KR101738414B1 - Apparatus for detecting vehicle accident and emergency call system using the same - Google Patents

Abstract

When an accident detection device installed in the vehicle recognizes a traffic accident, the accident data is automatically transmitted to the emergency call service control center. The accident detection apparatus tracks the three-dimensional motion of the vehicle, calculates the three-dimensional motion prediction range of the vehicle for a predetermined time from the current point of time, and determines that the actual three-dimensional motion tracking value is out of the prediction range. The accident data includes the above-mentioned vehicle position information, three-dimensional motion information, and image information.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a vehicle accident detection device,

The present invention relates to a vehicle accident detection apparatus and an emergency call system for automatically requesting a control center or receiving an accident when a traffic accident occurs using the apparatus.

In Europe, legislation on the Emergency Call system, called Ecall, is being promoted, and vehicles manufactured from March 2018 are expected to be equipped with an emergency call system.

The emergency call system judges that an accident has occurred when the driver directly presses the save button or develops the airbag as an example, and sends information necessary for acceptance of the accident together with the present location of the vehicle to the control center.

The driver may not be able to press the rescue button for reasons such as losing consciousness or the air bag may not be deployed properly. This situation may be caused by an emergency call system There is a problem that it can not.

An object of the present invention is to provide a vehicle accident detecting device and an emergency call system for accurately determining the occurrence of an accident and providing assistance to a driver in the event of a traffic accident.

Other objects of the present invention will be understood from the following description.

As mentioned in the Background section above, the occurrence of an accident needs to be precisely judged in order to provide the driver with the necessary assistance in an emergency. Situations where there are special events such as airbag deployment or excessive shock detection are easily diagnosed. However, in the absence of such an event, it is not easy to determine whether or not an accident occurred.

In the event of a traffic accident, a sudden stop, a sudden turn, etc. are usually accompanied, and various vehicle state values, such as acceleration, steering angle, roll, pitch, yaw angle, etc., . Therefore, it is possible to determine whether or not a vehicle accident has occurred based on the instant change rate of the vehicle state value.

If the instantaneous rate of change is very large above the threshold value, it is easy to make an accident determination. However, the types of traffic accidents are so diverse that it is not easy to determine whether or not an accident is occurring with uniform or several criteria. Assuming multiple accident situations, we can derive an accident decision algorithm that can be applied in those situations, but there are too many cases to consider and can not cover all situations.

Although the above-mentioned special event, a situation in which the instantaneous rate of change is very large beyond a threshold value, an accident determination algorithm capable of coping with a typical vehicle accident scenario and the like are not complete, in order to accurately and quickly determine a vehicle accident, Can be partially employed.

One of the main features of the present invention is to track the three-dimensional behavior of a vehicle and to judge an accident when it exhibits expected behavior and abnormal behavior. More specifically, the present invention tracks a three-dimensional motion of a vehicle to calculate a three-dimensional motion prediction range of the vehicle for a certain period of time from the current point of time, and when the actual three-dimensional motion tracking value is out of the prediction range, can do.

The method of predicting the three-dimensional motion of the vehicle based on the past three-dimensional motion and judging the accident when it is out of the prediction provides an opportunity to learn various vehicle accident situations and to upgrade the prediction model therefrom. This prediction model can be set strictly so that the risk of an accident is judged to be high in the past behavior of the vehicle, and an accidental judgment can be made even in a slightly more abnormal behavior of the vehicle.

The above prediction model predicts vehicle behavior for a certain period of time including errors. Therefore, the probability of misjudgment is smaller than that of all the accident situations based on the instantaneous rate of change. As mentioned above, there is no risk that an apparent accident situation will not be judged as an accident, since the above-mentioned special event, instantaneous change rate exceeding a threshold value can be partly employed in the accident detection apparatus according to the present invention.

The judgment data of the above-mentioned abnormal behavior of the vehicle is transmitted to the Ecall control center. Using this data, the control center can generate a 3D simulation image to check the abnormal behavior of the vehicle again, and can develop the Ecall service according to the severity of the accident that occurred.

Preferably, vehicle periphery image information before and after the depression time is also transmitted to the Ecall control center. The monitoring staff of the control center can check the received image information to accurately determine whether or not an accident occurred, and can respond promptly and appropriately to the accident.

A vehicle accident detecting apparatus according to the present invention includes: a GPS receiving unit for receiving position information of a vehicle; A collection unit for collecting vehicle state information propagated through an in-vehicle network; A sensor unit for sensing three-dimensional motion of the vehicle; An image generating unit for generating image information of the surroundings of the vehicle; An accident analyzing unit for determining whether a vehicle accident has occurred using information about the position, the state, and the three-dimensional motion of the vehicle; And a wireless communication unit for communication with an external device of the vehicle.

The above three-dimensional motion information may include 3D simulatable values including x, y, z, roll, pitch, and yaw angles. As is well known, x, y, and z correspond to orthogonal coordinate axes associated with the vehicle's direction of travel, and roll, pitch, and yaw are related to the tilted posture of the vehicle. A nine-axis sensor including acceleration, angular velocity, and geomagnetic sensor may be used for detecting three-dimensional motion information.

The accident analyzing unit tracks the three-dimensional motion of the vehicle every first period, and based on the past three-dimensional motion tracking data every second period, from the first predicted time point T to a future point in time T + Dimensional motion of the vehicle up to the first predicted range of the three-dimensional motion tracking value of the vehicle), and may determine that an accident is caused when the actual three-dimensional motion tracking value is out of the first predicted range.

Further, the second period is synchronized with the position information reception period of the GPS reception unit, and the accident analysis unit can correct the vehicle position at the time point T arriving every second period using the position information from the GPS reception unit.

In addition, the accident analysis unit may calculate the three-dimensional motion tracking values of the previous vehicle at a second predicted time point T '(T' is any point between the second period coming from the time point T and subsequently) T '+? T' (? T 'is a positive value greater than the first period and less than or equal to the second period) from the second predicted time point T' on the basis of the three- Dimensional motion of the three-dimensional motion of the object to be inspected may be calculated and an accident may be determined when the actual three-dimensional motion tracking value is out of the second prediction range.

By calculating the second predicted range at some point in the middle of the first predicted range in which vehicle behavior can be monitored based on the first predicted range as described above, A situation in which the prediction model is not prepared at the time point of T +? T when the first prediction range ends can be prevented.

In addition, the calculation of the second predicted range as described above prevents a vehicle accident from being misjudged by a momentary change in the vehicle state value. Frequent accident misjudgment is a problem for the operation of an emergency call system with numerous vehicles.

Also, the first prediction range includes three-dimensional spatial values set on the basis of a vector having a position at a first prediction time T as a starting point and a position at a future time T + t as an end point, Is the direction of movement of the shortest distance of the vehicle during Delta t, the size of the vector is the shortest moving distance of the vehicle during Delta t, and spatial values may be converged at the starting point and the end point, respectively.
According to the embodiment, the first prediction range may include three-dimensional space values set based on a vector having a position at a first prediction time T as a starting point and a position at a future time T + t as an end point, The second prediction range may include three-dimensional space values set based on a vector having a position at a second prediction time T 'as a start point and a position at a future time T' +? T 'as an end point.
Also, the three-dimensional space values of the first and second prediction ranges may be varied according to the risk based on the actual three-dimensional motion tracking value.

An emergency call system according to the present invention includes: a vehicle accident detection device having the above-described features; And a control center server for holding information on at least one of a vehicle equipped with the vehicle accident detection device and a driver of the vehicle.

When the accident is detected, the vehicle accident sensor transmits accident data for a certain period of time before and after the accident to the control center server. The accident data includes the position information of the vehicle, the three-dimensional motion information and the image information, , The three-dimensional motion information, and the image information may include synchronized or synchronizable time information.

The vehicle accident according to the present invention as described above can be accurately determined and the driver can be provided with assistance promptly and appropriately in accordance with the accident situation.

Further, the vehicle accident detection apparatus according to the present invention can utilize vehicle state information propagated through an internal network of the vehicle, and can independently perform an accident judgment, thereby coping with various accident situations.

In addition, the vehicle accident detection device and the emergency call system according to the present invention can be applied to a driver or an enterprise by analyzing an operation pattern to provide information for safe driving or reduction of fuel cost, And can be used for providing information.

1 is a schematic configuration diagram of a vehicle accident detection apparatus according to an embodiment of the present invention;
FIG. 2 is a schematic configuration diagram of an emergency call system according to an embodiment of the present invention;
3 to 5 are views showing an accident analysis model of a vehicle accident detection apparatus according to an embodiment of the present invention;
6 is a diagram schematically illustrating an example of a 3D simulation image generated using three-dimensional motion information of a vehicle according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to the accompanying drawings. In the drawings, the same components or parts are denoted by the same reference numerals as much as possible for convenience of description.

Referring to FIG. 1, an emergency call system according to an embodiment includes an accident detection device 10 installed in a vehicle 100 and a control center server 200.

When the accident data is received from the accident detection device 10 in the control center server 200, the monitor agent of the control center confirms the accident data and sends an emergency call service suitable for the accident situation to the vehicle registered in the control center server 200 (100) driver.

The control center server 200 holds registration information including at least any one of the vehicle 100 equipped with the accident detection device 10 and the driver of the vehicle 100. [ The accident data includes an identification code for identifying the vehicle or the driver, for example, the unique number of the accident detection device 10. [

2, the vehicle accident detecting apparatus 10 according to the embodiment includes an accident analyzing unit 11, a collecting unit 12, a sensor unit 13, a GPS receiving unit 15, an image generating unit 14, And a wireless communication unit 16.

The accident analysis unit 11 includes a control unit 11a and a storage unit 11b for processing data. The storage unit 11b may include a system memory and storage for storing data. Various kinds of raw data and processed data received by the accident analyzing unit 11 are stored in the storage.

The collecting unit 12 is connected to an in-vehicle network, such as CAN communication, and collects OBD (On Board Diagnosis) information of the vehicle. Most of the recently released domestic and international vehicles are equipped with an OBD-II port and can be connected to this port to collect various vehicle status information such as vehicle speed, RPM, steering wheel steering angle, acceleration and deceleration information, and deployment of airbags.

The sensor unit 13 uses a 9-axis sensor having acceleration, angular velocity and geomagnetic sensor for more accurate three-dimensional motion tracking of the vehicle. Of course, the use of a 3-axis acceleration sensor and a 6-axis gyro sensor having a 3-axis geomagnetic sensor is not excluded, and a 9-axis sensor is preferred for accuracy.

The three-dimensional motion information of the vehicle generated by the sensor unit 13 includes x, y, z, roll, pitch, and yaw angle so that 3D simulation is possible. x, y, and z are used to determine the direction and size of the vehicle, and the tilt and rotation angle of the vehicle through roll, pitch, and yaw angle. The vehicle state information collected by the collecting unit 12 may be utilized for three-dimensional motion tracking of the vehicle.

The GPS receiver 15 may be a global navigation system in common use. Receives signals from at least three GPS satellites, and calculates the position, speed, and direction of the vehicle to generate accurate position information of the vehicle.

The information detected by the collecting unit 12, the sensor unit 13, and the GPS receiving unit 15 is transmitted to the accident analysis unit 11 together with time information for accident analysis. The time information generated in each module may not coincide with each other, and different time information may be synchronized in the accident analyzing unit 11. [

The image generating unit 14 is for generating image information of the surroundings of the vehicle, and includes a front camera, a rear camera, and a lateral camera on the left and right of the vehicle. Some of these cameras, for example, front and rear cameras, may be incorporated in the accident detection device 10, and another part may be installed separately in the vehicle.

The image information generated by the image generating unit 14 is stored in the storage unit 11b of the accident analyzing unit 11 in chronological order. When the accident is determined, the accident analysis unit 11 transmits the image information to the control center server 200 together with other accident data. The control center can accurately grasp the accident situation through video information.

The vehicle accident detection device 10 can be utilized as a black box. It can be mounted on the indoor side of the windshield like a vehicle black box, the front camera generates an image of the front of the vehicle, and the rear camera can be used for video communication. In an emergency situation, the control center can quickly and accurately grasp the accident situation through a video call with the driver.

According to the embodiment, the vehicle accident detection apparatus 10 may include a display 17, a microphone 18, and a speaker 19. [

The wireless communication unit 16 may be a communication module capable of 3G or LTE communication, and may be a module integrated with the GPS receiving unit.

3 to 5, an accident analysis model or method according to an embodiment will be described.

3, the accident analysis unit 11 tracks the three-dimensional motion of the vehicle using the information received from the sensor unit 13, Dimensional motion prediction range 20 of the three-dimensional motion estimation value 20 is determined to be an accident when the actual three-dimensional motion tracking value R1 is out of the prediction range 20.

In order to track the three-dimensional motions of the vehicle, for example, x, y, and z, the position, direction of travel, and travel distance of the vehicle must be determined. The position information can be obtained from the GPS receiving unit 15, the traveling direction of the vehicle can be obtained from the geomagnetic sensor of the sensor unit 13, and the travel distance can be obtained from the acceleration sensor or the collecting unit 12 of the sensor unit 13.

The position information reception period of a typical GPS receiver is about 1 second (sec). The sensor unit 13 having an update period of about 100 ms may be used for three-dimensional motion tracking of the vehicle for a period of time between every second until the position information is received from the GPS receiving unit 15. [ The current position of the vehicle is updated or corrected every one second using the position information from the GPS receiving unit 15. [

The three-dimensional motion prediction range 20 includes an allowable error which is obtained by setting the three-dimensional motion of the vehicle, which can be predicted based on the current point of time T, on the basis of the past three-dimensional motion tracking data as range or space values . 3 to 5, the prediction ranges 20, 30, and 40 are simplified to a rectangular parallelepiped.

3, the position 22 of the vehicle at the future time point T + 1 is calculated, for example, one second after using the past three-dimensional motion tracking values at the vehicle position 21 of the current time point T , A vector P1 is obtained in which the current position 21 of the vehicle is the start point and the future position 22 is the end point.

The direction of the vector P1 represents the shortest distance traveled by the vehicle for one second, and the magnitude of the vector P2 represents the shortest travel distance of the vehicle for one second. The three-dimensional motion prediction range 20 of the vehicle can be represented by a spatial model from the start point to the end point of the up vector P1. Spatial values converge at the start and end points, respectively. The distance from the vector P1 to the boundary of the prediction range 20 is an error range.

When the instantaneous change rate of the vehicle state value is large, the three-dimensional motion prediction range 20 will be set to a shape in which the error range is reduced with reference to the vector P1. The error range varies depending on the speed of the vehicle and the degree of the accident risk. The error range is not always symmetrically reduced or expanded based on the vector P1.

The accident analysis unit 11 tracks the three-dimensional motion of the vehicle every 100 ms, and judges that the three-dimensional motion tracking values R1 are out of the prediction range 20 by an accident. Naturally, the accident analysis unit 11 also tracks the x, y, and z values as well as the roll, pitch, and yaw angle of the vehicle. The prediction range 20 may include an allowable three-dimensional attitude value of the vehicle within an error range.

FIG. 4 shows a second three-dimensional motion prediction range 30 generated at the current position 31 at time T + 0.5 after 0.5 second (sec) from the current position 21 of time T in FIG. 3 have.

The first three-dimensional motion prediction range 20 shown in FIG. 3 is generated every one second in accordance with the GPS reception period. The prediction of the vehicle behavior in a period of one second may cause a misjudgment in the error range. The accident analysis unit 11 generates a second three-dimensional motion prediction range 30 in a cycle of one second between generation cycles of the first three-dimensional motion prediction range 20. [

Referring to FIG. 4, the second 3D motion prediction range 30 is derived in the same manner as the first 3D motion prediction range 20. FIG. The accident analysis unit 11 calculates the three-dimensional motion prediction range 30 up to the future time point (T + 1.5) after one second using the past three-dimensional motion tracking values R1 at the current time point (T + 0.5) .

A vector P2 having a start point of the vehicle position 31 at the current point in time T + 0.5 as a start point and a position 32 as an end point in the future point of time T + 1.5 is derived, The spatial values of the prediction range 30 converge. The prediction range 30 may include an allowable three-dimensional attitude value of the vehicle within an error range.

The accident analysis unit 11 tracks the three-dimensional motion of the vehicle every 100 ms, and judges an accident if the tracking values P 2 are out of the prediction range 30.

5 shows a third 3D motion prediction range 40 generated at the current position 31 at time T + 1 after 0.5 second (sec) from the current position 21 at time T + 0.5 in FIG. 4 , The first and second prediction ranges 20 and 30, respectively.

Referring to FIG. 5, a third 3D motion prediction range 40 is derived in the same manner as the first 3D motion prediction range 20. FIG. The accident analysis unit 11 calculates the three-dimensional motion prediction range 30 up to the future time point (T + 1.5) after one second using the past three-dimensional motion tracking values R2 at the current time point T + 1 .

A vector P3 whose starting point is the vehicle position at the current point in time T + 1 and whose position is the end point in the future point of time T + 2 is derived, and the vector P3 is obtained at the both ends of the vector P3. The space value converges. The prediction range 40 may include an acceptable three-dimensional attitude value of the vehicle within an error range.

Also, the accident analysis unit 11 tracks the three-dimensional motion of the vehicle every 100 ms, and judges an accident if the tracing values R3 are out of the prediction range 40. [

When the accident is detected, the vehicle accident detection device 10 transmits accident data for a certain period of time before and after the accident to the control center server 200. The accident data includes position information of the vehicle, three-dimensional motion information, and image information, and the position information, the three-dimensional motion information, and the image information include synchronized or synchronizable time information.

The control center which receives the accident data from the vehicle accident detection device 10 can easily confirm the vehicle accident situation using the image information and outputs the 3D simulation image using the 3D motion information, The accident situation can be grasped more accurately.

FIG. 6 illustrates a 3D simulation image generated using the three-dimensional motion information of the vehicle described above.

As shown in FIG. 6, the three-dimensional posture and behavior of the vehicle can be grasped through a simulation image. When the location information and the surrounding image information are included in the 3D simulation imaging data, the cause of the accident can be more easily analyzed and the possibility of the dispute between the parties due to the traffic accident can be lowered.

The vehicle accident detection device 10 can classify the accident according to the danger level of the determined situation and inform the control center of the situation. The control center can provide an emergency call service of a predetermined process according to the risk level of the incident received. For example, in the case of a sudden overturning situation, it is possible to immediately report a request for an immediate response by the related authorities. In case of a general contact accident, the driver can directly call the driver.

While the invention has been shown and described with respect to the specific embodiments thereof, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit of the invention as set forth in the following claims.

10, 20, 30: 3D motion prediction range P1, P2: Vector

Claims (8)

A GPS receiver for receiving position information of the vehicle;
A collection unit for collecting vehicle state information propagated through an in-vehicle network;
A sensor unit for sensing three-dimensional motion of the vehicle;
An image generating unit for generating image information of the surroundings of the vehicle;
An accident analyzing unit for determining whether a vehicle accident has occurred using information about the position, the state, and the three-dimensional motion of the vehicle; And
And a wireless communication unit for communicating with an external device of the vehicle,
The three-dimensional motion information includes 3D-simulatable values including x, y, z, roll, pitch, and yaw angles,
The accident analysis department,
Dimensional trajectory of the vehicle every first period, and at each second period, from a first predicted time point T to a future point in time T +? T, where? T is greater than the first period Dimensional motion of the vehicle up to a positive value of the three-dimensional motion tracing value, and determines that the actual three-dimensional motion tracking value is out of the first prediction range as an accident,
From the second predicted time point T 'on the basis of the three-dimensional motion tracking values of the previous vehicle at a second predicted time point T' (where T 'is at any time between the second period coming from the time point T and subsequently) Calculating a second predicted range of the three-dimensional motion of the vehicle up to a point in time T '+ Δt' (Δt 'is a positive value larger than the first period and smaller than or equal to the second period) If the value is out of the second prediction range, it is judged as an accident,
The second cycle is synchronized with the position information reception cycle of the GPS receiver,
The accident analysis unit corrects the vehicle position at the time point T arriving every second cycle using the positional information from the GPS receiver,
The first prediction range includes three-dimensional spatial values set based on a vector having a position at a first prediction time T as a starting point and a position at a future time T + t as an end point,
The second prediction range includes three-dimensional spatial values set based on a vector having a position at a second prediction time T 'as a starting point and a position at a future time T' + Δt 'as an end point,
Wherein the three-dimensional space values of the first and second prediction ranges vary according to the risk based on the actual three-dimensional motion tracking value. delete delete delete delete A vehicle accident detection device according to claim 1; And
And a control center server which holds information on at least one of a vehicle equipped with a vehicle accident detection device and a driver of the vehicle,
In case of an accident, the vehicle accident sensor transmits accident data for a certain period of time before and after the accident to the control center server,
Wherein the accident data includes the vehicle position information, the three-dimensional motion information, and the image information, and the position information, the three-dimensional motion information, and the image information include synchronized or synchronizable time information. delete delete

Images