JP5446778B2 - Accident occurrence prediction device, accident occurrence prediction program, and accident occurrence prediction method - Google Patents

Description

  This case relates to an accident occurrence prediction device, an accident occurrence prediction program, and an accident occurrence prediction method.

  In the automobile industry, realizing an automobile that does not cause accidents is one of the ultimate goals. In order to predict whether or not the own vehicle will cause an accident with another vehicle in the near future, it is necessary to predict the situation of the own vehicle and the other vehicle in the near future. Conventionally, current positions and travel vectors (speed and travel direction) of the host vehicle and other vehicles are acquired, and positions of the host vehicle and other vehicles after a predetermined time are predicted from the travel vectors.

  However, the above estimation is based on the premise that the host vehicle and other vehicles continue to run with the same running vector. That is, when the own vehicle and another vehicle travel such that the travel vector is changed, the prediction is lost.

  On the other hand, Patent Document 1 discloses a technique for estimating the trajectory of an object (automobile) from map information. According to this technology, depending on the map information (road shape, etc.), such as straight movement when the road is a straight line, movement along the curve of the road when the road is a curve, and entering and moving to an intersection when the road is an intersection Thus, the future movement position of the target (car) is estimated. This makes it possible to enumerate one or a plurality of movements that the subject may perform after several seconds.

  Further, Patent Document 2 discloses a technique for storing traffic accident history information that has occurred in the past in a database and calculating the possibility of an accident by matching with the current situation of the vehicle.

JP 2005-165555 A Japanese Patent Laid-Open No. 11-120478

  However, even when the technique of Patent Document 1 is used, the movement of the vehicle that does not depend on the road shape cannot be estimated (predicted). For example, when the other vehicle A jumps out from the side of the other vehicle A when the own vehicle and the other vehicle A pass each other on the one-lane road, avoidance of the other vehicle A to avoid a collision with the other vehicle B There is a case where an action is suddenly performed. If this avoidance action is a protrusion to the lane in which the host vehicle is traveling, the host vehicle and the other vehicle A may collide (referred to as an accident). However, in the above-mentioned Patent Document 1, since it is impossible to estimate the movement of a vehicle that does not depend on the road shape such as the other vehicle A, it is impossible to avoid the above-mentioned accident.

  The technology of Patent Document 2 matches the situation of the vehicle (items such as location, terrain, travel speed, traffic volume, season, time, weather, temperature, humidity, etc.) with traffic accident history information to determine the possibility of an accident. It is a technique to calculate. However, since it is rare that all the above items match completely, even if the technique of Patent Document 2 is used, the accident prediction accuracy may not be so high. In other words, setting a large number of items used for matching for accident prediction in the dark clouds may cause an adverse effect on improving the accuracy of accident prediction.

  Therefore, this case has been made in view of the above problems, and an accident occurrence prediction device, an accident occurrence prediction program, and an accident occurrence prediction that can predict the occurrence of an accident caused by the movement of another vehicle that does not depend on the road shape. It aims to provide a method.

The accident occurrence prediction device described in the present specification indicates a state in which an accident may occur between the own vehicle and the other vehicle, and information on the relative positions of the own vehicle and the other vehicle not on the course of the own vehicle. , the holding which holds information of the movement vector of the vehicle and the other vehicle is not on the car course the free-and accidents definitions above defined information factors to vary the movement vector of the other vehicle is not on the host vehicle path And information on the position of the own vehicle and other vehicles not on the course of the own vehicle, the own vehicle and other vehicles not on the course of the own vehicle in a state where the own vehicle is actually driven. An acquisition unit that acquires information on movement vectors and information on factors that cause movement vectors of other vehicles not on the course of the host vehicle, information acquired by the acquisition unit, and the accident definition are compared. a determination unit which determines whether or not there is a possibility that an accident occurs, includes a path of the vehicle Factors for varying the movement vector of the other vehicle is not in is a factor to vary the movement vector of said no other vehicle on the vehicle of the path to the motion vector direction which enters in the path of the vehicle.

The accident occurrence prediction program described in the present specification, the computer, the information on the position of the own vehicle and other vehicles not on the course of the own vehicle in a state where the own vehicle is actually driven, the own vehicle and information of the movement vector of the other vehicle is not on the car course the free-and acquisition unit that acquires the information factors to vary the movement vector of the other vehicle is not on the host vehicle path, and acquired by the acquisition unit information and the conditions that might accident between the vehicle and the other vehicle is generated, the vehicle and the the free-wheel information of the relative position of the other vehicle not in the path, of the vehicle and the free-wheel information of the movement vector of the other vehicle is not on the path, and the compared accidents definitions and defined in the information factors to vary the movement vector of the other vehicle is not on the vehicle of course, the possibility of an accident occurring a certain whether determination unit which, Ru to function as a program, the vehicle Factors for varying the movement vector of the other vehicle not in the path is a factor to vary the movement vector of said no other vehicle on the vehicle of the path to the motion vector direction which enters in the path of the vehicle.

The accident occurrence prediction method described in the present specification includes information on the position of the own vehicle and other vehicles not on the course of the own vehicle in a state where the own vehicle is actually driven, the own vehicle, and the own vehicle. information of the movement vector of the other vehicle is not on the path of, and an acquisition step of acquiring the information factors to vary the movement vector of the other vehicle is not on the host vehicle path, an accident between the vehicle and the other vehicle state but that can occur, the vehicle and the the free-wheel information of the relative position of the other vehicle is not on the path, the vehicle and the free-wheel information of the movement vector of the other vehicle is not on the route, And whether there is a possibility that an accident will occur by comparing the accident definition defined by the information of the factor that fluctuates the movement vector of the other vehicle that is not on the route of the own vehicle with the information acquired in the acquisition process a determination step of determining whether a method executed by a computer, on the path of the vehicle Factors for varying the movement vector of the other vehicle have is a factor to vary the movement vector of said no other vehicle on the vehicle of the path to the motion vector direction which enters in the path of the vehicle.

  The accident occurrence prediction device, the accident occurrence prediction program, and the accident occurrence prediction method described in the present specification have an effect of being able to predict the occurrence of an accident caused by the movement of another vehicle that does not depend on the road shape.

It is a block diagram which shows the structure of the accident occurrence prediction apparatus which concerns on 1st Embodiment. 2A is a diagram showing a situation where an accident may occur, FIG. 2B is a diagram showing a state where an accident has occurred, and FIG. 2C is a diagram illustrating the first implementation. It is a figure which shows an example of the accident definition data used with a form. It is a flowchart which shows the process of the accident occurrence prediction apparatus in 1st Embodiment. It is a block diagram which shows the structure of the accident occurrence prediction apparatus which concerns on 2nd Embodiment. FIG. 5A is a diagram illustrating an example of a situation where an accident may occur, and FIG. 5B is a diagram illustrating accident definition data relating to the situation of FIG. FIG. 6A is a diagram illustrating an example of a situation where an accident may occur, and FIG. 6B is a diagram illustrating accident definition data relating to the situation of FIG. It is a block diagram which shows the structure of the accident occurrence prediction apparatus which concerns on 3rd Embodiment. FIG. 8A is a diagram illustrating an example of a situation where an accident may occur, and FIG. 8B is a diagram illustrating accident definition data relating to the situation of FIG. FIG. 9A is a block diagram illustrating a configuration of an accident occurrence prediction apparatus according to the fourth embodiment, and FIG. 9B is a diagram illustrating an example of accident definition data used in the fourth embodiment. is there. It is a block diagram which shows the structure of the accident occurrence prediction apparatus which concerns on 5th Embodiment. It is a block diagram which shows the structure of the accident occurrence prediction apparatus which concerns on 6th Embodiment. It is a block diagram which shows the structure of the accident occurrence prediction apparatus which concerns on 7th Embodiment.

<< First Embodiment >>
Hereinafter, a first embodiment of an accident occurrence prediction device will be described in detail with reference to FIGS. FIG. 1 is a block diagram showing an accident occurrence prediction apparatus 100A according to the first embodiment. The accident occurrence prediction device 100A is mounted on the automobile 50.

  As shown in FIG. 1, the accident occurrence prediction device 100A includes millimeter wave radars 10A and 10B, a vehicle speed pulse detection unit 11, a GPS device 14, a determination unit 16, a prediction status DB 18 as a holding unit, and a notification unit. 20 and LED22. In the present embodiment, the millimeter wave radars 10A and 10B, the vehicle speed pulse detection unit 11, and the GPS device 14 have a function as an acquisition unit that acquires various information.

  The millimeter wave radars 10A and 10B are provided on the front side and the rear side of the automobile 50, for example. The millimeter wave radars 10 </ b> A and 10 </ b> B detect the relative positions of objects (including automobiles, bicycles, pedestrians, buildings, etc.) existing around the automobile 50 with respect to the automobile 50. In other words, the millimeter wave radars 10A and 10B can acquire information on the relative position (position with respect to the own vehicle) of a car (also referred to as an accident direct object) that may become an object that will cause an accident in the near future. Further, the millimeter wave radars 10A and 10B can acquire information on the relative position (position with respect to the host vehicle) of an object (accident indirect target) that may be a factor that affects the traveling of the accident direct target.

  Furthermore, the millimeter wave radars 10A and 10B can obtain a change in the relative position of the accident direct target and the accident indirect target at a predetermined time. In other words, in the first embodiment, the millimeter wave radars 10A and 10B acquire the movement vector (movement speed and movement direction) of the accident direct target and the movement vector of the accident indirect target from the change in the relative position. It can be said that there is.

  Here, the accident direct object and the accident indirect object described above will be specifically described with reference to FIG. FIG. 2A schematically shows a road situation where an accident may occur in the near future. In FIG. 2A, the vehicle 300 and the other vehicle 400A are about to pass each other on the one-lane road, and the other vehicle 400B is stopped on the lane on which the other vehicle 400A travels. Show. In this situation, there is a possibility that the other vehicle 400A may protrude from the lane in which the host vehicle travels in an attempt to avoid the other vehicle 400B at a timing when the own vehicle 300 and the other vehicle 400A pass each other in the near future. In this case, as shown in FIG. 2B, the host vehicle 300 and the other vehicle 400A may collide. That is, when viewed from the own vehicle 300, the other vehicle 400A is an accident direct target, and the other vehicle 400B is an accident indirect target that changes the movement of the accident direct target (400A) (changes the movement vector).

  Returning to FIG. 1, the vehicle speed pulse detector 11 measures the vehicle speed pulse generated in proportion to the rotational speed of the axle of the automobile 50 and acquires the moving speed of the automobile 50. The GPS device 14 includes a GPS main device 12A, a GPS antenna 12B, and a map DB 12C. The GPS main device 12A acquires position information (latitude, longitude) of the automobile 50 from a GPS satellite via the GPS antenna 12B. Further, the GPS main device 12A acquires the road situation where the automobile 50 is located by using the acquired position information of the automobile 50 and the map data stored in the map DB 12C. This road situation includes road shapes and road structures such as intersections, straight roads, slopes, curves, presence / absence of crossings, and tunnel entrances. The moving direction of the automobile 50 can also be acquired from the GPS device 14. The moving direction is used for calculating the moving vector of the automobile 50 together with the moving speed acquired by the vehicle speed pulse detector 11. In addition, it is good also as providing the moving direction of the motor vehicle 50 using a gyro sensor, a direction detection sensor, etc. in the motor vehicle 50, using these sensors.

  The determination unit 16 acquires the relative positions and movement vectors of the accident direct target and the accident indirect target acquired by the millimeter wave radars 10A and 10B. Further, the determination unit 16 acquires the movement vector of the own vehicle acquired by the vehicle speed pulse detection unit 11 and the GPS device 14 and the road condition acquired by the GPS device 14. Then, the determination unit 16 compares each information with the accident prediction situation data stored in the prediction situation DB 18 (hereinafter, referred to as accident definition data), so that the vehicle and the accident direct target can be compared. Then, determine whether there is a possibility of an accident in the near future. The determination method in the determination unit 16 and the accident definition data stored in the prediction situation DB 18 will be described in detail later.

  When the determination unit 16 determines that an accident may occur in the near future, the notification unit 20 issues a warning to the driver of the automobile 50 via the LED 22 provided on the window frame or the like. In addition to the LED 22, the notification unit 20 uses a technology for projecting information by installing half mirrors or organic ELs at a plurality of locations, HUD (Head-Up Display), etc., on the windshield of the automobile 50, A warning may be displayed. Further, visual warning may be performed by other methods. Furthermore, the notification unit 20 is not limited to a visual warning, and may perform a warning through hearing using sound or the like, or a warning through touch using vibration or the like.

  FIG. 3 is a flowchart showing a processing procedure performed by the accident occurrence prediction apparatus 100A. Hereinafter, description will be made along FIG. 3 with reference to other drawings as appropriate.

  In step S10 of FIG. 3, the prediction situation DB 18 acquires accident definition data. As the accident definition data, data obtained by analyzing past accident cases, data on cases handled in defense driving or risk prediction driving, etc., which are compiled for safe driving education, can be used. For example, in the case of FIG. 2A, the accident definition data is as shown in FIG. In the accident definition data shown in FIG. 2C, information on the relative positions and movement vectors of the own vehicle, the accident direct object, and the accident indirect object is defined using the XY coordinates in FIG. Further, “straight road” is defined as the road structure information. A large number (in this case, N) of such accident definition data is prepared in various situations and stored in the prediction situation DB 18 in step S10. Various situations include the case where the accident indirect target is a bicycle, a pedestrian, an obstacle, and the case where the road structure is a slope, an intersection, a curve, or the like. Step S <b> 10 may be performed before the accident occurrence prediction device 100 </ b> A is mounted on the automobile 50. In addition, the accident definition data may be updated regularly or appropriately.

  Next, in step S12 of FIG. 3, the GPS device 14 acquires road structure information. Next, in step S14, the millimeter wave radars 10A and 10B, the vehicle speed pulse detection unit 11, and the GPS device 14 acquire information on the relative positions of the host vehicle and the surrounding objects, and information on movement vectors of the host vehicle and other vehicles. .

  Next, in step S16, the determination unit 16 sets a parameter n indicating the number of accident definition data to 1. Next, in step S18, the determination unit 16 acquires the relative position information, the movement vector information, and the road structure information acquired in steps S12 and S14, and these information and the nth (first) accident definition data. And compare. In this comparison, if the acquired information and the accident definition data substantially match, it means that the accident may occur in the near future. Here, the acquired information and the accident definition data “nearly match” is not limited to the case where the acquired information and the accident definition data completely match, and the error between the acquired information and the accident definition data is determined in advance. This includes cases where it falls within the predetermined range. For example, when the host vehicle 300 and the other vehicles 400A and 400B are actually in the situation as shown in FIG. 2 (a), the nth accident definition data is data as shown in FIG. 2 (c). Suppose there was. In this case, since the accident definition data in FIG. 2 (c) and the current situation substantially coincide, the determination unit 16 determines that the own vehicle 300 and the other vehicle (accident direct) as shown in FIG. 2 (b). Target) It is determined that an accident with 400A may occur in the near future.

  In addition, from the information acquired in step S14, it is unclear which car is the accident direct target and which car is the accident indirect target. Therefore, in the comparison in step S18, it is conceivable to make a brute force comparison of all possibilities, such as when the other vehicle 400A is directly subject to an accident or when the other vehicle 400B is directly subject to an accident.

  Next, in step S20, the determination unit 16 determines whether or not there is a possibility of an accident in the near future. When judgment here is affirmed, it transfers to step S22. In step S22, the notification unit 20 warns the driver who is driving the vehicle through the LED 22 that an accident may occur in the near future, and returns to step S12. The driver can avoid an accident beforehand by paying attention forward, stepping on the brake, returning the accelerator, or turning the steering wheel based on the warning.

  On the other hand, if the determination in step S20 is negative, the process proceeds to step S24. In step S <b> 24, the determination unit 16 determines whether n is N. Here, N means the total number of accident definition data stored in the prediction situation DB 18 as described above. Therefore, in step S24, it is determined whether or not the determination using all the accident definition data has been performed. Will be. If the determination here is negative, in step S26, n is incremented by 1 (n ← n + 1), and then the process returns to step S18. Thereafter, the determination unit 16 performs determination using the accident definition data until the determination in step S20 is affirmed or the determination in step S24 is affirmed. On the other hand, if the determination in step S26 is affirmative, the process proceeds to step S12.

  When the process returns to step S12, in a new situation, the acquired information is compared with the accident definition data in the same manner as described above, and the determination of the possibility of an accident in the near future is repeated.

  As described above in detail, according to the first embodiment, the determination unit 16 determines the state that the accident situation may occur between the own vehicle and the other vehicle held in the prediction situation DB 18. Accident definition data defined by information on the relative position of the vehicle and other vehicles, information on movement vectors of the own vehicle and other vehicles, and information on factors that cause movement vectors of other vehicles to change, and the vehicle is actually driven In this state, the millimeter wave radars 10A and 10B, the vehicle speed pulse detection unit 11, the position information of the own vehicle and other vehicles, the information on the movement vectors of the own vehicle and other vehicles, and the movement vector of other vehicles acquired by the GPS device 14 It is determined whether or not there is a possibility of an accident by comparing the information of the factor that causes fluctuation. Therefore, the determination unit 16 causes not only other vehicles that may cause an accident directly, but also factors that change the movement vector of other vehicles that may cause an accident (in the first embodiment, the other vehicle 400A). The other vehicle 400B) that changes the movement vector of the vehicle is also compared with the accident definition data to determine the possibility of the accident. Thereby, in this embodiment, it is possible to accurately predict an accident caused by the movement of another vehicle (movement vector fluctuation) that does not depend on the road shape.

  In the first embodiment, the notification unit 20 issues a warning to the driver via the LED 22 when the determination unit 16 determines that an accident may occur in the near future. Therefore, the driver can avoid an accident beforehand by paying attention to the front, stepping on the brake, returning the accelerator, or turning the steering wheel in accordance with the warning.

  In the first embodiment, the accident definition data is also defined in the road structure information where the own vehicle and the other vehicle are located, and the determination unit 16 includes the road structure information acquired by the GPS device 14, By comparing with accident definition data, it is determined whether or not there is a possibility of an accident. Thus, by further considering road structures such as intersections and curves, an accurate accident prediction can be performed.

  In the first embodiment, the execution order of steps S12 and S14 may be changed, or steps S12 and S14 may be performed simultaneously.

  In the first embodiment, the millimeter wave radar is used as a device for acquiring the relative position and the movement vector of the other vehicle. However, the present invention is not limited to this. For example, a stereo vision sensor using two cameras is used. It is also good. In the first embodiment, the case where the information on the relative position and the movement vector of the other vehicle is acquired by a common device has been described. However, the present invention is not limited to this, and the relative position and the movement vector are acquired by another device. You may do it.

  In the first embodiment, when the determination unit 16 determines that an accident is likely to occur, the determination unit 16 calculates the time until the accident occurs from the relative distance and relative speed between the host vehicle and the other vehicle. Can do. Therefore, the notification unit 20 may calculate the time until the accident occurs and vary the type of warning according to the time.

  In the first embodiment, in addition to calculating the time until the occurrence of an accident, information on how many seconds later the accident occurs can be associated with the accident definition data in advance. In this case, the type of warning can be made different without performing calculation or the like.

  In the process of FIG. 3, the case has been described where the accident prediction display is performed in step S22 when it is determined in step S20 that there is a possibility of an accident from one accident definition data. is not. For example, accident prediction using all accident definition data may be performed, and when it is determined that there is a possibility of an accident based on at least one accident definition data, the accident prediction display may be performed. In this case, it may be determined that there is a possibility of an accident in the determination using a plurality of accident definition data. In such a case, the notification unit 20 can preferentially display (warn) from the one with a short time until the occurrence of the accident.

<< Second Embodiment >>
Next, a second embodiment of the accident occurrence prediction device will be described based on FIGS. FIG. 4 shows a block diagram of an accident occurrence prediction apparatus 100B according to the second embodiment. As can be seen from a comparison between FIG. 4 and FIG. 1, the accident occurrence prediction device 100B of the second embodiment replaces the GPS device 14 of the accident occurrence prediction device 100A of the first embodiment with the camera 24 and the wind speed. A total of 26 is provided. In the second embodiment, it is assumed that the moving direction of the automobile 50 is acquired by a gyro sensor or a direction detection sensor (not shown). Hereinafter, the description of the same configuration as that of the first embodiment will be omitted or simplified.

  The camera 24 captures and acquires road surface conditions. The road surface condition includes, for example, at least one of the presence / absence of a puddle, freezing, rainfall and snowfall, dense fog, and the angle of the sun. The anemometer 26 acquires the wind direction and wind speed near the automobile 50. The anemometer 26 can calculate the actual wind speed by subtracting the moving speed in the traveling direction of the automobile from the detected wind speed and direction.

  In the second embodiment, a situation where an accident may occur in the future as shown in FIG. 5 (a) is stored in the prediction situation DB 18 as accident definition data as shown in FIG. 5 (b). I can keep it. Here, in FIG. 5 (a), since the cross wind is strong, the steering wheel is largely turned so that the other vehicle 400A does not collide with the other vehicle 400B, and the own vehicle 300 and the other vehicle 400A may collide with each other. Is shown. That is, in FIG. 5A and FIG. 5B, the factor that causes the other vehicle 400B and the wind vector (including the wind direction and the wind speed) to change the movement vector of the other vehicle 400A that is the accident direct target (accident indirect target). It becomes.

  In the second embodiment, accident definition data (see FIG. 6B) relating to the situation where there is no other vehicle 400B shown in FIG. 6A can also be created and stored in the predicted situation DB 18. . The accident definition data in FIG. 6B is accident definition data in which the accident indirect target is only the wind vector (including the wind direction and the wind speed).

  In addition, as with wind, water pools and freezing on the road surface can be subject to accidents. Therefore, the accident definition data may include the relative position of the water pool on the road surface or the frozen own vehicle instead of or together with the wind.

  In the second embodiment, the accident definition data as shown in FIGS. 5B and 6B is compared with information acquired by the millimeter wave radars 10A and 10B, the camera 24, the anemometer 26, and the like. Thus, as in the first embodiment, accident prediction is possible. In the second embodiment, by including a natural phenomenon as an accident indirect target, it is possible to accurately determine whether or not an accident may occur in the near future. Further, the notification unit 20 issues a warning to the driver according to the determination, so that the driver can avoid an accident beforehand.

<< Third Embodiment >>
Next, a third embodiment of the accident occurrence prediction device will be described with reference to FIGS. FIG. 7 shows a block diagram of an accident occurrence prediction apparatus 100C according to the third embodiment. As shown in FIG. 7, the accident occurrence prediction device 100C of the third embodiment has a configuration in which a GPS device 14 is added to the accident occurrence prediction device 100B of the second embodiment. Hereinafter, description of the same configuration as the first and second embodiments will be omitted or simplified.

  In the third embodiment, for example, the accident definition data as shown in FIG. 8B is added to the accident occurrence situation as shown in FIG. 8A (a situation where a cross wind is blowing at the exit of the tunnel). It can be stored in the prediction situation DB 18. In the accident definition data of FIG. 8B, the other vehicle 400B, the wind vector (including wind direction and wind speed), and the road structure (near the tunnel entrance) cause the movement vector of the other vehicle 400A that is the accident direct target to fluctuate. (Indirect accident target). Since the third embodiment has all the same configurations as those of the first and second embodiments described above, as shown in FIGS. 2 (c), 5 (b), and 6 (b). Accident definition data can also be stored in the prediction situation DB 18.

  In the third embodiment, the determination unit 16 acquires information on the relative positions of the host vehicle and other vehicles from the millimeter wave radars 10A and 10B. Further, the determination unit 16 acquires information on the movement vector of the own vehicle from the GPS device 14 and the vehicle speed pulse detection unit 11, and acquires information on the movement vector of the other vehicle from the millimeter wave radars 10A and 10B. Further, the determination unit 16 acquires a wind vector from the anemometer 26, a road surface condition from the camera 24, and acquires a road structure such as the presence or absence of a tunnel from the GPS device 14. Then, by comparing the acquired information and the accident definition data in the same procedure as in FIG. 3, it is determined whether or not there is a possibility of an accident in the near future.

  In this way, in the third embodiment, since natural phenomena and road structures are included as accident indirect targets, it is possible to accurately determine whether or not an accident may occur in the near future. Further, the notification unit 20 issues a warning to the driver according to the determination, so that the driver can avoid an accident beforehand.

<< Fourth Embodiment >>
Next, a fourth embodiment of the accident occurrence prediction device will be described with reference to FIG. FIG. 9A shows a block diagram of an accident occurrence prediction apparatus 100D according to the fourth embodiment. As shown in FIG. 9A, the accident occurrence prediction apparatus 100D of the fourth embodiment has a configuration in which the GPS apparatus 14 is excluded from the accident occurrence prediction apparatus 100A of the first embodiment. Hereinafter, the description of the same configuration as the first to third embodiments will be omitted or simplified.

  When the configuration as in the fourth embodiment is adopted, the relative position and movement vector of the own vehicle, the other vehicle directly subject to accident, the other vehicle subject to accident indirect, as shown in FIG. 9B are defined. The accident definition data is stored in the prediction situation DB 18.

  By using such accident definition data, it is possible to make an accident prediction that takes into account not only information on other vehicles directly subject to the accident but also information on other vehicles subject to accident indirect that fluctuate the movement vector of the other vehicles directly subject to the accident. . Thereby, it is possible to accurately determine whether or not there is a possibility of an accident in the near future. Further, the notification unit 20 issues a warning to the driver according to the determination, so that the driver can avoid an accident beforehand.

<< Fifth Embodiment >>
Next, a fifth embodiment of the accident occurrence prediction device will be described with reference to FIG. FIG. 10 shows a block diagram of an accident occurrence prediction apparatus 100E according to the fifth embodiment. As shown in FIG. 10, in the accident occurrence prediction device 100E of the fifth embodiment, the host vehicle control unit 30 is connected to the determination unit 16 of the accident occurrence prediction device 100A of the first embodiment. In the following description, the description of the same configuration as in the first to fourth embodiments is omitted or simplified.

  The own vehicle control unit 30 includes an engine control unit 32, a braking control unit 34, an airbag control unit 36, a seat belt control unit 38, and a door lock control unit 40. In the fifth embodiment, when the determination unit 16 determines that there is a possibility of an accident in the near future, the notification unit 20 issues a warning via the LED 22, and the own vehicle control unit 30. Control at least one of the components. Specifically, in the case of the engine control unit 32, the engine speed of the engine control unit 32 is changed and controlled in order to prevent an accident from occurring. Further, the brake control unit 34 automatically controls the brake in order to prevent an accident from occurring. Further, the airbag control unit 36 controls the airbag to operate in a short time in order to reduce the load on the driver or the like when an accident occurs. Further, the seat belt control unit 38 largely controls the tightening strength of the seat belt in order to reduce the load on the driver or the like when an accident occurs. Furthermore, if it is the door lock control part 40, in order to prevent a driver | operator etc. jumping out of a vehicle when an accident arises, control which locks a door lock automatically is performed.

  In this way, in the fifth embodiment, in addition to warning to the driver, control for avoiding an accident is automatically performed, or control for improving safety when an accident occurs is performed. Can be done automatically. Thereby, the avoidance of an accident and / or the damage by an accident can be reduced.

  In the fifth embodiment, the case where both the warning by the notification unit 20 and the LED 22 and the automatic control by the own vehicle control unit 30 are described has been described. However, the present invention is not limited thereto, and only automatic control may be performed. good. Moreover, the determination part 16 is good also as selecting whether warning or automatic control is performed according to the time estimated that an accident will generate | occur | produce.

  Moreover, although the said 5th Embodiment demonstrated the example in which the automatic control part 30 was connected with respect to the determination part 16 of 1st Embodiment, it is not restricted to this, The 2nd-4th Embodiment. A configuration in which the automatic control unit 30 is connected to the determination unit 16 can also be realized.

<< Sixth Embodiment >>
Next, a sixth embodiment of the accident occurrence prediction device will be described with reference to FIG. FIG. 11 shows a block diagram of an accident occurrence prediction apparatus 100F according to the sixth embodiment. As shown in FIG. 11, the accident occurrence prediction device 100F of the sixth embodiment includes an in-vehicle unit 80 mounted on the own vehicle and an infrastructure unit 90 mounted on infrastructure equipment on the road. . In the following description, the description of the same configuration as in the first to fourth embodiments is omitted or simplified.

  The in-vehicle unit 80 includes a determination unit 16, a prediction situation DB 18, a notification unit 20, an LED 22, a communication unit 42, and a communication antenna 44. The communication unit 42 has a function of transmitting information received via the communication antenna 44 to the determination unit 16. It is assumed that accident definition data as shown in FIGS. 2C, 5B, 6B, and 8B is stored in the prediction situation DB 18.

  A large number of infrastructure units 90 are arranged on a road (for example, at an intersection), and include a host vehicle / other vehicle state detection sensor 102, an anemometer 104, a road shape storage unit 110, and a communication module 108. Yes. These parts are assumed to be held on the road via the utility pole 120 or the like.

  The own vehicle / other vehicle status detection sensor 102 is a sensor that detects and acquires the position and speed of the own vehicle and other vehicles located near the place where the infrastructure unit 90 is installed. As the sensor 102, a millimeter wave radar, a stereo vision sensor, or the like can be used. The anemometer 104 acquires the wind vector (wind direction and wind speed) of the place where the infrastructure unit 90 is installed. The road shape storage unit 110 stores road shapes (intersections, curves, etc.) near the place where the infrastructure unit 90 is installed. The communication module 108 transmits information acquired by the own vehicle / other vehicle state detection sensor 102 and the anemometer 104 and information stored in the road shape storage unit 110 to the communication unit 42 of the vehicle located in the vicinity. To do.

  In the sixth embodiment, the determination unit 16 of the in-vehicle unit 80 stores the position and speed of the own vehicle and other vehicles, the wind vector, and the shape of the road received by the communication unit 42 in the prediction situation DB 18. Predict accidents in the near future compared with accident definition data. If the determination unit 16 determines that an accident may occur in the near future, the notification unit 20 issues a warning to the driver via the LED 22.

  As described above, according to the sixth embodiment, the determination unit 16 of the in-vehicle unit 80 uses the information transmitted from the infrastructure unit 90 side to predict the occurrence of an accident. There is no need to provide a sensor for accident determination. Thereby, the structure of the vehicle-mounted part 80 can be simplified.

  In the sixth embodiment, the case where a sensor or the like for acquiring information necessary for accident determination is not provided in the in-vehicle unit 80 has been described. However, the present invention is not limited to this. For example, at least one of the sensors described in the first to fifth embodiments (millimeter wave radar, vehicle speed pulse detection unit, GPS device, anemometer, camera, etc.) may be provided in the in-vehicle unit 80. In this case, you may abbreviate | omit a part structure of the infrastructure part 90 which has a function equivalent to the sensor etc. which were provided in the vehicle-mounted part 80 side. In the sixth embodiment, the case where the vehicle / other vehicle condition detection sensor 102, the anemometer 104, and the road shape storage unit 110 are provided in the infrastructure unit 90 has been described. However, the present invention is not limited to this. Absent. For example, the anemometer 104 and the road shape storage unit 110 may be omitted. The infrastructure unit 90 may be provided with a camera for acquiring the road surface condition. The prediction situation DB 18 only needs to store accident definition data that can be compared with information that can be acquired in either the infrastructure unit 90 or the in-vehicle unit 80.

  In the sixth embodiment as well, the vehicle control unit 30 described in the fifth embodiment may be connected to the determination unit 16. Thus, as in the fifth embodiment, it is possible to perform automatic control for avoiding accidents in the automobile 50.

<< Seventh Embodiment >>
Next, a seventh embodiment of the accident occurrence prediction device will be described with reference to FIG. FIG. 12 shows a block diagram of an accident occurrence prediction apparatus 100G according to the seventh embodiment. As shown in FIG. 12, the accident occurrence prediction device 100G of the seventh embodiment can communicate with the in-vehicle unit 180 mounted in the own vehicle (the own vehicle 300 in FIG. 12) and the in-vehicle unit 180. Server 200. It is assumed that the in-vehicle unit 180 is also mounted on the other vehicles 400A and 400B. In the following, the description of the same configuration as in the first to sixth embodiments is omitted or simplified.

  The in-vehicle unit 180 includes the vehicle speed pulse detection unit 11, the communication unit 42, the communication antenna 44, the notification unit 20, the LED 22, and the GPS device 14. In the in-vehicle unit 180, the vehicle speed pulse detection unit 11 detects the moving speed of the automobile 50, and the GPS device 14 detects the position of the automobile 50.

  The server 200 includes a communication unit 214, an acquisition unit 215, a determination unit 216, and a prediction situation DB 218. The communication unit 214 receives information on the moving speed and position of the automobile 50 transmitted from the communication unit 42 of the in-vehicle unit 180 via the communication antenna 44. The acquisition unit 215 acquires information received by the communication unit 214 and sends the information to the determination unit 216.

  The determination unit 216 compares the information on the own vehicle or other vehicle sent from the acquisition unit 215 with the accident definition data stored in the prediction situation DB 218, and an accident may occur in the near future. It is determined in the same manner as in the first to sixth embodiments. If the determination unit 216 determines that there is a possibility of an accident, the communication unit 214 mounts the determination result on the car corresponding to “own vehicle” or “accident direct target” in the accident. It transmits with respect to the vehicle-mounted part 180. FIG.

  On the in-vehicle unit 180 side, when the communication unit 42 receives a determination result that an accident may occur, the notification unit 20 issues a warning to the driver via the LED 22.

  As described above, according to the seventh embodiment, the server 200 that sends information on the position and movement vector of the host vehicle from the in-vehicle unit 180 to the server 200 and receives the information from a plurality of in-vehicle units causes an accident. Determine the possibility. Accordingly, as in the first to sixth embodiments, it is possible to accurately determine whether or not an accident may occur, and it is not necessary to provide a determination unit and a prediction situation DB in each in-vehicle unit 180. Simplification of the in-vehicle unit 180 can be achieved.

  In the seventh embodiment, the in-vehicle unit 180 may have the anemometer, the camera, or the like described in the second embodiment. In this case, by sending information acquired using an anemometer, a camera, or the like to the server 200, the server 200 can further determine the possibility of an accident using these information.

  In the seventh embodiment as well, the vehicle control unit 30 described in the fifth embodiment may be connected to the determination unit 16. Thus, as in the fifth embodiment, it is possible to perform automatic control for avoiding accidents in the automobile 50.

  In each of the above embodiments, a case has been described in which it is determined whether or not there is a possibility of an accident in the near future depending on whether or not the relative position and movement vector of the accident occurrence data are substantially the same. It is not limited to this. For example, accident definition data stored in the prediction situation DB may be defined as accident definition data of an accident that may occur after s (s is an arbitrary number) seconds. In this case, if the relative position or movement vector is approximately m times the relative position or movement vector defined in the accident definition data, it is determined that an accident may occur in (s × m) seconds. Can do.

  In each of the above embodiments, the case where the movement vectors of the own vehicle and the other vehicle are defined as the accident definition data is described. However, the present invention is not limited to this, and the relative movement vector of the own vehicle and the other vehicle is defined. May be. In defining the relative movement vector, it can be said that “information on movement vectors of own vehicle and other vehicles” is defined in the accident definition data. As described above, when defining the relative movement vector, the relative movement vector of the own vehicle and the other vehicle can be detected only by the millimeter wave radar or the stereo vision sensor without using the vehicle speed pulse detection unit. .

  In each of the above-described embodiments, the processing apparatus (computer) provided in the automobile 50 can be similarly executed by executing a program for realizing the functions of an acquisition unit, a determination unit, and a holding unit that acquire information. There is an effect. In this case, a program describing the processing contents of the functions that the processing apparatus should have is provided. By executing the program on a processing device (computer), the above processing functions are realized on the computer. The program describing the processing contents can be recorded on a computer-readable recording medium.

  When the program is distributed, for example, it is sold in the form of a portable recording medium such as a DVD (Digital Versatile Disc) or a CD-ROM (Compact Disc Read Only Memory) on which the program is recorded. It is also possible to store the program in a storage device of a server computer and transfer the program from the server computer to another computer via a network.

  The computer that executes the program stores, for example, the program recorded on the portable recording medium or the program transferred from the server computer in its own storage device. Then, the computer reads the program from its own storage device and executes processing according to the program. The computer can also read the program directly from the portable recording medium and execute processing according to the program. Further, each time the program is transferred from the server computer, the computer can sequentially execute processing according to the received program.

  Each embodiment mentioned above is an example of suitable implementation of the present invention. However, the present invention is not limited to this, and various modifications can be made without departing from the scope of the present invention.

In addition, the following additional notes are disclosed regarding the above description.
(Supplementary note 1) Information on the relative positions of the host vehicle and the other vehicle, information on the movement vectors of the host vehicle and the other vehicle, and a state where an accident may occur between the host vehicle and the other vehicle, and A holding unit that holds an accident definition defined by information on factors that cause movement vectors of other vehicles to change, information on the positions of the own vehicle and the other vehicle in a state where the own vehicle is actually driven, and the own vehicle The acquisition unit that acquires the information on the movement vector of the other vehicle and the information on the factor that causes the movement vector of the other vehicle to change, the information acquired by the acquisition unit, and the accident definition are compared. And a determination unit that determines whether or not there is a possibility of occurrence of an accident.
(Supplementary note 2) Supplementary note 1 further comprising at least one of a display unit that displays a determination result by the determination unit in the vehicle and a change unit that changes the state of the vehicle based on the determination result by the determination unit. The accident occurrence prediction device described.
(Supplementary note 3) The supplementary note 1 or 2, wherein the factor that fluctuates the movement vector of the other vehicle is at least one of an object that obstructs the path of the other vehicle and a natural phenomenon that affects the other vehicle. The accident occurrence prediction device described.
(Additional remark 4) The said accident definition is further defined by the information of the road structure where the said own vehicle and the said other vehicle are located, and the said acquisition part further acquires the information of the road structure where the said own vehicle and the said other vehicle are located The determination unit compares the information acquired by the acquisition unit with the accident definition to determine whether or not an accident may occur. The accident occurrence prediction device according to any one of the above.
(Additional remark 5) The said acquisition part is any one of the sensor mounted in the said own vehicle, the sensor mounted in the said other vehicle, the sensor installed outside the said own vehicle, and the said other vehicle. The accident occurrence prediction apparatus in any one of 1-4.
(Additional remark 6) At least 1 of the said holding | maintenance part, the said acquisition part, and the said determination part is provided outside the said own vehicle, The accident occurrence prediction apparatus in any one of Additional remarks 1-5 characterized by the above-mentioned.
(Supplementary Note 7) In a state where the host vehicle is actually driven, information on the position of the host vehicle and the other vehicle, information on the movement vector of the host vehicle and the other vehicle, and a movement vector of the other vehicle The acquisition unit that acquires information on factors that cause fluctuations, and the information acquired by the acquisition unit and the state in which an accident may occur between the host vehicle and the other vehicle. Is there a possibility that an accident will occur by comparing the position information, the information on the movement vectors of the host vehicle and the other vehicle, and the accident definition defined by the information on the factors that change the movement vector of the other vehicle? An accident occurrence prediction program that functions as a determination unit that determines whether or not.
(Supplementary Note 8) Fluctuating information on relative position of own vehicle and other vehicle, information on movement vector of own vehicle and other vehicle, and movement vector of other vehicle while own vehicle is actually driven An acquisition step of acquiring information on factors to cause, a state where an accident may occur between the own vehicle and the other vehicle, information on a relative position between the own vehicle and the other vehicle, the own vehicle, and the other Is there a possibility that an accident will occur by comparing the accident definition defined by the information on the movement vector of the vehicle and the information on the factor that fluctuates the movement vector of the other vehicle with the information acquired in the acquisition step? An accident occurrence prediction method including a determination step of determining whether or not.

10A, 10B Millimeter wave radar (acquisition unit)
11 Vehicle speed pulse detection unit (acquisition unit)
14 GPS device (acquisition part)
16 determination unit 18 prediction situation DB (holding unit)
100A-100F Accident occurrence prediction device

Claims (6)

The conditions that may accident occurs between the own vehicle and another vehicle, the vehicle and the the free-wheel information of the relative position of the other vehicle is not on the path, on the path of the vehicle and the free-wheel information of the movement vector without another vehicle, and said holding portion for holding an accident definitions defined in the vehicle factors of information to vary the movement vector of the other vehicle is not on the route,
In the state where the host vehicle is actually driven, information on the position of the host vehicle and other vehicles not on the course of the host vehicle, movement vectors of the host vehicle and other vehicles not on the course of the host vehicle , An acquisition unit that acquires information and information on factors that cause movement vectors of other vehicles that are not on the course of the vehicle ;
A determination unit that compares the information acquired by the acquisition unit with the accident definition to determine whether or not there is a possibility of an accident , and
The factor that fluctuates the movement vector of the other vehicle that is not on the course of the own vehicle is the factor that causes the movement vector of the other vehicle that is not on the course of the own vehicle to change to a movement vector in a direction of entering the course of the own vehicle. Accident occurrence prediction device characterized by being . The factors to vary the movement vector of the other vehicle is not on the host vehicle path, the influence on the object and the other vehicle is not on the vehicle of course interfere with the route of another vehicle is not on the vehicle of course 2. The accident occurrence prediction apparatus according to claim 1, wherein the accident occurrence prediction apparatus is at least one of natural phenomena. The acquisition unit is any one of a sensor mounted on the host vehicle, a sensor mounted on another vehicle not on the path of the host vehicle, and a sensor installed outside the host vehicle and outside the other vehicle. The accident occurrence prediction device according to claim 1 or 2.   The accident occurrence prediction device according to any one of claims 1 to 3, wherein at least one of the holding unit, the acquisition unit, and the determination unit is provided outside the host vehicle. Computer
Information on the position of the host vehicle and other vehicles not on the course of the host vehicle , information on movement vectors of the host vehicle and other vehicles not on the course of the host vehicle when the host vehicle is actually driven And an acquisition unit that acquires information on factors that cause movement vectors of other vehicles that are not on the route of the host vehicle, and an accident occurs between the host vehicle and the other vehicle with the information acquired by the acquisition unit. possible a condition, the vehicle and the the free-wheel information of the relative position of the other vehicle is not on the path, the vehicle and the free-wheel information of the movement vector of the other vehicle is not on the path, and the self Compare with the accident definition defined by the information of the factor that fluctuates the movement vector of other vehicles that are not on the course of the car , and function as a determination unit that determines whether there is a possibility of an accident ,
The factor that fluctuates the movement vector of the other vehicle that is not on the course of the own vehicle is the factor that causes the movement vector of the other vehicle that is not on the course of the own vehicle to change to a movement vector in a direction of entering the course of the own vehicle. Accident occurrence prediction program characterized by Information on the position of the host vehicle and other vehicles not on the course of the host vehicle , information on movement vectors of the host vehicle and other vehicles not on the course of the host vehicle when the host vehicle is actually driven And an acquisition step of acquiring information on factors that cause movement vectors of other vehicles not on the course of the host vehicle to fluctuate;
A state in which an accident may occur between the host vehicle and the other vehicle, information on the relative position of the host vehicle and the other vehicle that is not on the path of the host vehicle, the path of the host vehicle and the host vehicle The accident definition defined by the information on the movement vector of the other vehicle that is not present and the information on the factor that fluctuates the movement vector of the other vehicle that is not on the course of the host vehicle is compared with the information acquired in the acquisition step. The computer executes a determination process for determining whether or not an accident may occur ,
The factor that fluctuates the movement vector of the other vehicle that is not on the course of the own vehicle is the factor that causes the movement vector of the other vehicle that is not on the course of the own vehicle to change to a movement vector in a direction of entering the course of the own vehicle. A method for predicting the occurrence of an accident characterized by

Images