JP2009202720A - Communication system, its abnormality presumption method, and information reading device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a communication system capable of properly specifying an abnormality position of failure or the like, its abnormality estimation method and an information reading device. <P>SOLUTION: The communication system provided with a plurality of nodes for mutually performing CAN communication through a CAN communication line includes: a memory means for memorizing the detection number of error of the CAN communication by each of the plurality of nodes in every kind of errors; and an abnormality estimation means in which a sum of an accumulation value of the error detection number of bit error stored in the storage means and an accumulation value of the error detection number of form error estimates the maximum node or the CAN communication line connected to the node abnormal. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an abnormality detection technique for CAN (Controller Area Network) communication.

In a communication system, as conventional techniques for identifying the failure location, connection means for connecting to a communication line so that data can be transmitted and received, acquisition means for acquiring communication data sent to the communication line, and acquiring the communication data A timer unit that measures an elapsed time from the acquisition of the communication data to the next acquisition time, and a determination unit that determines that communication has been interrupted when the predetermined time has elapsed. Is known (for example, see Patent Document 1). According to this configuration, according to the present configuration, when communication data transmitted from another control device to a communication line with the shortest communication cycle cannot be acquired even after a determination time set in a short time has passed, Since it is determined that the control device is disconnected, it is described that the control device having an abnormality that interrupts the communication can be identified and the processing when the control device is stopped can be executed quickly. Yes.
JP 2003-143164 A

  By the way, in CAN communication, which is a communication method using a differential voltage between two communication lines, priority is set for messages. Each node connected by CAN communication starts communication asynchronously. When transmission is started at the same time, arbitration is performed, and a high priority message is obtained with a transmission right.

  However, with the conventional technology, when one of the two branch lines connecting the main line of the communication line and the node is disconnected, the node connected to the main line via the disconnected branch line normally communicates. Since it becomes impossible to detect the voltage, it is impossible to correctly recognize whether the transmission right can be acquired. As a result, when the node tries to communicate, the message on the communication line is destroyed via the branch line that is not disconnected. Therefore, the ECU that transmitted the destroyed message is normal as the ECU. Therefore, there is a risk of erroneous determination as a faulty ECU.

  Accordingly, an object of the present invention is to provide a communication system, an abnormality estimation method, and an information reading apparatus that can correctly identify an abnormal part such as a failure.

In order to achieve the above object, a communication system according to the present invention provides:
In a communication system including a plurality of nodes that perform CAN communication with each other via a CAN communication line,
Storage means for storing, for each error type, the number of times of CAN communication error detection detected by each of the plurality of nodes;
Abnormality estimation that estimates the node or the CAN communication line connected to the node having the largest sum of the accumulated value of the error detection times of bit errors and the accumulated value of the error detection times of form errors stored in the storage means as abnormal Means.

  Here, in the communication system, a node that detects at least one of a stuff error, an ACK error, and a CRC error without detecting either the bit error or the form error, or a CAN connected to the node. You may provide the normal determination means which determines a communication line as normal.

In order to achieve the above object, the abnormality estimation method according to the present invention includes:
An abnormality estimation method for a communication system including a plurality of nodes that perform CAN communication with each other via a CAN communication line,
Each time each node detects an error in CAN communication, the number of CAN communication errors detected in each node is accumulated for each error type, and the accumulated error detection count for bit errors and the accumulated error detection count for form errors The CAN communication line connected to the node having the largest sum or the node is estimated to be abnormal.

  Here, in the abnormality estimation method, a node that detects at least one of a stuff error, an ACK error, or a CRC error is detected without detecting either the bit error or the form error, or connected to the node. It may be determined that the CAN communication line is normal.

In order to achieve the above object, an information reading apparatus according to the present invention includes:
An information reading device capable of reading error information detected by a plurality of nodes performing CAN communication with each other via a CAN communication line,
Obtaining means for obtaining a cumulative value of the number of times of CAN communication error detection detected by each of the plurality of nodes based on the error information;
Abnormality estimation means for estimating the node having the maximum sum of the error detection count of bit errors acquired by the acquisition means and the cumulative value of form error error detection or the communication line connected to the node as abnormal Are provided.

  Here, in the information reading apparatus, the error information may be read through the CAN communication line. Further, a node that detects at least one of a stuff error, an ACK error, and a CRC error without detecting either the bit error or the form error based on the read error information or the node The CAN communication line connected to may be provided with a normal determination means for determining normal. Moreover, you may provide the output means which outputs the estimation result by the said abnormality estimation means.

  According to the present invention, an abnormal part such as a failure can be correctly identified.

  The best mode for carrying out the present invention will be described below with reference to the drawings. FIG. 1 is a configuration diagram of a vehicle communication system 100 according to an embodiment of the present invention. The vehicle communication system 100 is a CAN communication system communication system, and includes, for example, four electronic control devices 10, 20, 30, and 40 that are communicably connected to each other via a communication line. By configuring the communication line with a two-wire differential bus, it is possible to improve noise resistance of communication between ECUs. The communication line includes a main line 7 connected to two termination resistors 15 and 25 and branch lines 3, 4 and 5 connected to the main line 7. The main line 7 includes a main line 7H corresponding to a CAN-H line and a main line 7L corresponding to a CAN-L line. The termination resistor 15 (for example, 120Ω) is built in the ECU 10, and the termination resistor 25 (120Ω, which is the same as the termination resistor 15) is built in the ECU 20. The terminal resistor may be inserted between the main line 7H and the main line 7L without being built in the ECU. Branch line 3 (consisting of two lines 3H and 3L) connects main line 7 and ECU 30, branch line 4 (comprising two lines 4H and 4L) connects main line 7 and ECU 40, and branch line 5 (5H and 5L). The main line 7 and the failure diagnosis tool 50 are connected via a connector 60.

  Each ECU and the failure diagnosis tool 50 include a communication interface circuit for serial communication with each other via a communication line as a signal transmission / reception means. If the communication between the ECUs is CAN communication, a communication interface circuit capable of transmission and reception such as a CAN driver is connected to the communication line and provided in each ECU and the failure diagnosis tool 50. Each ECU and the failure diagnosis tool 50 transmit and receive communication data via a communication line, and control processing (for example, engine control processing, brake control processing, charging) that should be controlled based on the communication data received from the other party. Control processing, fault diagnosis control processing, etc.).

  FIG. 2 is a hardware configuration diagram of the ECU 10 incorporating the termination resistor 15. The ECU 20 including the termination resistor 25 has the same hardware configuration as that in FIG. 2, and the ECUs 30 and 40 and the failure diagnosis tool 50 are similar to those in FIG. 2 except that the termination resistor does not exist. Since it has a hardware configuration, the ECU 10 will be described as a representative.

  The ECU 10 is a control means including a microcomputer 12 having a memory for storing a control program and control data and a CPU (Central Processing Unit) for processing the control program.

  The communication interface circuit of the ECU 10 includes a microcomputer 12 having a CAN controller and a CAN transceiver (driver) 12 which is an interface IC between the communication lines.

  FIG. 3 is a block diagram of a CAN transceiver. The CAN transceiver is an IC for an interface between a microcomputer and a communication line. The terminal CANH is a bus output terminal for Hi level, and the terminal CANL is a bus output terminal for Lo level. The terminal Txd is a transmission data input terminal connected to the communication output terminal of the microcomputer, and the terminal Rxd is a reception data output terminal connected to the communication input terminal of the microcomputer. Vcc is a power supply terminal, and GND is a ground terminal. The transmitter outputs differential signal transmission data via the terminal CANH and the terminal CANL by switching the Hi side switch and the Lo side switch based on the input from the terminal Txd. The receiver receives differential signal transmission data from the terminal CANH and the terminal CANL and outputs the transmission data to the terminal Rxd.

  The ECU 10 includes a memory 13. The memory 13 may be a volatile memory, but is desirably a non-volatile memory when it is desired to retain the stored contents even after the power is turned off. A hard disk may also be used. Moreover, you may provide the input circuit into which the information obtained by the sensor is input according to the function given to ECU10, and the output circuit which outputs the calculation result by the microcomputer 12 as a control signal.

  As with the ECU 10, the ECUs 20, 30, and 40 are also provided with a microcomputer, a memory, a transceiver, and a termination resistor (if necessary).

  On the other hand, the failure diagnosis tool 50 also includes a CAN transceiver 51 and a microcomputer 52, as shown in FIG. The failure diagnosis tool 50 can be connected to the branch line 5 of the CAN communication line via a connector 60 provided in the vehicle. Further, the failure diagnosis tool 50 includes a display 58 as a means for providing information to the operator so that the processing result (for example, failure diagnosis result) of the microcomputer 52 can be visually recognized.

  The failure diagnosis tool 50 generates an abnormality by acquiring possession information of each ECU via a communication line composed of a main line and a branch line, or by acquiring possession information of an ECU removed from the communication line. The location is estimated, and the estimation result is provided to the operator via the display 52. When one of the two branch lines connecting the main line 7 of the CAN communication line and the ECU is broken, the failure diagnosis tool 50 breaks the branch line that connects the ECU to the main line 7 or malfunctions in the ECU. And estimate. For example, if any one of the branch line 3H connecting the main line 7H and the ECU 30 at the connection point a and the branch line 3L connecting the main line 7L and the ECU 30 at the connection point b is disconnected, the branch line 3 is disconnected or the ECU 30 is broken. Estimated. Similarly, if any one of the branch line 4H that connects the main line 7H and the ECU 40 at the connection point e and the branch line 4L that connects the main line 7L and the ECU 40 at the connection point f is disconnected, the disconnection of the branch line 4 or the ECU 40 Estimated failure. This will be described below. If one of the constituent lines 7H and 7L of the main line 7 is disconnected, communication via the communication line cannot be performed. However, by measuring the resistance value between 7H and 7L, the two termination resistors 15 , 25, only the resistance value on one side is measured, so that the disconnection of the main line 7 can be easily estimated. Further, even if both of the two branch lines connected to one ECU are disconnected, communication via the communication line excluding the ECU can be continuously executed although there is an abnormality of the disconnection.

  According to the CAN protocol, a communication error is detected in each ECU connected to a communication line composed of a main line and a branch line. In the CAN protocol, five types of errors (bit error, form error, ACK error, CRC error, and stuff error are defined. In short, a bit error has a different level from the bus level sent by itself. A form error is detected when a fixed format bit field is violated, an ACK error (acknowledge error) is detected when no ACK is sent, and a CRC error is A stuff error is detected when a bit stuffing rule is violated, and is detected when the CRC sequence added to the received message is different from the CRC calculated from the received data.

  The ECU that has detected the communication error accumulates the number of error detections for each error type. The accumulated value is stored in the memory of each ECU. The failure diagnosis tool 50 is connected to the vehicle-side connector 60, and all error types stored in each ECU and the cumulative value of the number of detections are read out. By using the failure diagnosis tool 50 deployed at a dealer in each region, the connector 60 for connecting the vehicle and the tool is set in the vehicle in advance, so that the number of times of error detection stored in the ECU can be easily read. it can.

  The failure diagnosis tool 50 estimates a failure location according to the following failure location estimation logic as an abnormality estimation means and a normality determination means.

(1) About ECU which detects only stuff error, ACK error, and CRC error It is considered that the ECU of detection content (1) receives an error message transmitted from another ECU and makes an error determination. Therefore, the failure diagnosis tool 50 determines that the detected content (1) ECU or the branch line connecting the detected content (1) ECU and the main line is normal.

(2) Regarding an ECU that detects at least one of a bit error and a form error As for the ECU of the detected content (2), two types of situations (A) and (B) can be assumed.

          (A) A situation in which a malfunction within the ECU of the detected content (2) or a disconnection of a branch line connecting the ECU of the detected content (2) and the main line has occurred. In the case of a failure in the situation (A), the transmission message cannot be accurately transmitted from the ECU connected to the disconnected branch line.

          (B) The detected content (2) is normal, but the communication is interrupted by another faulty ECU and those errors are detected.

Here, paying attention to the sum of the accumulated values of the number of detections of bit errors and form errors, between two ECUs arbitrarily selected from the ECUs of detection contents (2),
Sum of cumulative number of detections of faulty ECU in situation (A)
≧ A relation of the sum of accumulated values of the number of detections of the normal ECU in the situation (B) can be derived.

  This is because the situation (B) assumes a case where a faulty ECU has interfered with transmission during transmission of a normal ECU. A normal ECU completes transmission normally and does not perform error detection unless the faulty ECU interferes. However, if the faulty ECU performs transmission, a bit error or a form error is always detected.

If the faulty ECU always obstructs transmission during transmission of a single normal ECU, the sum of the cumulative number of bit error and form error detections
Sum of cumulative number of detections of faulty ECU in situation (A)
= The relationship is the sum of the cumulative values of the number of detections of the normal ECU in the situation (B). However, when the failure ECU transmits while a plurality of normal ECUs are transmitting to each other, the sum of the accumulated values of the number of bit error and form error detections,
Sum of cumulative number of detections of faulty ECU in situation (A)
This is because it is the sum of the cumulative values of the number of detections of the normal ECU in the situation (B).

  For example, in FIG. 1, it is assumed that a disconnection has occurred in the constituent line 3H of the branch line 3 between the terminal 30H of the ECU 30 and the connection point a. In this case, each of the ECUs 10, 20, and 40 detects at least one of a bit error and a form error, and the number of error detections stored in the memory of each ECU 10 Updated every time. However, since communication between the ECUs excluding the ECU 30 (that is, communication between the ECU 10 and the ECU 20, communication between the ECU 10 and the ECU 40, communication between the ECU 20 and the ECU 40) can be performed normally, it is connected to the disconnected branch line. The sum of the accumulated values of the bit error and form error detection times stored in the memory of the ECU 30 is the cumulative value of the bit error and form error detection times stored in the respective memories of the ECUs 10, 20, and 40. Therefore, it can be estimated that the ECU having the largest sum of the cumulative values is likely to be a faulty ECU. When there are a plurality of ECUs having the same cumulative value, it can be estimated that a faulty ECU exists among the plurality of ECUs.

  Therefore, the failure diagnosis tool 50 acquires the accumulated value of the number of times of error detection stored in the memory in each ECU, thereby accumulating the number of times of detection of the bit error and the form error in the ECU connected to the communication line. The branch line connected to the ECU having the highest possibility that the ECU having the memory storing the maximum value is in failure or having the memory storing the maximum value is It can be estimated that there is the highest possibility of disconnection. And it is estimated that the smaller the sum of the accumulated values of bit error and form error detection times, the lower the possibility that the ECU is out of order or the possibility that the branch line connected to the ECU is broken it can.

  In this way, a user such as a serviceman at a repair shop can easily check the failure of the communication line or ECU of the part where the possibility of the failure is high by following the failure estimation information of the failure diagnosis tool 50. As a result, repair can be easily performed.

  FIG. 4 is a memory flow executed by each microcomputer of each ECU. When each ECU detects a communication error (step 10), it stores a cumulative value of the number of error detections for each error type (step 12). FIG. 5 is a cumulative value output flow executed by the microcomputer of each ECU. When each ECU receives an output request for the cumulative value of the number of error detections from the failure diagnosis tool 50 or the like (step 20), the ECU outputs the cumulative value stored in its own memory via a communication interface circuit such as a CAN transceiver. (Step 22).

  FIG. 6 is a failure estimation flow executed by the microcomputer 52 of the failure diagnosis tool 50. The failure diagnosis tool 50 makes an output request for the accumulated value of the number of error detections according to the operation of the operator (step 30), and acquires the accumulated value of each ECU (step 32). The ECU having the maximum accumulated value among all ECUs is estimated as a faulty ECU, or the branch line connected to the maximum ECU is estimated to be disconnected (step 34). The estimation result is output and displayed on the display 52 (step 22).

  Therefore, according to the above-described embodiment, when the branch line of the communication line is broken on one side, the broken portion can be specified, and the estimation accuracy of the abnormal portion can be improved.

  The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the above-described embodiments, and various modifications and substitutions can be made to the above-described embodiments without departing from the scope of the present invention. Can be added.

  For example, the ECU is exemplified as the node connected to the communication line, but any node capable of CAN communication such as an intelligent sensor may be used. Further, the number of nodes is not limited to four, and three or five or more nodes can be considered similarly.

  Although the example in which the accumulated value of the number of error detections of the error detected by each ECU is stored in its own memory for each error type is shown, it is information on which ECU has detected even if it is not stored in its own memory May be stored in a memory other than itself (for example, a memory in the ECU having a built-in termination resistor) in a state where it can be specified (for example, provision of identification information such as an ID for specifying the ECU).

  Also, other communication lines such as IS0-K line and LIN, showing an example in which the stored information of each ECU is transmitted to the failure diagnosis tool 50 via the CAN communication line including the main line 7 and the branch lines 3, 4, 5. The stored information of each ECU may be transmitted to the failure diagnosis tool 50 via

  Further, the above-described failure location estimation logic may be incorporated in the node itself (ECU itself) connected to the communication line, and the ECU may estimate the failure location as abnormality estimation means and / or normality determination means. Each ECU needs to transmit the stored cumulative value of the number of error detections to the ECU that performs failure location estimation via the CAN communication line or another communication line. After the transmission, the failure location is estimated by the same estimation logic in the failure diagnosis tool 50 described above. The ECU that has estimated the failure location outputs the estimation result to an output unit such as a vehicle-mounted display or the display 58 of the failure diagnosis tool 50.

1 is a configuration diagram of a vehicle communication system 100 according to an embodiment of the present invention. It is a hardware block diagram of ECU10 which incorporates termination resistance 15. It is a block diagram of a CAN transceiver. It is the memory | storage flow which each microcomputer of ECU implements. It is a cumulative value output flow executed by the microcomputer of each ECU. It is a failure estimation flow executed by the microcomputer 52 of the failure diagnosis tool 50.

Explanation of symbols

3, 4, 5 Branch line 7 Main line 10, 20, 30, 40 ECU
11, 21, 31, 41, 51 CAN driver (CAN transceiver)
12, 52 Microcomputer 13 Memory 15, 25 Termination resistor 50 Fault diagnosis tool 58 Display 60 Connector 100 Vehicle communication system

Claims (8)

In a communication system including a plurality of nodes that perform CAN communication with each other via a CAN communication line,
Storage means for storing, for each error type, the number of times of CAN communication error detection detected by each of the plurality of nodes;
Abnormality estimation that estimates the node or the CAN communication line connected to the node having the largest sum of the accumulated value of the error detection times of bit errors and the accumulated value of the error detection times of form errors stored in the storage means as abnormal And a communication system.   Normal in which neither the bit error nor the form error is detected, and a node that detects at least one of a stuff error, an ACK error, and a CRC error or a CAN communication line connected to the node is determined to be normal The communication system according to claim 1, further comprising a determination unit. An abnormality estimation method for a communication system including a plurality of nodes that perform CAN communication with each other via a CAN communication line,
Each time each node detects an error in CAN communication, the number of CAN communication errors detected in each node is accumulated for each error type, and the accumulated error detection count for bit errors and the accumulated error detection count for form errors An abnormality estimation method for estimating a node having the largest sum of or a CAN communication line connected to the node as an abnormality.   A node that detects at least one of a stuff error, an ACK error, and a CRC error without detecting either the bit error or the form error, or a CAN communication line connected to the node is determined to be normal. The abnormality estimation method according to claim 3. An information reading device capable of reading error information detected by a plurality of nodes performing CAN communication with each other via a CAN communication line,
Obtaining means for obtaining a cumulative value of the number of times of CAN communication error detection detected by each of the plurality of nodes based on the error information;
Abnormality estimation means for estimating the node having the maximum sum of the error detection count of bit errors acquired by the acquisition means and the cumulative value of form error error detection or the communication line connected to the node as abnormal An information reading device comprising:   The information reading device according to claim 5, wherein the error information is read via the CAN communication line.   Based on the read error information, neither the bit error nor the form error is detected, and a node detecting at least one of a stuff error, an ACK error, and a CRC error is connected or connected to the node The information reading device according to claim 5, further comprising a normality determination unit that determines that the CAN communication line to be performed is normal.   The information reading apparatus according to claim 5, further comprising an output unit that outputs an estimation result obtained by the abnormality estimation unit.

Images