that the estimated torque calculated based on
the first equation approaches the inertial torque calculated based on the second equation.
[0008] Here, the inertial torque of the crank is calculated when the torque of the engine
is not transmitted to the drive wheel in the torque estimation system. In this case,
it can be considered that the inertial toque is equivalent to the torque generated
in the engine. Accordingly, the first equation is corrected such that the estimated
torque of the engine approaches the inertial torque of the crank as described above,
thereby allowing the value of the estimated torque calculated based on the first equation
to come close to the torque actually generated in the engine.
[0009] Moreover, since the inertial torque of the crank is calculated from the rotational
speed of the engine based on the second equation, the inertial torque of the crank
can be accurately calculated even when the rotational speed of the engine changes
in the torque estimation system. Thus, the transient characteristics of the inertial
torque when the rotational speed of the engine changes can be reflected in the correction
of the first equation. This allows the transient characteristics of the estimated
torque calculated based on the first equation to come close to the transient characteristics
of the torque actually generated in the engine.
[0010] Furthermore, the estimated torque of the engine and the inertial torque of the crank
are calculated based on the first equation and the second equation previously stored
in the storage in the torque estimation system. Accordingly, the estimated torque
and the inertial torque can be calculated without providing such a device as a torque
converter.
[0011] As a result, the torque of the engine can be estimated at high accuracy while the
vehicle can be reduced in cost and size.
[0012] According to the present invention, the first equation is corrected such that the
estimated torque of the engine approaches the inertial torque of the crank, thereby
allowing the value of the estimated torque calculated based on the first equation
to come close to the torque actually generated in the engine.
[0013] Moreover, since the inertial torque of the crank is calculated from the rotational
speed of the engine based on the second equation, the inertial torque of the crank
can be accurately calculated even when the rotational speed of the engine changes.
This allows the transient characteristics of the inertial torque when the rotational
speed of the engine changes to be reflected in the correction of the first equation.
As a result, the transient characteristics of the estimated torque calculated based
on the first equation can be brought close to the transient characteristics of the
torque actually generated in the engine.
[0014] Furthermore, the estimated torque of the engine and the inertial torque of the crank
are calculated based on the first equation and the second equation, so that the estimated
torque and the inertial torque can be calculated without providing such a device as
a torque converter.
[0015] As a result, the torque of the engine can be estimated at high accuracy while the
vehicle can be reduced in cost and size.
[0016] Further, the invention is advantageous as a vehicle including the inventive torque
estimation system is reduced in cost and size as no torque converter has to be provided
for estimating the actual output of the engine. In addition, when the output torque
of the engine changes, its transient characteristics are also reflected in the correction
of the torque map so that the torque generated in the engine is estimated at high
accuracy.
BRIEF DESCRIPTION OF THE DRAWINGS
[0017]
Fig. 1 is a schematic side view of a motorcycle.
Fig. 2 is a diagram showing the structures of a transmission and a shift mechanism.
Fig. 3 is a block diagram showing the structure of a transmission control system.
Fig. 4 is a flowchart showing a control operation of a CPU in gear-shifting.
Fig. 5 is a diagram showing one example of a relationship between an estimated torque
and an actual torque.
Fig. 6 is a diagram showing one example of the relationship between the estimated
torque and the actual torque.
Fig. 7 is a diagram showing one example of the relationship between the estimated
torque and the actual torque.
Fig. 8 is a diagram showing one example of the relationship between the estimated
torque and the actual torque.
Fig. 9 is a flowchart showing a correction operation of the CPU.
Fig. 10 is a flowchart showing a correction operation of the CPU.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0018] Hereinafter, description is made of a vehicle including a torque estimation system
according to an embodiment of the present invention while referring to the drawings.
Note that description is made of a motorcycle as an example of the vehicle in the
following paragraphs. In addition, description is made of a semi-automatic transmission
control system in which gear-shifting of a transmission is automatically performed
based on a shifting operation by a driver as an example of the torque estimation system.
(1) General Structure of Motorcycle
[0019] Fig. 1 is a schematic side view showing a motorcycle according to the present embodiment.
[0020] In the motorcycle 100 of Fig. 1, a head pipe 102 is provided at the front end of
a main body frame 101. A front fork 103 is provided at the head pipe 102 so as to
be able to swing from side to side. At the lower end of the front fork 103, a front
wheel 104 is rotatably supported. A handle 105 is provided at the upper end of the
head pipe 102.
[0021] The handle 105 is provided with an accelerator grip 106. An engine 107 is provided
at the center of the main body frame 101. A throttle body 108 is attached to an intake
port of the engine 107, and an exhaust pipe 109 is attached to an exhaust port of
the engine 107. The throttle body 108 is provided with a throttle valve 81.
[0022] A crankcase 110 is attached to the lower portion of the engine 107. A crank 2 (see
Fig. 2) of the engine 107 is housed in the crankcase 110.
[0023] A transmission case 111 is provided at the lower portion of the main body frame 101.
A transmission 5 (see Fig. 2) and a shift mechanism 6 (see Fig. 2), described below,
are provided in the transmission case 111. A shift pedal 112 is provided at the transmission
case 111.
[0024] Note that a clutch 3 (see Fig. 2) need not be disconnected by a driver for switching
a gear position of the transmission 5 in the present embodiment. In other words, a
semi-automatic transmission control system that automatically switches the gear position
of the transmission 5 based on a shifting operation by the driver is mounted in the
motorcycle 100 according to the present embodiment. Details of the transmission control
system will be described below.
[0025] A fuel tank 113 is provided above the engine 107 and a seat 114 is provided to the
rear of the fuel tank 113. An ECU (Electronic Control Unit) 50 is provided under the
seat 114. A rear arm 115 is connected to the main body frame 101 so as to extend to
the rear of the engine 107. A rear wheel 116 and a rear wheel driven sprocket 117
are rotatably held by the rear arm 115. A chain 118 is attached to the rear wheel
driven sprocket 117.
(2) Structures of the Transmission and the Shift Mechanism
[0026] Next, description is made of the transmission and the shift mechanism provided in
the transmission case 111 of Fig. 1.
[0027] Fig. 2 is a diagram showing the structures of the transmission and the shift mechanism.
[0028] As shown in Fig. 2, the transmission 5 includes a main shaft 5a and a drive shaft
5b. A multi-stage (e.g., five-stage) transmission gears 5c are attached to the main
shaft 5a, and a multi-stage transmission gears 5d are attached to the drive shaft
5b.
[0029] The main shaft 5a is coupled to the crank 2 of the engine 107 (Fig. 1) through the
clutch 3. The clutch 3 includes a pressure plate 3a, a plurality of clutch disks 3b
and a plurality of friction disks 3c. The clutch disks 3b are rotated by a torque
transmitted from the crank 2. Moreover, the friction disks 3c are coupled to the main
shaft 5a and rotate around the main shaft 5a as a rotation axis.
[0030] The friction disks 3c are biased by the pressure plate 3a in a direction in which
the friction disks 3c come into close contact with the clutch disks 3b. A state in
which the plurality of clutch disks 3b and the plurality of friction disks 3c are
in close contact with one another is referred to as a connection state of the clutch
3, and a state in which the plurality of clutch disks 3b and the plurality of friction
disks 3c are separated from one another is referred to as a disconnection state of
the clutch 3 in the following description. Although the torque of the crank 2 is transmitted
to the main shaft 5a through the clutch disks 3b and the friction disks 3c in the
connection state of the clutch 3, the torque of the crank 2 is not transmitted to
the main shaft 5a in the disconnection state of the clutch 3.
[0031] A push rod 5e is inserted into the main shaft 5a. One end of the push rod 5e is coupled
to the pressure plate 3a, and the other end thereof is coupled to an electric or hydraulic
clutch actuator 4.
[0032] In the present embodiment, the push rod 5e is pushed out to the side of the clutch
3 when the clutch actuator 4 is driven by control of the ECU 50. This causes the pressure
plate 3a to be pushed to cause the clutch disks 3b and the friction disks 3c to be
separated from one another. As a result, the clutch 3 is brought into the disconnection
state. Details of the control operation of the ECU 50 will be described below.
[0033] The torque transmitted from the crank 2 to the main shaft 5a when the clutch 3 is
in the connection state is transmitted to the drive shaft 5b through the transmission
gears 5c and the transmission gears 5d. The chain 118 of Fig. 1 is attached to the
drive shaft 5b. The torque of the drive shaft 5b is transmitted to the rear wheel
116 (Fig. 1) through the chain 118 and the rear wheel driven sprocket 117 (Fig. 1).
This causes the motorcycle 100 to drive.
[0034] A reduction gear ratio between the main shaft 5a and the drive shaft 5b is determined
by combination of the transmission gears 5c and the transmission gears 5d. Moreover,
the reduction gear ratio between the main shaft 5a and the drive shaft 5b is changed
by moving any transmission gears 5c, 5d of the plurality of transmission gears 5c,
5d. The transmission gears 5c, 5d are moved by operation of the shift mechanism 6.
[0035] The shift mechanism 6 includes a shift cam 6a. A plurality of cam grooves 6b (three
in Fig. 2) are formed in the shift cam 6a. A shift fork 6c is attached to each cam
groove 6b. The shift cam 6a is connected to an electric or hydraulic shift actuator
7 through a link mechanism that is not shown.
[0036] In the present embodiment, the shift cam 6a is rotated when the shift actuator 7
is driven by control of the ECU 50. This causes each shift fork 6c to move along the
cam groove 6b. As a result, any transmission gears 5c, 5d are moved to change the
gear position of the transmission 5.
(3) The Transmission Control System
[0037] Next, description is made of the transmission control system of the motorcycle 100.
[0038] Fig. 3 is a block diagram showing the configuration of the transmission control system
according to the present embodiment.
[0039] As shown in Fig. 3, the transmission control system 200 according to the present
embodiment includes a correction switch SW, an accelerator opening sensor SE1, a throttle
sensor SE2, an engine rotational speed sensor SE3, a shift cam rotation angle sensor
SE4, a drive shaft rotational speed sensor SE5, a shifting operation detection sensor
SE6, the ECU 50, the clutch actuator 4, the shift actuator 7 and a throttle actuator
8.
[0040] The correction switch SW is provided at the handle 105, for example. The correction
switch SW will be described below. The accelerator opening sensor SE1 detects an operation
amount of the accelerator grip 106 (Fig. 1) (hereinafter referred to as "an accelerator
opening") by the driver and applies the detected accelerator opening to the ECU 50.
The throttle sensor SE2 detects an opening of the throttle valve 81 (Fig. 1)(hereinafter
referred to as "a throttle opening") and applies the detected throttle opening to
the ECU 50. The engine rotational speed sensor SE3 detects a rotational speed of the
engine 107 (Fig. 1) and applies the detected rotational speed to the ECU 50. Note
that the engine rotational speed sensor SE3 detects an angular velocity of the crank
2 (Fig. 2) to detect the rotational speed of the engine 107 in the present embodiment.
[0041] The shift cam rotation angle sensor SE4 detects a rotation angle of a shift cam 6a
(Fig. 2) and applies the detected rotation angle to the ECU 50. The drive shaft rotational
speed sensor SE5 detects a rotational speed of the drive shaft 5b (Fig. 2) and applies
the detected rotational speed to the ECU 50.
[0042] The shifting operation detection sensor SE6 detects a direction in which the shift
pedal 112 (Fig. 1) is operated by the driver and applies a signal indicating the detected
direction of operation (a signal indicating up-shifting or a signal indicating down-shifting)
to the ECU 50. The shifting operation detection sensor SE6 is composed of a potentiometer,
a load sensor, a magnetostrictive sensor or the like, for example. Note that the shifting
operation detection sensor SE6 outputs the signal with a positive value when an up-shifting
operation is detected, and outputs the signal with a negative value when a down-shifting
operation is detected, for example.
[0043] The shift actuator 7 is electric or hydraulic, for example, and rotates the shift
cam 6a (Fig. 2) by control of the CPU 52, described below. The throttle actuator 8
includes an electric motor, for example, and adjusts the opening of the throttle valve
81 by control of the CPU 52.
[0044] The ECU 50 includes an interface circuit 51, a CPU (central processing unit) 52,
a ROM (read only memory) 53, and a RAM (random access memory) 54.
[0045] Output signals of the foregoing sensors SE1 to SE6 are applied to the CPU 52 through
the interface circuit 51. The CPU 52 adjusts the output of the engine 107 based on
results of detection of the sensors SE1 to SE6, described below. The ROM 53 stores
a control program of the CPU 52 and so on. The RAM 54 stores various kinds of data
and functions as a processing area of the CPU 52.
(4) Output Control of the Engine by the CPU
[0046] A stationary torque map and a torque estimation physical model are previously stored
in the ROM 53 (or the RAM 54) of the ECU 50 in the present embodiment. The stationary
torque map expresses a relationship among the rotational speed of the engine 107,
the throttle opening and the torque of the engine 107 measured in a standard driving
environment. The torque estimation physical model is mathematically expressed by the
following equation (1) and stored in the ROM 53 (or the RAM 54).
[0047] In the following equation (1), Tcal is an estimated torque of the engine 107, and
Tmap is a torque obtained from the stationary torque map based on the rotational speed
of the engine 107 and the throttle opening (hereinafter referred to as "a stationary
torque Tmap"). T1 is a time constant of a first-order delay element of the torque
generated in the engine 107, T2 is a dead time, and s is a Laplacean.

[0048] Note that the time constant T1 and the dead time T2 in the equation (1) above are
stored in the RAM 54 so as to be changed by a correction operation of the CPU 52.
Initial values of the time constant T1 and the dead time T2 are calculated by experiments,
simulations or the like, and previously stored in the RAM 54 (or the ROM 53). Note
that the time constant T1 and the dead time T2 are normalized by a cycle time period
of the engine 107 (a time period required for two rotations of the crank 2), for example,
and expressed by the product of the cycle time period of the engine 107 and a preset
coefficient. In this case, the time constant T1 of the same value and the dead time
T2 of the same value can be used regardless of the rotational speed of the engine
107.
[0049] In the present embodiment, the time constant T1 and the dead time T2 are corrected
by the CPU 52, so that an error between the estimated torque Tcal calculated based
on the equation (1) and the torque actually generated in the engine 107 can be reduced.
Details of the correction operation of the CPU 52 will be described below.
(a) Basic Operation in the Output Control
[0050] First, description is made of output adjustment of the engine 107 by the CPU 52 when
the shift pedal 112 (Fig. 1) is not operated by the driver.
[0051] In the present embodiment, the CPU 52 controls the throttle actuator 8 based on the
accelerator opening detected by the accelerator opening sensor SE1. Thus, the throttle
opening is adjusted and the output of the engine 107 is adjusted. Note that a relationship
between the accelerator opening and the throttle opening is previously stored in the
ROM 53 (or the RAM 54) of the ECU 50.
[0052] The CPU 52 calculates the stationary torque Tmap based on the throttle opening detected
by the throttle sensor SE2 and the rotational speed of the engine 107 detected by
the engine rotational speed sensor SE3. The CPU 52 calculates the estimated torque
Tcal based on the stationary torque Tmap and the equation (1). Then, the CPU 52 controls
the throttle actuator 8 based on the estimated torque Tcal.
(b) Output Control in Gear-Shifting
[0053] Next, description is made of the output adjustment of the engine 107 by the CPU 52
when the driver operates the shift pedal 112 for gear-shifting.
[0054] Fig. 4 is a flowchart showing the control operation of the CPU 52 in the gear-shifting.
[0055] As shown in Fig. 4, the CPU 52 first determines based on the output signal of the
shifting operation detection sensor SE6 (Fig. 3) whether the driver has performed
the shifting operation (Step S1). When the shifting operation has not been performed
by the driver, the ECU 50 waits until the shifting operation is performed.
[0056] When the shifting operation is performed by the driver, the CPU 52 disconnects the
clutch 3 (Fig. 2) by controlling the clutch actuator 4 (Fig. 2) (Step S2).
[0057] Next, the CPU 52 controls the throttle actuator 8 (Fig. 3) to adjust the throttle
opening, thereby increasing or decreasing the rotational speed of the engine 107 (Step
S3). Specifically, when the down-shifting operation is performed by the driver during
deceleration of the motorcycle 100, the CPU 52 causes the throttle opening to be larger
than a value determined based on the accelerator opening, for example. This causes
the rotational speed of the engine 107 to increase. When the up-shifting operation
is performed by the driver during acceleration of the motorcycle 100, the CPU 52 causes
the throttle opening to be smaller than the value determined based on the accelerator
opening, for example. This decreases the rotational speed of the engine 107.
[0058] Note that the process of Step S3 prevents the rotational speed of the main shaft
5a (the friction disks 3c (Fig. 2)) and the rotational speed of the clutch disks 3b
(Fig. 2) from being greatly different form each other at the time of connection of
the clutch 3 in Step S5, described below. Accordingly, an occurrence of a shift shock
in the motorcycle 100 is prevented.
[0059] Next, the CPU 52 controls the shift actuator 7 (Fig. 3) to rotate the shift cam 6a
(Fig. 2) (Step S4). Thus, the shift forks 6c (Fig. 2) are moved, thereby moving the
transmission gears 5c (Fig. 2) or the transmission gears 5d (Fig. 2). As a result,
the gear position of the transmission 5 is changed.
[0060] Thereafter, the CPU 52 controls the clutch actuator 4 to connect the clutch 3 (Step
S5). In this manner, the gear-shifting of the transmission 5 is finished.
(5) Correction of the Time Constant and the Dead Time
[0061] Next, description is made of a method of correcting the time constant T1 and the
dead time T2 in the equation (1).
[0062] In the present embodiment, the CPU 52 performs the correction operation when the
shifting operation is performed by the driver or when the correction switch SW (Fig
3) is turned on by the driver.
(a) Correction Operation in the case of the Down-Shifting
[0063] First, description is made of the correction operation performed in the case of the
down-shifting.
[0064] In the case of the down-shifting, the CPU 52 calculates an inertial torque Tr of
the crank 2 based on the following equation (2) previously stored in the ROM 53 (or
the RAM 54). Then, the CPU 52 performs the correction operation, described below,
assuming that the calculated inertial torque of the crank 2 as a torque actually generated
in the engine 107 (hereinafter referred to as "an actual torque Tr").

[0065] Note that in the equation (2), J is a moment of inertia of a body of rotation rotated
by the engine 107 at the time of disconnection of the clutch 3, and is preset based
on experiments, simulations or the like. The moment of inertia J is calculated by
adding the moments of inertia of a plurality of torque transmitting members (including
the crank 2 and the clutch disks 3b) between the crank 2 (Fig. 2) and the clutch disks
3b (Fig. 2), for example.
[0066] In the equation (2), (dω/dt) is an angular acceleration of the crank 2 in a period
during which the clutch 3 is disconnected. The angular acceleration (dω/dt) of the
crank 2 is detected by the engine rotational speed sensor SE3.
[0067] Figs. 5 and 6 are diagrams showing examples of a relationship between the estimated
torque Tcal calculated based on the equation (1) and the actual torque Tr calculated
based on the equation (2). Note that in Figs. 5 and 6, (a) is a graph showing change
with time of the actual torque Tr and the estimated torque Tcal, (b) is a graph showing
change with time of the rotational speed of the engine 107 detected by the engine
rotational speed sensor SE3, (c) is a graph showing change with time of the throttle
opening, and (d) is a graph showing change with time of a movement amount of the push
rod 5e (Fig. 2) (hereinafter referred to as "the clutch stroke").
[0068] In the example of Fig. 5, the clutch 3 is disconnected and the throttle opening is
increased from a value a to a value b at a time point t1. Thereafter, the actual torque
Tr is increased at a time point t2 and the estimated torque is increased at a time
point t3. In this example, a time period between the time point t1 at which the throttle
opening is increased and the time point t2 at which the increase of the actual torque
Tr is started is a dead time of the actual torque Tr, and a time period between the
time point t1 and the time point t3 at which the increase of the estimated torque
Tcal is started is a dead time of the estimated torque Tcal.
[0069] Accordingly, the dead time of the estimated torque Tcal is longer than the dead time
of the actual torque Tr in the example of Fig. 5. In such a case, the CPU 52 corrects
the dead time T2 of the torque estimation physical model (the equation (1)) such that
the dead time of the actual torque Tr and the dead time of the estimated torque Tcal
coincide with each other.
[0070] Note that for example, when values of the actual torque Tr and the estimated torque
Tcal reach a preset value at given time points, the CPU 52 determines that the foregoing
given time points are time points at which the increase of the actual torque Tr and
the estimated torque Tcal is started in the present embodiment. The above-mentioned
preset value is a value of 3 % to 10 % of a maximum value of the actual torque Tr
(a torque value at the time point t4 in Fig. 5), for example. Then, the dead time
T2 is corrected such that the time point at which the increase of the actual torque
Tr is started and the time point at which the increase of the estimated torque Tcal
is started coincide with each other. In this case, it is not determined that the increase
of the torques is started unless the values of the actual torque Tr and the estimated
torque Tcal attain or exceed the preset value. This prevents erroneous detection of
the time points at which the increase of the torques is started.
[0071] Meanwhile, after the throttle opening is increased at the time point t1, the actual
torque Tr and the estimated torque Tcal are nearly simultaneously increased at the
time pint t2 in the example of Fig. 6. That is, the dead times of the actual torque
Tr and the estimated torque Tcal are substantially equal. In the example of Fig. 6,
however, the estimated torque Tcal is increased more gently than the actual torque
Tr. In such a case, the CPU 52 corrects the time constant T1 of the torque estimation
physical model such that a rate of change of the estimated torque Tcal is equal to
that of the actual torque Tr.
[0072] In this manner, the time constant T1 and the dead time T2 of the torque estimation
physical model are corrected in the case of the down-shifting in the present embodiment.
This reduces the error between the estimated torque calculated from the torque estimation
physical model and the torque actually generated in the engine 107. Accordingly, the
engine 107 can be controlled at high accuracy.
(b) Correction Operation in the case of the Up-Shifting
[0073] Next, description is made of the correction operation performed in the case of the
up-shifting.
[0074] In the case of the up-shifting, the CPU 52 calculates the actual torque Tr based
on the equation (2) in the same manner as in the down-shifting, and performs the correction
operation described below.
[0075] Figs. 7 and 8 are diagrams showing examples of the relationship between the estimated
torque Tcal calculated based on the equation (1) and the actual torque Tr calculated
based on the equation (2). Note that in Figs. 7 and 8, (a) is a graph showing change
with time of the actual torque Tr and the estimated torque Tcal, (b) is a graph showing
change with time of the rotational speed of the engine 107 detected by the engine
rotational speed sensor SE3, (c) is a graph showing change with time of the throttle
opening, and (d) is a graph showing change with time of the clutch stroke.
[0076] Note that the time constant T1 in the equation (1) used in the correction operation
in the case of the up-shifting may be different from the time constant T1 in the equation
(1) used in the correction operation in the case of the down-shifting. Similarly,
the dead time T2 in the equation (1) used in the correction operation in the case
of the up-shifting may be different from the dead time T2 in the equation (1) used
in the correction operation in the case of the down-shifting.
[0077] In the example of Fig. 7, the clutch 3 is disconnected and the throttle opening is
decreased from a value c to a value d at a time point t1. Thereafter, the actual torque
Tr is decreased at a time point t2 and the estimated toque is decreased at a time
point t3. In this example, a time period between the time point t1 at which the throttle
opening is decreased and the time point t2 at which the decrease of the actual torque
Tr is started is a dead time of the actual torque Tr, and a time period between the
time point t1 and the time point t3 at which the decrease of the estimated torque
Tcal is started is a dead time of the estimated torque Tcal.
[0078] Accordingly, the dead time of the estimated torque Tcal is longer than the dead time
of the actual torque Tr in the example of Fig. 7. In such a case, the CPU 52 corrects
the dead time T2 of the torque estimation physical model (the equation (1)) such that
the dead time of the actual torque Tr and the dead time of the estimated torque Tcal
coincide with each other.
[0079] Meanwhile, after the throttle opening is decreased at the time point t1, the actual
torque Tr and the estimated torque Tcal are nearly simultaneously decreased at the
time point t2 in the example of Fig. 8. That is, the dead times of the actual torque
Tr and the estimated torque Tcal are substantially equal. In the example of Fig. 8,
however, the estimated torque Tcal is decreased more gently than the actual torque
Tr. In such a case, the CPU 52 corrects the time constant T1 of the torque estimation
physical model such that a rate of change of the estimated torque Tcal is equal to
that of the actual torque Tr.
[0080] In this manner, the time constant T1 and the dead time T2 of the torque estimation
physical model are corrected in the case of the up-shifting in the present embodiment.
This reduces the error between the estimated torque calculated from the torque estimation
physical model and the torque actually generated in the engine 107. Accordingly, the
engine 107 can be controlled at high accuracy.
(c) Correction Operation by the Correction Switch SW
[0081] Similarly to the cases of the down-shifting and the up-shifting, the CPU 52 corrects
the time constant T1 and the dead time T2 when a maintenance operator turns on the
correction switch SW.
[0082] Specifically, the CPU 52 disconnects the clutch 3 and increases the throttle opening
to increase or decrease the rotational speed of the engine 107 by 1000 rpm, for example.
Thereafter, the CPU 52 calculates the actual torque Tr and the estimated torque Tcal
when the rotational speed of the engine 107 is increased or decreased by 1000 rpm
in the same manner as that described above. Then, the CPU 52 corrects the time constant
T1 and the dead time T2 of the torque estimation physical model if the actual torque
Tr and the estimated torque Tcal are different from each other.
[0083] The CPU 52 performs the foregoing operation for every 1000 rpm in a range of the
rotational speed of the engine 107 from 1000 rpm to 9000 rpm, for example. This allows
the time constant T1 and the dead time T2 of the torque estimation physical model
to be corrected to suitable values in the wide range of rotational speeds of the engine
107. As a result, the engine 107 can be controlled at higher accuracy.
[0084] Note that in the case of setting the gear position of the transmission 5 in a neutral
position, the time constant T1 and the dead time T2 when the correction switch SW
is turned on may be corrected with the clutch 3 being connected. In this case, the
moment of inertia J in the equation (2) is calculated by adding the moments of inertia
of the plurality of torque transmitting members between the crank 2 (Fig. 2) and the
transmission gears 5c (Fig. 2).
(d) Control Flow
[0085] Figs. 9 and 10 are flowcharts showing the correction operation of the CPU 52.
[0086] As shown in Fig. 9, the CPU 52 first determines whether the motorcycle 100 is running
based on a detected value of the drive shaft rotational speed sensor SE5 (Step S11).
[0087] When the motorcycle 100 is running, the CPU 52 determines whether the shifting operation
is performed by the driver based on the output signal of the shifting operation detection
sensor SE6 (Step S12). When the shifting operation is not performed by the driver,
the CPU 52 returns to the process of Step S11.
[0088] When the shifting operation is performed by the driver, the CPU 52 waits until the
gear-shifting of the transmission 5 is finished (Step S13). Note that in Step S13,
it is determined that the gear-shifting is completed when the clutch stroke attains
not more than a preset value, for example. The clutch stroke may be calculated based
on a control amount applied from the CPU 52 to the clutch actuator 4 (Fig. 2), or
may be calculated by providing a detection sensor that detects the movement amount
of the push rod 5e (Fig. 2), for example.
[0089] After completion of the gear-shifting, the CPU 52 calculates the actual torque Tr
during disconnection of the clutch 3 and the dead time of the actual torque Tr (Step
S14). Note that the dead time of the actual torque Tr is calculated as a time period
from the increase or decrease of the throttle opening to the start of the increase
or decrease of the actual torque Tr (the time period between the time point t1 and
the time point t2 in Fig. 5), for example.
[0090] Next, the CPU 52 calculates a difference between the dead time of the actual torque
Tr calculated in Step S14 and the dead time T2 of the torque estimation physical model
(see the equation (1)) (Step S15).
[0091] Next, the CPU 52 determines whether the difference calculated in Step S15 is not
more than a preset first threshold value (T2/2, for example) (Step S16). When the
difference calculated in Step S15 is not more than the first threshold value, the
CPU 52 corrects the dead time T2 of the torque estimation physical model to the dead
time calculated in Step S14 (Step S17).
[0092] When the difference calculated in Step S15 is larger than the first threshold value,
the CPU 52 proceeds to the subsequent process with no correction of the dead time
T2 of the torque estimation physical model. Note that the process of Step S15 prevents
the dead time of the actual torque Tr from being reflected in the correction of the
dead time T2 of the torque estimation physical model even when the value of the actual
torque Tr is erroneously calculated due to measurement failure and so on. Accordingly,
the engine 107 can be controlled at higher accuracy.
[0093] Next, the CPU 52 calculates the time constant of the first-order delay element of
the actual torque Tr calculated in Step S14 (Fig. 9) as shown in Fig. 10 (Step S18).
Note that in Step S18, the time constant is calculated with the maximum value of the
actual torque Tr (the torque value at the time point t4 in Fig. 5) as the gain of
the first-order delay element in the down-shifting, and the time constant is calculated
with the minimum value of the actual torque Tr (the torque value at the time point
t4 in Fig. 7) as the gain of the first-order delay element in the up-shifting, for
example.
[0094] Next, the CPU 52 calculates a difference between the time constant of the first-order
delay element of the actual torque Tr calculated in Step S18 and the time constant
T1 (see the equation (1)) of the torque estimation physical model (the corrected torque
estimation physical model if the dead time T2 is corrected in Step S17 (Fig. 9)) (Step
S19).
[0095] Next, the CPU 52 determines whether the difference calculated in Step S19 is not
more than a preset second threshold value (T1/2, for example) (Step S20). When the
difference calculated in Step S19 is not more than the second threshold value, the
CPU 52 corrects the time constant T1 of the torque estimation physical model to the
time constant calculated in Step S18 (Step S21).
[0096] When the difference calculated in Step S20 is larger than the second threshold value,
the CPU 52 finishes the correction of the torque estimation physical model with no
correction of the time constant T1 of the torque estimation physical model. Note that
the process of Step S20 prevents the time constant of the first-order delay element
of the actual torque Tr from being reflected in the correction of the time constant
T1 of the torque estimation physical model even when the value of the actual torque
Tr is erroneously calculated due to measurement failure and so on. Accordingly, the
engine 107 can be controlled at higher accuracy.
[0097] As shown in Fig. 9, when it is determined in Step S11 that the motorcycle 100 is
not running, the CPU 52 determines whether the correction switch SW is turned on (Step
S22). When the correction switch SW is not turned on, the CPU 52 returns to the process
of Step S11.
[0098] When the correction switch SW is turned on, the CPU 52 controls the clutch actuator
4 and the shift actuator 7 to disconnect the clutch 3 and adjust the throttle opening,
thereby increasing or decreasing the rotational speed of the engine 107 (Step S23).
Thereafter, the same processes as Steps S14 to S21 are performed, so that the time
constant T1 and the dead time T2 of the torque estimation physical model are corrected.
[0099] Note that the rotational speed of the engine 107 is increased or decreased by 1000
rpm, for example, in Step S23 in the present embodiment. Then, the actual torque Tr
is calculated for each change of the rotational speed by 1000 rpm, and the time constant
T1 and the dead time T2 of the torque estimation physical model are corrected.
(6) Effects of the Present Embodiment
[0100] In the present embodiment, the time constant T1 and the dead time T2 of the torque
estimation physical model are corrected such that transient response characteristics
of the torque calculated based on the torque estimation physical model approach transient
response characteristics of the torque actually generated in the engine 107 in the
gear-shifting of the transmission 5. Accordingly, the engine 107 can be controlled
at high accuracy based on the torque estimation physical model.
[0101] The time constant T1 is corrected after the dead time T2 is corrected in the present
embodiment. In this case, the time constant T1 can be corrected while the dead time
of the actual torque Tr and the dead time T2 of the torque estimation physical model
are substantially equal to each other, thus allowing the correction of the torque
estimation physical model to be performed at high accuracy.
[0102] The torque estimation physical model is corrected while the motorcycle 100 is actually
running, thus allowing the transient response characteristics of the torque calculated
based on the torque estimation physical model to easily come close to the transient
response characteristics of the torque actually generated in the engine 107 at high
accuracy in the present embodiment.
[0103] In addition, the correction switch SW is turned on by the maintenance operator while
the motorcycle 100 is stopped, so that the torque estimation physical model can be
corrected in the wide range of rotational speeds of the engine 107 in the present
embodiment.
[0104] As a result, the torque generated in the engine 107 can be estimated at high accuracy
based on the torque estimation physical model.
(7) Other Embodiments
(a) Other Examples of the Correction Method
[0105] While the rotational speed of the engine 107 is increased or decreased by 1000 rpm
to correct the torque estimation physical model when the correction switch SW is turned
on by the maintenance operator in the foregoing embodiment, the increase or decrease
of the rotational speed of the engine 107 is not limited to the foregoing example.
For example, the rotational speed of the engine 107 may be increased or decreased
by less than 1000 rpm (500 rpm, for example) or larger than 1000 rpm in the correction
of the torque estimation physical model.
(b) Other Examples of the Motorcycle
[0106] While the shift pedal 112 is provided for the shifting operation by the driver in
the foregoing embodiment, a shift switch may be provided at the handle 105 for detecting
the shifting operation by the driver. In this case, the driver operates the shift
switch to easily perform the gear-shifting of the transmission 5.
[0107] While description is made of the motorcycle 100 as one example of the vehicle in
the foregoing embodiment, another vehicle such as a three-wheeled motor vehicle and
a four-wheeled motor vehicle may be applied.
(c) Other Examples of the Transmission Control System
[0108] While description is made of the semi-automatic transmission control system 200 that
automatically performs the gear-shifting of the transmission 5 based on the shifting
operation by the driver in the foregoing embodiment, the present invention can be
applied to a full automatic transmission control system.
[0109] In the full automatic transmission control system, up-shifting control and down-shifting
control may be started based on the torque calculated from the torque map, for example.
[0110] While the rotational speed of the engine 107 is adjusted by adjusting the throttle
opening in the foregoing embodiment, the rotational speed of the engine 107 may be
adjusted by adjusting an ignition timing of an air-fuel mixture or an amount of injected
fuel in the engine 107. Note that the CPU 52 can control an ignition plug, not shown,
to adjust the ignition timing of the air-fuel mixture. Moreover, the CPU 52 can control
a fuel injector, not shown, to adjust the amount of injected fuel.
[0111] While the one CPU 52 calculates the estimated torque, corrects the torque estimation
physical model, and adjusts the output of the engine 107 in the foregoing embodiment,
these operations may be performed by a plurality of CPUs 52.
(8) Correspondences between Elements in the Claims and Parts in Embodiments
[0112] In the following paragraph, non-limiting examples of correspondences between various
elements recited in the claims below and those described above with respect to various
preferred embodiments of the present invention are explained.
[0113] In the foregoing embodiments, the equation (1) is an example of a first equation,
the equation (2) is an example of a second equation, the RAM 54 or the ROM 53 are
examples of a storage, the CPU 52 is an example of an arithmetic processor and a controller,
the rear wheel 116 is an example of a drive wheel, the clutch 3, the clutch actuator
4, the transmission 5, the shift mechanism 6 and the shift actuator 7 are examples
of a transmitting mechanism, the clutch disk 3b, the friction disk 3c, the main shaft
5a, the drive shaft 5b, the transmission gear 5c and the transmission gear 5d are
examples of a plurality of rotation members.
[0114] As each of various elements recited in the claims, various other elements having
configurations or functions described in the claims can be also used.