JP6027743B2 - Vehicle behavior analysis device, vehicle behavior analysis program, and drive recorder - Google Patents

Description

  The present invention records the behavior and surroundings of a vehicle when an accident occurs or even if an accident such as a near-miss does not occur, and analyzes the cause of the situation after the fact. The present invention relates to a vehicle behavior analysis apparatus or a vehicle behavior analysis program that can be suitably performed.

  Recently, as a vehicle behavior data collection device, for example, images of the outside or inside of a driving vehicle (automobile) are automatically recorded, and the objective situation in the event of an accident or near-miss, and thus the driving tendency of the driver Vehicle-mounted drive recorders have been developed that enable post-mortem analysis. For example, in taxis, there is a movement to install this type of drive recorder for accident prevention measures by post-mortem analysis of daily driving and for objective evidence and investigation of the cause of accidents. Yes.

  Specifically, such a vehicle behavior data collection device records situation data such as internal / external image data, acceleration data, speed data, position data, etc. during running in a memory sequentially and chronologically. Yes. And it is comprised so that objective analysis, such as an accident, can be performed by referring the status data in the memory after that with another apparatus (refer patent document 1).

  However, when trying to analyze the situation data recorded in this way after the fact, these situation data are collected because they contain situation data indicating various behaviors in addition to specific vehicle behavior such as near-miss It is necessary to classify and extract situation data indicating desired vehicle behavior from a large number of situation data.

  Here, conventionally, a large number of collected situation data is visually confirmed one by one, and each situation data is classified into, for example, a collision accident, a near-miss, a sudden brake not related to a near-miss, or a simple noise. Is going.

  However, the task of confirming and classifying situation data one by one not only takes too much time to confirm one situation data, but can be classified accurately by arbitrary judgment of the user. There is a problem that it is difficult. In addition, there is a problem that the physical burden and the mental burden given to the user are large.

  On the other hand, as shown in Patent Document 2, there is an apparatus that automatically classifies situation data into situation data caused by noise and situation data used for behavior analysis such as near-miss to reduce the burden on the user. Yes. Specifically, this apparatus is configured to eliminate unnecessary situation data caused by noise based on the wave height and pulse width of the waveform of acceleration data.

  However, as in the case of the above-described apparatus, in the case of discriminating unnecessary situation data caused by noise and situation data used for behavior analysis such as near-miss using a threshold, the classification accuracy differs greatly depending on how the threshold is set. There's a problem. Moreover, even if the data resulting from noise can be removed using the threshold value, it is necessary to visually check the remaining situation data one by one and classify them.

  For example, if the threshold value is set low, the number of near-miss case candidates extracted from all the situation data increases, but the hit rate of near-miss cases in the extracted data decreases, and the proportion of unnecessary data included increases. On the other hand, if the threshold value is set high, the number of near-miss case candidates extracted from all the situation data will be reduced, but the probability of near-miss cases in the extracted situation data will be high, and the proportion of unnecessary data will be small. . However, in this case, the leak rate of the near-miss example becomes high. In other words, there is a trade-off between the hit rate and the leak rate in the method of uniformly separating by the threshold value. Also, just because the threshold value is raised and the hit rate is raised, the hit rate is much lower than that of visual classification and is not practical. Therefore, in order to correctly classify near-miss cases from situation data, it is still necessary to perform visual work.

JP 2007-11909 A Japanese Patent No. 4238293

  Therefore, the present invention has been made for the first time by paying attention to the fact that the characteristic amount of each acceleration data has a specific relative relationship for each behavior of the vehicle as a result of intensive studies by the present inventor. The main intended task is to automatically and reliably specify the vehicle behavior indicated by the situation data without classifying the behavior only by visual observation.

  In other words, the vehicle behavior analysis apparatus according to the present invention can include situation data including acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration acting on the vehicle, or roll angular acceleration, pitch angular acceleration, and yaw angular acceleration acting on the vehicle. A situation data receiving unit that receives situation data including at least one of situation data including angular acceleration data, and a relative relationship between feature amounts of each acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration in the received situation data; or Vehicle behavior that identifies the vehicle behavior indicated by the situation data using the relative relationship of at least one of the feature amounts of the angular acceleration data of roll angular acceleration, pitch angular acceleration, and yaw angular acceleration And a specific unit.

  If it is such, since the vehicle behavior can be specified by the relative relationship using the feature amount of each acceleration data or the feature amount of each angular acceleration data, the vehicle indicated by the situation data without depending on visual observation The behavior can be specified. Thereby, not only a user's arbitrary judgment can be excluded but a vehicle behavior can be specified objectively, and a user's time, physical, and mental burden can be reduced. Further, since the vehicle behavior can be specified by the relative relationship between the feature amount of each acceleration data or the feature amount of each angular acceleration data, it is possible to obtain a more reliable specification result than that determined by only the threshold value. .

Specifically, as a result of intensive studies by the inventor of the present application, in the following (1) to (3) vehicle behavior, there is a specific relative relationship in the feature values of the acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration. I understood that. That is, when the received situation data includes acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration, the vehicle behavior specifying unit determines a relative relationship between feature values of the acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration. It can be considered that the vehicle behavior indicated by the situation data is specified as one of the following (1) to (3).
(1) Collision accidents, near-misses that lead to collision accidents, or near-miss events that indicate other sudden braking.
(2) Behavior such as riding on a single wheel on a curb or taking off a wheel in a side groove.
(3) A bouncing behavior indicating bouncing caused by both wheels passing through the unevenness on the roadway.

  More specifically, when the feature amount of the longitudinal acceleration data is larger than the feature amount of the lateral acceleration data and the vertical acceleration data, the vehicle behavior specifying unit specifies the vehicle behavior indicated by the situation data as the near-miss behavior. If the feature amount of the left-right acceleration data is larger than the feature amount of the longitudinal acceleration data and the vertical acceleration data, the vehicle behavior indicated by the situation data is specified as the riding-up behavior, and the longitudinal acceleration data and the left-right acceleration data When the feature amount of the vertical acceleration data is larger than the feature amount, it is desirable to specify the vehicle behavior indicated by the situation data as the bound behavior.

  In the near-miss behavior, in order to further subdivide and specify a collision accident and a sudden brake other than near-miss, the situation data includes winker data indicating the operation information of the winker of the vehicle, and the vehicle behavior specifying unit When the blinker data is included in the situation data identified as the near-miss behavior or the like, it is desirable to identify the behavior as indicating a sudden brake for stopping at the road shoulder.

  Further, the vehicle behavior analysis program according to the present invention is a situation data including acceleration data of longitudinal acceleration, lateral acceleration and vertical acceleration acting on the vehicle, or roll angular acceleration, pitch angular acceleration and yaw angular acceleration acting on the vehicle. A situation data receiving unit that receives situation data including at least one of situation data including angular acceleration data, and a relative relationship between feature amounts of each acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration in the received situation data; or Vehicle behavior that identifies the vehicle behavior indicated by the situation data using the relative relationship of at least one of the feature amounts of the angular acceleration data of roll angular acceleration, pitch angular acceleration, and yaw angular acceleration The computer is provided with a function as a specific unit.

  The vehicle behavior analysis device acquires situation data from a drive recorder mounted on the vehicle and analyzes the situation data. The function of the vehicle behavior analysis device may be given to the drive recorder. That is, the drive recorder according to the present invention includes situation data including acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration acting on the vehicle, or roll angular acceleration, pitch angular acceleration, and yaw angular acceleration acting on the vehicle. Relative relationship between the situation data receiving unit that receives at least one of the situation data including the acceleration data and the characteristic amount of each acceleration data of the longitudinal acceleration, the lateral acceleration, and the vertical acceleration in the received situation data, or roll A vehicle behavior identifying unit that identifies a vehicle behavior indicated by the situation data using a relative relationship between at least one of the feature amounts of angular acceleration data of angular acceleration, pitch angular acceleration, and yaw angular acceleration. It is characterized by providing.

  If this is the case, it is possible to identify the vehicle behavior indicated by each situation data when the drive recorder stores the situation data in the memory. Can be reduced. Further, only a specific behavior can be selected and stored in the memory of the drive recorder, and the memory can be used efficiently.

  According to the present invention configured as described above, without using a method of visually classifying the behavior of the vehicle indicated by the situation data, or discriminating and classifying, for example, acceleration data included in the situation data using a threshold value, The vehicle behavior indicated by the situation data can be specified automatically and with high reliability.

FIG. 1 is a diagram schematically illustrating a vehicle behavior analysis system according to the present embodiment. FIG. 2 is a diagram showing components of the drive recorder of the same form. FIG. 3 is a functional configuration diagram of the vehicle behavior analysis apparatus according to the embodiment. FIG. 4 is a flowchart showing the operation of the vehicle behavior analysis apparatus of the same form. FIG. 5 is a diagram showing a correspondence relationship between the vehicle behavior and the feature amount of each acceleration in the same form.

DESCRIPTION OF SYMBOLS 100 ... Drive recorder V ... Vehicle 200 ... Vehicle behavior analysis apparatus 201 ... Situation data receiving part 202 ... Vehicle behavior specific | specification part

  An embodiment of a vehicle behavior analysis system according to the present invention will be described below with reference to the drawings.

<1. System configuration>
As shown in FIG. 1, the vehicle behavior analysis system according to the present embodiment acquires a drive recorder 100 that captures an image of the exterior front of an automobile (vehicle) V, moving image data captured by the drive recorder 100, and the like. And a vehicle behavior analysis apparatus 200 that specifies whether or not the automobile V exhibits a predetermined behavior based on the moving image indicated by the moving image data.

<2. Drive recorder>
The drive recorder 100 is adhered to the windshield, installed in the vicinity of the dashboard, or mounted at an appropriate position in the vehicle, and when an accident occurs or an accident occurs even if the accident does not occur. 2 is a vehicle-mounted type that records the behavior of the automobile V, surrounding conditions, and the like in a certain period of time before and after, etc., in a single or a plurality of casings, the basic components, that is, the detection means 3 shown in FIG. The unit 8 includes the means 8, the notification means 4, the input means 5, the communication means 6, the detachable recording means 7, and the like.

  The detection means 3 senses a situation related to the behavior of the automobile V, the surrounding situation, and the like, and outputs situation data indicating the situation. Here, three types of imaging means 31, an acceleration sensor 32, and a position sensor 33 are provided. At least. The imaging means 31 is, for example, a CCD camera that images a situation outside the vehicle ahead of the host vehicle and outputs situation data (moving image data) indicating the image. The acceleration sensor 32 is configured using, for example, a piezoresistive effect, senses three-dimensional acceleration acting on the vehicle, and outputs situation data (acceleration data) indicating the acceleration. Specifically, the acceleration sensor 32 senses longitudinal acceleration acting in the longitudinal direction of the vehicle V, lateral acceleration acting in the lateral direction of the vehicle V, and vertical acceleration acting in the vertical direction of the vehicle V. The position sensor 33 is, for example, a GPS receiver that detects radio waves from a plurality of satellites, senses the position of the vehicle V, and outputs status data (position data) indicating the position. The situation data includes other vehicle speed data transmitted from the vehicle speed sensor of the vehicle V, door opening / closing data indicating opening / closing of the door, brake data indicating ON / OFF of the brake, and blinker operation information of the vehicle V. And the like, and these are received via the connector CN. The connector CN is also used for a power source.

  The notification unit 4 includes an LED 41 that is a light emitter exposed on the surface of the casing, an audio output body (not shown) such as a buzzer or a speaker built in the casing, and the like.

  Here, the input means 5 is a button switch provided on the surface of the casing.

  Here, the communication means 6 refers to hardware for communication such as a wireless LAN or a mobile phone that is built in the casing and transmits / receives radio waves to / from a base station or a vehicle behavior analysis apparatus 200 described later.

  Here, the detachable recording means 7 is, for example, a CF memory card or an SD memory card that is detachably attached to a slot that opens to the side of the casing.

  The information processing means 8 is a so-called computer circuit having a CPU 81, an internal memory 82 (for example, a non-volatile memory), an I / O buffer circuit (which may include an AD converter) 83, etc. Built in the casing. The CPU 81 operates according to a program stored in a predetermined area of the memory 82 to perform control of each means and information processing.

  In brief, the CPU 81 constantly updates various status data during traveling, that is, acceleration data, position data, moving image data, etc., one after another in a temporary area set in the memory 82 (hereinafter also referred to as a temporary data storage unit). In the event that an event that indirectly indicates the occurrence of a near miss, an accident, an abnormality, or the like occurs, the status data over a certain period before and after that is stored in a normal area ( Hereinafter, it is also transferred to a regular recording data storage unit) and recorded.

  Examples of the event include a case where the acceleration (deceleration) indicated by the acceleration data exceeds a predetermined reference value, a case where the time continues for a certain period, a case where the door is opened and closed, a case where the vehicle is turned off, etc. It corresponds to it. Here, depending on the event that occurred, data was triggered only when the vehicle speed was higher than the upper limit speed, when the vehicle speed was lower than the acceleration / deceleration, or when some other conditions such as the presence or absence of braking were met. Recording is performed so that useless data is recorded as little as possible.

  Further, from the viewpoint of preventing useless data recording, it also has a learning function. In other words, before performing data recording, whether or not a near-miss or an accident has occurred is always notified to the driver by the notification means, and correct / incorrect input from the driver (for example, ON / OFF of the button switch 5) is made. I try to accept it. By repeating this, the driver's driving tendency is grasped to some extent, and for example, the driver's unique phenomenon that indirectly indicates an accident or the like is learned by changing a predetermined reference value of acceleration.

  Furthermore, the recorded situation data is weighted from the situation at the time of recording, and is classified according to the importance to be recorded. Then, when the memory capacity becomes full or the like, it is configured such that new status data is recorded by deleting from the less important ones.

  The situation data thus properly recorded is transmitted to an analysis center (not shown) wirelessly at a specific place, or transferred to the detachable recording means 7 and taken out to the analysis center. This is used for subsequent analysis using the vehicle behavior analysis apparatus 200.

<3. Vehicle Behavior Analysis Device>
The vehicle behavior analysis device 200 classifies the situation data group obtained by the drive recorder 100 mounted on a plurality of automobiles A for each predetermined vehicle behavior, and supports subsequent analysis. The specific device configuration of the vehicle behavior analysis apparatus 200 is a general purpose or dedicated computer equipped with a CPU, a memory, an input / output interface, an AD converter, etc., and a vehicle behavior analysis stored in a predetermined area of the memory. By cooperating the CPU, peripheral devices, etc. according to the program, functions as a situation data receiving unit 201, a situation data storage unit D1, a vehicle behavior specifying unit 202, an analysis data storage unit D2, etc., as shown in FIG. Demonstrate.

  Below, each part D1, 201, 202, D2 is demonstrated with operation | movement using FIG.

  The situation data receiving unit 201 receives the situation data including the moving image data and each acceleration data stored in the regular recording data storage unit of the drive recorder 100, and stores the situation data in the situation data storage unit D1 (FIG. 4, FIG. 4). Step S1). The situation data receiving unit 201 may be configured by a receiver that receives situation data transmitted using a wireless LAN or the like by the communication means (transmitter) 6 provided in the drive recorder 100. The situation data may be acquired via a CF card which is the detachable recording means 7 provided, for example.

  The situation data storage unit D1 stores and accumulates situation data including moving image data captured by the drive recorder 100 (step S2 in FIG. 4). In the present embodiment, the situation data storage unit D1 is configured to systematically store a plurality of situation data obtained by a plurality of automobiles V, for example, for each automobile V.

  The vehicle behavior specifying unit 202 acquires the situation data stored in the situation data storage unit D1, and calculates the feature amount of each acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration included in each situation data ( FIG. 4, step S3). The vehicle behavior specifying unit 202 of the present embodiment calculates, for example, standard deviation, which is a measure of the amplitude and variation of the acceleration waveform indicated by each acceleration data, as the feature amount of each acceleration data.

And the vehicle behavior specific | specification part 202 specifies the vehicle behavior which situation data shows using the relative magnitude relationship of the standard deviation which is the feature-value of each acceleration data (FIG. 4, step S4). Specifically, the vehicle behavior specifying unit 202 uses at least the following (1) to (1) to describe the vehicle behavior indicated by the situation data by using the relative magnitude relationship of the standard deviation of each acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration. It is specified as one of (3).
(1) Collision accidents, near-misses that lead to collision accidents, or near-miss events that indicate other sudden braking.
(2) Behavior such as riding on a single wheel on a curb or taking off a wheel in a side groove.
(3) A bouncing behavior indicating bouncing caused by both wheels passing through the unevenness on the roadway.

  In more detail, as shown in FIG. 5, the vehicle behavior specifying unit 202 has a case where the standard deviation of the acceleration waveform of the longitudinal acceleration data is larger than the standard deviation of the acceleration waveform of the lateral acceleration data and the standard deviation of the acceleration waveform of the vertical acceleration data. The vehicle behavior indicated by the situation data is identified as the near-miss behavior. This is based on the fact that the longitudinal acceleration changes most greatly among the accelerations due to the impact at the time of collision or the impact at the time of sudden braking, and the variation in the longitudinal acceleration waveform becomes the largest. In FIG. 5, a large number of situation data obtained by the drive recorder 100 is visually classified into the above (1) to (3), and the situation data is arranged for each behavior, and each acceleration of each situation data is obtained. It is the schematic diagram which showed the standard deviation of the waveform. The horizontal axis in FIG. 5 indicates each status data, and the vertical axis indicates the standard deviation of each acceleration waveform.

  Further, when the standard deviation of the acceleration waveform of the left / right acceleration data is larger than the standard deviation of the acceleration waveform of the longitudinal acceleration data and the standard deviation of the acceleration waveform of the vertical acceleration data, the vehicle behavior specifying unit 202 The vehicle behavior to be shown is specified as the behavior such as the ride (see FIG. 5). This is based on the fact that only one of the front wheels of the vehicle V changes abruptly in the behavior such as riding up, so that the left-right acceleration changes most greatly among the respective accelerations, and the variation in the left-right acceleration waveform becomes the largest. Yes.

  Further, when the standard deviation of the acceleration waveform of the vertical acceleration data is larger than the standard deviation of the acceleration waveform of the longitudinal acceleration data and the standard deviation of the acceleration waveform of the lateral acceleration data, the vehicle behavior specifying unit 202 indicates the situation data. The vehicle behavior is identified as the bound behavior (see FIG. 5). This is based on the fact that both the front wheels of the vehicle V fluctuate substantially simultaneously in the bounding behavior, that is, the vertical acceleration of the respective accelerations fluctuates the most and the variation in the vertical acceleration waveform becomes the largest. ing.

  In addition, the vehicle behavior specifying unit 202, when the winker data is included in the situation data specified as the near-miss behavior or the like, shows a behavior indicating a sudden brake for stopping at the road shoulder (road shoulder stop behavior). (See FIG. 5). By specifying in this way, it is possible to further automatically subdivide and classify near-miss behaviors. In FIG. 5, the presence / absence of a winker is indicated by a vertical thin line at the bottom of the figure. Moreover, the dense part of the blinker signal is shown by hatching.

  And the vehicle behavior specific | specification part 202 links | relates the behavior specific data which is the specific result with the situation data which specified the vehicle behavior as mentioned above, and stores it in the analysis data storage part D2 (FIG. 4, step S5).

  The analysis data storage unit D2 stores behavior specifying data corresponding to the situation data together with the situation data specified by the vehicle behavior specifying unit 202, systematically classified for each vehicle behavior. Specifically, the analysis data storage unit D2 stores corresponding situation data and behavior specifying data in a storage folder set for each behavior such as a near-miss behavior, a climbing behavior, and a bound behavior. For example, in the analysis data storage unit, a behavior folder that stores situation data specified as a near-miss behavior, a behavior folder that stores situation data specified as a behavior such as riding, a situation specified as a bound behavior Bound behavior folders for storing data are set, and corresponding situation data and behavior specifying data are stored in each folder.

  Further, in the analysis data storage unit D2, a road shoulder stop behavior folder for further classifying and storing the road shoulder stop behavior from the near hat behavior and the like is stored in a subdivided or hierarchical manner.

  Of the situation data stored in the analysis data storage unit D2, only the situation data classified as a specific vehicle behavior is selected by the operator operating the input means such as a keyboard or a mouse, for example, a display or the like. To the output means. Or only the situation data classified into the specific vehicle behavior is similarly selected and transferred to another analysis device, a memory, or the like.

<Effect of this embodiment>
According to the vehicle behavior analysis system according to the present embodiment configured as described above, the standard deviation of the longitudinal acceleration waveform, the standard deviation of the lateral acceleration waveform, and the vertical acceleration waveform indicated by the longitudinal acceleration data, the lateral acceleration data, and the vertical acceleration data. The vehicle behavior can be specified by the magnitude relationship of the standard deviation. Therefore, the vehicle behavior indicated by the situation data can be specified without relying on visual observation, and not only the user's arbitrary judgment can be excluded but the vehicle behavior can be specified objectively, as well as the user's time and body. Mental and mental burden can be reduced. Further, since the vehicle behavior can be specified by the magnitude relationship among the standard deviation of the longitudinal acceleration waveform, the standard deviation of the left-right acceleration waveform, and the standard deviation of the vertical acceleration waveform without using a threshold value, a highly reliable identification result is obtained. be able to.

<Other modified embodiments>
The present invention is not limited to the above embodiment.

  For example, although each acceleration data is used in the embodiment, if the drive recorder 100 has a gyro sensor, each angular acceleration of roll angular acceleration, pitch angular acceleration, and yaw angular acceleration acting on the vehicle is used. The vehicle behavior may be specified using the feature amount of the data.

  Further, in the above-described embodiment, the analysis data storage unit has a storage folder set for each behavior such as near-miss behavior, behavior such as climbing, and bound behavior. I can't. For example, the vehicle behavior analysis device may further include a situation data extraction unit that extracts situation data indicating a predetermined vehicle behavior from the situation data stored in the analysis data storage unit. The situation data extraction unit is configured to extract the situation data indicating the predetermined vehicle behavior selected by the operation of the input means such as the keyboard and the mouse by the operator based on the behavior specifying data given to the situation data. You may do it.

  Further, in the embodiment, the situation data obtained by the drive recorder 100 is collected in the vehicle behavior analysis apparatus, and then the vehicle behavior indicated by the situation data is specified by the vehicle behavior analysis apparatus. Similarly, at 100, the vehicle behavior indicated by the situation data may be specified. For example, the drive recorder 100 includes a situation data receiving unit and a vehicle behavior specifying unit, and only the situation data specified as a predetermined behavior (for example, a near miss behavior) by the vehicle behavior specifying unit is stored in a normal area ( It may be configured to be transferred to a regular recording data storage unit) and recorded.

  In addition, situation data classified as a near-miss behavior may be subdivided using a predetermined threshold. For example, it can be considered that the maximum value of the vertical acceleration is subtracted from the maximum value of the triaxial combined acceleration obtained by the acceleration sensor and the threshold value.

  In addition, in order to further identify collision accidents in near-miss behaviors, the feature values of left-right acceleration data, vertical acceleration data, and longitudinal acceleration data depend on the frictional force between the road surface caused by normal braking and the ground contact surface of the tire. It is desirable to specify that the behavior is a value that is significantly larger than the feature amount.

  In addition, the vehicle behavior analysis device may link the situation data with the map information. Specifically, the vehicle behavior analysis apparatus links the map information and the situation data using the position data included in the situation data. Thus, road analysis can be performed by estimating road information (for example, road surface deterioration state) from situation data identified as bound behavior.

  In addition, using the variation mode of the longitudinal acceleration data, it is possible to classify the behavior in the case of a collision with another vehicle or a structure and the behavior collided with the other vehicle among the collision accidents. . In other words, in order to specify a rear-end collision or a rear-end collision accident in the behavior such as the near miss, it is desirable to specify by the sign of the characteristic amount of the lateral acceleration data, the vertical acceleration data, and the longitudinal acceleration data.

  Moreover, in addition to the above-described embodiment, the situation data indicating the behavior such as riding may be as follows using the blinker data included in the situation data. That is, it is possible to determine whether the vehicle is turning left or right based on the blinker data, and to specify whether the vehicle V has climbed on a curb or the like when turning left or right.

In addition, to determine whether you have climbed the curb on the right turn or on the curb on the left turn, you can determine the order in which the left and right tires climbed by looking at the order of the left and right acceleration waveforms without the blinker data. Therefore, it can be judged.
As a result, it becomes possible to easily analyze the driving tendency of the driver, such as being easy to ride on the curb when turning left or being easy to get on the curb when turning right.

  Further, since it is possible to determine the U-turn by looking at the waveform of the lateral acceleration, it is possible to determine, for example, whether an accident or a near-miss occurs when the U-turn is made.

  The feature amount of the acceleration data in the embodiment is the standard deviation of the acceleration waveform, but other variations of the acceleration waveform may be used, or an average value of the acceleration may be used. In addition, a calculation value indicating a magnitude relationship peculiar to each behavior may be used for each acceleration.

  In addition, it goes without saying that the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.

  According to the present invention having such a configuration, the vehicle behavior indicated by the situation data can be specified automatically and with high reliability without classifying the behavior of the vehicle indicated by the situation data only by visual observation.

Claims (7)

Among the situation data including each acceleration data of longitudinal acceleration, lateral acceleration and vertical acceleration acting on the vehicle, or situation data including each angular acceleration data of roll angular acceleration, pitch angular acceleration and yaw angular acceleration acting on the vehicle A situation data receiving unit that receives at least one of the situation data;
The magnitude relationship between the waveform variations of the longitudinal acceleration, the lateral acceleration, and the vertical acceleration data in the received situation data, or the waveform variation of the angular acceleration data of the roll angular acceleration, the pitch angular acceleration, and the yaw angular acceleration. , at least one of using Baratsu-out of the magnitude relationship of the waveform, and a vehicle behavior identification unit for identifying a vehicle behavior in which the status data indicates among,
The vehicle behavior specifying unit uses the magnitude relationship of the one waveform variation , the collision accident, a near-miss indicating a near-miss that leads to a collision accident, or a near-miss that indicates a sudden brake, and a one-wheel ride on a curb or a side groove A vehicle behavior analysis device that identifies a behavior such as a ride-on behavior indicating that the wheel has been removed and a bounce behavior that indicates a bounce caused by the passage of both wheels through unevenness on the roadway. The vehicle behavior specifying unit is
If the variation of the wave form of the lateral acceleration data and the acceleration data before and after than the variation of the waveform of the vertical acceleration data is large, the vehicle behavior in which the status data indicates identified as the near miss or the like behavior,
If the variation of the waveform of the lateral acceleration data than the variation of the longitudinal acceleration data and vertical acceleration data of the waveform is large, the vehicle behavior in which the status data indicates identified with the rides like behavior,
The vehicle behavior analysis device according to claim 1, wherein when the waveform variation of the vertical acceleration data is larger than the waveform variation of the longitudinal acceleration data and the lateral acceleration data, the vehicle behavior indicated by the situation data is identified as the bound behavior. . The situation data includes turn signal data indicating operation information of the turn signal of the vehicle,
3. The vehicle behavior identifying unit identifies behavior indicating a sudden brake for stopping near a road shoulder when the winker data is included in the situation data identified as the near-miss behavior or the like. The vehicle behavior analysis apparatus described. The vehicle behavior analysis apparatus according to claim 1, wherein the variation in the waveform of each acceleration data or each angular acceleration data is a standard deviation value of an acceleration waveform indicated by each acceleration data or each angular acceleration data. Among the situation data including each acceleration data of longitudinal acceleration, lateral acceleration and vertical acceleration acting on the vehicle, or situation data including each angular acceleration data of roll angular acceleration, pitch angular acceleration and yaw angular acceleration acting on the vehicle A situation data receiving unit that receives at least one of the situation data;
The magnitude relationship between the waveform variations of the longitudinal acceleration, the lateral acceleration, and the vertical acceleration data in the received situation data, or the waveform variation of the angular acceleration data of the roll angular acceleration, the pitch angular acceleration, and the yaw angular acceleration. is at least one of using Baratsu-out of the magnitude relationship of the waveform, which equip the vehicle behavior identification unit for identifying a vehicle behavior in which the status data indicates, the function as the computer of,
The vehicle behavior specifying unit uses the magnitude relationship of the one waveform variation , the collision accident, a near-miss indicating a near-miss that leads to a collision accident, or a near-miss that indicates a sudden brake, and a one-wheel ride on a curb or a side groove A vehicle behavior analysis program characterized by specifying as one of a behavior such as a ride-on behavior indicating that the wheel is removed and a bounce behavior indicating a bounce caused by the passage of both wheels through unevenness on the roadway.   Situation data including acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration acting on the vehicle Or roll angular acceleration, pitch angular acceleration and yaw angular acceleration applied to the vehicle. Identify vehicle behavior indicated by at least one of the situation data including speed data A vehicle behavior identification method for
  Waves of acceleration data of longitudinal acceleration, lateral acceleration, and vertical acceleration in the situation data Variations in shape variation, or each of roll angular acceleration, pitch angular acceleration, and yaw angular acceleration The relationship between the variations in the waveform of the angular acceleration data, Using a magnitude relationship, a collision accident, a near-miss or other sudden braking that leads to a collision accident Near-miss behavior that indicates, and one-wheel climbing on curbstones or climbing that indicates wheel-removal to side grooves, etc. And bounce behavior that shows the bounce that occurs when both wheels pass through the unevenness on the roadway The vehicle behavior analysis method characterized by specifying as either of these. Among the situation data including each acceleration data of longitudinal acceleration, lateral acceleration and vertical acceleration acting on the vehicle, or situation data including each angular acceleration data of roll angular acceleration, pitch angular acceleration and yaw angular acceleration acting on the vehicle A situation data receiving unit that receives at least one of the situation data;
The magnitude relationship between the waveform variations of the longitudinal acceleration, the lateral acceleration, and the vertical acceleration data in the received situation data, or the waveform variation of the angular acceleration data of the roll angular acceleration, the pitch angular acceleration, and the yaw angular acceleration. , at least one of using Baratsu-out of the magnitude relationship of the waveform, and a vehicle behavior identification unit for identifying a vehicle behavior in which the status data indicates among,
The vehicle behavior specifying unit uses the magnitude relationship of the one waveform variation , the collision accident, a near-miss indicating a near-miss that leads to a collision accident, or a near-miss that indicates a sudden brake, and a one-wheel ride on a curb or a side groove A drive recorder that identifies either a behavior such as riding up that indicates a derailment of the vehicle or a bouncing behavior that indicates a bounce that occurs when both wheels pass through the unevenness on the roadway.

Images