JP2018109900A - In-vehicle electronic control apparatus and log storing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an in-vehicle electronic control apparatus and a log storing method capable of reducing a storage area required for recording an event.SOLUTION: An in-vehicle electronic control apparatus 100 has a log area 122 in which a log of an event that has occurred is recorded, a storage unit 112 for storing pattern information 112A including a normal order which is an occurrence order of normal events and a character string corresponding to the normal order, and a compression unit for compressing the log by recording the character string in place of the normal order in the log area.SELECTED DRAWING: Figure 2

Description

  The present invention relates to an on-vehicle electronic control device and a log storage method.

  Many electronic devices are mounted on recent vehicles. In order to investigate the cause of a problem occurring in a vehicle, it is important to analyze a log that is a record of the operation of these electronic devices. Patent Document 1 discloses a driving status information acquisition unit that acquires driving status information including one or both of an operating status and a vehicle status during driving of the vehicle, and the driving acquired during a period from ON to OFF of the vehicle main power source. A log generation unit that sequentially generates a vehicle driving log including at least a part of the situation information, and a vehicle driving log that is sequentially generated by the log generation unit using the driving status information belongs to the same travel. A log identity determination unit for determining whether or not, a log processing unit for connecting the vehicle operation logs based on a determination result by the log identity determination unit, and the vehicle operation log sorted in the same travel unit A vehicle operation log processing system including a log storage unit is disclosed.

JP 2010-152687 A

  In the invention described in Patent Document 1, the storage area required for event recording cannot be reduced.

The on-vehicle electronic control device according to the first aspect of the present invention includes a log area in which a log of an event that has occurred is recorded, a normal order that is the occurrence order of the event at normal time, and a character string corresponding to the normal order. A storage unit storing pattern information to be included, and a compression unit that compresses the log by recording the character string in the log area instead of the normal order.
The log storage method according to the second aspect of the present invention includes a pattern including a log area in which a log of an event that has occurred is recorded, a normal order that is the order of occurrence of the event at normal time, and a character string corresponding to the normal order A storage unit for storing information, and a log storage method executed in an in-vehicle electronic control device, wherein the in-vehicle electronic control device records the character string in the log area instead of the normal order. Compress the log.

  According to the present invention, the storage area required for event recording can be reduced.

The figure which shows the hardware constitutions of ECU The figure which shows the function structure of ECU in 1st Embodiment. The figure which shows an example of event information The figure which shows an example of the pattern information in 1st Embodiment Diagram showing an example of a temporary buffer Figure showing an example of the log area The figure which shows the operation example of the pattern recognition part in 1st Embodiment The flowchart which shows operation | movement of ECU in 1st Embodiment. The flowchart which shows operation | movement of ECU in 1st Embodiment. The figure which shows an example of the pattern information in the modification 1 The figure which shows an example of the normal order recorded in a tree format in the modification 2 The figure which shows the function structure of ECU in 2nd Embodiment. The figure which shows the function structure of ECU in 3rd Embodiment. The figure which shows an example of a time difference buffer The flowchart which shows operation | movement of ECU in 3rd Embodiment. The flowchart which shows operation | movement of ECU in 3rd Embodiment. The figure which shows the operation example in 3rd Embodiment

-First embodiment-
Hereinafter, a first embodiment of an ECU that is an in-vehicle electronic control device will be described with reference to FIGS. The ECU according to the present invention is mounted on a vehicle. Hereinafter, one ECU will be described, but the number of ECUs mounted on the vehicle is not limited to this.

(ECU hardware configuration)
FIG. 1 is a diagram illustrating a hardware configuration of the ECU 100 according to the first embodiment. The ECU 100 includes a CPU 11, a ROM 12, a RAM 13, and a nonvolatile memory 14. The CPU 11 develops a program stored in the ROM 12 and the nonvolatile memory 14 in the RAM 13 and executes it. The ROM 12 stores a program that does not need to be changed and constants used by the program, for example, information that defines a timeout. The RAM 13 is a volatile memory and is used for program development and temporary information storage used by the program. The nonvolatile memory 14 is a storage area in which information is stored even when power supply is cut off, and stores an operation log of the program and a program that may be updated, such as an operating system (OS).

(Functional configuration of ECU)
FIG. 2 is a diagram illustrating a functional configuration of the ECU 100. The ECU 100 includes a time register 111, a pattern recognition unit 112, event information 113, a temporary buffer 121, a log area 122, and a snapshot area 123.

  The time register 111 is a specific register built in the CPU 11. The time register 111 stores information indicating the current time, for example, “January 2, 2016 3: 45: 6.7890123456 seconds”. The value stored in the time register 111 is updated based on the operation of the oscillator built in the CPU 11, and the program executed by the CPU 11 can acquire the current time by referring to the time register 111.

  The pattern recognizing unit 112 recognizes the order of occurrence of events and determines whether or not the pattern matches a predetermined pattern. If the pattern recognition unit 112 determines that the pattern matches the predetermined pattern, an identifier corresponding to the corresponding pattern (hereinafter, “ Called "Pattern Identifier"). The predetermined pattern and the identifier corresponding to each pattern are stored in the ROM 12 or the nonvolatile memory 14 as the pattern information 112A. In the pattern information 112A, for each event included in the pattern, an assumed time interval from the occurrence of the previous event to the occurrence of the event is also stored. An example of the pattern information 112A will be described later. In the present embodiment, all the event patterns stored in the pattern information 112A are assumed to be the normal event order (hereinafter referred to as “normal order”). Furthermore, the pattern recognition unit 112 includes a not-illustrated notification unit, and when an abnormality is detected, the pattern recognition unit 112 notifies the occurrence of the abnormality using the notification unit. The notification unit may communicate with other devices connected to the ECU 100 to notify the abnormality, or may notify the abnormality by recording the occurrence of the abnormality in the RAM 13 or the nonvolatile memory 14.

  In the event information 113, for a plurality of events, an event ID that is an identifier of the event and identification information of the event are stored in association with each other. The event information 113 is a storage area secured in the nonvolatile memory 14. An example of the event information 113 will be described later.

  The temporary buffer 121 is an area in which event identification information and event occurrence time are temporarily recorded. However, an event ID may be recorded instead of the event identification information. The temporary buffer 121 is an area secured in the RAM 13. As will be described in detail later, the event recorded in the temporary buffer 121 is an event that may match any of the patterns described in the pattern information 112A. The event information recorded in the temporary buffer 121 is recorded in the log area 122 as it is or after being replaced. When event information is recorded in the log area 122, the event information recorded in the temporary buffer 121 is deleted. An example of the temporary buffer 121 will be described later.

  The log area 122 is an area in which event identification information and event occurrence time are recorded. However, an event ID or a pattern identifier may be recorded instead of the event identification information. The log area 122 is an area secured in the nonvolatile memory 14. The event information recorded in the log area 122 is read outside the ECU 100 via an interface (not shown). An example of the log area 122 will be described later.

  The snapshot area 123 is an area where the operation state of the ECU 100 when an abnormality of the ECU 100 is detected is recorded. The recorded operation state of the ECU 100 includes information recorded in the log area 122.

(event information)
FIG. 3 is a diagram illustrating an example of the event information 113. The event information 113 is composed of, for example, a plurality of records, and each record has a field in which an event ID that is an event identifier is stored and a field in which event identification information is stored. There is no limit to the number of records in the event information 113, that is, the number of events recorded in the event information 113. The event ID is, for example, a one-letter alphabet. In the present embodiment, A, B, C, X, Y, and Z are assigned to six events. The event identification information is various information for identifying an event. In other words, the “event” in the present embodiment is an event identified using the event information 113.

  The configuration of the event identification information is not limited, but may be a combination of an event property representing an event type and an event parameter representing a parameter value, for example. The event property is, for example, reception of a CAN frame, matching to a specific value of a variable in the program, or the like. The event parameter is, for example, a combination of a CANID value and a value stored in the data field when the event property is reception of a CAN frame. In the following description, an event whose event ID is A is also referred to as event A or the like. Since the CPU 11 can access information inside the CPU 11 and information stored in the RAM 13 and the nonvolatile memory 14, the event identification information can be acquired when an event occurs.

(Pattern information)
FIG. 4 is a diagram illustrating an example of the pattern information 112A. The pattern information 112A is composed of, for example, a plurality of records. Each record is a field in which pre-replacement information, which is a predetermined order of event occurrence and an event occurrence time interval, is stored, and an identifier corresponding to the event pattern. And a field for storing certain post-replacement information. There are no restrictions on the number of records in the pattern information 112A and the number of events constituting the pattern in each record.

  As illustrated in FIG. 4, the pre-replacement information field includes the first event to the fourth event in the order of occurrence, and the occurrence time interval assumed for each event. In the pattern information 112A, event identification information may be used as information for identifying an event, but an event ID is used here. For example, in the first example, the expected event occurrence order is A, X, Y, B, and in the second example, A, X, C. In FIG. 4, the time interval of occurrence assumed for each event is recorded in the field of “assumed interval”. The unit of time described in this field is, for example, nanoseconds or microseconds.

  For example, in the first example, “500” described between A of the first event and X of the second event is an estimated time interval from the occurrence of event A to the occurrence of event X. “500”. Similarly, the assumed time interval from the occurrence of event X to the occurrence of event Y is “2000”, and the assumed time interval from the occurrence of event Y to the occurrence of event B is “2000”. 300 ". In the post-replacement information field, P is described in the first example and Q is described in the second example as a pattern identifier that matches the pattern of the event that matches the pre-replacement information of the same record. Details of the replacement process will be described later.

(Temporary buffer)
FIG. 5 is a diagram illustrating an example of the temporary buffer 121. The temporary buffer 121 is composed of, for example, a plurality of records, and each record stores event identification information and a time stamp indicating the time when the event occurred.

(Log area)
FIG. 6 is a diagram illustrating an example of the log area 122. The log area 122 is composed of, for example, a plurality of records, and each record stores event identification information and a time stamp indicating the time when the event occurred. However, an event ID may be stored instead of the event identification information, or a pattern identifier included in the pattern information 112A may be stored.

(Operation of pattern recognition unit)
An operation and an operation example of the pattern recognition unit 112 will be described. When the order of events stored in the temporary buffer 121 matches one of the event patterns stored in the pattern information 112A, the pattern recognition unit 112 performs the following replacement process. That is, the event ID is replaced with the corresponding pattern identifier, and the time when the last event of the pattern occurs is set as the time when the event of the pattern identifier occurs. Also, the pattern recognition unit 112 does not match the event order stored in the temporary buffer 121 with any event pattern stored in the pattern information 112A, but matches any event pattern halfway. The occurrence of abnormality is notified using the notification unit without performing replacement. Furthermore, the pattern recognition unit 112 determines that the event sequence stored in the temporary buffer 121 does not match the pattern of any event stored in the pattern information 112A at all, that is, the first event stored in the temporary buffer 121 is the pattern information. If it does not match the first event of any pattern stored in 112A, neither replacement nor notification is performed.

  FIG. 7 is a diagram illustrating an operation example of the pattern recognition unit 112 in three cases of case 1 to case 3. FIG. 7 shows the event occurrence status and information stored in the log area 122 after the occurrence of all events for each case. In the event occurrence status column, event IDs are described in the order in which the events occurred, and the time when the occurrence of the event was observed is also written. In any case, it is assumed that nothing is recorded in the temporary buffer 121 when the first event occurs. In both cases, the time when the first event is observed is set to zero.

  In case 1, the order of events that have occurred is represented by event IDs “A, X, Y, B, C”. Since the four events from the top match the first pattern in the pattern information 112A, the pattern identifier P is recorded in the log area 122 instead of these four events. The time stamp records “2690”, which is the time when event B, which is the last event of this pattern, was observed. Since the fifth event C does not match any pattern in the pattern information 112A, the event ID and the observation time are recorded as they are.

  In case 2, the order of events that have occurred is represented by event IDs “A, X, Z”. This event pattern does not match any pattern in the pattern information 112A, but matches the first pattern partway. Specifically, the first two events “A, X” that have occurred coincide with the first two “A, X, Y, B” that is the first pattern of the pattern information 112A. For this reason, since the whole does not match, replacement is not performed, but an abnormality is reported because it partially matches. In the log area 122, the event ID of the event that has occurred and the observation time are described as they are.

  In case 3, only event C occurred. Since there is no pattern starting from the event C in the pattern information 112A, the pattern recognition unit 112 does not perform replacement. Therefore, in the log area 122, the event C that has occurred and the observation time of the event C are described as they are.

(flowchart)
8 and 9 are flowcharts showing the operation of the ECU 100 in the first embodiment. The CPU 11 executes a program whose operation is represented by FIGS. 8 and 9 every time an event occurs in the ECU 100. The execution subject of each step in the flowchart described below is the CPU 11 of the ECU 100.
In step S302, the CPU 11 executes the generated event. In subsequent step S <b> 304, the CPU 11 acquires information indicating the current time, that is, a time stamp, from the time register 111. Hereinafter, the time represented by the time stamp is also referred to as “event occurrence time” or “event observed time”. In subsequent step S306, the CPU 11 specifies the event ID of the event that has occurred based on the event information 113 and the event identification information of the event that has occurred.

  In the subsequent step S308, the CPU 11 determines whether or not the event pattern obtained by adding the event ID specified in step S306 to the event ID stored in the temporary buffer 121 is forwardly matched with any pattern of the pattern recognition unit 112. to decide. For example, when “A, X” is stored in the temporary buffer 121 and the event ID identified in step S306 is “Y”, the following determination is made. That is, “A, X, Y” obtained by adding “Y” to “A, X” matches the first three of “A, X, Y, B” which is the first pattern of the pattern recognition unit 112. Therefore, an affirmative determination is made for the forward match search. If the CPU 11 makes an affirmative determination, the process proceeds to step S310, and if the determination is negative, the CPU 11 proceeds to step S314. If nothing is recorded in the temporary buffer 121, whether or not there is a forward matching search using only the event ID specified in step S306, that is, there is a pattern heading the event ID specified in step S306. Make a decision.

  In step S310, the CPU 11 determines whether there is a timeout. That is, it is determined whether or not the difference between the occurrence time of the event that occurred immediately before and the occurrence time of the current event is longer than a predetermined time. The timeout time may be stored in the ROM 12, may be the same for all events, or may be set to a different value for each event. However, in this embodiment, it is assumed that there is no correlation between the timeout time and the assumed interval of the pattern recognition unit 112. If the CPU 11 determines that a timeout has not occurred, that is, the difference between the occurrence time of the event that occurred immediately before and the occurrence time of the current event is within a predetermined time, the CPU 11 proceeds to step S312 and the timeout has occurred. If YES, the process proceeds to step S314.

  In step S312, the CPU 11 stores the latest event ID and time stamp, that is, the event ID specified in step S306 and the time stamp acquired in step S304 in the temporary buffer 121, and the program whose operation is represented by FIGS. The execution of is terminated. In step S314, which is executed when a negative determination is made in step S308 or step S310, the CPU 11 determines whether something is recorded in the temporary buffer 121 or not. If the CPU 11 determines that nothing is recorded, the process proceeds to step S316. If the CPU 11 determines that some information is recorded, the process proceeds to step S330 in FIG. In step S316, the CPU 11 stores the latest event ID and time stamp, that is, the event ID specified in step S306 and the time stamp acquired in step S304 in the log area 122, and the program whose operation is represented by FIG. 8 and FIG. The execution of is terminated.

  If a negative determination is made in step S314 in FIG. 8, step S330 in FIG. 9 is executed. In step S330, the CPU 11 determines whether the ID string in the temporary buffer 121, that is, the event pattern stored in the temporary buffer 121, can be replaced based on the pattern information 112A. If it is determined that replacement is possible, the process proceeds to step S332, and if it is determined that replacement is not possible, the process proceeds to step S340.

  In step S332, the CPU 11 stores the pattern identifier obtained by the replacement determined to be replaceable in step S330 and the time stamp after the replacement in the log area 122. As described above, the time stamp after replacement is a time stamp corresponding to the time when the last event included in the corresponding pattern occurs. In the subsequent step S334, the CPU 11 wipes out the temporary buffer, that is, deletes all the information stored in the temporary buffer 121. In subsequent step S335, it is determined whether or not the latest event ID, that is, the event ID specified in step S306 coincides with the head of any of the patterns described in the pattern information 112A. The CPU 11 proceeds to step S336 when determining that it matches the head of any pattern, and proceeds to step S337 when determining that it does not match any of the patterns. In step S336, the latest event ID and time stamp, that is, the event ID specified in step S306 and the time stamp acquired in step S304 are stored in the temporary buffer 121, and the program whose operation is represented by FIGS. 8 and 9 is executed. Exit. In step S337, the latest event ID and time stamp are stored in the log area 122, and the execution of the program whose operation is represented by FIGS.

  In step S340, which is executed when a negative determination is made in step S330, the CPU 11 records all the information in the temporary buffer 121 in the log area 122 as it is, and deletes all the information in the temporary buffer 121. In subsequent step S342, the latest event ID and time stamp, that is, the event ID specified in step S306 and the time stamp acquired in step S304 are stored in the log area 122. In subsequent step S344, the CPU 11 notifies the abnormality using a not-illustrated notification unit. In subsequent step S346, the CPU 11 records all the information recorded in the log area 122 in the snapshot area 123, and ends the execution of the program whose operation is represented by FIGS.

(Operation example)
For each case shown in FIG. 7, the determination performed by the CPU 11 in the flowcharts shown in FIGS. 8 to 9 will be described. However, in the following, the determination at the branch of the flowchart will be mainly described, and the description of the part without the branch will be omitted. As described above, nothing is recorded in the temporary buffer 121 when the first event occurs in any case. In any case, the determination in step S310 in FIG. 8 is a time-out occurrence, that is, an affirmative determination.

  In the case 1 of FIG. 7, when the event A occurs, the CPU 11 makes an affirmative determination in step S308, and records the event A and its time stamp in the temporary buffer 121 in step S312. When any of events X, Y, and B occurs, the same determination as in event A is made. This is because “A”, “A, X”, “A, X, Y” and “A, X, Y, B” are all the first patterns of the pattern information 112A shown in FIG. This is because an affirmative determination is made in the forward matching search for “A, X, Y, B”.

  When event C finally occurs, in step S308, CPU 11 makes a negative determination for “A, X, Y, B, C” by a forward matching search for “A, X, Y, B”. In step S314, since the event “A, X, Y, B” is recorded in the temporary buffer 121, a negative determination is made, and the process proceeds to step S330. In step S330, since the event “A, X, Y, B” recorded in the temporary buffer 121 can be replaced with “P”, the CPU 11 makes an affirmative determination and proceeds to step S332. In step S332, the time stamp indicating the time when the pattern identifier P and event B occurred is stored in the log area 122, and the information in the temporary buffer 121 is deleted in step S334. In step S335, since the latest event C does not coincide with the head of any pattern in the pattern information 112A, the process proceeds to step S337, and the event C and its time stamp are recorded in the log area 122 in step S337.

  In case 2 of FIG. 7, the operation of the CPU 11 when event A and event X occur is the same as in case 1. When the event Z occurs, “A, X, Z” does not match any pattern of the pattern recognition unit 112, so the CPU 11 makes a negative determination in step S 308. In step S314, since there is information stored in the temporary buffer 121, the CPU 11 makes a negative determination and proceeds to step S330. In step S330, since “A, X” in the temporary buffer 121 cannot be replaced, the CPU 11 makes a negative determination, and records the event “A, X, Z” and its time stamp in the log area 122 in step S340 and subsequent steps.

  In case 3 of FIG. 7, the CPU 11 makes a negative determination in step S <b> 308, makes a positive determination in step S <b> 314, and records the event C and its time stamp in the log area 122 in step S <b> 316.

According to the first embodiment described above, the following operational effects are obtained.
(1) The ECU 100, which is an on-vehicle electronic control device, includes a log area 122 in which a log of an event that has occurred is recorded, a normal order that is a normal event generation order, and a pattern identifier that is a character string corresponding to the normal order. And a compression unit that compresses the log by recording the pattern identifier in the log area 122 instead of the normal order, that is, the pattern recognition unit 112.

  Information on events that occur in the normal order is not necessarily highly important in investigating the cause of the occurrence of a failure, but it is difficult to find the cause if no information on events that occur in the normal order is obtained. The ECU 100 stores the normal order and the character string corresponding to the normal order, so that the normal order can be determined. Furthermore, the ECU 100 can record the occurrence of events in the normal order using a small storage area by recording the pattern identifier instead of recording the normal order itself. That is, the ECU 100 can reduce the storage area required for event recording.

(2) The pattern recognition unit 112 records the event in the log area 122 without compressing the log when the occurrence order of the event does not match the normal order. For investigating the cause of the occurrence of a failure, information on events that occur in an unexpected order is more important than information on events that occur in an expected normal order. Therefore, events that occur in an unexpected order can be recorded in the pattern recognition unit 112 without being compressed, and can be used for investigating the cause of the failure.

(3) The pattern recognition unit 112 notifies the occurrence of an abnormality using a not-shown notification unit when the event generation order does not completely match the normal order and does not completely match (step S330 in FIG. 9: NO). (Step S344 in FIG. 9). The case where the order does not completely match the normal order but does not match the normal order is not clear because the case where the order does not match the normal order from the middle. Therefore, it can be clarified that abnormality has occurred by performing notification.

(4) Whenever the event occurs, the pattern recognition unit 112 determines whether the order of events that have occurred matches the normal order, and determines that the order matches the normal order (step S308 in FIG. 8: YES). The range to be temporarily stored is stored (step S312 in FIG. 8) to reduce the load of the next match determination.

  Therefore, the load on the pattern recognition unit 112 can be leveled and the real-time property of the ECU 100 can be ensured. For example, it is conceivable to perform a replacement of a certain amount after referring to the pattern information 112A after a certain amount of time has elapsed or after a predetermined amount of logs has been accumulated. The pattern recognition unit 112 uses many resources of the ECU 100 at the timing of processing. Therefore, at that timing, resources that can be used by other programs are reduced, and there is a possibility that the predetermined processing cannot be completed within a predetermined time, that is, the real-time property may not be ensured. As described above, since the ECU 100 can ensure real-time performance, the ECU 100 is suitable for a vehicle-mounted electronic control device that requires real-time processing.

(Modification 1)
The pattern information may store abnormal pattern information including an abnormal order that is the order of occurrence of events at the time of abnormality and an abnormal character string corresponding to the abnormal order. The abnormal pattern information is also composed of pre-replacement information and post-replacement information, similar to the pattern information 112A shown in FIG. When the order of events indicated in the pre-replacement information matches the order of events that have occurred, a pattern identifier is recorded in the log area 122 and notification of abnormality detection is also performed.

  FIG. 10 is a diagram illustrating an example of the pattern information 112B in the first modification. In the field constituting the pattern information 112B, a “normal / abnormal” field is added as compared with the pattern information 112A shown in FIG. This field indicates whether the record is a normal pattern or an abnormal pattern. That is, in the example shown in FIG. 10, the same two normal patterns and one abnormal pattern as the example of FIG. 4 are shown.

According to the first modification, the following operational effects can be obtained.
(5) The non-volatile memory 14 stores abnormal pattern information including an abnormal order that is the order of occurrence of events in the event of an abnormality and an abnormal character string corresponding to the abnormal order. When the event occurrence order matches the abnormal order, the pattern recognition unit 112 records an abnormal character string in the log area 122 instead of the abnormal order, and notifies the occurrence of abnormality using a not-illustrated notification unit. . Therefore, it is possible to reduce the storage area required for log recording by predetermining the abnormal order that is expected to occur.

(Modification 2)
In the first embodiment described above, the occurrence order of events included in the pattern information 112A, that is, pre-substitution information is recorded in a table format composed of a plurality of fields. However, it may be recorded in a tree format for easy tracing. In this case, the pattern recognition unit 112 records the current position on the node, and performs a sequential search that traces the tree one node each time an event occurs, whereby step S308 in FIG. 8 and step S330 in FIG. 9 are performed. Make a decision.

  FIG. 11 is a diagram showing an example of pre-replacement information of the pattern information 112A recorded in a tree format. The generation order of events shown in FIG. 11 is the same as the example of the pattern information 112A shown in FIG. N1 to N5 shown in FIG. 11 are given for convenience in order to identify the nodes constituting the tree. The tree shown in FIG. 11 spreads downward in the figure with N1 as a vertex. For example, it is possible to transition from N2 to N3 or N5, but from N3 to N4 only. For example, when three events occur in the order of A, X, and Y, the pattern recognition unit 112 recognizes that the current position on the tree is N3. When event C occurs next, “C” is not stored in N4 that can transition from N3, and therefore a negative determination is made in step S308 in FIG. A similar determination can be made in step S330 of FIG.

According to the second modification, the following operational effects can be obtained.
(6) The normal order is stored as a tree-structured database, and the pattern recognition unit 112 sequentially searches the tree-structured database every time an event occurs. Therefore, the load required for one search is leveled regardless of the current position in the normal order, which contributes to ensuring the real-time property of the ECU 100. This effect becomes more prominent as the normal order is longer.

(Modification 3)
In the first embodiment described above, when the ECU 100 detects an abnormality, all information recorded in the log area 122 is recorded in the snapshot area 123 (S346 in FIG. 9). The information may be saved. ECU 100 may transmit the information to a diagnostic device (not shown) provided in the same vehicle. Further, the ECU 100 may perform some countermeasure, for example, the operation of the braking device based on the detection of the abnormality.

(Modification 4)
In the first embodiment described above, the pattern recognition unit 112 transcribes the time stamp of the last event in the log area 122 when replacement is performed. However, the time stamps of all events may be posted. For example, in case 1 shown in FIG. 7A, the time stamp of event P may be recorded as “0, 410, 2490, 2690”.

(Modification 5)
In the first embodiment described above, an event ID is used as event identification information recorded in the log area 122. However, the event identification information shown in FIG. 4 may be used instead of the event ID, or information not shown in FIG. 4 may be recorded in the log area 122 as long as the information can identify the event. According to this modification, an unexpected event when an abnormality occurs can be recorded in the log area 122.

(Modification 6)
The event information 113 may not be directly saved in the nonvolatile memory 14 but may be created on the RAM 12 by executing a program stored in the nonvolatile memory 14. For example, an operating system (OS) may be stored in the non-volatile memory 14, and when the CPU 11 executes the OS, the OS may create the event information 113 on the RAM 12 as part of the startup process.

(Modification 7)
In the first embodiment, the occurrence of an abnormality is notified only when the order of events that have occurred does not completely match the normal order and does not match partway. However, the occurrence of an abnormality may be notified when it does not match the normal order regardless of whether or not it is partially matched. That is, steps S334 and S346 in FIG. 9 may be inserted immediately before or after step S316 in FIG.

(Modification 8)
In the first embodiment, the pattern information 112A may not include the “assumed interval” field.

-Second embodiment-
With reference to FIG. 12, a second embodiment of an ECU that is an in-vehicle electronic control device will be described. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and different points will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that the CPU does not need to specify an event ID in the program. Since the hardware configuration of the ECU in the second embodiment is the same as that of the first embodiment, description thereof is omitted. However, the program stored in the ROM is different from that in the first embodiment.

(Functional configuration of ECU)
FIG. 12 is a diagram illustrating a functional configuration of the ECU 100A in the second embodiment. In the ECU 100A, the event information 113 is removed from the configuration in the first embodiment. In the present embodiment, event IDs are not required because event IDs are assigned to all events that occur. The event ID is given by the OS stored in the nonvolatile memory 14, for example.

(ECU operation)
The operation of the ECU 100A in the second embodiment is the same as that of the first embodiment except for the following points. That is, the CPU 11 in the second embodiment does not have to execute step S306 in FIG. 8 of the first embodiment.

  According to the second embodiment described above, the ECU 100A does not need to store the event information 113, and does not have to specify the event ID based on the event identification information.

-Third embodiment-
A third embodiment of an ECU that is an in-vehicle electronic control device will be described with reference to FIGS. In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and different points will be mainly described. Points that are not particularly described are the same as those in the first embodiment. This embodiment differs from the first embodiment mainly in that not only event information but also time stamp compression is performed. Since the hardware configuration of the ECU in the third embodiment is the same as that of the first embodiment, the description thereof is omitted. However, the program stored in the ROM is different from that in the first embodiment.

(Functional configuration of ECU)
FIG. 13 is a diagram illustrating a functional configuration of the ECU 100B according to the third embodiment. In the ECU 100B, the event information 113 is removed from the configuration in the first embodiment, and a time difference buffer 124 is added. The time difference buffer 124 is a temporary storage area secured in the RAM 13 and is used to calculate a second-order difference described later. In the present embodiment, as in the second embodiment, event IDs are not required because event IDs are attached to all the events that occur. The event ID is given by the OS stored in the nonvolatile memory 14, for example.

(ECU operation)
When replacing the event pattern, the CPU 11 in the present embodiment records the second-order difference described below instead of recording the time stamp of the time when the last event of the pattern occurred. The second floor difference is a difference between the actual time difference and the assumed time difference. The actual time difference is the difference between the times when two events occur and is the difference between the actually observed times. The assumed time difference is a difference in time when two events occur, and is a difference in expected time. The assumed time difference is a predetermined value and is stored in the ROM 12 or the nonvolatile memory 14. The assumed time difference is recorded as “assumed interval” in the pattern information 112A, for example.

For example, when the occurrence times of event A and event B are observed as “January 2, 2016 3: 00: 0.000000000” and “January 2, 2016 3: 01: 2.33344555”, respectively. In addition, the actual time difference is “1 minute 2.333444555”. When the assumed time difference is “1 minute 2.33344333 seconds”, the difference between the actual time difference and the assumed time difference is “0.0000000022 seconds”. That is, since it can be expressed as “2.22 × 10 ⁻⁷ seconds” or “2.22E-7”, the second-order difference has a smaller data amount than the actual time difference. Note that the higher the time resolution and the smaller the difference between the assumed time difference and the actual time difference, the greater the difference in the data amount between the actual time difference and the second-order difference.

(Time difference buffer)
FIG. 14 is a diagram illustrating an example of the time difference buffer 124. The time difference buffer 124 is composed of, for example, a plurality of records, and each record is assumed to be an inter-event field that indicates an interval between events, and a real time difference field that is a difference between actually observed times. And a second-order difference field in which a difference between the value of the actual time difference field and the value of the assumed time difference field is stored.

  The occurrence order of events and the actual time difference in FIG. 14 are based on the top four events in case 1 shown in FIG. That is, the order in which the events occurred is “A, X, Y, B”, and the respective time stamps are “0, 410, 2490, 2690”. Therefore, the actual time difference is “410, 2080, 200”. The assumed time difference in FIG. 13 is “500, 2000, 300” which is the assumed interval of the pattern information 112A shown in FIG. The second-order difference shown in FIG. 14 is a value obtained by subtracting the value of the assumed time difference from the value of the actual time difference, and thus is “−90, +80, −100” in order.

(flowchart)
15 and 16 are flowcharts showing the operation of the ECU 100B in the third embodiment. However, the operation common to the first embodiment is denoted by the same step number and the description thereof is omitted. The CPU 11 executes a program whose operation is represented by FIGS. 15 and 16 every time an event occurs in the ECU 100B. The execution subject of each step in the flowchart described below is the CPU 11 of the ECU 100B.

  When the event occurs, the CPU 11 executes the event (step S302) and obtains a time stamp from the time register 111. Next, step S308 is executed. If an affirmative determination is made, the process proceeds to step S310. If a negative determination is made, the process proceeds to step S314. If an affirmative determination is made in step S310, the process proceeds to step S400, and if a negative determination is made, the process proceeds to step S314. In step S <b> 400, the CPU 11 determines whether something is recorded in the temporary buffer 121. If the CPU 11 determines that there is something recorded in the temporary buffer 121, the process proceeds to step S402. If the CPU 11 determines that nothing is recorded in the temporary buffer 121, the process proceeds to step S312. In step S402, the estimated time difference corresponding to the latest event recorded in the temporary buffer 121 and the event executed in step S302 is read from the ROM 12 or the nonvolatile memory 14 and written to the time difference buffer 124. For example, when the latest event recorded in the temporary buffer 121 is A and the event executed in step S302 is X, if the pattern information 112A is as shown in FIG. It is written in the time difference buffer 124.

  In the subsequent step S404, the second-order difference is calculated using the estimated time difference read in step S402 and recorded in the time difference buffer 124. That is, first, the difference between the time stamp of the latest event recorded in the temporary buffer 121 and the time stamp acquired in step S304 is calculated and recorded in the real time difference field of the time difference buffer 124. Next, the difference between the actual time difference and the assumed time difference in the time difference buffer 124 is calculated, and this difference is written in the time difference buffer 124 as a second-order difference. Next, the process proceeds to step S312.

  When an affirmative determination is made in step S400, and in step S312 executed after step S404, the latest event ID and time stamp are recorded in the temporary buffer 121. That is, the event ID of the event executed in step S302 and the time stamp acquired in step S304 are recorded in the temporary buffer 121. Then, the execution of the program whose operation is represented by FIGS. 15 and 16 is terminated.

  In step S <b> 314 that is executed when a negative determination is made in step S <b> 308 or step S <b> 310, the CPU 11 determines whether something is recorded in the temporary buffer 121. If the CPU 11 determines that nothing is recorded, the process proceeds to step S316, and if it determines that any information is recorded, the process proceeds to step S330 in FIG. In step S316, the CPU 11 stores the latest event ID and time stamp, that is, the event ID specified in step S306 and the time stamp acquired in step S304 in the log area 122, and the program whose operation is represented by FIG. 15 and FIG. The execution of is terminated.

  In step S330 of FIG. 16, the CPU 11 determines whether or not the ID string in the temporary buffer 121, that is, the event pattern stored in the temporary buffer 121, can be replaced based on the pattern information 112A. If it is determined that the replacement is possible, the process proceeds to step S412. If it is determined that the replacement is not possible, the process proceeds to step S340. In step S412, the CPU 11 stores the pattern identifier obtained by the replacement determined to be replaceable in step S330 and the second-order difference value of the time difference buffer 124 in the log area 122. In the subsequent step S414, the CPU 11 deletes all information stored in the temporary buffer 121 and the time difference buffer 124. The subsequent processing from step S335 is the same as that of the first embodiment. The latest event ID and time stamp are stored in the temporary buffer 121 or the log area 122, and the operation of the program whose operations are represented by FIGS. End execution.

  In step S340, which is executed when a negative determination is made in step S330, the CPU 11 records all the information in the temporary buffer 121 in the log area 122 as it is, and deletes all the information in the temporary buffer 121. In the subsequent step S416, all information in the time difference buffer 124 is deleted, and the process proceeds to step S342. Since the process after step S342 is the same as that of the first embodiment, the description thereof is omitted.

(Operation example)
FIG. 17 is a diagram showing an event occurrence state in this operation example and information recorded in the log area 122 after all events have occurred. When the first four events are observed by the ECU 100B, the time difference buffer 124 stores information shown in FIG. When the last event C occurs, a negative determination is made in steps S308 and S314 in FIG. 15, and an affirmative determination is made in step S330 in FIG. 16, so step S412 is executed. Therefore, the pattern identifier P is recorded as event identification information in the log area 122, and “−90, +80, −100” is recorded as a time stamp. In the first embodiment, the time stamp corresponding to the observation time of the last event among the series of events corresponding to the pattern to be replaced as shown in FIG. 7A is recorded. In the embodiment, the time interval at which each event occurs is recorded.

According to the third embodiment described above, the following operational effects can be obtained.
(7) The pattern information 112A includes an assumed time difference that is a difference between occurrence times assumed in advance for two consecutive events included in the pattern information, that is, an assumed interval. The pattern recognition unit 112 calculates a real time difference, which is a difference between actual times and a time between successive events included in the pattern information, and is a difference between the real time difference and the assumed time difference. Further record the second floor difference in the log area.
Therefore, the ECU 100B can record an interval at which an event has occurred using a small storage area. This effect becomes more remarkable as the resolution of the time counted by the time register 111 is higher and the actual time difference is closer to the assumed time difference.

(Modification 1 of 3rd Embodiment)
The ECU 110B may include the event information 113, and the CPU 11 may specify the event ID. That is, in FIG. 15, step S306 may be executed after step S304, and then the process may proceed to step S308.

(Modification 2 of the third embodiment)
In step S412 of FIG. 16, not only the value of the time difference buffer 124 but also the observation time of any event to be replaced may be recorded in the log area 122 together. According to this modification, the time at which each event has occurred can also be specified using the information recorded in the log area 122.

The present invention also includes the following log storage method.
(1) a log area in which a log of an event that has occurred is recorded, and a storage unit in which pattern information including a normal order that is the occurrence order of the event at a normal time and a character string corresponding to the normal order is stored A log storage method executed in an in-vehicle electronic control device, wherein the in-vehicle electronic control device compresses a log by recording the character string in the log area instead of the normal order.
(2) In the log storage method, when the occurrence order of the events does not match the normal order, the on-vehicle electronic control device records the event in the log area without compressing the log. .
(3) In the log storage method, the vehicle-mounted electronic control device notifies the occurrence of an abnormality when the occurrence order of the events does not completely match the normal order but matches partway.
(4) In the log storage method, each time the event occurs, the in-vehicle electronic control device determines whether the order of the events generated and the normal order match, and determines that the normal order matches halfway, A log storage method for temporarily storing a range in which the in-vehicle electronic control device matches.
(5) In the log storage method, the normal order is stored as a tree-structured database, and the in-vehicle electronic control device sequentially searches the tree-structured database every time the event occurs.
(6) In the log storage method, the pattern information includes an assumed time difference that is a difference between occurrence times assumed in advance for two consecutive events included in the pattern information. A log storage method for calculating a real time difference, which is a difference between actual occurrence times of the two events, and further recording a difference between the real time difference and the assumed time difference in the log area.
(7) In the log storage method, the storage unit further stores abnormal pattern information including an abnormal order that is the order of occurrence of the event at the time of a specific abnormality, and an abnormal character string corresponding to the abnormal order, When the event occurrence order matches the abnormality order, the in-vehicle electronic control device records the abnormality character string in the log area instead of the abnormality order, and notifies the occurrence of abnormality .

The ECU 100 may include an input / output interface (not shown), and a program may be read from another device via an input / output interface and a medium that can be used by the ECU 100 when necessary. Here, the medium refers to, for example, a storage medium that can be attached to and detached from the input / output interface, or a communication medium, that is, a wired, wireless, or optical network, or a carrier wave or digital signal that propagates through the network. Also, part or all of the functions realized by the program may be realized by a hardware circuit or FPGA.
The above-described embodiments and modifications may be combined.
Although various embodiments and modifications have been described above, the present invention is not limited to these contents. Other embodiments conceivable within the scope of the technical idea of the present invention are also included in the scope of the present invention.

14 ... Nonvolatile memory
112 ... Pattern recognition unit 112A, 112B ... Pattern information
113… Event information
121… Temporary buffer
122 ... Log area
123 ... Snapshot area
124 ... Time difference buffer

Claims (14)

A log area where logs of events that occur are recorded,
A storage unit for storing pattern information including a normal order that is the order of occurrence of the events at a normal time and a character string corresponding to the normal order;
An in-vehicle electronic control device comprising: a compression unit that compresses the log by recording the character string in the log area instead of the normal order. The in-vehicle electronic control device according to claim 1,
The compression unit is an in-vehicle electronic control unit that records the event in the log area without compressing the log when the occurrence order of the event does not match the normal order. The in-vehicle electronic control device according to claim 1,
It further includes a notification unit that notifies the occurrence of an abnormality,
The compression unit is an in-vehicle electronic control device that uses the notification unit to notify the occurrence of an abnormality when the occurrence order of the events does not completely match the normal order and does not completely match. The in-vehicle electronic control device according to claim 1,
The compression unit performs a determination of matching between the order of the events that occur each time the event occurs and the normal order, and if it is determined that the order matches the normal order, temporarily stores the matching range. Electronic control device. The in-vehicle electronic control device according to claim 4,
The normal order is stored as a tree-structured database,
The compression unit is an on-vehicle electronic control device that sequentially searches the tree-structured database every time the event occurs. The in-vehicle electronic control device according to claim 1,
The pattern information includes an assumed time difference that is a difference between occurrence times assumed in advance for two consecutive events included in the pattern information,
The in-vehicle electronic control device, wherein the compression unit calculates a real time difference that is a difference between actual occurrence times of the two events, and further records a difference between the real time difference and the assumed time difference in the log area. The in-vehicle electronic control device according to claim 1,
It further includes a notification unit that notifies the occurrence of an abnormality,
The storage unit further stores abnormal pattern information including an abnormal order that is the order of occurrence of the event at the time of a specific abnormality, and an abnormal character string corresponding to the abnormal order,
The compression unit records the abnormal character string in the log area instead of the abnormal order when the occurrence order of the events matches the abnormal order, and notifies the occurrence of abnormality using the notification unit. In-vehicle electronic control device. An in-vehicle electronic device comprising: a log area in which a log of an event that has occurred is recorded; and a storage unit in which pattern information including a normal order that is the order in which the event occurs and a character string that corresponds to the normal order is stored A log storage method executed in a control device,
A log storage method in which the in-vehicle electronic control device compresses a log by recording the character string in the log area instead of the normal order. The log storage method according to claim 8,
A log storage method in which the in-vehicle electronic control device records the event in the log area without compressing the log when the event occurrence order does not match the normal order. The log storage method according to claim 8,
A log storage method in which the in-vehicle electronic control device notifies the occurrence of an abnormality when the occurrence order of the events does not completely match the normal order and does not completely match. The log storage method according to claim 8,
When the event occurs, the in-vehicle electronic control device performs a match determination between the order of the events generated and the normal order. Log storage method to temporarily store. The log storage method according to claim 11,
The normal order is stored as a tree-structured database,
A log storage method in which the in-vehicle electronic control device sequentially searches the tree-structured database every time the event occurs. The log storage method according to claim 8,
The pattern information includes an assumed time difference that is a difference between occurrence times assumed in advance for two consecutive events included in the pattern information,
A log storage method in which the in-vehicle electronic control device calculates a real time difference which is a difference between actual occurrence times of the two events, and further records a difference between the real time difference and the assumed time difference in the log area. . The log storage method according to claim 8,
The storage unit further stores abnormal pattern information including an abnormal order that is the order of occurrence of the event at the time of a specific abnormality, and an abnormal character string corresponding to the abnormal order,
When the event occurrence order matches the abnormality order, the in-vehicle electronic control device records the abnormality character string in the log area instead of the abnormality order, and notifies the occurrence of abnormality .

Images