JP2009289204A - In-vehicle data recording system and in-vehicle data recording method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an in-vehicle data recording system more flexibly switching a plurality of recording modes of which the sorts of data to be recorded and sampling rates are different from each other. <P>SOLUTION: The in-vehicle data recording system 100 having both of a first mode for continuously recording data output from an in-vehicle apparatus 2 of a first group at a first sampling rate and a second mode for recording data output from an in-vehicle apparatus 3 of a second group at a second sampling rate higher than the first sampling rate for a predetermined period when a prescribed state is detected includes: a data abnormality generation mechanism analysis means 13 for analyzing a mechanism for generating data abnormality based on the data recorded in the first mode; and a recording mode switching means 12 for switching the first mode to the second mode before the generation of data abnormality based on an analysis result of the data abnormality generation mechanism analysis means 13. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention has a function of continuously recording data output from an in-vehicle device at a predetermined sampling rate and a function of recording data output from the in-vehicle device at a higher sampling rate for a predetermined period only when a predetermined event occurs. And an in-vehicle data recording method used by the in-vehicle data recording system.

  Conventionally, it operates as a tachograph that continuously records the running speed at regular time intervals during normal operation, and operates as an event data recorder that records more types of data related to vehicle behavior at shorter time intervals when an accident occurs A state recording device is known (for example, refer to Patent Document 1).

  This operation state recording device includes a first recording mode for recording a traveling speed at intervals of 0.5 seconds, a peripheral monitoring device, a GPS device, a gyro sensor, a steering wheel angle sensor, a wheel rotation speed sensor at intervals of 0.1 seconds. A second recording mode for recording data output from the brake pedal pressure sensor and the ABS, and the first recording mode is detected when a speed over, sudden braking, wheel lock, insufficient distance between vehicles, or sudden steering is detected. The recording mode is switched to the second recording mode so that data necessary for each of the operation management and the accident analysis can be recorded.

In addition, the operation state recording device sets the recording mode to the first recording mode when the abnormal state as described above is removed even after the recording mode is switched from the first recording mode to the second recording mode. In addition, the data recorded in the second recording mode can be recorded by transmitting the data recorded in the second recording mode to the communication center and then deleting the data, or deleting the data without transmitting to the communication center. To effectively use the storage area.
JP 2002-42288 A

  However, since the operation state recording device described in Patent Document 1 only starts the second recording mode only when it is detected that a predetermined event registered in advance has occurred, appropriate data is transmitted at an appropriate timing. In some cases, recording cannot be performed at an appropriate sampling rate.

  In view of the above, an object of the present invention is to provide an in-vehicle data recording system and an in-vehicle data recording method that can more flexibly switch between a plurality of recording modes having different types of data to be recorded and different sampling rates.

  In order to achieve the above object, an in-vehicle data recording system according to a first aspect of the present invention detects a first mode for recording data output by a first group of in-vehicle devices at a first sampling rate, and a predetermined state. An in-vehicle data recording system having a second mode for recording data output by the second group of in-vehicle devices at a second sampling rate higher than the first sampling rate. A data abnormality occurrence mechanism analyzing means for analyzing a mechanism of occurrence of the data abnormality based on the obtained data, and the first mode before the occurrence of the data abnormality based on the analysis result by the data abnormality occurrence mechanism analyzing means. Recording mode switching means for switching to the second mode.

  The second invention is an in-vehicle data recording system according to the first invention, wherein the data abnormality occurrence mechanism analysis means sets the first mode based on the data recorded in the first mode. A condition for switching to the second mode is generated.

  The third invention is an in-vehicle data recording system according to the first or second invention, based on data recording means for recording failure analysis data in the first mode and outputs of various in-vehicle devices. Accident occurrence predicting means for predicting the probability of occurrence of an accident, and when it is determined that the probability of occurrence of an accident is high, the recording mode switching means switches the first mode to the second mode, The data recording means records accident analysis data in the second mode.

  The on-vehicle data recording method according to the fourth aspect of the invention includes a first mode for recording data output by the first group of in-vehicle devices at a first sampling rate, and a second group when a predetermined state is detected. An in-vehicle data recording method having a second mode for recording data output by an in-vehicle device at a second sampling rate higher than the first sampling rate, the data being based on the data recorded in the first mode Data abnormality occurrence mechanism analysis step for analyzing the mechanism of occurrence of abnormality, and recording for switching the first mode to the second mode before occurrence of data abnormality based on the analysis result in the data abnormality occurrence mechanism analysis step And a mode switching step.

  Further, a fifth invention is an in-vehicle data recording method according to the fourth invention, wherein a recording for generating a condition for switching the first mode to the second mode based on the data recorded in the first mode. The method further includes a mode switching condition generation step.

  The sixth invention is an in-vehicle data recording method according to the fourth or fifth invention, based on a data recording step of recording failure analysis data in the first mode, and outputs of various in-vehicle devices. An accident occurrence prediction step for predicting the probability of occurrence of an accident, and when the probability of occurrence of an accident is determined to be high in the accident occurrence prediction step, the first mode is the second mode in the recording mode switching step. The mode is switched, and accident analysis data is recorded in the second mode in the data recording step.

  By the means described above, the present invention can provide an in-vehicle data recording system and an in-vehicle data recording method that can more flexibly switch between a plurality of recording modes having different types of data to be recorded and different sampling rates.

  Hereinafter, the best mode for carrying out the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram illustrating a configuration example of an in-vehicle data recording system according to the present invention. An in-vehicle data recording system 100 includes a control device 1, a failure analysis device 2, an accident analysis device 3, and a storage device 4. Each component is connected to each other via an in-vehicle LAN such as a CAN (Controller Area Network) or a LIN (Local Interconnect Network).

  The control device 1 is a computer including a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), etc., for example, a data recording means 10, an accident occurrence prediction means 11, a recording mode switching. While storing programs corresponding to each of the means 12 and the data abnormality occurrence mechanism analyzing means 13 in the ROM, the CPU is caused to execute processing corresponding to each means.

  The failure analysis device 2 is a device that outputs data (hereinafter referred to as “failure analysis data”) used to analyze the failure content of a vehicle or an in-vehicle device, and directly outputs raw data, for example. And a secondary device that performs a predetermined calculation based on the output of raw data by the primary device and outputs a level signal indicating a predetermined state and a trigger signal indicating that the predetermined state has been reached. Including.

  The primary device is a radar sensor or clearance sonar for measuring the distance to an object around the vehicle, an acceleration sensor for measuring acceleration in the three longitudinal directions of the vehicle in the longitudinal direction, the vertical direction, and the horizontal direction. Steering angle sensor for measuring rotational speed and rotational angle, throttle opening sensor for measuring throttle opening, vehicle speed sensor for measuring wheel rotational speed and vehicle running speed, accelerator pedal or brake pedal Treading pressure sensor for measuring the treading pressure of the engine, water temperature sensor for measuring the water temperature of the engine cooling water, oil temperature sensor for measuring the oil temperature of the transmission oil, vehicle position (latitude, longitude, altitude) A GPS (Global Positioning System) sensor for measuring the rotation speed or a rotation sensor for measuring the engine speed And the like.

  The secondary device is an engine ECU (Electronic Control Unit) that outputs a signal representing the operating state of the engine while controlling the throttle opening, brake pedal pressure, etc. with an actuator based on the output of the accelerator pedal depression pressure sensor, or An inter-vehicle distance control ECU or the like that outputs a signal indicating the inter-vehicle distance state between the preceding vehicle or the following vehicle and the host vehicle while controlling the throttle opening, brake pedal pressure, etc. with an actuator based on the output of the radar sensor .

  In addition, the secondary device includes an ABS (Antilock Brake System) that outputs data related to the wheel lock state, a TRC (Traction control) device that outputs data related to the traction state, and a VSC (Vehicle) that outputs data related to the side slip state of the vehicle. It may further include a Stability Control) device, or a VDIM (Vehicle Dynamics Integrated Management) device that outputs data relating to the state of motion of the vehicle in the front, rear, left, and right directions.

  The accident analysis device 3 is a device that outputs data (hereinafter referred to as “accident analysis data”) used for analyzing the accident contents, and outputs, for example, data relating to the attachment / detachment state of the seat belt. It includes a seat belt ECU, an airbag ECU that outputs data related to the open / close state of the airbag, an image sensor (camera) that outputs image data when an accident occurs, an in-vehicle microphone that outputs data related to sound when the accident occurs.

  The storage device 4 is a device for recording data received from the failure analysis device 2 or the accident analysis device 3 by the control device 1, for example, a non-volatile storage medium such as a hard disk or a flash memory. It is arranged with various measures such as heat resistance and water resistance at a position in the passenger compartment where it is not affected as much as possible by impact at the time of occurrence, heat from fire, or contamination by foreign matter such as water, mud or dust.

  Next, various units included in the control device 1 will be described.

  The data recording means 10 is means for recording data received by the control device 1 from the failure analysis device 2 or the accident analysis device 3. For example, the failure analysis data is recorded at a low sampling rate (for example, 0.5 Low sampling mode for recording at a second interval), high sampling mode for recording accident analysis data at a high sampling rate (for example, at intervals of 0.1 second), and medium for sampling failure analysis data (For example, at intervals of 0.3 seconds) The medium sampling mode is recorded.

  Further, the data recording means 10 continuously performs recording in the low sampling mode, and performs recording in the high sampling mode when it is determined by the accident occurrence predicting means 11 described later that the probability of the occurrence of the accident is high. .

  The accident occurrence predicting means 11 is a means for predicting the occurrence of an accident. For example, when the inter-vehicle distance between the preceding vehicle or the succeeding vehicle and the host vehicle is rapidly reduced based on the output of the radar sensor, the ABS is used. When it is detected that the wheels are locked by the above, it is determined that the probability of the occurrence of an accident is high when a side slip of the host vehicle is detected by the VSC device or when sudden braking or sudden steering is executed.

  Further, the accident occurrence prediction means 11 includes the contents of the accident (for example, contact between the preceding vehicle and the own vehicle, contact between the following vehicle and the own vehicle, collision with an obstacle due to a lane departure on a curve, etc.). Accordingly, the configuration of the accident analysis data recorded in the data recording means 10 may be changed.

  For example, when the accident occurrence prediction unit 11 predicts contact between the preceding vehicle and the host vehicle, the accident occurrence prediction unit 11 records image data output by a camera that captures the front and predicts contact between the subsequent vehicle and the host vehicle. In this case, image data output by a camera that captures the rear side is recorded.

  In addition, the accident occurrence prediction means 11 has confirmed that a predetermined time has elapsed since it was determined that the probability of the occurrence of the accident is high, and that the accident did not actually occur based on the outputs of various on-vehicle devices. In this case, this is notified to the recording mode switching means 12 described later.

  Further, the accident occurrence prediction means 11 is a case where a predetermined time has elapsed from the time when it is determined that the probability of occurrence of the accident is high, and when it is confirmed that the accident has actually occurred based on the outputs of various on-vehicle devices. The data recording may be stopped by outputting a control signal to the data recording means 10. This is to prevent a large amount of useless data from being recorded by continuing recording in the high sampling mode even after a predetermined time has elapsed since the occurrence of the accident.

  The recording mode switching unit 12 is a unit for switching the recording mode in the data recording unit 10. For example, when the accident occurrence prediction unit 11 determines that the probability of occurrence of an accident is high, the recording mode switching unit 12 A control signal is output to switch the recording mode from the low sampling mode to the high sampling mode.

  The recording mode switching means 12 determines that the probability of the occurrence of the accident is high by the accident occurrence prediction means 11 and switches the recording mode from the low sampling mode to the high sampling mode. When a notification that no occurrence has occurred is received, a control signal is output to the data recording means 10 to return the recording mode from the high sampling mode to the low sampling mode.

  The recording mode switching means 12 is determined to have a high probability of occurrence of an accident by the accident occurrence prediction means 11 and the accident occurrence prediction means even after a predetermined time has elapsed after the recording mode is switched from the low sampling mode to the high sampling mode. When no notification is obtained from 11 or when a notification that the accident has actually occurred is received, a control signal is output to the data recording means 10 to stop data recording.

  The recording mode switching means 12 changes the recording mode from the low sampling mode to the medium sampling mode when a condition for switching the recording mode newly set by the data abnormality occurrence mechanism analyzing means 13 described later is satisfied. In addition, after the recording mode is switched from the low sampling mode to the medium sampling mode, the recording mode is changed from the low sampling mode to the low sampling mode when a predetermined time newly set by the data abnormality occurrence mechanism analysis unit 13 described later has elapsed. Return to sampling mode.

  The data abnormality occurrence mechanism analyzing means 13 is a means for analyzing the mechanism in which the data abnormality occurs, for example, reading past data recorded in the low sampling mode from the storage device 4 and Extract data anomalies caused by failures (the range of normal values of data output by various in-vehicle devices is defined in advance) and analyze the mechanism (regularity) in which the data anomalies occur .

  The data abnormality occurrence mechanism analysis unit 13 outputs, for example, another vehicle-mounted device at a predetermined time preceding each of a plurality of time points at which data that is considered abnormal among data output by a specific vehicle-mounted device is acquired. Data to be read out, and common conditions (the range that can be taken by data output by various in-vehicle devices, the range that the change rate (differential value) of the data can take, etc.) are derived.

  On that basis, the data abnormality occurrence mechanism analysis means 13 is, for example, that the output value of the radar sensor becomes abnormal when the steering wheel is rotated to the right by Y degrees or more while traveling at X km / hour or more, or the headlight Regularity is found such that the output value of the engine ECU becomes abnormal when the set temperature of the air conditioner is set to be equal to or higher than Z ° C. in a state where is turned on.

  The data abnormality occurrence mechanism analyzing means 13 generates a condition for switching the recording mode based on the found regularity (hereinafter referred to as “recording mode switching condition”), and the recording after the switching is performed. The mode sampling rate and the structure of data to be recorded are determined. This is because it is possible to record necessary and sufficient data corresponding to individual data abnormalities.

  Note that the control device 1 displays a correspondence table that indicates at what sampling rate the data output from any other in-vehicle device should be recorded when an abnormality is found in the data output from one in-vehicle device. It is assumed that it is registered in the ROM in advance.

  Next, referring to FIG. 2, the in-vehicle data recording system 100 predicts the occurrence of an accident based on a predetermined condition and switches the recording mode from the low sampling mode to the high sampling mode (hereinafter referred to as “accident prediction recording mode switching”). Process ”) will be described.

  FIG. 2 is a flowchart showing the flow of the accident prediction recording mode switching process. The in-vehicle data recording system 100 repeatedly executes this process at a predetermined cycle while recording in the low sampling mode. To do.

  First, the control device 1 determines whether or not there is a high probability that an accident will occur by comparing the output of the radar sensor or the like with a threshold value registered in advance by the accident occurrence predicting means 11 (step S1). When it is determined that the probability of occurrence is high (YES in step S1), the recording mode switching unit 12 outputs a control signal to the data recording unit 10 to switch the recording mode from the low sampling mode to the high sampling mode (step S1). S2).

  Note that the control device 1 continues recording in the low sampling mode as it is without switching the recording mode from the low sampling mode to the high sampling mode unless it is determined that the probability of an accident occurring is high (NO in step S1). .

  Next, referring to FIG. 3, the in-vehicle data recording system 100 predicts the occurrence of an accident and switches the recording mode from the low sampling mode to the high sampling mode, and then returns the recording mode to the low sampling mode according to the situation. The process to be performed (hereinafter referred to as “record mode switching process after accident prediction”) will be described.

  FIG. 3 is a flowchart showing the flow of the recording mode switching process after the accident prediction. The in-vehicle data recording system 100 predicts the occurrence of the accident and switches the recording mode from the low sampling mode to the high sampling mode. This processing is repeatedly executed at a predetermined cycle until the recording mode is returned to the low sampling mode or until a predetermined time has elapsed and the recording is stopped.

  The control device 1 measures the elapsed time from when the recording mode is switched from the low sampling mode to the high sampling mode (step S11), and when the predetermined time has not elapsed (NO in step S11). The recording in the high sampling mode is continued as it is without returning the recording mode from the high sampling mode to the low sampling mode.

  When the predetermined time has elapsed (YES in step S11), the control device 1 causes the accident occurrence prediction unit 11 to actually generate an accident that is considered to have a high probability of occurrence based on the output of various in-vehicle devices. Confirm whether or not, and notify the confirmation result to the recording mode switching means 12 (step S12).

  When receiving a notification that an accident that is highly likely to occur has not actually occurred (YES in step S12), the recording mode switching means 12 outputs a control signal to the data recording means 10 for recording. The mode is returned from the high sampling mode to the low sampling mode (step S13).

  On the other hand, when receiving a notification that an accident with a high probability of occurrence has actually occurred, or when not receiving any notification (NO in step S12), the recording mode switching means 12 performs data recording. A control signal is output to the means 10 to stop recording (step S14).

  Thus, the in-vehicle data recording system 100 can record an appropriate amount of necessary data without recording a large amount of unnecessary data.

  Next, referring to FIG. 4, based on data recorded in the past by the in-vehicle data recording system 100, an analysis is made of the mechanism that causes a data abnormality (for example, an abnormality caused by a failure of the vehicle or the in-vehicle device). In order to investigate the cause of the data abnormality, the recording mode is switched to the medium sampling mode before the next data abnormality occurs, so that data before and after the data abnormality occurs can be recorded with higher density. Processing for newly generating a recording mode switching condition (for example, a pre-registered condition such as “the ABS outputs data indicating that the wheel has been locked”) The “recording mode switching condition generation process” will be described.

  FIG. 4 is a flowchart showing the flow of the recording mode switching condition generation process. The in-vehicle data recording system 100 executes this process every time the vehicle travels a predetermined distance (for example, 100 km). And

  First, the control device 1 reads the data recorded in the low sampling mode from the storage device 4 by the data abnormality occurrence mechanism analysis means 13 (step S21), and determines whether or not the occurrence of the data abnormality is regular. Analysis is performed (step S22).

  Thereafter, when the occurrence of data abnormality is regular by the data abnormality occurrence mechanism analysis means 13, the control device 1 generates a recording mode switching condition based on the regularity (step S23).

  Further, the control device 1 identifies the in-vehicle device that has output the abnormal data by the data abnormality occurrence mechanism analysis unit 13 and refers to the correspondence table stored in the ROM, and is suspected of malfunctioning. It is determined at what sampling rate the data output from any other in-vehicle device related to the in-vehicle device is recorded, and the medium sampling mode is defined (step S24).

  As a result, the in-vehicle data recording system 100 can dynamically define another recording mode switching condition similar to the recording mode switching condition registered in advance as used by the accident occurrence predicting means 11, and data to be recorded A plurality of recording modes having different types and sampling rates can be used more flexibly.

  Next, referring to FIG. 5, when a newly generated recording mode switching condition is satisfied in order to record data before and after the occurrence of data abnormality with higher density in order to investigate the cause of data abnormality, A process in which the data recording system 100 switches the recording mode from the low sampling mode to the medium sampling mode (hereinafter referred to as “data abnormality prediction recording mode switching process”) will be described.

  FIG. 5 is a flowchart showing the flow of the recording mode switching process at the time of data abnormality prediction. The in-vehicle data recording system 100 repeatedly executes this process at a predetermined cycle while recording in the low sampling mode. And

  First, the control device 1 determines whether or not the recording mode switching condition newly set by the data abnormality occurrence mechanism analysis unit 13 is satisfied based on the outputs of various in-vehicle devices (step S31). When it is determined that the set recording mode switching condition is satisfied (YES in step S31), the recording mode switching unit 12 outputs a control signal to the data recording unit 10 to change the recording mode from the low sampling mode to the medium sampling mode. (Step S32).

  Note that the control device 1 uses the low sampling mode without switching the recording mode from the low sampling mode to the medium sampling mode unless it is determined that the newly set recording mode switching condition is satisfied (NO in step S31). Continue recording.

  Finally, referring to FIG. 6, after the in-vehicle data recording system 100 predicts the occurrence of data abnormality and switches the recording mode from the low sampling mode to the medium sampling mode, the recording mode is changed from the medium sampling mode according to the situation. Processing for returning to the low sampling mode (hereinafter referred to as “recording mode switching processing after data abnormality prediction”) will be described.

  FIG. 6 is a flowchart showing the flow of the recording mode switching process after data abnormality prediction, and the in-vehicle data recording system 100 predicts the occurrence of the data abnormality and switches the recording mode from the low sampling mode to the medium sampling mode. Thereafter, this processing is repeatedly executed at a predetermined cycle until the recording mode is returned to the low sampling mode.

  The control device 1 measures the elapsed time from when the recording mode is switched from the low sampling mode to the medium sampling mode (step S41). If the predetermined time has not elapsed (NO in step S41), the recording is performed. Recording in the middle sampling mode is continued without changing the mode from the middle sampling mode to the low sampling mode.

  When the predetermined time has elapsed (YES in step S41), the control device 1 outputs a control signal to the data recording means 10 by the recording mode switching means 12 to return the recording mode from the medium sampling mode to the low sampling mode ( Step S42).

  With the above configuration, the in-vehicle data recording system 100 switches the recording mode based on a recording mode switching condition that can be determined in advance (for example, a condition for switching the recording mode immediately before an accident occurs). In addition, a recording mode switching condition that cannot be determined in advance (for example, a condition for switching the recording mode immediately before the occurrence of a data abnormality having a different generation mechanism for each vehicle) is generated. Since the recording mode is switched based on the dynamically generated recording mode switching condition, a plurality of recording modes having different types of data to be recorded and different sampling rates can be switched more flexibly.

  In addition, the in-vehicle data recording system 100 can perform failure diagnosis of the vehicle or the in-vehicle device in more detail and more efficiently.

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

  For example, in the above-described embodiment, the in-vehicle data recording system 100 records the failure analysis data in the high sampling mode when the occurrence of the accident is predicted while continuously recording the failure analysis data in the low sampling mode. Start recording of failure analysis data in the medium sampling mode when the occurrence of data abnormality is predicted, but if it is predicted that no data abnormality will occur, the failure analysis data will be recorded at a lower sampling rate ( For example, another recording mode may be used in which recording is performed at 1 second intervals.

It is a block diagram which shows the structural example of the vehicle-mounted data recording system which concerns on this invention. It is a flowchart which shows the flow of the recording mode switching process at the time of accident prediction. It is a flowchart which shows the flow of the recording mode switching process after accident prediction. It is a flowchart which shows the flow of a recording mode switching condition production | generation process. It is a flowchart which shows the flow of the recording mode switching process at the time of data abnormality prediction. It is a flowchart which shows the flow of the recording mode switching process after data abnormality prediction.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Control apparatus 2 Failure analysis apparatus 3 Accident analysis apparatus 4 Storage apparatus 10 Data recording means 11 Accident occurrence prediction means 12 Recording mode switching means 13 Data abnormality occurrence mechanism analysis means 100 In-vehicle data recording system

Claims (6)

The first mode for recording the data output by the first group of in-vehicle devices at the first sampling rate, and the data output by the second group of in-vehicle devices when a predetermined state is detected than the first sampling rate. A vehicle-mounted data recording system having a second mode for recording at a high second sampling rate,
A data abnormality occurrence mechanism analysis means for analyzing a mechanism in which a data abnormality occurs based on the data recorded in the first mode;
Recording mode switching means for switching the first mode to the second mode before the occurrence of data abnormality based on the analysis result by the data abnormality occurrence mechanism analysis means,
An in-vehicle data recording system comprising: The data abnormality occurrence mechanism analysis means generates a condition for switching the first mode to the second mode based on the data recorded in the first mode.
The in-vehicle data recording system according to claim 1. Data recording means for recording failure analysis data in the first mode;
Accident occurrence prediction means for predicting the probability of occurrence of an accident based on the output of various in-vehicle devices,
When it is determined that the probability of occurrence of an accident is high, the recording mode switching means switches the first mode to the second mode, and the data recording means records accident analysis data in the second mode. ,
The in-vehicle data recording system according to claim 1 or 2, wherein The first mode for recording the data output by the first group of in-vehicle devices at the first sampling rate, and the data output by the second group of in-vehicle devices when a predetermined state is detected than the first sampling rate. An in-vehicle data recording method having a second mode for recording at a high second sampling rate,
A data abnormality occurrence mechanism analysis step for analyzing a mechanism in which the data abnormality occurs based on the data recorded in the first mode;
A recording mode switching step for switching the first mode to the second mode before occurrence of data abnormality based on the analysis result in the data abnormality occurrence mechanism analysis step,
An in-vehicle data recording method comprising: A recording mode switching condition generating step for generating a condition for switching the first mode to the second mode based on the data recorded in the first mode;
The in-vehicle data recording method according to claim 4. A data recording step of recording failure analysis data in the first mode;
An accident occurrence prediction step for predicting the probability of occurrence of an accident based on the output of various in-vehicle devices,
When it is determined in the accident occurrence prediction step that the probability of occurrence of an accident is high, the first mode is switched to the second mode in the recording mode switching step, and the accident analysis data is stored in the data recording step. Recorded in two modes,
The in-vehicle data recording method according to claim 4 or 5, characterized in that.

Images