JP2016097917A - Vehicle ecu - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce time which a vehicle ECU such as an air bag ECU requires for self diagnosis.SOLUTION: A vehicle ECU modifies a program which a CPU is made execute, uses vacant time during diagnosis of an ASIC, and diagnoses sensors (third and fifth sensors) which are not coupled to the ASIC. Thereby, a time for diagnosis of both sensors can be reduced. Therefore, even when a hardware is a conventional one, by only modifying a diagnosis program which the CPU is made execute, time which the vehicle ECU requires for self diagnosis can be reduced.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a vehicle ECU having a plurality of sensors and a processor and having a self-diagnosis function. The vehicle ECU of the present invention can also be applied to, for example, an automobile airbag ECU, an ECU that performs a determination of deploying pedestrian protection means such as a pop-up hood in an automobile having a hood.

  In Patent Document 1, a plurality of satellite sensor units connected to an air bag ECU as an ECU for a vehicle by two bus lines separated on the left and right sides of the vehicle, respectively, can be checked in a shorter time whether there is any abnormality. An invention to diagnose is disclosed. However, this document does not seem to mention the diagnosis of the sensor equipped in the airbag ECU.

  Therefore, a procedure conventionally performed in diagnosing various sensors (hereinafter simply referred to as “sensors”) incorporated in an airbag ECU as a vehicle ECU will be described. For example, let us say that there are five sensors, and each of them is referred to as a first sensor to a fifth sensor, regardless of whether or not they are connected to an ASIC as a first processor. Then, in the conventional airbag ECU, as shown in FIG. 8, when the ignition switch is turned on, the ASIC initial diagnosis is performed prior to the sensor check, and all sensors are diagnosed in order after the completion of the ASIC diagnosis. To go.

Japanese Patent Laying-Open No. 2005-212617

  However, in the above-described background art, since the procedure for sequentially diagnosing all the sensors is performed after the completion of the ASIC diagnosis, it takes some time to complete the inspection of the ASIC and all the sensors.

  The time required for the diagnosis of the vehicle ECU including the airbag ECU increases the number of sensors connected to the ECU, such as sensors and satellite sensors built in the ECU, for safety reasons. It tends to take longer.

  On the other hand, the diagnosis of the airbag ECU is performed immediately after the ignition switch is turned on in an automobile, but the time required for the self-diagnosis of the airbag ECU is required to be further shortened by the automobile manufacturer. ing. This request can be said to be a matter of course, considering the user psychology that it is desired to bring the car into a state where it can run in as short a time as possible. In addition, vehicles such as vehicles other than automobiles, aircraft, ships, and the like will have similar requirements if they are equipped with airbags or other personnel protection means.

  By the way, it is normal that there is a waiting time for various diagnosis commands from the CPU to be returned from the ASIC when the ASIC is diagnosed. Therefore, it is considered that the waiting time is considered as the CPU idle time, and the sensor diagnosis is performed in parallel with the TSS method during the idle time, but in practice, it often does not go well. That is, if a sensor connected to the ASIC is diagnosed during the ASIC diagnosis, a diagnosis impossible state or a misdiagnosis may frequently occur.

  Alternatively, even if the CPU is configured to allow SMT (simultaneous multi-threading) that can be processed in parallel or has a multi-core configuration, it is the same as diagnosing a sensor connected to the ASIC during ASIC diagnosis. Inconvenience occurs. In addition, an SMT-capable CPU or a multi-core CPU is more expensive than a single-task CPU, and thus it is inevitable that the airbag ECU becomes expensive.

  Therefore, the present invention provides a vehicular ECU that can shorten the time required for the self-diagnosis of the ECU including the diagnosis of a processor such as an ASIC and the diagnosis of a sensor without basically changing hardware. This is a problem to be solved.

  In this section, the main configuration of the present invention for solving the above-described problems and the effects brought about by the configuration will be briefly described.

(Main configuration of the present invention)
The vehicle ECU (100) of the present invention includes a plurality of sensors (1: 11-15), a second processor (2) connected to each of the plurality of sensors, and a part of these sensors ( 11, 12, 14) and a first processor (3) connected to the second processor. The second processor has a sensor diagnostic function (SC) for diagnosing each of these sensors and a processor diagnostic function (AC) for diagnosing the first processor.

  Here, the plurality of sensors are a first sensor group (G1) composed of sensors (11, 12, 14) connected to the first processor and a second sensor composed of the remaining sensors (13, 15). It consists of a group (G2). The second processor has a parallel processing function of processing at least a part of sensor diagnosis functions among the sensors belonging to the second sensor group in parallel with the processor diagnosis function.

(The effect)
Then, in the vehicle ECU of the present invention, the diagnosis is performed in parallel with the diagnosis of the first processor among all the sensors, even though it is connected to the second processor. At least some of the sensors in the second sensor group that are not. Therefore, even if the diagnosis of the first processor and the diagnosis of these sensors (not connected to the first processor) are performed at the same time, there is no functional interference between the two, and the diagnosis of both of them is surely performed. Can be processed in parallel.

  During diagnosis of the first processor, these sensors are diagnosed using the idle time of the second processor until a response signal is returned from the first processor. The time required for is almost unchanged. On the other hand, since the sensor diagnosis is completed during the diagnosis of the first processor, the number of sensors to be diagnosed after the diagnosis of the first processor is reduced accordingly. As a result, the time required for the diagnosis of the first processor and the diagnosis of the sensor is shortened, and the time from when the ignition switch is turned on until the initial diagnosis of the vehicle ECU is completed is also shortened. Similarly, even if the initial diagnosis is not performed, the time required for the self-diagnosis of the vehicle ECU is shortened.

  Moreover, the present invention can be implemented only by changing part of the self-diagnosis software processed by the second processor, and it is not necessary to change or add to the hardware built in the ECU. Basically, there will be no design change. Therefore, it is not necessary to increase the manufacturing cost. However, if the capacity of the ROM or the like incorporating the software of the second processor is not enough to store software that increases somewhat with the implementation of the present invention, the compatible product having the necessary storage capacity for the ROM. There may be minor hardware modifications that can be replaced.

  Therefore, according to the vehicle ECU of the present invention, the time required for the diagnosis of the built-in first processor and the diagnosis of the sensor can be shortened with basically no hardware change and almost no cost increase. . As a result, there is an effect that the time required for self-diagnosis such as initial diagnosis of the ECU is shortened accordingly.

(Remarks)
Here, the scope of the parallel processing function varies depending on the configuration of the second processor. However, even if the second processor is, for example, an inexpensive single-task CPU as in the past, it can be implemented by the TSS (Time Shelling System) method. Therefore, since it is basically not necessary to change the vehicle ECU hardware configuration to implement the present invention, the vehicle ECU of the present invention can be introduced with almost no cost increase. If a CPU such as an SMT (simultaneous multi-threading) type or an MC (multi-core) type is standardly adopted as the second processor used in the vehicle ECU, such a CPU as the second processor may be used. The present invention can be implemented.

  In the following drawings, an embodiment of a vehicle ECU according to the present invention and an airbag ECU as a modified embodiment and an airbag ECU (FIG. 8) as a conventional technique will be described.

1 is a block diagram showing a configuration of an airbag ECU as Embodiment 1 of the present invention. Timing chart showing diagnosis procedure of ASIC and each sensor in embodiment 1 FIG. 5 is a timing chart showing the same diagnostic procedure in Modification 1 of Example 1. Timing chart showing diagnosis procedure of ASIC and each sensor in embodiment 2 The timing diagram which shows the same diagnostic procedure in the deformation | transformation aspect 1 of Example 2. FIG. Timing chart showing diagnosis procedure of ASIC and each sensor in embodiment 3 Timing chart showing the same diagnostic procedure in variant 1 of Example 3 Timing chart showing diagnostic procedure of ASIC and each sensor in prior art

  Taking an airbag ECU as a specific example of the vehicle ECU according to the present invention, configurations and operational effects of a plurality of embodiments and modifications will be described below. In these embodiments and modifications thereof, an airbag ECU is illustrated, but the present invention can also be applied to other vehicle ECUs such as an ECU that controls a pop-up hood for protecting pedestrians.

(Constitution)
The vehicular ECU as the first embodiment of the present invention is an air bag ECU 100 whose main configuration is shown in FIG. The airbag ECU 100 is an ECU that controls an occupant protection airbag. A communication bus 4 to be connected is incorporated.

  The CPU 2 is connected to all the sensors 11-15 and the ASIC 3, and includes an ASIC diagnostic function AC as a processor diagnostic function for diagnosing the ASIC 3, a sensor diagnostic function SC for each sensor 11-15, and a collision determination function (not shown). Etc. On the other hand, the ASIC 3 is connected to some of the sensors 11, 12, and 14, and has a control function (not shown) of the sensors 11, 12, and 14, a collision determination function (not shown), and the like.

  All of the sensors 11-15 are connected to the CPU 2, but some of them are connected to the ASIC 3, and some are not connected to the ASIC 3. Therefore, the sensor 11-15 is connected to the two sensors depending on whether it is connected to the ASIC 3. Divide into groups G1 and G2.

  That is, a sensor group including the sensors 11, 12, and 14 connected to both the CPU 2 and the ASIC 3 is referred to as a first sensor group G1. On the other hand, a sensor group consisting of the remaining sensors 13 and 15 connected to the CPU 2 but not connected to the ASIC 3 will be referred to as a second sensor group G2.

  The CPU 2 has a sensor diagnostic function SC for diagnosing each sensor 11-15 and an ASIC diagnostic function AC for diagnosing the ASIC 3. The sensor diagnostic function SC is a first group diagnostic function SC1 that diagnoses each of the sensors 11, 12, and 14 that belong to the first sensor group G1, and a second that diagnoses each of the sensors 13 and 15 that belong to the second sensor group G2. It has a group diagnosis function SC2. That is, each sensor diagnostic function for each sensor 11, 12, 13, 14, 15 is roughly divided into a first group diagnostic function SC1 and a second group diagnostic function SC2 depending on the group to which the sensor to be diagnosed belongs. .

  The CPU 2 has a parallel processing function for processing the second group diagnostic function SC2 for diagnosing the sensors 13 and 15 belonging to the second sensor group G2 in parallel with the ASIC diagnostic function AC. At this time, the first group diagnosis function SC1 and the second group diagnosis function SC2 are processed with a time lag. That is, the first group diagnosis function SC1 and the second group diagnosis function SC2 are not processed in parallel. More specifically, the process of the sensor diagnosis function for each of the sensors 11, 12, 14 belonging to the first sensor group G1 and the process of the sensor diagnosis function for each of the sensors 13, 15 belonging to the second sensor group G2 are as follows: Processed at different times. That is, any one of the sensors 11, 12, and 14 belonging to the first sensor group G1 and any one of the sensors 13 and 15 belonging to the second sensor group G2 are not diagnosed in parallel. Yes.

Here, the types of the sensors 11-15 (first sensor to fifth sensor) are as follows, and the sensor 1 and the sensor 2 are one-chip sensors 10 formed in the same chip.
First sensor 11: High range G sensor in X axis direction Second sensor 12: High range G sensor in Y axis direction Third sensor 13: Low range G sensor in X axis direction Fourth sensor 14: Roll rate (X axis Angular velocity around sensor / Fifth sensor 15: Low range G sensor in the Z-axis direction

(Function and effect)
Since the airbag ECU 100 of the present embodiment is configured as described above, it operates as follows.

  That is, as shown in FIG. 2, the diagnosis is performed in parallel with the diagnosis of the ASIC 3 among all the sensors 1 (11-15) by the airbag ECU 100, even though it is connected to the CPU 2. Only the third and fifth sensors that have not been performed. The third sensor and the fifth sensor are the sensors 13 and 15 (see FIG. 1) of the second sensor group G2. Therefore, the diagnostic function AC of the ASIC 3 and the diagnostic function SC2 of the sensors 13 and 15 (second sensor group G2) not connected to the ASIC 3 can be processed in parallel by the CPU 2.

  This is because there is no functional interference between the second sensor group G2 and the ASIC 3 between the two diagnostic functions AC and SC2, and the diagnosis of both the three and G2 can be reliably processed in parallel.

  Since these sensors are diagnosed using the CPU idle time until the response signal is returned from the ASIC in the ASIC diagnosis, the time required for the ASIC diagnosis hardly changes. On the other hand, since the diagnosis of these sensors is completed during the ASIC diagnosis, the number of sensors to be diagnosed after the ASIC diagnosis is reduced accordingly. As a result, the time required for diagnosis of the built-in ASIC and sensor is shortened.

  Thereafter, the sensors 11, 12, and 14 (first sensor, second sensor, and fourth sensor) of the first sensor group G1 are sequentially diagnosed by the CPU 2 using the ASIC 3 as well.

  As a result, compared with the check time in FIG. 8 showing the prior art, the time required for the check is shortened by the time corresponding to two frames in the left-right direction corresponding to the time axis in FIG. .

  The check time of the third sensor 13 and the fifth sensor 15 corresponding to the shortening time is about 20 to 30% compared to the overall check time according to the conventional technique (see FIG. 8). In other words, in the diagnosis of all the sensors 1 (11-15) and the ASIC 3, it is possible to expect this time reduction.

  In addition, the present embodiment can be realized only by modifying a part of the software of the CPU 2, and it is not necessary to change or add hardware, so that a design change is unnecessary. That is, since it can be implemented with the hardware of the conventional product (even if it is rare if the capacity of the ROM (not shown) is increased), there is almost no increase in cost in the mass-produced product.

  Therefore, according to the airbag ECU 100 of the present embodiment, the time required for the diagnosis of the built-in ASIC 3 and the diagnosis of the sensor 1 (11-15) (check time in FIG. 2) is reduced with almost no cost increase. There is an effect that can be.

(Modification 1)
As a modification 1 of the present embodiment, as shown in FIG. 3, an airbag ECU 100 (see FIG. 1) can be implemented in which the check procedure is replaced before and after the first embodiment (see FIG. 2). In this modification, the CPU 2 performs the processing of the diagnostic function SC1 (see FIG. 1) of the sensors 11, 12, and 14 (first sensor, second sensor, and fourth sensor) belonging to the first sensor group G1 as the ASIC diagnostic function. Complete before starting the AC process.

  That is, as apparent from comparing FIG. 3 with FIG. 2 described above, the programs for executing the internal diagnostic function of the airbag ECU 100 by the CPU 2 are the parallel diagnostic programs of the ASICs 3 and G2, and the diagnostic program of the first sensor group G1. This order is reversed from that of the first embodiment. In this case, the diagnostic results of the sensors 11, 12, and 14 (first, second, and fourth sensors) belonging to the first sensor group G1 may be affected by the malfunction of the ASIC 3, but in any case, the first A malfunction is detected in the sensor group G1 or the ASIC 3, and a malfunction of the airbag ECU 100 is warned.

  Therefore, the same effect as that of the first embodiment can be obtained also by this modification.

(Constitution)
As shown in FIG. 1 again, the airbag ECU 100 according to the second embodiment of the present invention has the same hardware configuration as that of the first embodiment.

  The difference between the present embodiment and the first embodiment is that the first group diagnosis function SC1 is a combination of sensors 11, 12, and 14 belonging to the first sensor group G1 that do not interfere with each other, as shown in FIG. It has a function to process some of them in parallel. In the present embodiment, in addition to the parallel diagnosis function in the first embodiment, the diagnosis of the sensors 11 and 12 (first sensor and second sensor) is performed in parallel after the diagnosis of the ASIC 3, and then the sensor 14 (fourth sensor). ) Is diagnosed.

  That is, although it is housed in the one-chip 10, the acceleration in the X-axis direction and the acceleration in the Y-axis direction are independent from each other, so that there is no mutual interference during sensor diagnosis. The CPU 2 diagnoses both the sensors 11 and 12 in parallel. That is, the two sensors 11 and 12 in the one-chip sensor 10 are a combination in which the diagnostic functions for the sensors 11 and 12 do not interfere with each other, and thus the diagnostic functions for the sensors 11 and 12 are processed in parallel with each other.

(Function and effect)
Since the airbag ECU 100 of the present embodiment is configured as described above, the following operational effects are exhibited.

  That is, in this modification (see FIG. 4), the check time is further shortened by one frame as compared with the first embodiment (see FIG. 2). As a result, compared with the prior art (see FIG. 8), the time required for diagnosis of the ASIC 3 and the sensors 1 can be expected to be approximately 30 to 40%.

(Modification 1)
As a modification of the present embodiment, as shown in FIG. 5, the CPU 2 processes the diagnosis procedure in the procedure of diagnosing the first sensor group G1 prior to the parallel diagnosis of the second sensor group G2 and the ASIC 3. Implementation of the airbag ECU 100 is possible. This modified embodiment with respect to the above-described second embodiment corresponds to the modified embodiment 1 with respect to the first embodiment.

  Also according to this modification, the same effect as that of the second embodiment can be obtained.

(Constitution)
As shown in FIG. 1, the airbag ECU 100 according to the third embodiment of the present invention has the same hardware configuration as that of the first embodiment.

  As shown in FIG. 6, this embodiment differs from the first and second embodiments described above in that the first group diagnosis function SC1 interferes with each other among the sensors 11, 12, and 14 belonging to the first sensor group G1. It is to have a function to process all of the combinations that are not in parallel. That is, in addition to the parallel diagnosis function in the first embodiment, the diagnosis of the three sensors 11, 12, 14 (first, second, and fourth sensors) is performed in parallel after the diagnosis of the ASIC 3.

  That is, in addition to the sensors 11 and 12 (X-axis and Y-axis acceleration sensors) in the one-chip sensor 10, the sensor 14 (roll rate sensor) is also diagnosed in parallel. If it writes along with FIG. 6, a 1st sensor, a 2nd sensor, and a 4th sensor will be diagnosed in parallel.

(Function and effect)
Since the airbag ECU 100 of the present embodiment is configured as described above, the following operational effects are exhibited.

  That is, in this modification (see FIG. 6), the check time is further shortened by one frame as compared with the second embodiment (see FIG. 4). As a result, compared with the prior art (see FIG. 8), the time required for diagnosis of the ASIC 3 and the sensors 1 can be expected to be approximately 30% to 40%.

(Modification 1)
As a modification of the present embodiment, as shown in FIG. 7, the CPU 2 processes the diagnosis procedure in the procedure of diagnosing the first sensor group G1 prior to the parallel diagnosis with the second sensor group G2 and the ASIC 3. Implementation of the airbag ECU 100 is possible. This modified embodiment with respect to the third embodiment corresponds to the modified embodiment 1 with respect to the first embodiment and the modified embodiment 1 with respect to the second embodiment.

  Also by this modification, the same effect as Example 3 can be obtained.

100: Airbag ECU (as the vehicle ECU of the present invention)
1: Multiple sensors 10: One-chip sensor (11, 12)
11, 12, 14: Sensors connected to ASIC 13, 15: Sensors not connected to ASIC G1: First sensor group (Sensors 11, 12, 14)
G2: Second sensor group (sensors 13 and 15)
2: CPU (as second processor)
3: ASIC (custom IC: as first processor) 4: Communication bus AC: ASIC diagnostic function (as processor diagnostic function) SC: sensor diagnostic function SC1: first group diagnostic function SC2: second group diagnostic function

Claims (7)

A plurality of sensors (11-15);
A first processor (3) connected to some (11, 12, 14) of the plurality of sensors;
A second processor (2) connected to each sensor and the first processor and having a sensor diagnostic function (SC) for diagnosing each sensor and a processor diagnostic function (AC) for diagnosing the first processor;
In the vehicle ECU (100) having
The plurality of sensors include a first sensor group (G1) composed of sensors (11, 12, 14) connected to the first processor and a second sensor group (G2) composed of the remaining sensors (13, 15). )
The second processor processes at least a part of sensor diagnostic functions among the sensors belonging to the second sensor group in parallel with the processor diagnostic function.
Vehicle ECU (100). The second processor performs parallel processing of at least a part of a sensor diagnosis function with a combination of the sensors belonging to the same group of the first sensor group and the second sensor group without interference with each other. To
The vehicle ECU according to claim 1. The second processor is
A first group diagnostic function (SC1) for diagnosing each sensor belonging to the first sensor group; and a second group diagnostic function (SC2) for diagnosing each sensor belonging to the second sensor group;
The first group diagnosis function and the second group diagnosis function are processed with a time shift from each other,
The vehicle ECU according to any one of claims 1 and 2. The second processor performs processing of a sensor diagnostic function for each of the sensors belonging to the first sensor group and processing of a sensor diagnostic function for each of the sensors belonging to the second sensor group at different times. Characterized by the
The vehicle ECU according to any one of claims 1 to 3. The second processor completes processing of at least a part of the sensor diagnosis function among the sensors belonging to the first sensor group before starting the processing of the processor diagnosis function.
The vehicle ECU according to any one of claims 1 to 4. Each of the first processor and the second processor has a collision determination function,
The vehicle ECU according to any one of claims 1 to 5. A plurality of sensors belonging to any one of the first sensor group and the second sensor group G2 among the plurality of sensors are formed as one-chip sensors,
At least one of the combinations in which the diagnostic functions for each of the plurality of sensors do not interfere with each other is processed in parallel,
The vehicle ECU according to any one of claims 1 to 6.

Images