EP1573579A2

EP1573579A2 - Diagnostic method for an electronic systems unit

Info

Publication number: EP1573579A2
Application number: EP03799675A
Authority: EP
Inventors: Samuel Boutin
Original assignee: Renault SAS
Current assignee: Renault SAS
Priority date: 2002-12-20
Filing date: 2003-12-19
Publication date: 2005-09-14
Also published as: JP2006515089A; US20060136155A1; KR101002379B1; US7418321B2; FR2849236B1; FR2849236A1; WO2004059519A2; KR20050094821A; WO2004059519A3

Abstract

The invention relates to a diagnostic method for the default operation of an electronic systems unit, comprising components (Ani; Cni; UCEn; B), producing and using data, at least one datum of which (xi)can take a pre-determined fixed value (xip), after the occurrence of an erroneous functioning of at least one of the components (Ani; Cri; UCEn; B) of said unit. The method is characterised by: i) classification of (a), during a design phase of the architecture of said systems, the particular values following the corresponding fault and recording said classification in a computer memory, for example in the form of a consultation table.

Description

Method for diagnosing a set of electronic systems

The present invention relates to a method for diagnosing a set of electronic systems comprising components (A ⁿ , - C ⁿ i, - UCE _n ; B), producing and consuming data, at least one of said data (xi) being capable of taking a particular predetermined value (Xi _P ), following the appearance of a malfunction of at least one of the components (A ⁿ i, - C ⁿ i, - UCE _n ; B) of said set .

There are known sets of electronic systems of this type, designed in particular to equip motor vehicles. Such a vehicle commonly comprises several systems each ensuring the execution of a service such as controlling the engine propelling the vehicle, managing the air conditioning of the passenger compartment, managing the connections of the vehicle to the ground (eg braking, suspension ), and management of telephone communications. The hardware components of all of these systems are shown diagrammatically in FIG. 1 of the attached drawing, for example for a vehicle. These components essentially comprise electronic control units or "computers" UCE _m , each computer possibly being connected to sensors C ⁿ i and to actuators A ^m _j , all the computers being connected to at least one same bus B for transmitting therein. or receive information, for example multiplexed, from or to other computers connected to bus B. This multiplexing is obtained in particular, as is well known for the CAN bus for example, by introducing the information in question in messages materialized by digital signal frames.

As an illustrative example, the S ₂ system of "engine control" includes the ECU computer, several sensors C i sensitive to quantities such as the engine speed, for internal combustion for example, the pressure at the intake manifold of this engine, the pressure of the outside air, engine coolant water temperature, air temperature, battery charge status, and several Aj actuators. The ECU ₂ computer is duly programmed to execute several engine control functions such as: idling control, regulation of the richness of the air / fuel mixture, adjustment of the ignition advance of this mixture and re -exhaust gas circulation. To do this, the ECU ₂ computer uses information from the aforementioned sensors C i and develops control signals for the actuators A constituted by an additional air control valve and a spark plug ignition coil for the "idle speed regulation" function. ", a fuel injector for the" richness regulator "function, the same ignition coil for the" ignition advance "function and a valve for the" exhaust gas recirculation "function.

The other "services" mentioned above, e.g. "air conditioning of the passenger compartment", "connection with the ground", are performed by systems of architecture similar to that presented above for controlling the engine.

All these systems put in communication by the same bus B constitute a multiplex network. It can therefore be seen that several functions belonging to different systems can exploit information from the same sensors, for example, which avoids costly redundancies in the structure of all the systems. The use of a multiplex network also makes it possible to significantly reduce the length of the electrical lines interconnecting the various elements of the assembly. Such a multiplex assembly also allows the installation of non-classical and possibly complex functions, sometimes involving several systems and therefore called "transversal". As an illustrative and nonlimiting example, the perception of the information "air bag (or" airbag ") triggered", significant that the vehicle has been impacted, can then be treated so as to control the issuance of a call for help by a mobile telephone device on board the vehicle. We know from French patent application N ° FR 01 15819, the concept of particular value and its use in a method of diagnosing operating faults of a set of electronic systems.

However, this method does not make it possible to select a type of fault for the diagnosis. For example, we do not know how to search only for breakdowns coming from connectors for example, or to look for the most probable breakdowns first, this probability resting on a knowledge of the design techniques implemented.

To remedy these drawbacks, the present invention relates to a method for diagnosing a malfunction of a set of electronic systems, said systems comprising components (A ⁿ i, - C ⁿ ι; UCE _n ; B), producing and consuming data, at least one of said data (xi) being capable of taking a particular predetermined value (x_ _p ), such as for example an indication of fault stored in a memory of a computer, following the appearance of a malfunction of at least one of components _(. i a ⁿ - C ⁿ j; UCE _n; B) of said set, this method being characterized in that: i) class (a), during a design phase of architecture of said set of systems, the particular values according to the types of associated faults and said classification is recorded in a computer memory, for example in the form of a look-up table; ii) a diagnostic tool is connected, during a diagnostic phase, to said set of electronic systems, said tool having access to said classification; iii) removing (b) the particular values corresponding to types of component faults predefined in said classification as particularly reliable; iv) selecting (c) the data (xi) having taken a particular value (xι _p ); v) there is automatically calculated (d), for each data item (xi) selected in step (iv), a group (X∞j _. ) of the data likely to be at the origin of the particular values (xι _p ) taken by the data (xi); vi) one automatically establishes (e) a list (X ") of the data contained in the intersection of said groups (X∞i) of data, and vii) one records the particular values and their propagation on a storage means for a tool intended for the diagnosis of said set of electronic systems.

Thanks to this method, it is possible to search for current or potential failures according to their type, which makes it possible to shorten the time for searching for faults by retaining only the most probable faults. It will be understood that the term electronic system covers all electronic and electrical systems producing and consuming data.

We can exclude (g) from said list (X∞) faults whose consequences have not been observed. This reduces the time to troubleshoot.

It is possible, if no defect remains in step (vi), to start again in step (ii) taking into account (j) the classes of defect which had been previously eliminated. It is possible, if a fault remains in step (vi), to verify that one of the faults identified in phase (g) is indeed the cause of the problem which led to launching said diagnostic phase and if this is not the case, we resume step h).

It is possible to analyze said list (X∞) to identify the component or components of the assembly for which a malfunction is the cause of the particular values (xi _P ) taken by said data (xi). Types of faults can belong to at least one of the categories listed below: the values created following the unavailability of data sent by a function, - the particular values created following the detection of a fault in a sensor or actuator,

- the particular values created following a faulty connection, at the level of a connector or a wire,

- the particular values created following a computer fault,

- the particular values created following a failure to execute a program on a microcontroller and

- the particular values created following a fault in a communication network.

We can automatically determine a probability for each data to take a particular value according to the category to which it belongs and that we modify step (b) by first taking into account

1 'step (j).

Thanks to these provisions, a fault finding can be guided by first taking into account the most probable breakdowns. Said set of electronic systems may include a set of systems for equipping a vehicle.

The method may include a step of analyzing the feasibility and / or fallibility of said set of electronic systems and establishing an output indicating said feasibility and / or fallibility.

The present invention also provides an article of commerce comprising a memory readable by a computer, a program executable by a computer being recorded on said memory for diagnosing malfunctions of a set of electronic systems, characterized in that said program includes coding for: i) classifying (a), during a design phase of the architecture of said set of systems, the particular values according to associated types of defect and said classification is recorded in a computer memory, for example under the form of a consultation table; ii) connecting, during a diagnostic phase, a diagnostic tool to said set of electronic systems, said tool having access to said classification; iii) delete (b) the particular values corresponding to types of component faults predefined in said classification as particularly reliable; iv) select (c) the data (xi) having taken a particular value (xi _P ); v) automatically calculate (d), for each datum (xi) selected in step (iv), a group

(X∞) data likely to be at the origin of the particular values (x-. _P ) taken by the data (xi); vi) automatically establish a list (X∞) of the data contained in the intersection of said groups (X∞i) of data, and vii) record particular values and their propagation on a storage means for a tool intended for the diagnosis of said set of electronic systems.

The present invention also provides a computer tool programmed for the diagnosis of a set of electronic systems using steps of the method of the present invention or programmed using a trade item of the present invention.

Said set of electronic systems may include a set of systems for equipping a motor vehicle.

Other objects, characteristics and advantages of the present invention will appear by way of example on reading the description which will follow and on examining the appended drawing in which: FIG. 1 is a diagram of a set of systems devices that it is proposed to provide with means of diagnosing operating faults according to the present invention, this assembly being described in the preamble to this description and FIG. 2 is a block diagram describing the application of the invention in a method diagnostic of electronic system failure.

The particular values of the data produced by the components of a set of systems can be categorized, in particular during the design phase, according to the types of defects that they reflect.

We can distinguish the particular functional values which are linked to the sensors, actuators and functions and the particular operational values which come from a particular embodiment of the functions by the computers, the data buses, the wire connections and connectors.

Among the particular functional values, for a stream of input data of a function, there are different categories of particular values, in particular those determining an invalid value of a data item, and those determining a value outside the domain of definition of the data set by the designer.

Among the specific functional values, a distinction is also made between the specific values created following the failure of a sensor or an actuator, the detection of the fault being typically done by a function which controls the sensor or the actuator.

Among the particular operational values, that is to say coming from a particular implementation of the functions by the computers, the communication buses between the computers and a wiring, a distinction is made in particular: the particular values diagnosing a fault on a wired link , in particular short circuits to ground and open circuits; the particular values relating to the detection of a communication fault on a multiplex bus, originating in particular from an absent data frame; - the particular values relating to the absence of a computer on a network, one can then distinguish the particular values produced within the isolated computer and the particular values produced by computers observing the isolated computer. This category of particular values is produced for example when data frames of said network are absent, that is to say that they are expected and not received; - the particular values relating to a failure to execute on a computer, for example memory corruption, i.e. an accidental inversion of at least one bit in memory, detected by a CRC (Cyclic Redundancy Check ) or cyclic redundancy test of a re-initialization of said computer following the detection of an anomaly by a program on board said computer; the particular values relating to a supply fault; and the particular values characterizing a combination of defects.

We know from the French patent application N ° FR 01

15819, filed by the applicant and incorporated here by reference, a method of diagnosing malfunctions of a set of electronic systems producing and consuming data, at least one of said data (xi) being capable of taking on a particular value (xι ) predetermined, following the appearance of a malfunction of at least one of the components of said assembly, this method comprising the following steps: a) at the launch of a diagnostic phase, the data (xi) are selected which have a particular value (xι _p ), b) for each datum (i) thus selected, we search for the group (X∞j _. ) of the data likely to be at the origin of the particular value (xi _P ) taken by the data (xi), c) the list (X∞) of the data belonging to the intersection of said groups (X∞i) of data is established, and d) the said list (X∞) is analyzed to identify the components of the set of which a malfunction is at the origin of the particular values (Xi _P ) taken by said data (xi). By implementing the present invention, the process summarized above is improved to take into account the categories of particular values of the present invention. A particular embodiment of this improved process is detailed with reference to FIG. 2. It contains the following steps:

an initialization step 210 of known type,

a step a) 215 at the launch of a diagnostic phase, during which the data (xi) which have a particular value (xι _p ) are selected;

- during a step b) 220, this data is excluded from this list belonging to categories deemed more reliable by the person skilled in the art, for example only the particular values relating to connection faults are initially retained , or, if the searches have yielded nothing at this point, we add particular values of categories previously excluded;

- during a step c) 225, each datum (xi) which has a particular value (xι _p ) is selected; - during a step d) 230, given the data (xi) thus selected in step a), the group (X∞i) of the data likely to be at the origin of the particular value is sought (xι _p ) taken by the data (xi);

- during a step e) 235, the list (X∞) of data belonging to the intersection of said groups is established

(X∞i) of data;

- during a step f) 240, said list (X∞) is analyzed to identify the component or components of the assembly of which a malfunction is the cause of the particular values (xι _p ) taken by said data (i);

- during a step g) 245, given the faults envisaged in step f) 240, all the particular values corresponding to faults which have not been observed on the system are excluded, whether by l skilled in the art after precise verification or by observation of the operation of certain components. For example, a lamp supply wire cannot be shorted if the lamp can be turned on and off. Therefore, if a particular value characterizing the said short-circuit was envisaged, it can be excluded.

- a step j) 255, during which, in case, after step g) 245, there remains no particular value to be examined, this may mean that one has excluded too many possibilities at the beginning of the analysis, and the process is repeated from step c) 225, but retaining particular categories of value excluded hitherto in the analysis;

a step h) 250, during which the results of step g) 245 are examined and if, after examination, the envisaged faults do not allow a fault to be identified, the process is repeated in step c) 225 by extending the scope of the research to new categories of particular values during step j) 255.

If we consider, for each category, probabilities of occurrence, drawn from empirical measures for example, we can assign to the different particular values probability of occurrence. When a particular value covers several faults, possibly coming from different categories, it inherits the sum of the probabilities of occurrence of each of the faults.

Then, the process described above is taken again according to the probabilities of occurrence of the different categories of particular values. For example "we add to the list first the most probable elements then, if step b) is repeated, we add to the list, elements a little less probable and so on until it at least one particular value remains at the end of step h) and / or until the cause of the fault which motivated the diagnosis is detected. The set of electronic systems may include a set of systems for equipping a motor vehicle. The method of the invention may include a step of analyzing the feasibility and / or fallibility of said set of electronic systems and a step of establishing an output indicating said feasibility and / or fallibility. It can help accepting or rejecting a set of systems, for example proposed for a car.

This process can be implemented using a computer tool allowing the editing of the various objects necessary for the design and the partial automation of the various stages of the process constituted by the invention.

The method of the invention can be programmed on an article of commerce comprising a memory readable by a computer, for example a CD, DVD or equivalent or on the hard disk of a computer. Such a program for the method will be executable by a computer, and will be recorded on said memory to be executed by said computer.

Claims

1. Method for diagnosing a malfunction of a set of electronic systems, said systems comprising components (A ⁿ i, - C ⁿ ι; UCE _n ; B), producing and consuming data, at least one of said data ( xi) being capable of taking a particular predetermined value (ι _p ), such as for example an indication of fault stored in a memory of a computer, following the appearance of a functional fault of at least one of the components (A ⁿ ι; C ⁿ i; UCE _n ; B) of said set, this process being characterized in that: i) we classify (a), during a design phase of the architecture of said set of systems, the particular values according to associated types of fault and the said classification is recorded in a computer memory, for example in the form of a look-up table; ii) a diagnostic tool is connected, during a diagnostic phase, to said set of electronic systems, said tool having access to said classification; iii) removing (b) the particular values corresponding to types of component faults predefined in said classification as particularly reliable; iv) selecting (c) the data (xi) having taken a particular value (xι _p ); v) a group is automatically calculated (d), for each datum (xi) selected in step (iv)

(X∞i) data likely to be at the origin of the particular values (xι _p ) taken by the data (xi); vi) a list (X∞) of the data contained in the intersection of said groups (X∞i) of data is automatically established, and vii) a step of recording the particular values and their propagation on a storage means for a tool provided for the diagnosis of said set of electronic systems.

2. Method according to claim 1, characterized in that one excludes (g) from said list (X∞) the defects whose consequences have not been observed.

3. Method according to claim 1 or 2, characterized in that, if no defect remains in step (vi), we go back to step (ii) taking into account (j) classes of defect which had been ruled out before.

4. Method according to claim 3, characterized in that, if a defect remains in step (vi), it is verified that one of the faults identified in phase (g) is indeed the cause of the problem which led to launching said diagnostic phase and if this is not the case, step 1 h) is repeated.

5. Method according to any one of claims 1 to 4, characterized in that one analyzes said list (X∞) to identify the component or components of the assembly of which a malfunction is at the origin of the values. particular (Xi _P ) taken by said data (xi).

6. Method according to any one of claims 1 to 5, characterized in that types of fault belong to at least one of the categories listed below: - the values created consecutively to the unavailability of data sent by a function, the particular values created following the detection of a sensor or actuator fault, - the particular values created following a connection fault, at a connector or a wire, the particular values created following a fault in a computer, - the particular values created following a failure to execute a program on a microcontroller and the particular values created following a fault in the level of a communication network;

7. A diagnostic method according to any one of claims l to 6, characterized in that one automatically determines a probability for each datum to take a particular value according to the category to which it belongs and which one modifies step (b) by first taking into account step (j).

8. A diagnostic method according to any one of claims l to 7, characterized in that said set of electronic systems comprises a set of systems for equipping a vehicle.

9. A diagnostic method according to any one of claims 1 to 8, characterized in that it comprises a step of analyzing the feasibility and / or fallibility of said set of electronic systems and the establishment of an output indicating said feasibility and / or fallibility.

10. An article of commerce comprising a memory readable by a computer, a program executable by a computer being recorded on said memory for diagnosing malfunctions of a set of electronic systems, characterized in that said program includes coding for : i) classify (a), during a design phase of the architecture of said set of systems, the particular values according to the types of associated faults and said classification is recorded in a computer memory, for example in the form of a consultation table; ii) connecting, during a diagnostic phase, a diagnostic tool to said set of electronic systems, said tool having access to said classification; iii) delete (b) the particular values corresponding to types of component faults predefined in said classification as particularly reliable; iv) select (c) the data (xi) having taken a particular value (xι _p ); v) automatically calculate (d), for each data item (xi) selected in step (iv), a group (X∞i) of the data likely to be at the origin of the particular values (xι _p ) taken by the given (Xi); vi) automatically establish (e) a list (X∞) of the data contained in the intersection of said groups (X∞i) of data, and vii) record particular values and their propagation on a storage means for a planned tool for the diagnosis of said set of electronic systems.

11. A computer tool programmed for diagnosis of a set of electronic systems using steps of the method according to any one of Claims 1 to 9 or programmed using a trade item according to Claim 10.