KR20060053662A - Accident memory device of vehicle and method thereof and method of accident memory analysis - Google Patents

Abstract

The present invention is equipped with a removable memory in the data collection device for collecting the driving history, if the accident occurrence is detected from the impact amount from the signal of the acceleration sensor during driving, and stores the data on the accident occurrence in the removable memory, removable data mounted on the data collection device After the memory was removed, it was installed in a personal communication terminal or personal PC, and the remote diagnosis center was connected for communication so that the analysis of accident record data could be requested.

The remote diagnosis center analyzes the vehicle trajectory and driver's driving intention from the transmitted accident record data to track the cause of the accident, and stores the analyzed result in a database to use as a reference for disputes between the offender and the victim. In other words, it can be used as a data to solve the problems of the chronic vehicle design.

Accident record, impact amount, telemetry center, driving trajectory, coordinate transformation

Description

Accident recorder and method of vehicle and analysis method of accident record {ACCIDENT MEMORY DEVICE OF VEHICLE AND METHOD THEREOF AND METHOD OF ACCIDENT MEMORY ANALYSIS}

1 is a view showing an accident recording apparatus of a vehicle according to the present invention.

2 is a flowchart of an embodiment of executing an accident detection recording in an accident recording apparatus of a vehicle according to the present invention;

3 is a flow diagram of one embodiment for performing an analysis of an incident record at a remote diagnostic center in accordance with the present invention.

4 is a view showing the direction of the force applied to the roll error correction in the accident record analysis according to the present invention.

5 is a view showing the direction of the force applied to the pitching error correction in the accident record analysis according to the present invention.

6 is a view applied to the pitching angle calculation in the accident record analysis according to the present invention.

Figure 7 is applied to the coordinate transformation in the accident record analysis according to the present invention.

8 is a view applied to the trajectory calculation in the accident record analysis according to the present invention.

The present invention relates to an accident recording apparatus of a vehicle, and more particularly, automatically detects an accident occurrence using an impact amount and stores the accident in a mobile memory, and installs the mobile memory in a personal communication terminal or a personal PC and then communicates with a remote diagnosis center. The present invention relates to a vehicle accident recorder and a method and an accident record analysis method for requesting analysis of accident record data, and tracking a cause of an accident by analyzing a vehicle trajectory and driving intention of a driver.

According to the statistics of traffic accidents, an average of about 300,000 traffic accidents occur each year, causing about 10,000 deaths and about 300,000 injuries.

The main causes of such traffic accidents are the failure of safe driving such as central line violation, signal violation, speeding driving, drunk driving, and drowsy driving.

Human and property damages resulting from such accidents and disputes between the offender and the victim are also becoming social problems.

Accordingly, the vehicle is provided with various devices for detecting the deceleration occurring while driving to determine whether an accident has occurred and accordingly to protect the driver and the occupant, and record the driving history of the vehicle at the time of the accident to generate an accident It is equipped with a recording device that can be used as data for analysis.

The accident history recording device mounted on the vehicle is equipped with an acceleration sensor having a predetermined value in front of the vehicle, typically a 50G acceleration sensor, and when a constant acceleration value, that is, an acceleration value of 50G or more is detected from the acceleration sensor, an accident occurs. It recognizes and records the driving history for that point so that it can be used as a data for future accident analysis.

In addition, the accident occurrence is tracked by recording the vehicle operation information and the front image information of the driver and analyzing the vehicle operation information and the image information of the driver recorded when the accident occurs.

In addition, when an accident is detected, a method of automatically notifying a remote place of data of an accident occurrence time stored in an accident history recording apparatus using a CDMA terminal mounted in a vehicle has been applied.

However, when a large amount of accident occurrence information is transmitted to a remote location using a CDMA terminal, a large amount of time is required for data transmission, resulting in an excessive expense of communication fee.

In addition, when a crash is reported using a CDMA terminal fixedly mounted in a vehicle, there is a possibility that the CDMA terminal may be damaged at the time of the accident, and when the damage occurs, the CDMA terminal may not be notified of the accident. have.

In addition, when a large capacity memory is installed in order to record accident information in a vehicle, there is a problem of causing an increase in production cost due to the memory price.

The present invention has been invented to solve the above problems, the object of which is to install a removable memory in the data collection device for collecting the driving history, if the accident occurrence is detected from the impact amount while driving the data about the accident occurrence in the removable memory It was to be saved.

In addition, the removable memory mounted on the data collection device is removed and then mounted on a personal communication terminal or personal PC to connect to the remote diagnostic center to request the analysis of accident record data.

In addition, the remote diagnosis center analyzes the vehicle trajectory and driver's driving intention from the transmitted accident record data to track the cause of the accident, and stores the analyzed result in a database for reference to disputes between the offender and the victim. It can be used as a material to solve the problems of the chronic vehicle design.

The present invention for realizing the above object is connected to a variety of controllers mounted on the vehicle in the can communication line and K-communication line data collection device for collecting and storing the self-diagnostic data and control data; An acceleration sensor mounted on the vehicle body and transmitting information on the amount of impact generated during a collision to the data collection device; A removable memory detachable from the data collection device and configured to store a history of the accident and information immediately before and immediately after the accident; A personal communication terminal that connects and detaches a removable memory from the data collection device and transmits the stored accident occurrence history and data about immediately before and after the accident to a remote analysis center by wireless communication to request an analysis; The accident record comprising a personal PC which is connected to the removable memory from the data collection device, and transmits the stored accident occurrence history and data about immediately before and after the accident to a remote analysis center through internet communication to request an analysis. Provide the device.

In addition, the present invention comprises the steps of determining whether the impact amount of the reference value or more is detected by reading information of the acceleration sensor while driving; Determining that an accident occurs when an impact amount equal to or greater than a reference value is detected, extracting only a vehicle vibration frequency component from a vehicle behavior frequency, a vehicle vibration frequency, and a noise frequency present in a signal of an acceleration sensor, and sampling at a predetermined time interval; Storing the accident occurrence data detected by the sampling in a temporary memory; Extracting vehicle control information data before and after the occurrence of the accident when a predetermined time elapses after the occurrence of the accident; Storing the accident occurrence data and vehicle control information data before and after the occurrence of an accident in a removable memory; Separating the removable memory from the data collection device equipped with a personal communication terminal and a PC to transmit the information stored in the portable memory to the remote analysis center requesting the analysis of the accident record information comprising the step of Provide a method of recording.

The present invention also provides a method of determining whether an analysis of an accident record is requested to a remote analysis center through wireless communication or Internet communication; If the analysis of the accident record is requested, converting the progress direction and lateral acceleration signal values into digital signals and then filtering them; Correcting a zero level offset value between the longitudinal acceleration sensor value, the lateral acceleration sensor value, and the directional angular velocity sensor value; Correcting the roll error, bank running error and pitching error; Correcting the noise of the sensor and the sensitivity of the sensor; Converting coordinates from the data extracted through the correction and calculating a driving trajectory; Extracting and storing the accident type and the behavior of the vehicle in a database by analyzing the calculated driving trajectory; It provides a vehicle accident record analysis method comprising the step of providing to the driver the results of the analyzed accident type and the results of the behavior of the vehicle.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

As shown in FIG. 1, an accident recording apparatus for a vehicle according to the present invention includes a data collecting device 100, an acceleration sensor 200, a can communication line 300, a K communication line 400, and a removable memory 500. , A personal communication terminal 600, a personal PC 700, and a remote diagnostic center 800 are configured.

The data collection device 100 is equipped with a removable memory 200 made of an SD card, connected to a controller for controlling an engine, a transmission, and a chassis electric field through a can communication line 300, and a chassis through a K communication line 400. It is connected to the controller of the battlefield and body battlefield, and collects and stores various data provided by each controller.

In addition, the data collection device 100 is connected to the acceleration sensor 200, when the shock amount is detected from the acceleration sensor 200 or more than the set amount of the deceleration is determined that the accident occurs, the removable memory 500 is mounted to the history of the accident occurrence Store in

The acceleration sensor 200 is preferably mounted on the vehicle body so that the amount of impact during a collision can be efficiently transmitted to the data collection device 100.

The personal communication terminal 400 is a Bluetooth phone or a PCS phone, and connects the removable memory 500 to the data collection device 100 and then accesses the remote diagnosis center 800 by wireless communication to analyze the recorded accident data. request.

Personal PC 700 is connected to the removable memory 500 from the data collection device 100 and then connected to the remote diagnosis center 800 over the Internet to request the analysis of the recorded accident data, the remote diagnosis center 800 Receive and display the analyzed result information provided by).

Referring to the operation of the automatic storage of the accident record and the analysis using the remote diagnosis center 800 by using the present invention comprising the function as described above are as follows.

In the state where the removable memory 500 is mounted on the data collection device 100, the data collection device 100 detects power-on, and then, via can communication and K-line communication, the engine controller, the transmission controller, the chassis controller, and the body length controller. It collects and stores various self-diagnosis data in communication with the back.

In the process of collecting and storing the self-diagnosis data as described above, information of the acceleration sensor 200 is read (S102), and it is determined whether an impact amount equal to or greater than the set deceleration is detected (S103).

The detection of the impact amount is calculated by multiplying the vehicle mass by the acceleration and setting the amount of impact generated when the vehicle collides directly with the fixed object at a speed of 7 MPH.

If an impact amount equal to or greater than the deceleration set in the determination of S103 is detected, it is determined that an accident occurs, and only components of the vehicle vibration frequency from the vehicle behavior frequency, the vehicle vibration frequency, and the noise frequency included in the information of the acceleration sensor 200 through filtering. It extracts (S104).

Then, since the vehicle body vibration is terminated around 100msec at the time of collision of the vehicle, in order to extract the acceleration value, the sampling process is performed in a unit of 2msec minimum component of the extracted body vibration frequency (S105).

Incident occurrence data detected and detected as described above is stored in the temporary memory (S106), and it is determined whether a predetermined time, preferably 3 seconds has elapsed after the detection of the accident (S107).

If the predetermined time after the accident detection has not elapsed, the collection and storage of the accident data is repeated, and if it is determined that the predetermined time elapses, the data collection device 100 for a predetermined time before the accident, preferably The data of the engine controller, the transmission controller, the chassis controller, and the body length controller collected by CAN communication and K-line communication for 10 seconds before the occurrence of the accident are extracted (S108).

Then, the data of the engine controller, the transmission controller, the chassis controller, and the body length controller collected by CAN communication and K-line communication for a predetermined time after the accident occurrence time, preferably 3 seconds after the accident occurrence time are extracted ( S109).

As described above, when data before the time of accident occurrence and data extraction after the occurrence of the accident are completed, the accident occurrence data stored in the temporary memory is accessed and the removable memory together with the extracted data about a predetermined time based on the time of the accident occurrence. 500) (S110).

The data before and after the occurrence of the accident collected by the CAN communication and K-line communication is the longitudinal acceleration (2G), lateral acceleration (2G), longitudinal acceleration (50G), lateral acceleration (50G), engine speed (RPM), vehicle speed, It contains information such as steering angle, contact of inhibit switch, contact of brake switch, headlight switch contact, turn signal switch contact, seat belt switch and so on.

When the data on the occurrence of the accident is stored in the removable memory 500 as described above, the driver or accident handling personnel separates the removable memory 500 mounted on the data collection device 100 and then personal communication terminal 600 or personal Mounted on the PC 700 (S111), and connected to the remote analysis center 800 by wireless or Internet communication and transmits the collected accident occurrence data to request the analysis (S112).

The accident analysis analysis algorithm, data graphic, vehicle trajectory reproduction algorithm on the two-dimensional plane, report preparation algorithm are set, and the remote analysis center 800 where the database is installed requires analysis of the accident record data by wireless or internet communication. It is determined whether is detected (S201).

When the analysis request of the accident record data is detected in the determination of S201, the sensor signal value of the record data, that is, the traveling direction, the lateral acceleration signal, the yaw rate signal, and the like are converted into a digital signal (S202).

Thereafter, unit conversion is performed on the sensor signal value converted into the digital signal by applying the following equation (S203).

X = (X _{A / D} -X _offset ) × (5/555) × (Sensor sensitivity) × (Converted value of each unit)

Thus, acceleration a _x = [G _x 5-2.5 x (5/555)] x (5/555) x (1 / 0.4) x 9.8 (m / sec ² )

= [G _x 50-2.5 × (5/555)] × (5/555) × (1 / 0.04) × 9.8 (m / sec ² )

In addition, acceleration a _y is also the same as said Formula, and yaw rate r is as follows.

r = [r-2.5 × (5/555)] × (5/555) × (1 / 0.02) × (π / 180) [rad / sec]

When unit conversion is completed through the application of Equation 1, filtering is performed on the corresponding values (S204), and zero-level correction and error correction are performed on the filtered values (S205).

The offset value is set so that the integral value of the longitudinal acceleration is close to the speed of the transmission. The output of the acceleration sensor and the speed of the transmission have the following relationship.

From the above equation

In the above equation, X _{A / D} is the digital conversion of the sensor output, Offset is the sensor zero offset, S is the scalar factor, and V _TM is the speed of the transmission.

In the case of equation (1), if the speed of transmission is updated every second, discontinuity occurs and accuracy is not provided in the calculation.However, equation (2) and equation (3) should obtain the same result theoretically. Errors occur in numerical integration, resulting in slightly different results.

For example, using the formula (2) to calculate the offset

In the above equation, the ± sign is a problem of the sensor mounting direction and is negative if the output is positive during acceleration and positive if it is negative output during acceleration.

In addition, zero level offset correction of the yellow acceleration sensor is performed as follows.

This is corrected by using the transmission speed and steering angle sensor. The following relationship is established between lateral acceleration, speed and steering angle in normal turning.

Therefore, the lateral acceleration offset is calculated by the following equation.

In the above formula, A is a stabilization factor value (0.0003), L is a wheelbase, S is a scalar factor, and δ is a steering angle.

In the above equation, the ± sign is a problem in the sensor mounting direction, and is positive when the output is positive in the right direction, and positive when the output is negative in the right direction.

In addition, the zero level offset correction of the directional angular velocity sensor is performed as follows.

The offset is set to match the zero point of the direction angular velocity sensor estimated using the speed of the transmission and the steering angle sensor.

For normal turning, the following relationship holds between yaw rate and speed steering angle.

Therefore, the yaw rate offset is as follows.

In the above formula, A is a stabilization factor value (0.0003), L is a wheelbase, S is a scalar factor, and δ is a steering angle.

In the above equation, the ± sign is a problem in the sensor mounting direction, and is positive when the output is positive in the right direction, and positive when the output is negative in the right direction.

When the zero level correction and the error correction are completed through the above process, the roll error correction is performed to solve the problem of sensor measurement and calculation occurring in the process of assuming the three-dimensional vehicle motion as the two-dimensional plane (S206). ).

Lateral acceleration is affected by the force of gravity when a roll or a vehicle generated when the vehicle is turning or the vehicle is traveling on a road surface (bank driving) inclined laterally.

Therefore, in the locus reproduction logic, such acceleration is reproduced as if it is a lateral slip, so correction for this is required.

Referring to FIG. 4, the following relationship is established between the measured lateral acceleration and the actual acceleration, the measured yaw rate and the actual yaw rate.

Therefore, the roll angle in the above formula

이며,

Is,

The actual lateral acceleration and the actual yaw rate are determined as follows.

In the above, G _yS is the measurement acceleration, G _y is the actual acceleration, r _S is the measurement yaw rate, r is the actual rate, θ is the roll angle, G is the gravitational acceleration, V is the speed in the direction of travel do.

In the case where the lateral acceleration occurs in the state where there is almost no yaw rate, the lateral acceleration is set to 0 in this case because the lateral acceleration occurs.

It is necessary to distinguish between the change in the lateral acceleration due to the roll and the change in the lateral acceleration due to the actual lateral slip or rotation of the vehicle.

In the case of side slip or spin, the lateral acceleration is smaller than the acceleration generated by the centripetal force during rotation.

Therefore, the calculation by the above equation produces (-) rolls at the (+) rate for rolls, and (+) rolls at (+) yaw rate for side slip or spin. In this case, cancel the roll error correction and use the measured acceleration as the true value.

When the correction of the roll error is completed through the above process, pitching occurs in the vehicle during rapid acceleration or sudden braking, and in this case, the influence of gravity acts on the longitudinal acceleration sensor. It is calculated faster or slower than speed.

Therefore, in order to solve this problem, a simple model of a vehicle is set as shown in FIG. 5, and a dynamic equation is derived based on this, and the pitching angle of the vehicle is calculated using FIG. 6 to correct the error (S207).

The relationship between the actual measured acceleration and the actual acceleration is as follows.

In the above formula, G _xs is the measurement acceleration, Gx is the actual acceleration, θq is the pitching angle, and G is the gravity acceleration.

Therefore, if a dynamic equation is established using the above-described FIG. 5 and converted into a state equation, it is derived as in the following equation.

In the above formula, M is the vehicle mass, l _y is the vehicle pitching moment at inertia, and G _x is the vehicle longitudinal acceleration.

The parameters used above are nonlinear and have a problem that measurement is difficult.

Therefore, the following general quadratic model is established and the parameter is extracted using the experimental data as shown in the following equation.

The parameters used for A and B are calculated using the parameter estimating technique from the test data.

When the pitching error is compensated by applying the above procedure, the noise on the signal of the sensor is firstly removed using a hardware filter, and then the software is secondarily removed by applying a software filter.

In the case of the acceleration sensor, the unit output for the input is 200mV per 1G, and the yaw rate sensor has the sensitivity of the sensor having a nominal value of 20mV per 1deg / sec, but an error occurs in the value. In operation S208, the error error is compensated.

Typically, sensors provide a self-diagnosis terminal, which calculates the sensitivity of the sensor by outputting a constant voltage.

Thereafter, coordinate transformation is performed on the data of each detected sensor (S209).

For example, as shown in FIG. 7, when the angle of the interior frame XIYI and the body frames xB and yB is q, a vector represented by the body frame and a frame represented by the interframe are shown. The following equation holds.

Therefore, the acceleration of the vehicle and the acceleration of the body frame in the inter-frame are expressed by the following equation.

The coordinate system of the vehicle applies the SAE protocol. In FIG. 7 and equations, XI and YI denote interior frames, xB and yB denote body frames, and ax denotes longitudinal acceleration in body frames. Ay stands for Lateral Acceleration in Body Frame, AX stands for X-directional Acceleration in Inertial Frame, and AY stands for Interior Frame Is the Y-directional Acceleration in Inertial Frame, vx is the Longitudinal Velocity in Body Frame, and vy is the Lateral Velocity in Body Frame. VX stands for X-directional Velocity in Inertial Frame, and VY stands for Y-directiona in Interior Frame. Velocity in Inertial Frame where r is Yaw Velocity, q is yaw angle and d is steering angle.

When the coordinate transformation is completed for each sensor data by applying the above equation, the driving trajectory of the vehicle is calculated (S210).

In calculating the travel trajectory, the yaw angle is extracted by applying the following equation.

In addition, the acceleration in the interior frame is extracted by applying the following equation,

By integrating the acceleration in the interior frame, the velocity in the interior frame is extracted as in the following equation.

When the speed in the interior frame is extracted by applying the above equation, the vehicle position in the interior frame is extracted by applying the following equation.

In each equation for driving trajectory calculation, the initial speed Vx [0] uses the speed of the transmission, and V _y [0], q [0], P _x [0], and P _y [0] are all ' It is set to 0 '.

In addition, in order to describe the motion of the vehicle on the plane, the acceleration of the X and Y axes and the yaw rate of the Z axis are required.

Therefore, in order to calculate the driving trajectory for the accident record, the yaw rate sensor is generally used, but the yaw rate sensor acts as a factor to increase the manufacturing cost of the vehicle as an expensive component.

Therefore, the present invention applies a method for calculating the vehicle driving trajectory using only the information of the acceleration sensor for cost reduction.

This assumes that the vehicle behaves so as to ignore side slips in the two-dimensional plane as shown in FIG. 8.

In the figure, it is assumed that the tangential direction of the vehicle is t, the vertical direction is n, and the side slip is ignored, so that the direction angle of the vehicle coincides with the tangential direction.

When the acceleration of the vehicle is represented by the t-axis and the n-axis, the acceleration vector of the vehicle is expressed by the following equation.

Since the side slip is assumed to be ignored, the direction angle of the vehicle is equal to the rotation angle θ of the vehicle.

Assuming that the rotational angular velocity, which is a time derivative of the rotational angle, is ω, the following relation holds.

Since V is the connection speed in the above equation, the yaw rate can be extracted through integration of the tangential acceleration as in the following equation.

According to the above procedure, when the driving trajectory of the vehicle is extracted, the type of accident occurrence is extracted by analyzing it (S211), and data about the extracted accident occurrence is stored in a database (S212).

In addition, by providing information on the type of accident occurrence stored in the database according to the driver's request, it can be used as evidence in the dispute about the accident (S213).

In addition, through the analysis of the type of accident occurrence stored in the database is used as a data to solve the problem of the vehicle design.

As described above, the present invention stores various state information of the vehicle and data on the occurrence of an accident in a large capacity mobile memory, and mounts the portable memory to a Bluetooth phone or a personal PC to remotely record the accident record data. By transmitting to the analysis center for scientific cause identification, clear evidence can be used in disputes about traffic accidents, providing convenience and reliability in analyzing accident records.

Claims (6)

A data collection device connected to various controllers mounted on the vehicle through the CAN communication line and the K-communication line to collect and store self-diagnostic data and control data; An acceleration sensor mounted on the vehicle body and transmitting information on the amount of impact generated during a collision to the data collection device; A removable memory detachable from the data collection device and configured to store a history of the accident and information immediately before and immediately after the accident; A personal communication terminal that connects and detaches a removable memory from the data collection device and transmits the stored accident occurrence history and data about immediately before and after the accident to a remote analysis center by wireless communication to request an analysis; The accident record comprising a personal PC which is connected to the removable memory from the data collection device, and transmits the stored accident occurrence history and data about immediately before and after the accident to a remote analysis center through internet communication to request an analysis. Device. The method of claim 1, The personal communication terminal and the personal PC are accident recording apparatus of the vehicle, characterized in that receiving a result of the analysis of the accident record data in a conversation form with the remote diagnostic center. Determining whether an impact amount equal to or greater than a reference value is detected by reading information of the acceleration sensor while driving; Determining that an accident occurs when an impact amount equal to or greater than a reference value is detected, extracting only a vehicle vibration frequency component from a vehicle behavior frequency, a vehicle vibration frequency, and a noise frequency present in a signal of an acceleration sensor, and sampling at a predetermined time interval; Storing the accident occurrence data detected by the sampling in a temporary memory; Extracting vehicle control information data before and after the occurrence of the accident when a predetermined time elapses after the occurrence of the accident; Storing the accident occurrence data and vehicle control information data before and after the occurrence of an accident in a removable memory; Separating the removable memory from the data collection device equipped with a personal communication terminal and a PC to transmit the information stored in the portable memory to the remote analysis center requesting the analysis of the accident record information comprising the step of Recording method. The method of claim 3, The vehicle control information data before the accident occurs to extract the vehicle control information data for 10 seconds before the accident, and the vehicle control information data after the accident occurs to extract the vehicle control information for 3 seconds after the accident How to record an accident on a vehicle The method of claim 3, The vehicle control information data before and after the accident is 2G, 50G longitudinal acceleration, 2G, 50G lateral acceleration, engine speed, vehicle speed, steering angle, inhibitor switch contact, brake switch contact, headlight switch, turn signal switch contact And the seat belt switch contact information. Determining whether an analysis of an accident record is requested to a remote analysis center through wireless communication or Internet communication; If the analysis of the accident record is requested, converting the progress direction and lateral acceleration signal values into digital signals and then filtering them; Correcting a zero level offset value between the longitudinal acceleration sensor value, the lateral acceleration sensor value, and the directional angular velocity sensor value; Correcting the roll error, bank running error and pitching error; Correcting the noise of the sensor and the sensitivity of the sensor; Converting coordinates from the data extracted through the correction and calculating a driving trajectory; Extracting and storing the accident type and the behavior of the vehicle in a database by analyzing the calculated driving trajectory; And providing the driver with the result of the analyzed accident type and the result of the behavior of the vehicle.

Images