United States Patent [19]
`Peters
`
`IlllllllllllllIllll|||llllllllllllllllllllllllllllillllllllllllllllllllllil
`USOO5592614A
`[11] Patent Number:
`5,592,614
`[45] Date of Patent:
`Jan. 7 1997
`
`9
`
`[54] FAULT IDENTIFICATION SYSTEM 7
`
`[57]
`
`ABSTRACT
`
`[75] Inventor: Michael Peters, Mdnch?ster England
`’
`[73] Assignee; Gem-ad Limited, London, England
`_
`
`_
`[211 App 1' No" 754,899
`[22] Filed:
`Sep. 4, 1991
`[30]
`Foreign Application Priority Data
`
`United Kingdom ................. .. 9019718
`
`Sep. 8, 1990 [GB]
`6
`[51] Int. C1. .................................................... .. G06F 11/22
`[52] U-S- 0
`395/183-02
`[58] Field of Search ................................ .. 371/251, 15.1;
`395/61, 916, 183-02
`
`[56]
`
`References Cited
`U_S_ PATENT DOCUMENTS
`
`4,267,569
`4,847,795
`
`364/431
`5/1981 Nygaard et a1.
`7/ 1989 B?k?l' 6i‘. a1. .......................... .. 364/579
`
`Primary Examiner—Robert W. Beausoliel, Jr.
`Assistant Examiner—S. Elmore
`Attorney, Agent, or Firm— Nixon & Vanderhye RC.
`
`A system for identifying faults in an electrical circuit such as
`a circuit of a road vehicle. The electrical circuit is intended
`to provide a plurality of outputs such as lamp illumination in
`response to predetermined circuit conditions. Data describ~
`ing individual components of the electrical circuit and the
`interconnections therebetween is stored. Data identifying a
`fault symptom is input to the system by a technician to
`indicate that for one said predetermined set of circuit con
`ditions the intended output is not provided. Stored data is
`accessed which describes selected components which if
`faulty could be the cause of the fault symptom. Values of
`electrical variables to be expected at predetermined mea_
`surement points in a sub-circuit de?ned by the selected
`components and the connections therebetween are calcu
`lated. Measurement points are selected at which points the
`electrical variables are to be measured and the technician is
`instructed to selectively input measurement data indicating
`the electrical variables measured at the selected measure
`ment points. The rnput measurement data is compared with
`the calculated electrical variables and faulty components are
`identi?ed in accordance with that Comparison data_
`
`11 Claims, 15 Drawing Sheets
`
`Obtain target
`101
`\ vehicle build
`specification
`
`Decide which
`102
`\ vehlcl‘= SXSlP'HS
`need testing
`
`SPECIFICATION
`~
`
`.
`
`_/-c|ncu|T
`
`Obtain symptoms
`103
`\ of vehicle
`malfunction
`
`Determine possible
`104
`\\ causes of
`symptoms
`
`Final tests to
`105
`\ confim or deny
`each fault
`
`\‘SYMPTOMS
`
`FAULTS
`
`‘
`
`-
`
`TESTS
`
`“Select and perform k ‘
`106
`\ best next lest
`stop
`E
`
`Evaluate test
`107
`\\ results
`
`10B
`
`Issue
`Diagnosis
`
`— RESULT
`
`FAULTS
`
`IPR2013-00417 - Ex. 1021
`Toyota Motor Corp., Petitioner
`1
`
`

`
`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 1 of 15
`
`5,592,614
`
`Obtain target
`101
`\ vehicle build
`specification
`
`SPECIFICATION
`
`~
`
`__/ —CIRCUIT
`
`\ "SYMPTOMS
`
`r
`
`FAULTS
`
`TESTS
`
`——
`
`~RESULT
`
`FAULTS
`
`102
`Decide \ivhich
`\ vehidcle szistems
`
`es mg
`
`nee
`
`Obtain symptoms
`103
`\ of vehicle
`malfunction
`
`Determine possible
`\ causes of
`symptoms
`
`Final tests to
`105
`\ confim or deny
`each fault
`
`Select and ‘perform
`106
`\ best next test
`stop
`
`Evaluate test
`107
`\ results
`
`Issue
`Diagnosis
`
`FIG.1
`
`
`
`2
`
`

`
`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 2 0f 15
`
`5,592,614
`
`295cm
`650
`
`wmEuoE
`
`
`
`EmQ mBEm>
`
`cEoEzoQ.
`
`836m B
`mmmcoqwmm
`
`20Em>
`
`mmtmzO
`
`méUI
`
`
`
`
`
`2:82 8.60588 202%
`
`ma 7N2.
`
`5-.
`
`
`
`3
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 3 0f 15
`
`5,592,614
`
`....
`
`\\,\
`
`\\
`
`\wé n 25.
`
`X55255 " 2E
`
`Non
`
`>EDXD4 n EFF
`
`mdI
`
`Pom >
`
`Om m?rr - Dom
`200 mu
`
`/
`
`
`
`4
`
`

`
`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 4 of 15
`
`5,592,614
`
`Be
`
`Now
`
`new
`
`ecu
`
`new
`
`new
`
`new
`
`¢.o_m
`
`
`
`
`
`323%m>Eo-._mm_.=s-¢w:_m>zo.mw_s_mz<E
`
`
`
`
`
`..m.S.IQsv.QIE-O2Eon».ZOPODQMQ
`
`
`
`_uoE._=:oom>_mo-oz<:.:._o_mo:_m>mo_w-m>_mn
`
`
`
`
`
`
`
`.H.S<lmQEat2oE.._.mmm<
`
`
`
`
`
`..m.Sm-2<m<7m0-mo.QI.TO2EotZOPODQMQ
`
`
`
`Uw_u..mom=um>_mo-oz<:-.Em._m....m>mo_m-m>_mn
`
`
`
`
`
`
`
`..m.Smhm.>.:E.<U.2<n:\fiEotZOPODQNQ
`
`
`
`
`
`
`
`
`
`_umU._mom_Umutt._.OZw:_m>»m>._<»<o
`
`...~..S<lmQEoh2o:mmmm<
`
`
`
`UmE.._.Eoon_m._.._..u_m:_m>._.w>._<._.<U
`
`
`
`
`
`
`
`
`
`Um_u._mom=uZ<n_<..Eooxomw:_m>.EvE<s_=<
`
`
`
`
`
`
`
`.H.5<.¢mQEatZOPODQMQ
`
`
`
`UwE..=:ouZ<n_<_..m:_m>EvE<s_Al
`
`
`
`
`
`EmmaEohm2o:mmmm<
`
`
`
`5
`
`
`
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 5 of 1s
`
`5,592,614
`
`Got-mamm
`
`UZ_mmm>mE
`
`m6;
`
`Q‘
`O
`In
`
`new New
`
`Pom
`
`UZEUDH-(ZEMFXU
`UEZOEDZOO-EE
`
`
`
`
`
`UZFIQJh-(ZIP-b: JOmPZOO mQDmO
`
`
`
`6
`
`

`
`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 6 of 15
`
`5,592,614
`
`9mmmi
`
`
`
`mEw_._GOu_-._.ZOm..._
`
`
`
`
`
`OHM.mE.u_._zm:._.Em<N<_._
`
`Sm
`
`
`
`
`
`%.mm..9..._m_>>30>on.®Z_._.I0_u_._<zmmEm_+0flomamm:oE>>\
`
`
`
`
`
`
`
`_nm5m._mmm._._._0_._n_<m_._J,
`
`6
`
`
`
` 9m:._o_._wou_-..._<mm
`
`
`
`m.:._..o_._0z_mmm>mm
`
`
`
`m:._w.._wz_v_m<._
`
`
`
`
`
`o_m,..o_Nm..:_o_._no.6
`
`92“.
`
`
`
`mszfizoo89¢.558:8mzoswam
`
`
`
`7
`
`

`
`U
`
`a
`
`5,592,614
`
`S.2:.
`
`\....:n_
`
`U
`
`am...
`
`n.%mzmoozM.m~.ozm.o§-wm_;M 9S-mm;>
`SmoozWmmooE<o.3-woo_oDm E-moo_o
`7,Wmnoz<.3-moo_o
`Dwnzm.o&._.wm_>>
`
`S.~5-.:m_._mw.wzE.~..3._w:w
`
`..\\.\\\\\\\\
`
`:+jm:w nwu.m>Emoa.zmo:H.N.mzxmooz MW.2_a.:.3._m:m/7u...
`
`
`
`
`mu.m::.moozn Wm>_»<owz.zmo:% zmo:
`
`
`
`munm_mm<:o.zE.uz_m<o-zmo_.3._m:mlull:oz_m<o-zmo_._-jm:mm_mm<_._o
`
`Mmw._._oE.zoo-zmo:
`
`
`
`..mEucmaecoo«mpsmaco§Smo.=o:&.mo.Eum<m«mac:ABEmmm<W_mu.Eom\&
`
`
`
`
`
`tonmomtmfic*0_.E:o..mmQ:m@£mnmmmxxé
`
`
`
`
`
`
`
`
`ummnoz._<z_sEm»mo_>mnzofiomzzoo
`
`
`
`8
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 8 of 15
`
`5,592,614
`
`wd‘I
`
`A @E. Q? vow/WA \w .
`
`\ \ A ..
`
`Pom
`
`
`
`9
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 9 of 15
`
`5,592,614
`
`2N5 W EEcoo \
`060 E Q
`
`
`
`mammtzmmnw 2052 , k
`>>oc¢oEmcoo
`
`\
`
`
`?nznomwmmob .
`N8
`56769200
`
`EEcoo ~ .
`
`?nzaomwzgmu
`
`QmEE=m
`
`i EEcoo
`a“ 7
`
`
`
`hQOOU-UCZEMDV .
`
`0.01
`
`
`
`
`
`3.5505 zzmm E wocmgcoo 85865!
`
`282w
`982w 4
`
`Fem
`
`mEwmmbg
`3532a
`
`mczwmxm
`
`56m 6
`
`
`
`10
`
`

`
`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 10 of 15
`
`5,592,614
`
`
`
`
`
`
`
`._.5om_ozmaoan>mEmohomzzooIU._._>>w>mm.E:oo.Em._._.zOm_u_:_m_mm<:o.z_._
`
`
`
`
`
`
`
`Sow
`
`
`
`mzamm£.E....mQ
`
`
`
`to20:52:o:.>>wEu:>mmE:oomoonw.mmczmmm<n_
`
`
`
`
`
`WEDNKOs..E.~WEflmmol
`
`
`
`
`
`
`
`.._:o...__ozmao2.>mEmopomzzoo_._o.:>>m>mm.E:ooEm._._.zomu_E_-zE
`
`
`
`
`
`
`
`
`
`
`
`.._._DOm=Ozm._o2.>9:EOHUMZZOUmn_m-mom:ommo..<._om_>mm.E<mET2_n_
`
`
`
`
`
`
`
`
`
`
`
`.::om_oZmn_O2.>mEEOHONZZOOm>_.F<wmz>mmF<mE_.-z_n_
`
`
`
`
`
`
`
`
`
`...5om_ozmmo2.>mES8.:mmmo_.:m.._E$-mm=s
`
`
`
`.::om_ozmmo3>55S8.:mmmm:._m:Em..mm_>>
`
`
`
`
`
`mtzmkmE.E:E.6m.Eo:oQEoU$8301
`
`
`
`
`
`::om_ozmao2.>mEmohomzzoom>Emo._>mmt.<mE72?.
`
`
`
`
`
`
`
`
`
`::om_ozmmo2.>mEmopomzzoomn_m-a<o._mo»<..om_>mm.E<mE_.-z_n_
`
`
`
`
`
`
`
`
`
`
`
`::om_uzmao2.>mEEOEOMZZOU.=._<._>mmE:oo.F._m._Emm._.DO-Z_n_
`
`
`
`
`
`
`
`
`
`
`
`::om_ozmmo2.>9:mopomzzoun=>_<a_>mm._.m:ooEm...Emmzz_-zE
`
`
`
`
`
`
`
`
`
`
`
`::om_uzmao2.>mEn_s_<._>mmE=oo.Em._
`
`
`
`
`
`
`
`._._D0_...=0zmmo2.>mES8..._mmmo:._m_._:_o~.mm=s
`
`
`
`
`
`
`
`._.__..o_..__ozmao2.>mES8-_._mmmo:._m::_&.mm=s
`
`
`
`
`
`
`
`::om_ozmao2.>mES8-_._mamass...:_wmm=s
`
`
`
`
`
`
`
`.::om_ozmao2.>mES8._._mmmo:._m_._:_m.mm=s
`
`
`
`
`
`.:.o_u_
`
`11
`
`
`
`11
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 11 of 15
`
`5,592,614
`
`Q2<z>n A V éocmzkonci
`293182
`oz_zow<mm
`
`2.:
`
`ma:
`
`55. Na 2. -
`
`Now
`
`o§<z>a
`
`Jun-OE
`
`
`o§<z>o 8.55am:
`/w 3.
`.6528 $53525 5m?
`
`
`
`232 5:20:06 52.
`
`
`
`12
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 12 of 15
`
`5,592,614
`
`EEEEQiéEEBg
`
`
`
`52 2:9 cozmmmaoi
`
`EEEEEQAQEEBK
`
`HV~HV
`
`asénéiéézgz
`
`223525
`
`:ééucmisé?nz
`
`\8 Bags.
`
`o?EEEEg
`
`HVHHV
`
`b=m> hem tm§ :23.
`
`
`
`
`bmcmae?mm 9m
`
`$.01
`
`
`
`
`
`‘.2 2:9. cozmmmaoi
`
`lllllillllllllllllllllllllllllllll'llllllll'll
`
`
`
`5* 2:9 cozmmuno?
`
`E52?
`HV>HV
`
`08,52»;
`
`225mg
`
`HV>HV
`
`55%.“;
`
`
`
`13
`
`

`
`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 13 of 15
`
`5,592,614
`
`mo:Sm_9BSoEmcoaEoo
`
`mcozocé__:w=:o.__oSn
`
`
`
`
`Bmumm._<zmm._km
`
`0M.mgoOM..\mm:m__..E.Jbgml
`
`
`
`
`
`:o=ocE_mE._.=>>=:o._0
`
`
`m.EmEm<m®C\QEO0m:E.5XNNDK
`
`
`
`co_~o_._E_mE><_2::E_0
`
`Nu;.xEs.3ES
`
`
`
`mmcm.mo:m._m_2cm_mmo
`‘Vam»xvm.vEm..m>m
`
`,,cm/>om5oomc_
`
`xm_aE_m\Emsmu”.McmEmSmmm_>_
`
`s..
`.
`
`-mo:Sm_98SoEmcon_Eo0»..wW¢«¢«¢«o«+«¢«¢«M«M
`
`
`
`mcozocé__=m~_3U.__OSn
`
`
`
`
`
`:o=o:E_mE><_2EEO
`
`
`
`
`
`co=o::_=mE._.=>>=:o.__O
`
`25:...
`
`14
`
`
`
`14
`
`
`
`
`
`
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 14 0f 15
`
`5,592,614
`
`FIG/M
`
`
`
`15
`
`

`
`US. Patent
`
`Jan. 7, 1997
`
`Sheet 15 of 15
`
`5,592,614
`
`8
`
`mm
`
`:3
`23
`
`0
`
`op
`
`S
`
`ND
`
`\ \ m
`
`33
`
`NC;
`
`m3
`
`3
`
`13
`
`mu
`
`N3
`
`3
`
`D
`
`
`
`mJOmE>w Also-GO.“
`
`E5.‘ 5 muim»
`252 o
`
`
`
`2526
`.650 2x24 4
`
`=92; 9 EEQER
`
`cosomccoO
`
`DZwmumJ
`
`A a
`2
`. 2
`
`m3
`.33
`
`E820 QBEQR
`.2526 Q
`
`Q3
`
`juoowhcg?mh: n U.
`
`8m 2550 n .
`ncnokmxomm u a
`ucaohmmhom u @
`
`:E 660500 m
`
`
`
`QES bmzmm @
`
`co=>>m @
`wt; »
`
`EEEEP 0
`
`
`
`16
`
`

`
`5,592,614
`
`1
`FAULT IDENTIFICATION SYSTEM
`
`BACKGROUND OF THE INVENTION
`
`5
`
`3O
`
`35
`
`40
`
`45
`
`50
`
`55
`
`1. Field of the Invention
`The present invention relates to a system for identifying
`faults in an electrical circuit such as the electrical circuit of
`a road vehicle.
`2. Related Art
`Modern road vehicles have complex electrical circuits
`performing many functions such as engine, climate, instru
`mentation and lighting control. Such an electrical circuit
`provides many user perceivable outputs, and malfunctions
`of these outputs are easy to identify in a general manner, e. g.
`“the internal lights do not work”, or “the engine runs in an
`irregular manner”. The faults which cause these malfunc
`tions are often simple matters, e.g. a connector going open
`circuit, or burn-out of a component. Generally speaking
`however it is not easy to diagnose the underlying fault from
`the symptom that the perceived output malfunction repre
`sents due to the extensive interconnection between circuits
`supplying different output devices and the complex nature of
`some of the components used, e.g. wiring harnesses and
`microprocessors.
`Fault identi?cation has traditionally been the responsibil
`ity of skilled service technicians provided with simple
`voltage and current monitoring tools and circuit diagrams of
`the vehicle in question. As the complexity of vehicle wiring
`has increased, and the differences between variations of the
`same vehicle model have increased, it has become progres
`sively more di?icult for technicians to diagnose electrical
`faults.
`Diagnostic equipment is known which can be plugged
`into an appropriate socket provided in a motor vehicle. The
`socket is connected to sensors mounted on the vehicle, and
`these sensors deliver output signals to the diagnostic equip
`ment which handles any appropriate signal processing. This
`approach has two major lirrritations however. Firstly, the
`equipment is generally only capable of monitoring a small
`number of variables. Secondly, the diagnostic equipment
`must be closely matched to the vehicle and cannot therefore
`be used as a general purpose tool.
`U.S. Pat. Speci?cation No. 4,267,569 describes a system
`in which diagnostic equipment is connected to a vehicle that
`carries an on-board microprocessor. In order to increase the
`diagnostic capability, the on-board microprocessor executes
`?rst and second subprograms under the control of the
`diagnostic equipment. The ?rst subprograrn gives actual
`system operational data to the equipment, whereas the
`second simulates values for speci?c vehicle data under
`various operating conditions and compares the simulated
`values with the actual values. Although this system does
`provide greater ?exibility in that part of the diagnostic
`function is performed on the vehicle, the diagnostic equip
`ment must still be matched to the vehicle. Thus the system
`can provide no help with vehicles not speci?cally designed
`for diagnosis by such equipment. Furthermore, the system is
`primarily concerned with identifying the existence of mal
`functions in particular sections of the system, e.g. a fuel
`injector control circuit, rather than pinpointing faults caus
`ing those malfunctions. The ?nal fault identi?cation proce
`dure is thus still left to the technician.
`
`SUMMARY OF THE INVENTION
`
`It is an object of the present invention to obviate or
`mitigate the problems outlined above.
`
`2
`According to the present invention there is provided a
`system for identifying faults in an electrical circuit intended
`to provide a plurality of outputs each in response to at least
`one predetermined set of circuit conditions, including
`a) apparatus storing data describing individual compo
`nents and interconnections between said components
`that together de?ne the electrical circuit,
`b) apparatus for inputting data to the system identifying a
`fault symptom indicating that for a said one predeter
`mined set of circuit conditions the intended output is
`not provided,
`c) apparatus for accessing stored data describing selected
`components a fault in any one of which could be the
`cause of the said fault symptom,
`d) apparatus for calculating electrical variables to be
`expected at predetermined measurement points in a
`sub-circuit de?ned by said selected components and the
`connections therebetween given the said set of circuit
`conditions,
`e) apparatus for selecting measurement points in said
`sub-circuit at which points said electrical variables are
`to be measured,
`f) apparatus for selectively inputting measurement data
`indicating the electrical variables measured at said
`selected measurement points,
`g) apparatus for comparing input measurement data with
`said calculated electrical variables, and
`h) apparatus for identifying at least one faulty component
`from the comparison between said input measurement
`data and the said calculated electrical variables.
`Preferably, the system includes apparatus for identifying
`a single component responsible for the said intended output
`which is not provided given said one predetermined set of
`circuit conditions, apparatus for inputting data to the system
`identifying further symptoms each indicating the output
`provided by said single component for a respective one of a
`plurality of further conditions of the said circuit, apparatus
`for listing components of said sub-circuit potentially con
`tributing to each said symptoms, and apparatus for identi
`fying components of the sub-circuit which are most likely to
`be the cause of said fault symptom from a comparison of the
`component listings.
`The component identifying apparatus may be arranged to
`identify as a ?rst set listed components common to all fault
`systems. These components are those most likely to be
`causing the output malfunction. Where there is a non-fault
`symptom, any components listed for such a symptom are
`preferably excluded from the ?rst set. A second set of
`components may be listed for any components listed for any
`fault symptoms and not within the ?rst set. Thus the ?rst set
`represents a foreground of potential faults and the second set
`represents a background which may be assessed if testing
`fails to locate a fault in the ?rst set.
`The system may further comprise apparatus for identify~
`ing each set of components in said sub-circuit which de?nes
`a series circuit path extending from the said single compo
`nent to a reference voltage source, and apparatus for iden
`tifying active circuit paths which control conditions at the
`said single component by reference to the series impedance
`of the components making up the respective set of compo
`nents, components not contributing to the active circuit
`paths being excluded from said component listings.
`The active circuit path identifying apparatus may further
`comprise apparatus for calculating the voltage change along
`each circuit path, means for calculating the series resistance
`along each circuit path, and apparatus for selecting circuit
`paths having relatively low series resistances.
`
`
`
`17
`
`

`
`5,592,614
`
`3
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`An embodiment of the present invention will now be
`described, by way of example, with reference to the accom
`panying drawings, in which:
`FIG. 1 is a schematic illustration of a systematic approach
`followed in an embodiment of the present invention;
`FIG. 2 is a schematic representation of the ?ow of data in
`a ?rst stage of the embodiment of the present invention;
`FIG. 3 illustrates an interlock constraint example;
`FIG. 4 illustrates an interlock rule trace example;
`FIG. 5 illustrates a functional area hierarchy example;
`FIG. 6 illustrates a functional area query example:
`FIG. 7 illustrates an example of a representation of circuit
`components;
`FIG. 8 illustrates initial faults population partitioning;
`FIG. 9 illustrates a fault reassignment activity;
`FIG. 10 illustrates a fault population display example;
`FIG. 11 illustrates a test generation module;
`FIG. 12 illustrates a constraint propagation example;
`FIG. 13 illustrates test limit calculations;
`FIG. 14 illustrates the handling of intermittent and user
`errors; and
`FIG. 15 illustrates a display produced by the system for a
`courtesy light circuit analysis.
`
`DETAILED DESCRIPTION OF EXEMPLARY
`EMBODIMENTS
`
`FIG. 1 is a schematic illustration of the systematic
`approach to fault diagnosis followed by one system operat
`ing in accordance with the present invention. There are eight
`basic stages 101-108 which are generally executed sequen
`tially although in some circumstances it is necessary to loop
`back in the sequence to re?ne the process. Each of these
`eight stages is described in general terms below, and a
`speci?c example of the analysis of one fault is then
`described in greater detail.
`A major advantage of the present invention is that
`although a direct connection to the vehicle could be made
`this is not necessary. Accordingly the system can be used for
`diagnosis of any vehicle for which a full circuit description
`is available. This does mean however that the system must
`be informed as to which vehicle (the target vehicle) is to be
`tested. The ?rst stage 101 of the sequence is accordingly
`concerned with basic target vehicle identi?cation to access
`the target vehicle build speci?cation.
`The system stores comprehensive data fully describing
`every component of the electrical circuit of each potential
`target vehicle, and all the interconnections between every
`component in each potential target vehicle. A component
`could be a simple two terminal device, e. g. a switch, or a part
`of a sub-assembly, e.g. a wire in a harness made up from a
`number of wires. The ?rst stage is thus concerned with
`identifying the full circuit speci?cation of the target vehicle
`from the stored data describing all potential target data.
`The user could be presented with a series of questions to
`enable vehicle identi?cation, e.g.:
`Maker? (e.g. Ford)
`Model? (e.g. Testerosa)
`Market (e.g. Norway/Sweden)
`Trim? (e.g. GTi)
`Model Year? (e.g. 1992)
`
`10
`
`15
`
`25
`
`30
`
`35
`
`45
`
`55
`
`65
`
`4
`
`Bodyshell? (e.g. 4-door)
`Drive-Side? (e.g. right hand drive)
`Such an approach would be valid, as precise target vehicle
`data could be obtained. In many cases however, some of the
`questions outlined above would be redundant. For example,
`there is no need to identify drive-side if diagnosis of a fault
`does not require this information as the fault is in a sub
`system which is the same for left and right hand drive
`vehicles. To avoid redundant questions, the system is
`arranged such that the ?rst stage feeds data to subsequent
`stages only when this is necessary. Thus the ?rst stage is in
`effect consulted only when it is necessary for data to be
`provided for use in subsequent stages.
`An outline of the vehicle speci?cation module 101 is
`shown in FIG. 2. Generic vehicle data 201 is the component
`and build data for all potential target vehicles. Interlock
`constraint data 202 describes invalid combinations e. g. right
`hand side drive for USA market. Vehicle speci?c data 203
`describes the current system knowledge of the target vehicle.
`The interlock constraint data enables a reduction of the
`number of questions asked by the vehicle speci?cation
`manager 204 where the answer to one question enables the
`answer of the subsequent questions to be determined. For
`example, it is not necessary to ask the drive side if the
`market has already been identi?ed and the market de?nes
`the answer, e.g. market USA predeterrnines that the drive
`side is left hand. Thus the number of questions asked of the
`vehicle or service technician 205 may be reduced to a
`minimum.
`The second stage 102 determines the vehicle system
`requiring testing. In combination with the ?rst stage 101,
`various vehicle system possibilities may be readily elimi
`nated. For example, FIG. 3 shows an interlock constraint set
`example relating bodyshell (bodystyle) 301, trim 302 and
`wiper 303 options. From FIG. 3 it is apparent that for the
`saloon version rear wash-wipe is not an option even with
`luxury trim. Thus in this case rear~wash wipe circuitry will
`be omitted from the target vehicle speci?cation and no
`subsequent queries will be made with regard to rear-wash
`wipe circuitry. FIG. 4 illustrates an interlock rule trace in
`greater detail. Thus, once the user indicates at 401 that the
`market is Japan, the system automatically con?rms 402-404
`that the target vehicle is right hand drive, discards the
`possibility of four wheel drive at 405, and con?rms that a
`catalyst is ?tted at 406-407. The system thus acquires
`knowledge incrementally, requesting data from the user only
`when needed.
`Turning now to the second stage 102 in greater detail,
`FIG. 5 illustrates one example of a vehicle functional area
`hierarchy 501. The user selects the functional area in which
`a problem has been perceived, e.g. the head lights, by
`sequentially selecting “lighting” at 502, “external lighting” >
`at 503, “head” at 504. Selection may be achieved for
`example by positioning a cursor on a screen displaying the
`selection options, or by use of a touch screen. FIG. 7
`illustrates the screen 701 perceived after “head” has been
`selected. Thus the user can rapidly isolate that section of the
`target vehicle speci?cation of interest.
`Once a speci?c vehicle area such as Head Lights has been
`identi?ed, the system consults its database to obtain the
`relevant circuit details. As described above, the database
`contains descriptions of all components for all known
`vehicles.
`Vehicle components are stored in a two-level representa
`tion: the higher level corresponds to items such as harnesses,
`lamps and relays which may be ?tted to or omitted from
`speci?c vehicles: the lower level corresponds to basic elec
`
`
`
`18
`
`

`
`5,592,614
`
`20
`
`25
`
`5
`trical building blocks, such as wires and external connection
`points (e.g. connector pins), within the higher level items.
`An example is illustrated in FIG. 7.
`Each higher-level item is tagged with a description of
`which vehicles it is ?tted to and how many instances are
`?tted to those vehicles (e. g. a particular courtesy light switch
`might only be ?tted to cars with central locking and there
`may be either two or four needed depending upon how many
`doors the car has).
`Each lower-level item is tagged with a description of
`which vehicle systems that component participates in. Thus,
`the vehicle battery will probably be tagged with virtually
`every system whereas the courtesy light switches will only
`be tagged with the courtesy light system.
`Using these tags, it is possible to identify:
`higher-level components which are ?tted to the vehicle
`Lower-level components within these which are relevant
`to the circuit under analysis
`links from the circuit under analysis to other vehicle
`circuits (necessary for diagnosing some short-circuit
`faults)
`It may be necessary to ask further questions about the
`vehicle build speci?cation at this stage in order to select the
`appropriate components.
`The net result is a stream of lower-level component
`descriptions which is passed to the next stage.
`All connectors form part of named interfaces. It is there
`fore possible to determine which connector components
`should be joined together to form a complete circuit. The
`named interfaces use the tagging scheme described previ
`ously so further vehicle speci?cation questions may be
`asked at this stage.
`Once the connector components have been matched, it is
`possible to allocate appropriate graphical representations for
`each of them (e.g. mated pin/socket, unmated socket,
`unmated pin). Connector components and splices have sche
`matic positional information associated with them (again
`using the tagging system). Thus as described below, the full
`circuit of interest may be displayed on a screen, the display
`being laid out automatically using a set of heuristic layout
`rules.
`The third stage 103 is concerned with obtaining symp
`toms of the vehicle mal?rnction to be investigated. It is
`assumed that all electrical anomalies are propagated through
`output transducers. For those output transducers which the
`vehicle user might be expected to perceive directly, such as
`lamps, a menu is generated asking which transducers are
`believed to malfunction. These directly-identi?ed output
`transducers, in conjunction with any transducers which
`cannot be perceived directly (such as fuel injectors) are each
`used as the starting point of a qualitative circuit analysis.
`This analysis aims to identify the valid states of the target
`output transducer and relate these to input transducer states
`and electrical circuit paths. Thus a particular headlamp bulb
`may have states OFF and BRIGHT in one circuit but, in
`another circuit, it could also be DIM. An example of such an
`analysis is described in greater detail below.
`A menu is created asking which states are causing prob
`lems. If there is more than one combination of input trans
`ducer states which should result in the problem output
`transducer state and the input transducers involved are
`manually-operable then a further menu is created which
`seeks information about those input transducers.
`The symptom menus offer a three-way choice for each
`option:
`GOOD—de?nitely no problem
`BAD——problem de?nitely present
`
`6
`UNKNOWN-problem may be present
`This allows incomplete information to be entered and
`reduces the incentive to guess.
`The outcome of this phase is a set of symptoms of alleged
`incorrect behaviour (or a set of symptoms of unknown
`behaviour) possibly supplemented by a set of symptoms of .
`alleged correct behaviour.
`A symptom is a declaration of how a vehicle function
`performs under a de?ned set of conditions. Symptoms have
`two prime attributes, that is conditions and their results.
`Conditions are related to vehicle system inputs, e.g. switch
`position, and results are related to vehicle system outputs,
`e.g. lamp state “bright” (seen), heater state “011" (not seen).
`Relevant output transducers are identi?ed (by query if
`necessary). Electrical paths to relevant transducers and con
`ditions are identi?ed. Expected responses of output trans
`ducers to conditions are calculated. The technician is asked
`to classify each condition as good, bad or unknown. Que~
`rying is hierarchic and expands bad results only. For
`example, if the problem is with the courtesy lights, the
`following queries may be made:
`Courtesy Lights: where is the problem?
`
`LEFT Courtesy Lamp
`RIGHT Courtesy Lamp
`
`Unknown Bad Good
`Unknown Bad Good
`
`Assuming that the technician selects “Bad” for “Right
`Courtesy Lamp”, then:
`Does the Right Courtesy Lamp have a problem?
`
`with its OFF state?
`with its BRIGHT state?
`
`Unknown Bad Good
`Unknown Bad Good
`
`Assuming that the technician selects “Bad” for “with its
`Bright state” then:
`Does the Right Courtesy Lamp BRIGHT problem occur?
`
`when the DRIVER’S DOOR is open‘!
`when the PASSENGER'S DOOR
`is open?
`
`Unknown Bad Good
`Unknown Bad Good
`
`45
`
`SO
`
`55
`
`65
`
`The technician will then select the appropriate output, e. g.
`“Bad” for “when the passenger’s door is open”. For any
`question to which no reply is given, a default answer of
`“unknown” is assumed.
`Having identi?ed the relevant symptoms, the possible
`causes of these symptoms must be identi?ed in the fourth
`stage 104. Each symptom is considered individually. The
`appropriate output transducer is asked to account for the
`symptom, which it does by identifying any appropriate
`internal failure mechanisms and/or proposing external
`causes and asking other devices connected to its terminals
`how they might account for those causes. In this way, all
`possible causes are generated. A cause consists of a device
`(e.g. wire), a failure mode (e.g. open-circuit) and a notional
`likelihood of occurrence (e,g. 5, 10 or 50).
`Any cause which is proposed by all bad symptoms and no
`good symptoms is given a high priority. All other causes
`arising from bad symptoms are given a low priority. Sub
`sequent diagnosis initially concentrates exclusively on high
`priority causes. Only if all high priority causes are disproved
`are the low priority causes considered. This ensures that
`simple faults are diagnosed quickly and e?iciently whilst
`allowing complex faults to be diagnosed eventually.
`For example, if there are two door switches controlled
`respectively by the passenger and driver side doors and
`
`
`
`19
`
`

`
`5,592,614
`
`7
`controlling two courtesy light bulbs in parallel, and the
`passcnger’s door switch will not turn the right (drivers)
`courtesy light on, but the driver’s door switch will turn the
`courtesy light on, the passenger’s door switch subcircuit is
`assumed to be high priority, and the right courtesy lamp
`subcircuit to be low priority.
`The vehicle is then set up for testing. A bad symptom is
`selected. The vehicle is placed in the allegedly failing state
`and the output transducer state checked. If all appears well,
`a warning of possible intennittency is issued and, if further
`bad symptoms are present, the process is repeated using a
`different bad symptom. If no bad symptom is reproducible
`then the diagnosis process is aborted. Otherwise it proceeds
`to the test generation phase.
`The next (?fth) stage 105 is test generation. A rigorous
`analysis of exactly what is the “best” test to perform next is
`generally possible but the computing power required for all
`but the most trivial of circuits becomes immense. Given that
`the data upon which test selection is based (e.g. component
`failure rates) is subject to considerable inaccuracy, it is
`reasonable to use heuristics to achieve a good, if not
`absolutely optimum, selection of tests (c.f. chess playing
`computer programs).
`The approach adopted is to treat the tester, vehicle and
`service technician as state machines. Initially, only compos
`ite states similar to the existing state are considered. Tests
`are generated using each of these states. The test generation
`method used determines how every observable circuit
`parameter (e. g. current, voltage, output transducer state) will
`vary in the presence of each high-priority cause. Any param
`eter which differs measurably depending upon which high
`priority faults are present can form the basis of a useful test
`step.
`The quality of these potential tests is assessed heuristi
`cally, using estimates of the following parameters:
`Direct Cost-how much service technician time the test
`step will take
`Amortised Cost-how much of the cost of a test step may
`be spread over several test steps (e.g. removing a trim
`panel may be time-consuming but could allow several
`subsequent tests to be performed cheaply.
`Bene?t—the amount of information which can be
`expected from the test step
`Scope for Error—-how di?cult the test step is to perform
`correctly
`Prohibitions-—e.g. not probing close to the fan when the
`engine is running
`Timing Constraints—minimum and maximum inter-step
`delays
`Functional Constraints—e.g. avoid unmating connectors
`which may be failed open circuit since this may inad
`vertently remove the fault
`Strategic Considerations——restrict the number of test
`strategies being performed in parallel, both to ease the
`computational overhead and to prevent the appearance
`of having no coherent strategy
`Desirability of State Change-whether the test step would
`result in a composite state which is more or less useful
`than the present composite state
`A heuristic assessment is made of the likelihood of
`obtaining a more satisfactory test if a greater range of
`composite states were to be considered. This may lead to
`further test generation cycles. When no further cycles are
`deemed useful, the most promising test candidate is selected
`for execution.
`The next stage 106 is selection and performance of the
`“best” test step. The test step selected is presented to the
`
`10
`
`25
`
`35
`
`40
`
`45
`
`55
`
`60
`
`8
`service technician. The service technician may decline this
`test step, in which case the sy