Vehicle-mounted control system

Abstract

An in-vehicle control system according to one embodiment of the present invention is provided with: a storage device for use in vehicle control; a fault omen diagnosis device which diagnoses the fault omen of the storage device; and a frequency changing circuit that changes the operating frequency of the storage device to a second frequency that is higher than a first frequency that is a frequency used during normal control of the vehicle when the fault omen diagnostic device diagnoses the storage device.

Description

Technical Field

[0001] This invention relates to an in-vehicle control system for predicting and diagnosing faults in in-vehicle semiconductor storage devices that cause delays in operating speed due to aging over the years. Background Technology

[0002] Due to the advancements and expansion of autonomous driving, advanced driver assistance technologies, and the widespread adoption of car sharing and automated logistics, vehicle uptime is expected to increase several times to tens of times in the near future. This significant increase in uptime is anticipated to shorten the time to failure for semiconductor devices used in automobiles. However, ensuring long-term reliability several times to tens of times higher than before in semiconductor devices remains challenging in terms of both cost and technology. Therefore, the method of maintaining vehicle functionality and performance by replacing components or units is expected to become the mainstream approach in the future.

[0003] Against this backdrop, there is a need for highly accurate detection of early warning signs of semiconductor device failures and for notifying the driver. Specifically, by monitoring the status of each semiconductor device installed in the vehicle, calculating the time it takes for each to reach a failure, identifying the components that need replacement, and replacing only the affected components, replacement costs can be reduced compared to replacing them at fixed intervals.

[0004] Semiconductor memory devices are widely used in advanced driving systems that perform high-level control and are required to have high reliability. Aging phenomena in semiconductor memory devices include increased retention charge leakage due to aging of the insulating film of the memory cells, changes in threshold voltage, and read / write time delays due to increased wiring resistance, which can ultimately lead to insulating film damage or malfunction. By monitoring these aging phenomena of semiconductor memory devices, failures can be predicted. For example, Patent Document 1 provides a technique for predicting such semiconductor memory device failures. According to Patent Document 1, the threshold voltage of the non-volatile memory is gradually changed while monitoring the threshold distribution of the memory, and the degree of aging is determined based on changes in the threshold distribution. Existing technical documents Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2013-122793 Summary of the Invention The problem the invention aims to solve

[0006] As mentioned earlier, in addition to changes in threshold voltage, aging modes of semiconductor memory devices also include operating time delays caused by factors such as increased wiring resistance. In particular, memory devices such as DRAM and SRAM operate at high frequencies and are prone to malfunctions due to operating time delays caused by aging over time. This invention addresses these issues and aims to provide a technique for predicting failures caused by operating time delays resulting from aging over time. Technical means to solve the problem

[0007] To address the aforementioned issues, an in-vehicle control system according to an embodiment of the present invention includes: a storage device used in vehicle control; a fault prediction diagnostic device for diagnosing fault predictions of the storage device; and a frequency changing circuit that, when the fault prediction diagnostic device diagnoses the storage device, changes the operating frequency of the storage device to a second frequency that is higher than a first frequency used during normal vehicle control. The effects of the invention

[0008] According to the present invention, by performing diagnostics at a higher frequency than when the vehicle control system is actually operating, it is possible to detect storage areas that are slowing down due to aging, even during normal operation of the vehicle control system, thereby enabling the prediction of malfunctions. Attached Figure Description

[0009] Figure 1 This is a block diagram illustrating the general outline of the vehicle control system involved in Embodiment 1. Figure 2 This is a graph showing the relationship between the frequency used for diagnosis and the bad bit rate (number) of the aging memory device in Example 1. Figure 3 This is a block diagram illustrating the general outline of the vehicle control system involved in Embodiment 2. Figure 4 This is a graph showing the increase in bit defects in the storage device over time in Example 2. Figure 5 This is a flowchart illustrating the process of predicting storage device failures in Example 2. Figure 6 This is a block diagram illustrating the general outline of the vehicle control system involved in Embodiment 3. Figure 7 This is a graph showing the relationship between the frequency used for diagnosis and the bad bit rate (number) of the aging memory device in Example 3. Figure 8 This is a graph showing the relationship between the voltage used for diagnosis and the bad bit rate (number) of the aged memory device in Example 3. Figure 9 This is a block diagram illustrating the general outline of the vehicle control system involved in Embodiment 4. Figure 10 This is a graph showing the increase in bit defects in the storage device over time in Example 4. Figure 11 This is a flowchart illustrating the process of predicting storage device failures in Example 4. Figure 12 This is a block diagram illustrating the general outline of the vehicle control system involved in Embodiment 5. Figure 13 This is a graph showing the increase in bit defects in the storage device over time in Example 5. Figure 14 This is a flowchart illustrating the process of predicting storage device failures in Example 5. Figure 15 This is a block diagram illustrating the general outline of the vehicle control system involved in Embodiment 6. Figure 16 This is a graph showing the increase in bit defects in the storage device over time in Example 6. Figure 17 This is a flowchart illustrating the process of performing storage device failure prediction in Example 6. Detailed Implementation

[0010] The embodiments are described below using the accompanying drawings.

[0011] [Example 1] Figure 1 This is a block diagram illustrating the configuration of an on-board control system (hereinafter referred to as the "System") according to Embodiment 1 of the present invention. In this embodiment, the System, for a semiconductor memory device 1, comprises a fault prediction and diagnostic device 2 for the semiconductor memory device 1 and a frequency conversion circuit 3 for changing the operating frequency of the semiconductor memory device. The semiconductor memory device 1 is, for example, composed of SRAM or DRAM. The fault prediction and diagnostic device 2 is composed of an arithmetic element having a CPU (Central Processing Unit) and ROM (Read Only Memory). The frequency conversion circuit 3 will be described later.

[0012] In this embodiment, the fault prediction diagnostic device 2 performs tests such as ALL=1 / 0 and checkerboard tests on each memory cell of the semiconductor memory device 1, and reads the values ​​written to the memory cells as READ results through the characteristic monitoring unit 2-1. The characteristic diagnostic unit 2-2 confirms whether the read results of the characteristic monitoring unit 2-1 are consistent with the values ​​written to each memory cell, and determines the PASS / FAIL of each memory cell. The determination results are statistically analyzed in the fault information statistics unit 2-3 for the number of faulty bits or the faulty bit rate. The fault prediction determination unit 2-4 determines the aging state of the semiconductor memory device 1 to be diagnosed based on the value of the faulty bit rate (number). Then, it calculates the time until the vehicle control system malfunctions due to aging, and if the predicted time for the semiconductor memory device 1 to operate normally is shorter than a preset time, it issues an alarm to the driver as a notification.

[0013] The feature monitoring unit 2-1 may also monitor the ECC (Error Collection Code) frequency or ECC error rate instead of the READ results of the above tests. Furthermore, this fault prediction diagnosis is preferably performed before the vehicle is driven.

[0014] The frequency changing circuit 3 has the function of switching the operating frequency when the car is running and when it is being diagnosed. When diagnosing, the memory is diagnosed at a higher frequency than when the car is running. Figure 2 This indicates the frequency dependence of the defective bit rate (number) of a semiconductor memory device. Solid lines represent unaged semiconductor memory devices, dashed lines represent slightly aged semiconductor memory devices, and dotted lines represent semiconductor memory devices undergoing further aging. For example... Figure 2 As shown, any device does not malfunction within the frequency range used during vehicle operation (normal control). However, due to factors such as increased wiring resistance caused by aging, the read time from the memory cell becomes longer. By performing diagnostics at a higher frequency (b) than the frequency range during vehicle operation (a), defective bits in the aging device can be detected. That is, even when no abnormalities occur during vehicle operation (normal control), a slower operating speed can be detected. Furthermore, when performing diagnostics at a higher frequency, the rate of detected defective bits is larger, thus allowing for significant detection of changes and identifying potential fault signs.

[0015] [Example 2] Next, the vehicle control system according to Embodiment 2 of the present invention will be described. Furthermore, descriptions of previously explained functions, etc., will be omitted below. Figure 3 This is a block diagram illustrating the system configuration involved in Embodiment 2. For example... Figure 3As shown, the system according to Embodiment 2 further includes a fault information holding unit 2-5 in the fault prediction diagnostic device 2 of Embodiment 1. The fault information holding unit 2-5 stores values ​​statistically analyzed by the fault information statistics unit 2-3 and the elapsed time since the semiconductor memory device was first used in its internal non-volatile memory. The data stored by the fault information holding unit 2-5 is... Figure 4 The bit defect information shown is the time-series data of the defective bit rate (number) of the storage device. Figure 4 The fault precursor value shown is the rate of increase in the bad bit rate calculated by the lifetime prediction formula described later. The precursor fault time refers to the predicted time when a fault will occur if the storage device continues to be used from the time of final diagnosis.

[0016] In this embodiment, the fault prediction determination unit 2-4 applies a predetermined calculation formula to the most recently obtained value in the data stored by the fault information storage unit 2-5 to interpolate a fault prediction value. Then, if the interpolated value exceeds a predetermined value, the determination device will malfunction and issue an alarm to the driver.

[0017] Figure 5 This describes the process of using time-series data on the defective bit rate to determine fault precursors. The characteristic monitoring unit 2-1 sets the voltage and frequency of the storage device and performs tests, reading the values ​​written to the storage cells (steps 1 and 2). Next, the characteristic diagnostic unit 2-2 performs a memory test by determining whether the reading results from the characteristic monitoring unit 2-1 match the actual values ​​written to the storage cells (step 3). The defect information statistics unit 2-3 compiles statistics on the measured defect information (defective bit rate in this embodiment) (step 4).

[0018] The defective information retention unit 2-5 stores the defective bit rate and elapsed time measured in the memory test results in the memory (step 5). Then, the fault prediction unit 2-4 generates a predicted lifetime formula based on the time series shift of the defective bit rate, calculates the time before failure, and counts the elapsed time to date (steps 6-8). Here, the lifetime prediction formula is a calculation formula that predicts how long the memory device will remain in use before failure based on the measured defective bit rate and usage time of the memory device.

[0019] If the time before the fault arrives is less than the standard value ("Yes" in step 9), an alarm is issued and the driver is notified of the time before the fault arrives (step 10). If the time before the fault arrives is greater than the standard value ("No" in step 9), normal operation continues.

[0020] According to this embodiment, it is possible to determine the aging effects of stress differences such as usage frequency and temperature of semiconductor devices in automobiles. Furthermore, the defect information storage unit 2-5, which holds the time-series data, does not necessarily have to be an on-board non-volatile memory; it can also save the data by sending the data to an external computer server via OTA (Over-The-Air) technology. In addition, fault prediction can be processed in an external computer, and the results can be notified to the vehicle via OTA.

[0021] [Example 3] Next, the vehicle control system relating to Embodiment 3 of the present invention will be described. Figure 6 This is a block diagram illustrating the configuration of the vehicle control system involved in Embodiment 3. For example... Figure 6 As shown, the system in this embodiment, in Figure 3 The system shown further includes a voltage changing circuit for changing the voltage that enables the semiconductor memory device to operate. Figure 7 Is with Figure 2 Similarly, a graph showing the frequency dependence of the rate (number) of bad bits in aging memory devices, and such as Figure 8 As shown, the same principle applies to voltage. That is, the lower the operating voltage of a semiconductor memory device is set during diagnostics compared to the voltage used when a car is running, the more likely it is to detect faulty bits in aging memory devices. In this way, by diagnosing semiconductor memory devices under low voltage conditions in addition to high frequencies, it is possible to detect not only faults caused by aging operating speeds but also faults caused by reduced charge in memory cells and changes in threshold voltage.

[0022] [Example 4] Next, the vehicle control system involved in Embodiment 4 will be described. The system of Embodiment 4 is as follows: Figure 9 As shown, it is relative to Figure 3 The configuration of Embodiment 2 shown is modified so that the bad information retention unit is changed to a configuration that only retains the bad information rate at the time of the previous diagnosis, i.e., the most recent bad information rate, in the retention unit 2-6.

[0023] Figure 10 This is a diagram used to illustrate the method for predicting fault precursors in this embodiment. For example... Figure 10 As shown, the rate of change of aging is calculated and the signs of failure are predicted based on the results of the last two diagnoses, namely the latest diagnoses and the adverse information in the previous diagnoses, as well as the time between them.

[0024] Figure 11 This is the fault prediction process in this configuration. Figure 11Steps 1 to 3 are the same as in the above embodiment. In step 4, the previous fault information retention unit 2-6 only retains the fault bit rate. This is because the elapsed time is only used when accumulating and retaining fault information. Then, the fault prediction determination unit 2-4 calculates the increment of the fault bit rate in the last diagnosis and the previous diagnosis in step 5, counts the time during this period in step 6, and calculates the time before the fault bit rate exceeds the fault determination benchmark for determining that the system is malfunctioning based on these values. If the time before the fault occurs is below the standard value ("Yes" in step 8), an alarm is issued and the driver is notified of the time before the fault occurs (step 9). If the time before the fault occurs is greater than the standard value, it is used normally ("No" in step 8).

[0025] According to this embodiment, the capacity of the non-volatile memory used to maintain the defect rate of the storage cell can be reduced, and the circuit of the fault prediction unit can also be reduced.

[0026] [Example 5] Next, the vehicle control system according to Embodiment 5 will be described. The system configuration of this embodiment is as follows: Figure 12 As shown, it is the same as in Embodiment 1. In this embodiment, however, it does not have the defective information retention section found in Embodiments 2-4, but rather... Figure 13 As shown, if the final diagnostic result exceeds a predetermined defect rate threshold (defective bit determination benchmark value), a fault precursor is notified. That is, a predetermined lifetime prediction formula is applied to the final diagnostic result, and if the calculation result exceeds the predetermined threshold, a fault precursor is diagnosed. However, in this embodiment, since time-series data of bit defect information as in the above embodiments cannot be used, the lifetime prediction formula uses a formula pre-set for each memory device.

[0027] Figure 14 The above-described determination process is illustrated. Steps 1 to 3 are the same as in the above embodiment. In step 4, the fault prediction determination unit 2-4 measures the number of defective bits. If this number becomes higher than the standard value in terms of bit rate, in step 5, the difference from the standard value is calculated. From this point in time, the increase in defect rate relative to time is considered a constant to calculate the time before the fault, a fault prediction is performed, and the calculated predicted time to the fault is communicated (step 6).

[0028] According to this embodiment, since the system does not have a memory to store time-series data of bad information, the fault prediction diagnostic unit can be made smaller.

[0029] [Example 6] Next, the vehicle control system involved in Example 6 will be described. Figure 15 This is a block diagram illustrating the system overview of this embodiment. For example... Figure 15 As shown, the system in this embodiment is relative to... Figure 3 The configuration of Embodiment 2 shown further includes an environmental monitoring device 5 and a correction unit 2-9 that corrects the diagnostic results of the semiconductor memory device based on the environmental monitoring values. Furthermore, a statistics unit 2-10 and a storage unit 2-11 correspond to the corrected information, respectively.

[0030] The aging process of semiconductor devices is affected by factors such as temperature, current, and voltage during use. For example... Figure 16 As shown by the black dot (e), the rate of change of the original defect rate over time varies due to the influence of the usage environment. Therefore, using this data for fault prediction will result in greater errors. By using environmental monitoring values ​​such as temperature, current, and voltage during use to correct for the aging of semiconductor memory devices over time, it is possible to... Figure 16 Like the white dot (f), it improves the accuracy of predicting changes. Specifically, for example, when the temperature during diagnosis is lower than the average temperature, it can be assumed that more adverse effects will occur in the higher temperatures typically used, thus... Figure 16 As shown, a correction is performed that further increases the bit defect rate (number).

[0031] The diagnostic process for predicting temperature based on environmental monitoring values ​​is shown below. Figure 17 Steps 1-4 are the same as in the above embodiment. In step 5, the characteristic diagnostic unit 2-2 measures the temperature of the memory. In step 6, the correction unit 2-9 performs correction of the defective bit rate based on the temperature measurement result. In the next step 7, the corrected defective information statistics unit 2-10 counts the operating time, and in step 8, the corrected defective information retention unit stores this information in the memory. In step 9, the fault prediction unit 2-12 predicts the fault time based on the corrected defective bit rate and the operating time. If the time before the fault occurs is below a predetermined value ("Yes" in step 10), an alarm is issued and the driver is notified (step 11). If the time before the fault occurs is greater than the predetermined value, the memory is used normally ("No" in step 10).

[0032] According to this embodiment, more accurate diagnosis can be achieved by calibrating the measurement conditions.

[0033] In addition, in this embodiment, the environmental monitoring device 5 is configured to monitor the temperature during use, but it is not limited to this. The environmental monitoring device 5 may also monitor the ambient temperature of the semiconductor memory device or the voltage that enables the semiconductor memory device to operate.

[0034] Furthermore, in the above embodiments, the frequency during semiconductor memory device diagnostics is higher than the frequency of electronic control devices during vehicle operation, and it is the frequency at which the defective bit rate begins to increase in the frequency dependence of the defective bit rate when the semiconductor device leaves the factory.

[0035] Based on the embodiments of the present invention described above, the following effects can be obtained. (1) An embodiment of the present invention relates to an in-vehicle control system comprising: a storage device used in vehicle control; a characteristic monitoring unit for monitoring the characteristics of the storage device; a frequency changing unit for changing the operating frequency of the storage device to a second frequency higher than a first frequency used in normal vehicle control when the characteristic monitoring unit monitors the characteristics of the storage device; and a determination unit for determining a malfunction indication of the storage device based on the monitoring result of the characteristic monitoring unit.

[0036] With the above configuration, even during normal operation of the vehicle control system, it is possible to detect storage areas that have slowed down due to aging, thereby providing early warning of malfunctions.

[0037] (2) The fault prediction diagnostic device also includes: a fault information statistics unit, whose statistical characteristic monitoring unit outputs fault information including any one of the number of faulty bits, ECC frequency, or ECC error number; and a fault information holding unit, which holds the fault information, and the determination unit determines the fault prediction of the storage device based on the time series data held in the fault information holding unit. By using such specific time series data, (1) can be achieved efficiently.

[0038] (3) It includes: a defective information statistics unit, whose statistical characteristic monitoring unit outputs defective information including any one of defective bit count, ECC frequency, or ECC error count; and a defective information retention unit, which retains the defective information, and the determination unit determines the fault signs of the storage device based on the time-series data retained in the defective information retention unit. By using such specific time-series data, (1) can be achieved efficiently.

[0039] (4) It also includes a voltage conversion unit that, while the characteristic monitoring unit monitors the characteristics of the storage device, converts the power supply voltage of the storage device to a second voltage that is lower than the first voltage used during normal vehicle control. In this way, by using voltage in addition to frequency for diagnosis, more comprehensive diagnostics can be achieved.

[0040] (5) The determination unit determines the fault signs of the storage device based on the difference between the fault information held in the fault information holding unit based on the previous monitoring result and the fault information held in the most recent monitoring result, as well as the working time from the previous monitoring to the most recent monitoring. Therefore, only the two most recent monitoring results need to be held, thus reducing the memory capacity.

[0041] (6) If the most recent bad information is interpolated to a predetermined warning value, and the result exceeds a predetermined threshold, a fault warning for the storage device is determined. Therefore, there is no need to retain previous data, and thus no need to prepare a storage device for data retention.

[0042] (7) It also includes: an environmental monitoring device that monitors the working environment data when the characteristic monitoring unit is monitoring; the fault prediction diagnosis device further includes: a correction unit that corrects the output result of the characteristic monitoring unit based on the monitoring result of the environmental monitoring device; and a fault information holding unit that holds the fault information corrected by the correction unit. Thus, more accurate diagnosis can be performed based on the corrected monitoring result.

[0043] (8) The operating environment data is the ambient temperature or voltage of the storage device. In this invention, this data can be well utilized.

[0044] (9) The frequency conditions used by the characteristic monitoring unit when monitoring the characteristics of the storage device are the frequency conditions under which the storage device malfunctions at the time of manufacture. By using such conditions, it is possible to identify aging storage devices.

[0045] Furthermore, the present invention is not limited to the above embodiments and various modifications are possible. For example, the above embodiments have been described in detail for ease of understanding, and the present invention is not necessarily limited to having all the described configurations. Additionally, some configurations of one embodiment may be replaced with configurations of other embodiments. Furthermore, configurations of other embodiments may be added to the configuration of one embodiment. Moreover, some configurations of each embodiment may be deleted, added / replaced with other configurations. Symbol Explanation

[0046] 1… Semiconductor memory device, 2… Fault prediction diagnostic device, 2-1… Characteristic monitoring unit, 2-2… Characteristic diagnostic unit, 2-3… Malfunction information statistics unit, 2-4… Fault prediction judgment unit, 2-5… Malfunction information retention unit, 2-9… Correction unit, 3… Frequency changing circuit, 4… Voltage changing circuit, 5… Environmental monitoring device.

Claims

1. A vehicle-mounted control system, characterized in that, have: Storage devices used in automotive controls; A fault prediction diagnostic device that diagnoses fault predictions of the storage device; as well as A frequency changing circuit, which, when the fault prediction diagnostic device diagnoses the storage device, changes the operating frequency of the storage device to a second frequency higher than a first frequency, the first frequency being the frequency used during normal control of the vehicle.

2. The vehicle control system according to claim 1, characterized in that, The fault precursor diagnostic device includes: A characteristic monitoring unit is used to monitor the characteristics of the storage device; and The determination unit determines the potential fault signs of the storage device based on the monitoring results of the characteristic monitoring unit.

3. The vehicle control system according to claim 2, characterized in that, The fault precursor diagnostic device includes: The defect information statistics unit counts the defect information output by the characteristic monitoring unit, which includes any one of the following: defect number, ECC frequency, or ECC error number; and The harmful information retention unit retains the harmful information. The determination unit determines the fault signs of the storage device based on the time-series data held in the bad information holding unit.

4. The vehicle control system according to claim 2, characterized in that, It also has: A voltage changing circuit, when the characteristic monitoring unit monitors the characteristics of the storage device, changes the power supply voltage of the storage device to a second voltage that is lower than a first voltage, the first voltage being the voltage used during normal control of the vehicle.

5. The vehicle control system according to claim 3, characterized in that, The determination unit determines the fault signs of the storage device based on the difference between the bad information held in the bad information holding unit based on the previous monitoring result and the bad information based on the most recent monitoring result, and the working time from the previous monitoring to the most recent monitoring.

6. The vehicle control system according to claim 3, characterized in that, The determination unit interpolates a predetermined warning value to the most recent bad information, and determines a fault warning of the storage device if the result exceeds a predetermined threshold.

7. The vehicle control system according to claim 3, characterized in that, It also has: An environmental monitoring device monitors the working environment data monitored by the characteristic monitoring unit. The fault prediction diagnostic device also includes a calibration unit, which corrects the output of the characteristic monitoring unit based on the monitoring results of the environmental monitoring device. The defective information retention unit retains the defective information after it has been corrected by the correction unit.

8. The vehicle control system according to claim 7, characterized in that, The operating environment data is the ambient temperature or voltage of the storage device.

9. The vehicle control system according to claim 2, characterized in that, The frequency conditions used by the characteristic monitoring unit when monitoring the characteristics of the storage device are the frequency conditions under which the storage device malfunctions at the time of manufacture.

Images