DE102004060459A1 - Data bus system for motor vehicle, has main bus subscriber detecting variation of stored message text by comparing with previous message circulation to store missing message text of bus subscriber in recent message circulation - Google Patents

Abstract

The system has a main bus subscriber (13) defined by a bus subscribers (6) in a set of subscribers (2-6). The main subscriber has a message detection unit to determine a message circulation of the subscribers and to store a defined message text from the circulation. The main subscriber detects a variation in the stored text by comparing with a previous circulation to store missing text of the subscribers in recent circulation. The missing text of the subscriber is stored in an error memory. An independent claim is also included for a method for facilitating diagnostics of faults of bus subscriber of a data bus system.

Description

The The invention relates to a data bus system for motor vehicles with a Plurality of electronic bus subscribers communicating with each other via a bus line are connected, so over These bus line messages are interchangeable. The invention relates Continue a procedure to facilitate the diagnosis of errors of bus subscribers of such a bus system and a computer program product.

One Such data bus system is, for example, by the CAN bus (controller Area Network) for Data communication, in particular of control units in motor vehicles, known. To such a data bus is a plurality of electronic bus subscribers, especially control devices a motor vehicle, connected. These bus subscribers can in addition to the connection to the data bus further inputs and outputs, e.g. to connect of sensors and actuators or to connect the power supply, exhibit. This known data bus system works according to the multi-master principle, which means that all bus subscribers of the bus system have equal rights are. When the data bus is free, each bus user can send a message which is also referred to as a data telegram transferred. The message of a sending bus participant is received by all other bus subscribers, but only evaluated by such bus subscribers, the data of this Require data telegram.

Becomes in the data system uses network management, e.g. according to OSEK standard, sends every one of the bus subscribers participating in the communication periodically one Message indicating whether he is maintaining communication want. Each of the bus subscribers has a unique identifier, e.g. an address, wherein sending a message, for example in ascending or descending order of identification could be done. As a result, e.g. an annular communication, wherein the last, a message sending, bus participants as the next recipient the first names a message sending bus subscriber. Submitting from messages from the first to the last bus participant will too referred to as message circulation.

One Data bus system for Today, motor vehicles have a large number of bus subscribers. Due to errors in the hardware or in the software can be a Bus participant for short or longer Time to fail. The failure of a bus participant is in the above described communication such that a bus subscriber, which expects a message from the failing bus participant, the one from depending on the information functionalities stop executing can. To give an indication of the cause of this malfunction to receive in an existing memory in him an entry on the Missing message generated. Because the communication chain in However, a data bus system is dynamic, e.g. due to one over several Participant-reaching communication chain, when reading such a Error memory entry can not be readily determined Which of the bus participants originally caused the error entry. Especially in data bus systems with a large number of bus users can due to sporadic errors occur a variety of such Fault memory entries be present in a majority of the bus subscribers, so that the real cause for the error memory entry is no longer traceable. Furthermore there is the problem that the capacity of the fault memory of the bus participants for space and / or cost reasons is often limited, thereby providing relevant information for the elucidation of the cause of the error can be lost quickly.

It is therefore an object of the present invention, a data bus system and to provide a method by which to diagnose errors of Bus participants a bus system for motor vehicles facilitated becomes.

This object is achieved with a data bus system according to the features of claim 1 and with a method according to the features of claim 7 and a computer program product for carrying out the method according to the features of the claim 10 solved. Advantageous embodiments are given in the respective dependent claims.

Of the The basic idea of the present invention is monitoring the bus participant on the basis of a regularly monitored by all Bus participants sent message. Based on the observation of a or multiple message rounds can faulty bus participants identified and stored for further evaluation become.

In accordance with the invention, the data bus system has a bus subscriber defined as a main bus subscriber, which is set up to detect a message circulation of the bus subscribers in a normal operation of the data bus system and to buffer a defined message content from each message of a transmitting bus subscriber, whereby each message circulates in one message circulating the intact bus subscriber of the data bus system sends exactly one message, so that a number of message contents corresponding to the number of intact bus subscribers is buffered. The bus subscriber defined as the main bus subscriber is also set up to detect a change in the cached message contents compared to at least the previous message cycle in order to store the message content of a bus subscriber missing in the most recent message cycle or the missing message contents of several bus users in a fault memory , As a result, sporadically or permanently faulty bus subscribers can be identified in a simple manner, which is determined, for example, by comparing two or more message circulations of the bus subscribers. The comparison takes place on the basis of the observation of a change in the buffered message contents of the message circulations, or the comparison of the message circulations with a desired state stored in the main bus subscriber. It is also conceivable to control the duration of the missing messages. In particular, it is provided that the message content is represented according to an expedient embodiment by a unique identifier a bus subscriber, the evaluation of errors is facilitated.

In another variant is provided that by the main bus subscriber detected message represents a network management message, which includes a status entry indicating the operability represents the message sending bus station. Such a network management message becomes dynamic set the order of message circulation, i. it will defines the order in which the acknowledgment messages are sent to the bus be delivered. This approach has the advantage that the Identification of one or more faulty bus participants not independent, separate messages need to be generated, but that instead used the network management messages already used in a multi-master network can be. This is the identification of a faulty bus participant without change the functionality the bus participant possible.

Conveniently, the fault memory is arranged in the main bus subscriber, whereby a overall cost-effective Realization is possible because only in one of the multiplicity of bus subscribers an enlarged memory must be provided. In one embodiment, it is provided that the fault memory is dimensioned such that per connected Bus subscriber to the bus line two error entries are stored.

In a further advantageous embodiment It is envisaged that the saving of the missing or changed Message content or the missing or changed message content one or more bus subscribers after passing through a predetermined Number of message rounds he follows. For this purpose, in the main bus subscriber a buffer be provided in the form of a ring memory whose size is according to the number of bus subscribers and the number of monitored ones News Rounds measured. Waiting for a predetermined number of message cycles points to the creation of an error entry in the error memory the advantage that false alarms can be avoided. hereby is reliable ensures that the fault memory is not just temporary, one-off failures of Messages are written that have no cause whatsoever on the functionality of the bus subscriber and thus of the data bus system. In this Design is thus assumed to be a message must fail more than once before closing on a mistake becomes.

With the method according to the invention The benefits are the same as those mentioned above with the data bus system according to the invention have been described.

So is in the inventive method provided that the main bus subscriber a message circulation the bus participant determines and from each message of a sending Busteilnehmers caches a defined message content, in a message circulation, each intact bus subscriber of the Data bus system sends exactly one message, so that one of the number the intact bus user corresponding number of message content is cached. In a further step, the main bus subscriber will join every message circulating a change in the cached News content at least the previous message circulation to the most recent news circulation missing or changed Message content of a bus device or the missing or changed Message contents of several bus participants in a fault memory save.

In In one embodiment, the main bus subscriber determines from each of a message sent to a bus subscriber and save this between.

In a further advantageous embodiment the saving of the missing or changed message content takes place or the missing message content of one or more bus subscribers after passing through a predetermined number of message cycles.

The Invention will now be described with reference to an embodiment in the drawing explained in more detail. It demonstrate:

1 the structure of a data bus system according to the invention, and

2 data stored in a buffer which lead to an entry in a fault memory of the data bus system.

In 1 are in the data bus system according to the invention 1 bus users 2 . 3 . 4 . 5 . 6 on a bus line 7 connected. The data bus can be designed, for example, as a CAN bus (Controller Area Network) or as a FlexRay bus, in which all bus subscribers have equal rights. This is called a so-called multi-master principle. In addition to a respective connection to the bus line 7 can the bus participants 2 . 3 . 4 . 5 . 6 have additional additional connections, eg for connection to respective sensors, actuators or for connection to a power supply. Although only 5 bus users are shown in the figure, has a data bus system 1 in a car application in practice a much larger number. Each of the bus participants 2 . 3 . 4 . 5 . 6 is provided in a known manner with at least one application function and a network management, wherein the at least one application function tasks such as the monitoring of sensors or the control of actuators, while the network management for the control or maintenance of communication via the bus 7 is responsible for the other bus subscribers.

In principle, the data transmission takes place in such a way that the message of a sending bus subscriber is received by all other bus subscribers. However, an evaluation is generally carried out only by those bus subscribers who need data from this message. After switching on and initializing the data bus system 1 transmit all bus participants 2 . 3 . 4 . 5 . 6 regularly a message 10 over the data line 7 , In the exemplary embodiment, this may be, for example, the reference numeral 3 be provided bus subscriber, which in turn another message 10 over the bus line 7 transmits, in which in turn is determined which of the other bus participants 4 . 5 . 6 to send a confirmation message about receiving this message. Provided all the bus participants 2 . 3 . 4 . 5 . 6 of the data bus system 1 are intact and work correctly, in the present embodiment with five bus subscribers thus a total of five organisationally related messages 10 over the bus line 7 transfer. Such a process is referred to as message circulation, wherein the duration of such a message circulation is fixed. The time duration is determined eg in the OSEK network management by the number of participants and the distance of the messages.

The bus participant 6 is in the present embodiment as a main bus subscriber 13 defined, which distinguishes itself from the other bus subscribers in that it, in addition to the application functionality and the network management functionality via a buffer 8th , a fault memory 9 as well as a message collecting means 12 includes. The message collecting device 12 is designed to be used by the bus subscribers to the bus line 7 transmitted messages 10 to determine and identify a defined message content from this or determine. This message content is preferably an identifier identifying the transmitting bus subscriber and possibly a status entry. Both information is usually stored in the header of such messages at predetermined locations. At the news 10 they are preferably network management messages in which the status entry is used to define the dynamic circulation of the message delivery. Such a procedure must therefore be dynamic since the number of participants in a communication ring can vary. This is possible on the one hand because of functional aspects of the data bus system and on the other hand because of errors.

The invention exploits the presence and dispatch of these network management messages by the main bus user 13 be monitored. In this case, upon detection of such a network management message, the message content, at least consisting of its identification and optionally its status entry, in the buffer 8th cached.

The ones in the cache 8th cached entries and their temporal change can be from the 2 be better seen. The during a news round 14 message contents stored in the cache 11 are shown. In a message circulation i will be on the bus line 7 pending messages by the message collecting means 12 determined and the specified message content 11 identified and stored. In the present embodiment, the message content includes 11 an identification of the message sending bus station and a status entry 15 , In message circulation i, a total of five entries are made, which is the number of communications system communication 1 appropriate Number of bus participants corresponds.

Based while of the message circulation i determined information is not yet conclusion possible, whether there is an error in one of the bus users of the data bus system. Therefore, a predetermined number of message cycles, e.g. three, stored in the cache, and finally a Comparison to a change with at least the previous message circulation.

For the first time, a change occurs in the message circulation j, which only comprises four entries, it being assumed that the bus subscriber 5 should have an error. After this has a not only unique error, it can no longer send messages to the bus during the message circulation j, and the further message cycles j + 1, j + 2, etc. 7 send, so that in these said message circulations a the bus subscriber 5 missing entry. By way of example, after three message rounds (j, j + 1, j + 2) deviate from the previous message rounds, an error is detected, resulting in the error memory 9 the identification of the relevant bus subscriber, in the present embodiment, that of the bus subscriber 5 , is registered. As the other bus participants 2 . 3 . 4 . 6 operate according to their intended purpose, these are located in the fault memory 9 no defect entry.

The fault memory 9 is dimensioned in the exemplary embodiment such that two entries can be made for each of the bus users. These could be, for example, an entry for sporadic errors and an entry for long-term errors. In principle, both the preferably constructed as a ring buffer memory 8th as well as the fault memory 9 have any size. The memory organization can in principle deviate from the form described here.

With in the embodiment described configuration, which requires only a small memory has a reliable one permanent storage of error entries possible, resulting in a defective Bus participants are purposefully identified and exchanged or can be repaired.

Of the Advantage of the present invention is that the review of Bus participants on intended function on the basis of network management messages, especially in their Header information entered (identification of the bus participant and status entry). Remains the network management message one Bus participant for longer Time off, a fault memory entry is made in the main bus subscriber.

By Sufficient dimensioning of the fault memory can the triggering an error Bus participants can be identified even after a long time. The fault memory serves exclusively for storing error conditions and does not record general data.

in principle Is it possible, the error of a bus participant in the time scale a message circulation and store. To avoid however, it is preferable for "false alarms" that an error entry occurs only when the relevant network management message of the relevant bus subscriber fails more than once.

The Troubleshooting for a faulty bus subscriber becomes particular Significantly easier with sporadic bus user errors. These can be over a longer period of time be stored and remain so comprehensible.

1

data bus system

2

bus users

3

bus users

4

bus users

5

bus users

6

bus users

7

bus line

8th

cache

9

error memory

10

message

11

message content

12

News detecting means

13

main bus station

14

News circulation

15

status entry

Claims (10)

Data bus system ( 1 ) for motor vehicles having a plurality of electronic bus subscribers ( 2 . 3 . 4 . 5 . 6 ), which are connected via a bus ( 7 ) are connected to each other, so that via this bus line messages ( 10 ) are replaceable, the data bus system ( 1 ) one as the main bus subscriber ( 13 ) defined bus subscribers ( 2 . 3 . 4 . 5 . 6 ), which is adapted to - in a normal operation of the data bus system ( 1 ) a message circulation of the bus subscribers ( 2 . 3 . 4 . 5 . 6 ) and from each message ( 10 ) of a sending bus subscriber a defined message content ( 11 ), whereby in a message circulation each intact bus subscriber ( 2 . 3 . 4 . 5 . 6 ) of the data bus system ( 1 ) sends exactly one message, so that one of the number of intact bus subscribers corresponding number of message contents ( 11 ) is cached, and - a change in the cached message content every time a message circulates ( 11 ) to at least the previous message circulation to the message content missing in the most recent message circulation ( 11 ) of a bus participant or the missing message content ( 11 ) of several bus users in a fault memory ( 8th ) store. Data bus system according to claim 1, characterized in that the information provided by the main bus subscriber ( 13 ) message ( 10 ) represents a network management message comprising a status entry indicating the operability of the bus user transmitting the message ( 2 . 3 . 4 . 5 . 6 ). Data bus system according to claim 1 or 2, characterized in that the fault memory ( 8th ) in the main bus subscriber ( 13 ) is arranged. Data bus system according to claim 3, characterized in that the error memory ( 8th ) is dimensioned such that per connected bus subscriber ( 2 . 3 . 4 . 5 . 6 ) to the bus line ( 7 ) two error entries are storable. Data bus system according to one of the preceding claims, characterized in that the storing of the missing or changed message content ( 11 ) or the missing or modified message content ( 11 ) of one or more bus subscribers ( 2 . 3 . 4 . 5 . 6 ) after passing through a predetermined number of message cycles. Data bus system according to one of the preceding claims, characterized in that the message content ( 11 ) is represented by a unique identifier a bus subscriber. Method for facilitating the diagnosis of errors of bus users of a bus system ( 1 ) for motor vehicles, the data bus system ( 1 ) a plurality of electronic bus subscribers ( 2 . 3 . 4 . 5 . 6 ), which are connected via a bus ( 7 ) are connected to each other, so that via this bus line messages ( 10 ) are interchangeable, and this one being a main bus subscriber ( 13 ) defined bus subscribers ( 2 . 3 . 4 . 5 . 6 ), in which - the main bus subscriber ( 13 ) a message circulation ( 14 ) of the bus participants ( 2 . 3 . 4 . 5 . 6 ) and from each message ( 10 ) of a sending bus subscriber a defined message content ( 11 ), whereby in a message cycle each intact bus subscriber ( 2 . 3 . 4 . 5 . 6 ) of the data bus system ( 1 ) sends exactly one message, so that a number of message contents corresponding to the number of intact bus subscribers is buffered, and - the main bus subscriber ( 4 ) with each message circulation ( 14 ) a change in the cached message content ( 10 ) to at least the previous message circulation to the missing or changed message content of a bus station in the most recent message circulation or the missing or changed message contents of multiple bus users in a fault memory ( 8th ) store. Method according to Claim 7, characterized in that the main bus subscriber ( 13 ) from each of a bus subscriber ( 2 . 3 . 4 . 5 . 6 ) sent message ( 10 ) determines and caches a tag identifying it. Method according to claim 7 or 8, characterized in that the storing of the missing or changed message content or the missing message contents of one or more bus users ( 2 . 3 . 4 . 5 . 6 ) after passing through a predetermined number of message cycles ( 14 ) he follows. Computer program product for carrying out the Method according to one of the claims 7 to 9.