`Onari et al.
`
`III USOO5189621A
`
`5,189.621
`Feb. 23, 1993
`
`11)
`45
`
`Patent Number:
`Date of Patent:
`
`54 ELECTRONIC ENGINE CONTROL
`APPARATUS
`75 Inventors: Mikihiko Onari, Kokubunji;
`Motohisa Funabashi, Sagamihara;
`Teruji Sekozawa, Kawasaki; Makoto
`Shioya, Tokyo, all of Japan
`73 Assignee: Hitachi, Ltd., Tokyo, Japan
`(21) Appl. No.: 622,217
`22 Filed:
`Dec. 3, 1990
`
`63
`
`Related U.S. Application Data
`Continuation of Ser. No. 233,209, Aug. 17, 1988, aban
`doned, which is a continuation-in-part of Ser. No.
`46,388, May 6, 1987, Pat. No. 4,853,720.
`Foreign Application Priority Data
`30
`Aug. 19, 1987 JP
`Japan ................................ 62-204006
`Oct. 28, 1987 JP
`Japan ................................ 32-270202
`51
`Int. Cl....................... G06F 15/20; FO2D 41/00;
`B6OK 41/OO
`52 U.S. Cl. .......................... 364/431,04; 364/431.03;
`364/148; 123/480; 395/905
`58 Field of Search ...................... 364/431.01, 431.03,
`364/43.04, 431.05, 424.01, 148, 150, 151, 152,
`154; 123/480, 350; 395/905
`References Cited
`U.S. PATENT DOCUMENTS
`4,058,796 l/1977 Oishi et al. .......................... 340A679
`4,346,776 8/1982 Taplin ..........
`... 123/342
`4,439,824 3/1984 Mayer ......
`... 364/424.01
`4,597,049 6/1986 Murakami ....
`... 364/43.07
`4,727,838 3/1988 Oshiage et al. ..................... 123/399
`4,729,356 3/1988 Kaneko et al. ...................... 123A399
`
`(56)
`
`4,747,055 5/1988 Eto et al. ........................ 364/424.01
`4,763,745 8/1988 Eto et al. ............................ 364/74
`4,773,010 9/1988 Suzuki et al...
`. 364/424.05
`4,773,012 9/1988 to et al. ......................... 364/424.0
`4,829,434 5/1989 Karmel et al. ................... 364/424.
`4,853,720 8/1989 Onari et al. .................... 364/43.07
`FOREIGN PATENT DOCUMENTS
`0059586 2/1982 European Pat. Off. .
`0144608 6/1985 European Pat. Off. .
`3715423 11/1987 Fed. Rep. of Germany .
`2047361 11/1980 United Kingdom .
`2151049 7/1985 United Kingdom .
`OTHER PUBLICATIONS
`U.S. Ser. No. 55,530, filed May 29, 1987.
`U.S. Ser. No. 155,391, filed Feb. 12, 1988.
`Primary Examiner-Vincent N. Trans
`Attorney, Agent, or Firm-Fay, Sharpe, Beall, Fagan,
`Minnich & McKee
`ABSTRACT
`57
`An electronic engine control apparatus includes: a plu
`rality of first sensors for detecting the driving action
`taken in accordance with a driver's intent; a plurality of
`second sensors for detecting the operating conditions of
`a vehicle and an engine; a plurality of actuators for
`controlling the engine; a unit for discriminating the
`driver's intent of how to drive the vehicle based on
`output signals from the first and second sensors; and a
`unit for controlling the engine to match the driver's
`intent by selectively adjusting at least one of the actua
`tors, in accordance with the discriminated driver's in
`te.
`
`59 Claims, 17 Drawing Sheets
`
`MEASURED
`WAR1ABE
`
`MEASURED
`VALUE
`
`MEMBERSHIP
`FUNCON
`
`FREQUENCY
`DATA (Fki)
`
`WEGHT OF
`coy, Trix
`
`EST MATION OF
`DRIVING COND.
`
`
`
`WECE
`SPEED
`W
`
`dNAdt
`
`9th
`
`A 6th
`
`GEAR
`POSION
`GP
`
`SEERING
`
`BMW1056
`Page 1 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`VIS9l4d
`
`SY3AING LNSWNOXMIANS
` Nveun|AN@WNOMIANGOND|AN@WNOMIANGONDBONSY3SA3Yd
`
`
`
`
`ONIAIO4%SSONSYSASYdSYAAINdNOLNS0N3d30SNOLLOVONIAINGSILSINSLOVYVHO
`AYMHOIH|GvoyNivINnOW!1EuLSALHOdSAWWHON
`
`
`
`
`(MH)(4W)(4S)7a8)(ON)
`
`
`
`OuLHOTS.HLM
`
`Sheet 1 of 17
`
`5,189,621
`
`BMW1056
`Page 2 of 36
`
`BMW1056
`Page 2 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 2 of 17
`
`5,189,621
`
`O
`
`=O
`
`
`
`ANZA©AWaNALHSITISOX1I3HI1
`
`
`
`
`
`(7W)(4S)
`
`L33u1S
`
`©ca
`AYMHOIHNVaun
`
`
`ANAWNOYIANSONIAIYC
`(ON)(39)
`
`STLN39S|aNdSwan|S,|aNdSwan|g|Old
`
`ALYOdS“IWWYON
`
`3IGGIW (99)(ds)
`
`aIGGIN
`
`BMW1056
`Page 3 of 36
`
`BMW1056
`Page 3 of 36
`
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 3 of 17
`
`5,189,821
`
`
`
`
`
`E TO || H.B.A
`AJO S N 3'S
`(13 Bd S
`
`BMW1056
`Page 4 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 4 of 17
`
`5,189.621
`
`F. G. 3
`
`INTERRUPTION
`
`4.
`
`TO MEASURE
`
`42
`
`TO DIFFERENTATE N-43
`
`TO CALCULATE
`MEMBERSHIP FUNCTION
`
`44
`
`TO ACCUMULATE VALUES
`OF MEMBERSHIP FUNCTION
`46
`
`45
`
`SED Y
`
`ES
`FREQUENCY (Fki)
`= ACCUMULATED VALUE/x
`
`49
`
`47
`
`L = L +
`
`L = O
`ACCUMULATED VAYE 5O
`
`C END D-48
`
`TO ESTMATE DRIVING
`CONDITIONS
`(DRIVING ENVIRONMENT)
`Dji = Mjk Fki
`
`5
`
`TO DETERMINE DRIVING
`CONDITION VARABLE
`
`52
`
`
`
`
`
`TO CHANGE DRIVING
`CHARACTERISTICS
`
`53
`
`BMW1056
`Page 5 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 5 of 17
`
`5,189.621
`
`
`
`QE E dS
`
`ETOIHBA
`
`
`NOI. LISOd
`}}\/E19
`d5)
`
`9NINEELLS
`
`GEBdS
`
`BMW1056
`Page 6 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 6 of 17
`
`5,189.621
`
`
`
`VEHICLE Ev
`
`6O (Km/h)
`
`dN / dt
`
`A6th
`
`BMW1056
`Page 7 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 7 of 17
`
`5,189.621
`
`F. G. 6
`
`
`
`
`
`4, 9th
`
`
`
`2o
`
`I/O
`CIRCUIT
`
`19
`
`
`
`REPULSION FORCE
`CONTROL UNIT
`
`
`
`BMW1056
`Page 8 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 8 of 17
`
`5,189.621
`
`
`
`/ ( 9 | -}
`
`
`
`
`
`BMW1056
`Page 9 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 9 of 17
`
`5,189.621
`
`F. G. 9
`
`
`
`74.7717/74727.4777,747/774777,227,777
`
`BMW1056
`Page 10 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 10 of 17
`
`5,189,621
`
`
`LINNTOY¥LNODZJ"
`
`ONIWILHYVdS
`NOWOSCN|—aa
`
`VOIOla
`
`
`
`
`TNd3YZ|LINA“OY8LNODaa30x04NOIS
`
`
`
`BOYO4NOISINdaY
`
`+J3dY4OS
`
`NOISS3ud30
`
`Cd:
`
`
`
`ONIWILHYVdS
`
`TOYLNOS
`
`
`
`a1VYNOILWYS7AIO”V
`
`LANL
`
`LINAONILLAS
`
`BMW1056
`Page 11 of 36
`
`BMW1056
`Page 11 of 36
`
`
`
`
`
`
`
`
`U.S. Patent
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 11 of 17
`
`5,189.621
`5,189,621
`
`(fW)
`
`
`
`SWIiL— ANIVA
`
`JONOILVWILS3
`“GNODSONIAIYG
`
`
`XIYLVW“ANOD
`JOLHOISM
`AONaNDSYSi
`(443)viva
`dIHSYSeWaw
`dSyunNsv3aw
`Q3ayuNSVAaW
`
`NOILONNA
`
`J1ISVIYVA
`
`JSDIHSA
`
`Aasad$
`
`+P/NP
`
`We
`
`WAv
`
`yV3a9
`
`ONIMAALS
`
`G33dS
`
`BMW1056
`Page 12 of 36
`
`BMW1056
`Page 12 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 12 of 17
`
`5,189,821
`
`FIG. I2
`
`a
`
`42
`55
`
`TO JUDGE
`DRIVING CONDITION
`
`SET INTIAL VALUES
`
`56
`
`WITHIN RANGE
`TO DIFFERENTIATE-43
`
`L = O
`ACCUMULATED wye
`TO CAL CULATE
`MEMBERSHIP FUNCTION N44 C E N D D
`
`OUTSDE
`RANGE
`
`C E N D D
`
`TO ACCUMULATE VALUES 45
`OF MEMBERSHIP FUNCTION
`46
`
`47
`
`L = L +
`
`<GdN,
`
`YES
`FREQUENCY (Fki)
`=ACCUMULATED VALUE / LX
`
`49
`
`5
`
`48
`
`
`
`
`
`TO ESTIMATE DRIVING
`CONDITIONS
`Dji = Mjk Fki
`
`TO DETERMINE DRIVING
`CONDITION VARABLE
`Djs mpx Dji
`
`TO CHANGE DRIVING
`CHARACTERISTICS
`
`BMW1056
`Page 13 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 13 of 17
`
`5,189.621
`
`F. G. 3
`
`TIMER
`INTERRUPTION
`
`4.
`
`TO MEASURE v,6ac-42
`
`57
`
`NO
`
`YES
`
`58
`
`NO SD
`
`A6ac(ti) = 9ac (ti) - 9ac(ti-)n- 43
`
`TO BLOCK 44
`
`56
`
`L = O
`ACCUMULATEDVAE
`
`BMW1056
`Page 14 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 14 of 17
`
`5,189.621
`
`F. G. 4A
`
`SPORTY
`(Sp)
`
`NORMAL
`(No)
`
`9
`CC
`S
`
`GENTLE
`(Ge)
`
`t
`
`t2 tip
`
`ti- ti
`
`O
`
`TME
`
`F. G. 4B
`
`GENTLE
`(Ge)
`
`
`
`O
`
`8 oc (ti) - 8 ac (ti-I)
`
`BMW1056
`Page 15 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 15 of 17
`
`5,189.621
`
`F. G. 4C
`
`Sp
`
`NO
`
`Ge
`
`S
`
`M
`
`L
`
`F. G. 4D
`
`
`
`Ge
`
`NO
`
`Sp
`
`BMW1056
`Page 16 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 16 of 17
`
`5,189.621
`
`F. G. 5A Go-------------
`
`SPORTY
`
`G
`G
`
`t
`
`F. G. 5B
`
`G
`G
`
`NORMA
`1.-----------
`
`t
`
`G
`
`tip to its
`
`to es is
`
`is
`
`is a
`
`ties with
`
`F. G. 5 C
`
`
`
`Go
`
`GENTLE
`
`t
`
`F. G. 5D
`
`
`
`G
`
`F. G. 5E
`
`Go- - - - - - -
`
`
`
`LINEAR
`
`BMW1056
`Page 17 of 36
`
`
`
`U.S. Patent
`
`Feb. 23, 1993
`
`Sheet 17 of 17
`
`5,189,621
`
`91914d
`
`Cd.
`
`L359yvL
`
`
`
`
`
`ONINILHYVdS31VvyNOILWYSTOO”
`
`
`
`
`
`
`
`LINATOYULNODLINNONILLAS
`
`08
`
`LINNTOYLNOS
`
`ANIWA
`
`c8
`
`
`
`NOLLVYSTS90V
`
`
`
`TINDW)31VY
`
`INANOLLDS)
`
`BMW1056
`Page 18 of 36
`
`BMW1056
`Page 18 of 36
`
`
`
`
`
`
`
`
`O
`
`15
`
`30
`
`1
`
`ELECTRONIC ENGINE CONTROL APPARATUS
`
`5
`
`This application is a continuation application of U.S.
`Ser. No. 07/233,209, filed Aug. 17, 1988 (now aban
`doned), which is a continuation-in-part application of
`U.S. Ser. No. 07/046,338, filed May 6, 1987 (now U.S.
`Pat. No. 4,853,720).
`BACKGROUND OF THE INVENTION
`The present invention relates to an electronic engine
`control apparatus, and more particularly to an elec
`tronic engine control apparatus whereby the driving
`characteristics of a vehicle are adjusted so as to match a
`driver's intent of how to drive the vehicle. Namely, the
`driving characteristics are adjusted in accordance with
`a driving environment of a vehicle and/or a driver's
`preference to the way he or she drives the vehicle.
`Conventional engine control electronic apparatus of
`commercial vehicles have the engine control character
`20
`istics adjusted to match the driving characteristics
`which are most preferred by vehicle users (drivers and
`fellow persons) under the driving (running) environ
`ments most frequently encountered. A so-called great
`est common divisor type matching has been adopted for
`25
`the engine control characteristics. The term "match
`ing' herein means adjusting an engine control unit so as
`to meet the requirements of a vehicle driver.
`In a known engine control apparatus for electroni
`cally controlling a throttle valve, a method of electroni
`cally controlling the sensitivity of opening a throttle
`valve relative to an amount of accelerator pedal depres
`sion in accordance with the accelerator pedal depres
`sion amount and speed is adopted. This method is dis
`closed in U.S. Pat. No. 4,597,049.
`35
`For the conventional electronic engine control, how
`ever, a method has been adopted whereby the control is
`conducted based on the engine conditions during sev
`eral engine strokes. In other words, the main function of
`such engine control is to control the engine in accor
`40
`dance with the measured results obtained for a short
`period. Thus, the engine control apparatus is not
`equipped with a unit for discriminating driving environ
`ments and drivers preferences in accordance with the
`measured results obtained for a long period of several
`45
`tens to hundreds engine strokes. Such a long measuring
`period for discrimination is herein defined as an evalua
`tion period.
`The above-mentioned greatest common divisor type
`matching does not meet all driving environments and
`50
`various driver's preferences (feelings). Also, the above
`mentioned method whereby a change in the sensitivity
`of opening a throttle valve relative to an amount of
`acceleration pedal depression, as is the case of the
`above-described U.S. Patent, poses a problem when a
`55
`sensitivity change occurs because of racing or double
`clutching.
`SUMMARY OF THE INVENTION
`It is an object of the present invention to satisfy vehi:
`cle users concerned with driveability and provide an
`engine control apparatus which is arranged to adjust or
`change the driving characteristics in accordance with
`the driver's intent of how to drive a vehicle, i.e., in
`accordance with the driving environment and/or driv
`65
`er's preference.
`The above object can be achieved by 1.) measuring
`the driver action for a predetermined period with the
`
`5, 189, 621
`2
`aid of a computer built in the engine control apparatus,
`2.) discriminating and categorizing the driving environ
`ments and/or the driver's preferences based on the
`measured results, and 3.) causing the engine control
`characteristics to match the driving environments and
`/or the driver's preferences in accordance with the
`categorized results. To this end, the driving conditions
`and the driver's action are measured a plurality of times
`for a period longer than several cycles, and the mea
`sured results are calculated and processed.
`Generally, a vehicle is driven by a driver in accor
`dance with the driver's intent of how to drive it under
`a given driving environment. The driver's preference
`being included in such an intent. On the other hand, the
`driving conditions of a vehicle reflect the driving envi
`ronment and driver's preference. In consideration of
`these points, according to the present invention, the
`category of environments and preferences under differ
`ent driving conditions is identified to determine the
`control parameters in conformity with the identified
`category. The measuring variables for discriminating
`the driver's intent, i.e., driving environments (i.e., run
`ning environments) and/or driver's preferences, include
`accelerator pedal position (angle) and depression speed,
`vehicle velocity, engine speed (revolutions of engine
`per unit time), acceleration rate, gear position (manual
`transmission (MT) vehicle), shift operation, steering
`speed, break occurrence frequency and the like.
`Included in these measuring variables are the prefer
`ences of a driver as to how to drive the vehicle in a
`given driving environment.
`The qualitative and empirical analysis on the influ
`ence of driving environments and driver's preferences
`upon the measuring parameters presented the results as
`shown in FIGS. 1A and 1B. The "driving conditions”
`are herein used to define the state where a driver drives
`a vehicle in a given driving environment in accordance
`with his or her preference.
`Therefore, the driving conditions represent the driv
`ing environment, the driver's preference, or the combi
`nation thereof as the case may be. Thus, as the candi
`dates for the driving condition variables to be described
`later, variables representative of the driving environ
`ments and variables representative of the driver's pref.
`erences are used.
`-
`According to the present invention, in order to cate
`gorize the driving conditions in accordance with the
`analyzed results, membership functions of the Fuzzy
`theory for example are used with respect to the measur
`ing variables. A membership function which is used in
`the Fuzzy theory expresses the degree of participation
`of an element (variable) to a concept by a numeral value
`of 0 to 1.
`Each of the measured variables during a predeter
`mined period (evaluation period) is classified into three
`types, large, middle or small by using membership func
`tions. The occurrence frequencies of respective types
`are also counted. With this method, the driving condi
`tions and the driver actions during the evaluation period
`can be statistically recognized. The results give basic
`data. The data is the basis for discriminating the driving
`conditions.
`The occurrence frequency of each measuring vari
`able is converted into a driving condition through con
`version matrix calculation so that the occurrence proba
`bility of each driving condition can be obtained. If the
`maximum occurrence probability of a driving condition
`exceeds a standard level, this maximum occurrence
`
`BMW1056
`Page 19 of 36
`
`
`
`O
`
`15
`
`5,189,821
`4.
`3.
`FIG. 4 illustrates the concepts of the engine control
`probability can be used as an estimated value of the
`process contents shown in FIG. 3;
`driving condition during the evaluation period.
`FIGS. 5A to 5F show examples of the membership
`The conversion matrix values for calculating the
`functions shown in FIG. 4;
`driving condition from the occurrence frequency data
`FIG. 6 shows an example of a control mechanism for
`are pre-set in accordance with the occurrence probabili
`controlling the depression repulsion force of an acceler
`ties of each driving condition corresponding to the
`ator pedal;
`large, middle and small frequencies of each measuring
`FIG. 7 illustrates the concepts of the engine control
`variable. Like the conversion matrix values, occurrence
`process contents according to a second embodiment of
`probability values may be set to a value of 1 or 0 de
`the invention;
`pending upon whether the measuring variable contrib
`FIG. 8 shows another example of a control mecha
`utes to estimating the driving condition. Weighted val
`nism for controlling the depression repulsion force of an
`ues may be used for those measuring variables which
`accelerator pedal;
`contribute more to estimate the driving conditions than
`FIG. 9 illustrates a case where the present invention
`the other measuring variables or classification types.
`is applied to an electronic throttle valve;
`Desired conversion matrix values enable the system
`FIGS. 10A and 10B are functional block diagrams
`to positively utilize the characteristics of respective
`illustrating the controls made with respect to FIGS. 6
`measuring variables and discriminate the driving condi
`and 8 and FIG. 9, respectively;
`tions. In addition, by adding the estimated value of a
`FIG. 11 illustrates the concepts of the engine control
`driving condition obtained for each measuring variable
`process contents according to a third embodiment of
`20
`to a proper measuring variable, the driving condition
`the invention;
`can be discriminated more accurately.
`FIG. 12 is a flow chart showing the engine control
`After identifying a driving condition, a correspond
`operation according to a fourth embodiment of the
`ing control method is selected to change engine control
`invention;
`parameters. Ignition timings, amount of injected fuel
`FIG. 13 is a flow chart showing an example of the
`25
`and the like are chosen as the control parameters.
`driving condition discrimination unit shown in FIG. 12;
`Apart from the above, according to the studied and
`FIGS. 14A to 14D illustrate the process contents
`categorized results, the driver's intents/preferences
`shown in FIG. 12;
`include three modes: (1) Sporty, (2) Normal and (3)
`FIGS. 15A to 15E show various patterns of target
`Gentle. There is a tendency that where a vehicle
`acceleration rate characteristics; and
`30
`matched to the average preference is used, a driver with
`FIG. 16 is a block diagram for explaining the engine
`"Sporty" preference more abruptly depresses the accel
`control operation according to a fifth embodiment of
`erator pedal and brake pedal than a driver with "Nor
`the invention.
`mal' preference, and a driver with "Gentle' preference
`DESCRIPTION OF THE PREFERRED
`does so more gently than a driver with "Normal prefer
`EMBODIMENTS
`eCe.
`Such tendency is not always present, but may change
`The embodiments of the electronic engine control
`with the road conditions and the fellow persons. Even if
`apparatus according to the present invention will be
`the driver's intent/preference are expressed indefinitely
`described below with reference to the accompanying
`for a time, by time sequentially measuring and evaluat
`drawings.
`ing the motions of accelerator pedal and brake pedal it
`FIG. 2 shows the structure of a typical example of the
`become possible to categorize the driver's intent into
`electronic engine control apparatus. In this embodi
`one of the above-mentioned three modes. In particular,
`ment, the driving environments are discriminated based
`the motions of both the accelerator and brake pedals are
`on the measured values of each measuring variable
`continuously measured at the start, acceleration, con
`using the structure shown in FIG. 2. The engine control
`45
`stant speed running, reduced speed and stop, and other
`parameter value is changed in accordance with the
`conditions, and the measured results are categorized
`discrimination result.
`using membership functions.
`In the figure, reference numeral 1 denotes an acceler
`According to the categorized results, the sensitivity
`ator pedal, 2 a brake pedal, and 3 denotes a steering
`of depressing an accelerator pedal for example is made
`wheel. Reference numerals 4 and 5 denote rheostats
`SO
`high for a driver with "Sporty” preference, and is low
`whose sliding contacts move by the amounts corre
`ered for a driver with "Gentle' preference.
`sponding to the depression amounts of the accelerator
`The engine characteristics are controlled in accor
`and brake pedals 1 and 2, respectively. An output volt
`age 6b of the rheostat 5 represents the depression angle
`dance with the driving environments and/or the driv
`er's preferences as described above so that comfortable
`of the brake pedal, and an output voltage 6th of the
`55
`driveability and riding comfort are assured as well as a
`rheostat 4 represents the depression angle of the accel
`vehicle matched to the driver's preference.
`erator pedal, i.e., the opening angle of a throttle valve 7.
`Similarly, reference numeral 6 denotes a rheostat whose
`BRIEF DESCRIPTION OF THE DRAWINGS
`sliding contact moves by the amount corresponding to
`the rotary angle of the steering wheel which is repre
`FIGS. 1A and 1B show the relation between measur
`ing variables and the driving environments and driver's
`sented by an output voltage 6 of the rheostat 6. In the
`preferences;
`embodiment shown in FIG. 2, the throttle valve 7
`FIG. 2 schematically shows the structure of a typical
`which mechanically couples to the accelerator pedal 1
`example of an engine control apparatus according to the
`is used by way of example. Reference numeral 8 denotes
`a gear position detector which detects the gear position
`present invention;
`of a shift lever, the output signal Gp therefrom being
`FIG. 3 is a flow chart for explaining the engine con
`representative of the gear position. Reference numeral 9
`trol operation according to a first embodiment of the
`denotes an angular position sensor which is mounted
`present invention;
`
`35
`
`BMW1056
`Page 20 of 36
`
`
`
`10
`
`15
`
`25
`
`5,189.621
`6
`5
`speed N and A6th are small, and the brake operation
`facing a gear 34 interlocking with a crank shaft 32 and
`frequency is high.
`outputs a position signal Ps at a predetermined number
`There are less junctions on a suburban road than on
`of rotations of the gear, e.g., every one rotation. Refer
`an urban street so a vehicle generally runs near a limited
`ence numeral 12 denotes a vehicle speed sensor whose
`output signal v represents vehicle speed. Reference
`speed. Therefore, if the vehicle speed maintains a lim
`ited speed and the frequency and range of steering oper
`numeral 13 denotes an air flow meter which measures
`ation is small, it can be discriminated as a suburban road
`the air flow amount into an intake manifold. Reference
`running.
`numeral 14 denotes a fuel injection valve, and 15 an
`There are many junctions on an urban street. There
`ignition plug. Reference numeral 10 denotes a control
`fore, the operation frequency of both the brake pedal
`unit which processes inputted signals v, Ps, 8b, 6th, 8st,
`and steering wheel is high. The vehicle speed v and
`Gp and Qa and supplies control signals to the fuel injec
`tion valve 14 and the ignition plug 15. The control unit
`acceleration rate AN (=dN/dt) take middle values.
`10 includes an I/O circuit 20, a central processing unit
`There are many curves on a mountain road so accel
`eration and deceleration are repeated and the steering
`(CPU) 22, a read-only memory (ROM) 24, a random
`wheel is frequently operated. Therefore, in general, if
`access memory (RAM) 26, and amplifiers 28 and 30.
`A6th and dN/dt are large, if the frequency of change in
`The measured values of each measuring variable are
`time sequentially processed by the control unit 10. The
`depressing the brake and accelerator pedals is high, and
`if the frequency of operating the steering wheel is high,
`process contents differentiate representative measured
`values. The differentiated values of representative mea
`it can be discriminated as a mountain road running.
`20
`suring variables are given by the following equations
`In this embodiment, for example, if the vehicle speed
`v is high, it is discriminated as a highway running. If the
`where t, t-i-1, t+2, ... are sample timings at the mea
`vehicle speed v is low and the change rate dN/dt of the
`surement.
`engine speed is small, it is discriminated as a congested
`road running. If both A6th and dN/dt are large, it is
`discriminated as a mountain road running.
`The difference between the running conditions on a
`suburban road and an urban street is small so the dis
`crimination therebetween is less significant. Therefore,
`in this embodiment, only the urban street is to be dis
`criminated. The urban street running is discriminated if
`the vehicle speed v and acceleration rate AN take mid
`dle values.
`The suburban road and the urban street may be cate
`gorized and discriminated dependent upon the environ
`ments and countries where a vehicle is used. It is not
`necessary to discriminate the mountain road running in
`the district where there is no mountain. Categorization
`of driving environments and measuring variables for
`such categorization are properly selected dependent
`upon a particular environment where a vehicle is used.
`In this embodiment, it is therefore assumed that the
`driving environments are categorized into the above
`mentioned four types and discriminated based on, the
`vehicle speed v, engine speed change rate (acceleration
`rate) dN/dt, and throttle opening degree change rate
`A6th among various measuring variables. Thus, the
`membership functions and the like for the gear position
`and steering speed are omitted in FIG. 4. FIG. 4 illus
`trates the functions performed by blocks 44 to 52 which
`will be described below.
`The program shown in FIG. 3 starts upon a timer
`interruption at block 41. The measuring variables repre
`sentative of the driving environments and the driver's
`preference are measured by various sensors at block 42.
`Measured signals Ps, v and 8th are inputted to the I/O
`circuit 20. The engine speed N is calculated using Ps.
`The measuring variables whose change with time is
`requested to be obtained, (i.e., N and 6th) are subjected
`to differentiation at block 43 to obtain the differentiated
`values (i.e., AN (= dN/dt) and A8th). The time-sequen
`tial values v, AN and A6th are categorized for each
`measuring variable, preferably into large (L), middle
`(M) or small (S) groups, by using the membership func
`tions (FIG. 4, FIGS.5A, 5B and 5D) at block 44. The
`probability of categorization of each measured value
`into a particular group is given by a membership func
`tion value.
`
`30
`where N represents an engine speed (revolutions of
`engine per unit time) which is given, e.g., by the number
`of inputted signals Ps's during a unit time.
`The control unit 10 causes the measured values, in
`35
`cluding the differentiated values, to be processed with
`membership functions of the Fuzzy theory so that the
`occurrence frequencies of each measuring variable are
`obtained. The driving environment is estimated by sub
`jecting the occurrence frequencies to the conversion
`matrix. Then, the control parameters, changed in accor
`dance with the estimated value of the driving environ
`ment are obtained. The fuel injection amount and the
`spark timings by way of example are adjusted in accor
`dance with the changed control parameters.
`45
`The engine control operation of the control unit 10
`will be described with reference to the flow chart of
`FIG. 3. FIG. 4 shows the engine control concepts.
`FIGS. 5A to SF shows examples of membership func
`tions.
`50
`The membership functions shown in FIGS. 5A to 5F
`are the examples for the membership functions shown in
`FIG. 4, obtained using a certain type of a vehicle.
`For example, by categorizing the driving environ
`ments into a congested road, an urban street, a suburban
`55
`road, a mountain road, and a highway, the respective
`running features can be grasped.
`The feature of each driving environment will be de
`scribed below which is shown in FIGS. 1A and 1B, and
`FIGS. SA to SF.
`Generally, the vehicle speed v is high and gear posi
`tion is changed rarely during a highway running.
`Therefore, if the vehicle speed is high and a top gear is
`used with a low frequency of a shift operation, it can be
`discriminated as a highway running.
`65
`Generally, a forward operation and a stop operation
`are repeated on a congested road. Therefore, the vehi
`cle speed is low, the rate of change dN/dt of the engine
`
`BMW1056
`Page 21 of 36
`
`
`
`5, 189,821
`8
`7
`for Mt environment is the largest among the calculated
`The membership function values at each timer inter
`ruption are accumulated for each measuring variable at
`values
`block 45.
`At block 46, the number 1 of timer interruptions is
`compared with a predetermined number lx to check if
`the former becomes equal to the latter or not.
`The timer interruption period multiplied by lx is the
`period for evaluating the driving environments. If the
`timer interruption number 1 is smaller than lx, the value
`1 is incremented by 1 at block 47 to follow block 48.
`If the timer interruption number 1 becomes lx, the
`accumulated value of membership function values for
`each measuring variable is divided by lx at block 49 to
`obtain occurrence frequencies Fki of each of the large,
`middle and small groups (k) for each measuring variable
`(i).
`At block 50, the timer interruption number 1 and the
`accumulated values are initialized.
`The following equation (1) is used to perform matrix
`calculation for estimating the driving environment at
`block 51.
`
`O
`
`15
`
`20
`
`X Di
`
`for CG, St, Mt, and Hw so that the value Dj is deter
`mined as the driving condition variable for the moun
`tain road environment (Mt). Thus, the driving condition
`is discriminated as the mountain road (Mt). In this case,
`the value Dj may be determined as the driving condi
`tion variable on condition that it exceeds a predeter
`mined value, e.g., 2.0.
`The driving characteristics are changed in accor
`dance with the determined driving condition variable
`Dj at block 53.
`In the example shown in FIG. 4, the values of Dji
`obtained only from the vehicle speed v and the values of
`Dji obtained from the vehicle speed v, the engine speed
`change rate dN/dt and the throttle valve opening de
`gree change rate A6th are shown for the comparison
`sake. The latter values are shown more effective for
`determining the value Dj.
`In order to change the driving characteristics, the
`following known elementary technology may be uti
`lized. In the case where the accelerator pedal and the
`throttle valve are directly coupled via a linkage mecha
`nism, a method of changing a depression repulsion force
`of the accelerator pedal, a method of changing an air
`fuel ratio (A/F) by changing the coefficient (e.g., km) of
`the equation Ti=kM-kQa/N for calculating the injected
`fuel amount, and a method of changing the ignition or
`spark timings are utilized. In the case where a vehicle
`adopting an electronic throttle is used, a method of
`changing the transmission characteristic from the accel
`erator pedal to the throttle valve (as disclosed, e.g., in
`U.S. Pat. No. 4,597,049), a method of servo-controlling
`so as to obtain a desired acceleration rate pattern of a
`vehicle which has been obtained based on the motion of
`the accelerator pedal are utilized. For an A/T vehicle,
`a method of changing the characteristic of a torque
`converter is utilized.
`In this embodiment, the method of changing the igni
`tion timings was chosen to give the following descrip
`tion of changing the driving characteristics.
`The ignition or spark timings 6ad are determined by
`the following equation (3) at block 53.
`
`6.ad = f(Di, Dimax, 6a)
`(3)
`The spark timing 6ad at the maximum torque is deter
`mined by the above equation based on a deviation of the
`value Dj from a value Djmax which is a value Djcorre
`sponding to the spark timing 6ad at the maximum
`torque. In the case where the range of changing the
`spark timings can be narrowed, the following equation
`(4) is used as an approximate equation.
`
`(1)
`
`25
`
`Di = i Mik. Fki
`where
`Fkis occurrence frequency of a measuring variable;
`Mjk=conversion matrix for conversion from occur
`rence frequency to driving condition: weight for
`30
`selecting driving condition;
`Dji=value indicating a possibility of each driving
`condition (environment) (j);
`i=number of measuring variable: vehicle speed (v),
`dN/dt, 6th, A6th, Gear, engine speed (n), etc.;
`35
`j= number of driving condition (environment): con
`gestion (CG), urban street (ST), mountain road
`(Mt), and highway (HW); and
`k= group number of measured value: small (S), mid
`dle (M) and large (L).
`The measuring variables i are dN/dt and A6th in this
`embodiment.
`The driving environment is determined during the
`evaluation period as in the following. The values of Dji
`for necessary measuring variables (i) are added to
`45
`gether. Namely, the values of Dji for v, dN/dt and A6th
`are added together to obtain
`
`X Dji
`
`SO
`
`for each environment. The driving environment is de
`termined as the driving condition (j) which takes a larg
`est value among the added results Dji. The maximum
`55
`value Dj is given by the following equation (2).
`
`Di = maxx, Dji
`j
`
`(2)
`
`The value Djis called as a driving condition variable.
`In particular, in the example shown in FIG. 4, the
`value of
`
`Dji
`
`- D -
`6ad = Dimax