JP2017162075A - Vehicle vicinity monitoring device, vehicle vicinity monitoring method, vehicle vicinity monitoring program, and record medium thereof - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a vehicle vicinity monitoring device capable of detecting a state of an object vehicle, and determining a possibility of collision of the object vehicle onto own vehicle based on the state and capable of making a wake-up call or controlling own vehicle, and to provide a vehicle vicinity monitoring method, and a vehicle vicinity monitoring program and a record medium thereof.SOLUTION: The vehicle vicinity monitoring device calculates a movement estimation area of an object vehicle from a camera image signal of the object vehicle, and calculates movement estimation route of own vehicle from the sensor signal. The vehicle vicinity monitoring device determines whether the movement estimation route of own vehicle enters movement possible area of the object vehicle after a certain time. When determining the possibility of entering, the vehicle vicinity monitoring device outputs an alarm signal to prevent a collision.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a vehicle periphery monitoring device, a vehicle periphery monitoring method, and a vehicle periphery monitoring program for calling attention or performing braking such as braking in order to prevent a collision between the host vehicle and a vehicle in the vicinity of the host vehicle. And its recording medium.

  As a prior art of such a vehicle periphery monitoring device, by using the position and relative speed of the target vehicle, when the danger from the rear of the vehicle is evaluated to be greater than the danger in front of the vehicle, the increased braking force There is a control method and device for a brake device that performs a decrease or restricts an increase (see, for example, Patent Document 1).

  In addition, among the straight lines that make up the boundary line between the range that can be seen and the range that cannot be seen, the alerting position that is the end point on the current position side of the intersecting straight line that intersects when the moving body travels in the movement direction is identified. There is an alerting device corresponding to the distance between the current position and the alerting position in the sound image localization direction where the alerting position is located when viewed from the driver of the vehicle, that is, the approach of the target moving object (for example, Patent Document 2) reference).

Japanese Patent No. 4709360 Japanese Patent Laying-Open No. 2015-056786

  However, in both Patent Documents 1 and 2, if the target vehicle enters a positional relationship that cannot be stopped by braking the host vehicle due to sudden start, the subject vehicle may not be able to decelerate in time and may collide. There is.

  The present invention has been made in order to solve the above-described problem. A vehicle periphery monitoring device, a vehicle periphery monitoring method, and a vehicle periphery monitoring method capable of eliminating the possibility of the target vehicle colliding with the host vehicle from the state of the target vehicle. An object is to provide a vehicle periphery monitoring program and its recording medium.

  In order to achieve the above object, a vehicle periphery monitoring apparatus according to the present invention includes a detection unit that calculates a movement estimation area of a target vehicle from a camera video signal, and an own vehicle that calculates a movement estimation route of the host vehicle from a sensor signal. When it is determined that the vehicle state recognition unit, the risk determination unit that determines whether or not the movement estimation route of the own vehicle enters the movement estimation area of the target vehicle after a certain time, and the risk determination unit enter, A warning unit for outputting a warning signal.

  In the present invention, in order to achieve the above object, a movement estimation area of the target vehicle is calculated from the camera video signal, a movement estimation route of the own vehicle is calculated from the sensor signal, and the movement estimation area of the target vehicle is calculated. A vehicle periphery monitoring method is provided that determines whether or not the movement estimation route of the host vehicle enters after a predetermined time, and outputs a warning signal when it is determined that the vehicle enters.

  In order to achieve the above object, the present invention provides a vehicle periphery monitoring program that causes a computer to execute the vehicle periphery monitoring method described above.

  In order to achieve the above object, the present invention provides a computer-readable storage medium storing the vehicle periphery monitoring program described above.

  According to the vehicle periphery monitoring device, the vehicle periphery monitoring method, the vehicle periphery monitoring program, and the recording medium thereof according to the present invention, the movable area of the target vehicle is calculated from the camera video signal of the target vehicle, and is automatically detected from the sensor signal. Since the vehicle movement estimation route is calculated and the vehicle is determined to enter the movement estimation area of the target vehicle after a certain time, a warning signal is output. For example, a collision between the vehicle and the target vehicle during parking Can be prevented in advance.

It is a block diagram which shows the structure common to each embodiment of the vehicle periphery monitoring apparatus which concerns on this invention, the vehicle periphery monitoring method, a vehicle periphery monitoring program, and its recording medium. 2 is a flowchart showing an example of the operation of the vehicle periphery monitoring apparatus, the vehicle periphery monitoring method, the vehicle periphery monitoring program, and the recording medium shown in FIG. 1 according to the first embodiment. It is a top view for demonstrating roughly the relationship between the vehicle periphery monitoring apparatus which concerns on this invention, the vehicle periphery monitoring method, a vehicle periphery monitoring program, and the target vehicle in the recording medium, and the own vehicle. 6 is a flowchart illustrating an example of operation of the vehicle periphery monitoring apparatus, the vehicle periphery monitoring method, the vehicle periphery monitoring program, and the recording medium thereof according to Embodiment 2 shown in FIG. 6 is a flowchart showing an example of operation according to Embodiment 3 of the vehicle periphery monitoring device, the vehicle periphery monitoring method, the vehicle periphery monitoring program, and the recording medium thereof shown in FIG.

  Embodiments of the present invention will be described below with reference to the accompanying drawings. In each embodiment, the vehicle periphery monitoring device will be mainly described. However, the present invention is similarly applied to a vehicle periphery monitoring method, a vehicle periphery monitoring program, and a recording medium thereof. Moreover, although the object vehicle has taken the case where it is located in a parking lot as an example, it is not limited to this case.

Embodiment 1 FIG.
1 includes a camera 2, a microphone 3, a speed sensor 4, a yaw rate sensor 5, an HMI (Human Machine Interface) device 6, and a vehicle. The control unit 7 and the car navigation system 8 are connected. In addition, as will be described later, these constituent elements are not essential for all the embodiments, and are used according to each embodiment.

  Here, the vehicle periphery monitoring device 1 detects the state of the parked vehicle, determines a relative relationship with the host vehicle, and issues a warning. The camera 2 acquires a surrounding image including the front of the host vehicle. The microphone 3 is attached to the host vehicle and records surrounding environmental sounds. The speed sensor 4 detects the speed of the host vehicle. The yaw rate sensor 5 detects the yaw rate of the host vehicle. The HMI device 6 is a speaker, a display, or the like that alerts the driver when a vehicle approach is detected. The vehicle control unit 7 performs vehicle brake control and the like based on the vehicle control signal. The car navigation system 8 provides information on the location where the vehicle is present from GPS or map information.

  The vehicle periphery monitoring device 1 further includes a detection unit 11 that performs image processing on the video acquired by the camera 2, an acoustic processing unit 12 that analyzes environmental sound acquired by the microphone 3, speed information from the speed sensor 4, and From the output information of the own vehicle state recognition unit 13 that estimates the driving state of the own vehicle from the vehicle operation information such as the yaw rate information from the yaw rate sensor 5, the detection unit 11, the acoustic processing unit 12, and the own vehicle state recognition unit 13, A risk determination unit 14 that determines the risk of collision of the host vehicle with the target vehicle, a warning unit 15 that alerts the driver via the HMI device 6 from the determination result of the risk determination unit 14, The control signal generation unit 16 generates a control signal to the vehicle control unit 7 such as a brake device based on the determination result of the danger determination unit 14. Note that output information of the car navigation system 8 is also given to the danger determination unit 14.

  Hereinafter, the operation of the first embodiment will be described based on the flowchart shown in FIG. 2 and the situation shown in FIG. 3 where the vehicle periphery monitoring apparatus is actually assumed to be used. In the first embodiment, an operation in the case where only the detection by the camera 2 is configured will be described.

  The detection part 11 acquires the image | video of the vehicle periphery from the camera 2 attached to the own vehicle 20 (step S1). Here, as shown in FIG. 3, a situation is assumed in which a parked vehicle, that is, a target vehicle 30 exists in the vicinity of the parking lot. Video from the camera 2 is continuously acquired at regular intervals by the detection unit 11 having an image processing function. The period here is preferably an interval that can alert the driver early and can perform image analysis in more detail, such as several tens to several hundreds msec.

  The detection unit 11 further detects the presence / absence of the parked vehicle 30 from the image acquired from the camera 2 (step S2), and notifies the risk determination unit 14 of the position and direction thereof. The detection method in this case can be detected by pattern matching based on pixel values, luminance gradient information and edge information, or may be a learning method or a method using pixel value changes between frames. Alternatively, a method using stereo parallax may be used, both of which are well known in the art.

  Further, when the target vehicle 30 is detected (Yes in step S2), the state of the target vehicle 30 is detected (step S3). That is, here, it is determined whether there is a passenger in the vehicle 30, the wiper attached to the vehicle 30 is operating, the headlight is lit, or the vehicle 30 is vibrating. Whether or not there is a passenger in the target vehicle may be determined by an infrared sensor or the like in the target vehicle. If at least one of these determinations is applicable, it can be determined that the target vehicle 30 is in a movable dangerous state (Yes in step S3), and this is notified to the risk determination unit 14.

  In the danger determination unit 14, the target vehicle 30 in a state notified from the detection unit 11 may start due to sudden acceleration. Therefore, a dangerous area estimated to move from the position and orientation of the target vehicle 30 is selected. Calculate (step S4). The danger area may be a fixed value, or may be dynamically calculated according to the operation state of the target vehicle 30. In the case of the parking lot as shown in FIG. 3, the target vehicle 30 moves straight or moves with its handle cut within a certain angle range, so the danger area is rectangular or fan-shaped. Become.

Further, the danger area of the target vehicle 30 may be changed dynamically from the positional relationship with the host vehicle 20.
For example, when the distance between the host vehicle 20 and the target vehicle 30 is x and the vehicle speed of the host vehicle is v, the time t until the target vehicle 30 is reached is t = x / v.

Assuming for accelerating the target vehicle 30 at a constant acceleration a, the vehicle 20 by using the time t until reach the rear of the target vehicle 30, the movement amount d of the target vehicle 30, d = at ^2/2 It becomes.
Further, when the steering angle of the target vehicle 30 is defined within a certain range. A sector having the radius of d is the danger area of the object.

  On the other hand, the host vehicle state recognition unit 13 acquires the driving state of the host vehicle 20 from various sensors mounted on the host vehicle 20, for example, the speed sensor 4 and the yaw rate sensor 5, as shown in FIG. 14 (step S5). The danger determination unit 14 calculates a movement estimation route of the host vehicle 20 from the driving state information of the host vehicle 20 sent from the host vehicle state recognition unit 13 (step S5).

Then, based on the information on the danger area of the target vehicle 30 and the estimated travel route of the host vehicle 20, the danger determination unit 14 estimates that the target vehicle 30 enters the danger area, that is, enters, after a certain time. (Step S6), and when it is determined that the vehicle will enter after a certain time, the warning unit 15 is notified.
For example, the movement distances x (t) and y (t) that travel after a certain time from the vehicle speed and the yaw rate can be estimated by the following equations.

Receiving this notification, the warning unit 15 performs output such as sound and display via the HMI device 6 to alert the driver (step S7).
Here, not only the warning to the driver but also the control of the host vehicle may be performed by decelerating the vehicle 20 by the vehicle control unit 7 or instructing a steering operation through the control signal generation unit 16.

In the present embodiment, the present invention can be applied not only in the front direction but also in multiple directions, for example, when observing the rear side with the image of the rear camera when moving backward.
Further, when the host vehicle is parked, the danger area may be calculated in consideration of not only the possibility that the target vehicle moves forward but also the possibility that the target vehicle spreads laterally by opening the door.

  Further, in this embodiment, when detecting the state of the target vehicle, the risk level is quantified based on the detection reliability and change timing, and the size of the danger area and the volume of the warning sound are determined according to the risk level. The start timing and strength of the automatic brake may be changed. For example, the time until the predetermined time is numerically expressed, and the shorter the time, the higher the risk can be set.

  Further, for example, it may be excluded from the target vehicle by detecting from the output image of the camera 2 that the passenger in the target vehicle is facing the own vehicle.

  Further, in the present embodiment, the camera 2 has been described using a visible light camera, but a method of detecting a target vehicle and a passenger from a temperature using a far-infrared camera may be employed.

  Through the above processing, whether the target vehicle is movable is detected using a camera, and the possibility of a collision with the own vehicle is obtained to alert the driver, and if necessary, It is possible to ensure safety by performing control.

Embodiment 2. FIG.
The flowchart of FIG. 4 shows an operation example of the vehicle periphery monitoring apparatus 1 according to the second embodiment. In this embodiment, in addition to the detection by the camera 2, ambient sounds are collected by the microphone 3 (step S11). ). Then, it is determined whether or not there is an engine sound in the direction of the target vehicle detected in step S2 (step S12). Thus, the presence of the target vehicle is detected and the reliability is improved.

  Thus, not only the result of the image processing but also the amount of change in the engine sound included in the sound around the host vehicle is analyzed by the acoustic processing unit 12 in the vehicle periphery monitoring device 1, and when the engine sound is present ( In step S12 (Yes), the subsequent steps S4 to S7 are executed for the first time.

  In this case, the time interval at which the acoustic processing unit 12 acquires the ambient sound from the microphone 3 is also an interval in consideration of being able to warn the driver at an early stage and performing more detailed acoustic analysis, such as several tens to several hundreds msec. Is desirable.

Embodiment 3 FIG.
FIG. 5 is a flowchart showing activation conditions as a third embodiment of the vehicle periphery monitoring device according to the present invention. The GPS information and map information that the car navigation system 8 has, that is, the current location information of the host vehicle. (Step S21), and only when the host vehicle is present in the parking lot from the current location information (Yes in step S22), the vehicle periphery monitoring device according to the first and second embodiments shown in FIGS. It starts (step S23).

Instead of using the apparatus for starting up the apparatus, for example, the apparatus may be operated at all times, and the fact that it is a parking lot may be used for calculation of the risk level of the risk determination unit 14.
Moreover, you may set so that a danger determination part may be started only at the time of automatic driving | running | working driving | running | working.

  In each of the above embodiments, the movement estimation area of the target vehicle is calculated from the camera video signal, the movement estimation route of the own vehicle is calculated from the sensor signal, and the own vehicle is added to the movement estimation area of the target vehicle. It is determined whether or not the movement estimation route of the vehicle enters after a certain time, and when the risk determination unit determines that the vehicle enters, a vehicle periphery monitoring method for outputting a warning signal is included.

  Moreover, each said embodiment is provided as a vehicle periphery monitoring program which makes a computer perform the vehicle periphery monitoring method as described above.

  Each of the above-described embodiments is also provided as a computer-readable storage medium that stores the vehicle periphery monitoring program described above.

In order to achieve the above object, a vehicle periphery monitoring device according to the present invention includes a detection unit that calculates a movement estimation area of a target vehicle from a camera video signal, and a host vehicle that acquires a driving state of the host vehicle from a sensor signal. A state recognition unit calculates a movement estimation route of the host vehicle from the driving state of the host vehicle, and determines whether the movement estimation route of the host vehicle enters the movement estimation area of the target vehicle after a predetermined time. A risk determination unit, and a warning unit that outputs a warning signal when the risk determination unit determines that the vehicle is entering, and the risk determination unit has the detection unit having vibration of the target vehicle. When it is detected that the target vehicle has performed a wiper operation, or when the temperature in the target vehicle is detected, the movement estimation route of the host vehicle is calculated .
In addition, a vehicle periphery monitoring apparatus according to the present invention for achieving the above object acquires a driving state of the host vehicle from a detection unit that calculates a movement estimation area of the target vehicle from a camera video signal and a sensor signal. Whether or not the own vehicle state recognition unit calculates the movement estimation route of the own vehicle from the driving state of the own vehicle and whether or not the movement estimation route of the own vehicle enters the movement estimation area of the target vehicle after a predetermined time. A risk determination unit for determining, and a warning unit that outputs a warning signal when the risk determination unit determines that the vehicle is approaching. The risk determination unit includes a wiper operation of the target vehicle. When it is detected that the vehicle has traveled, the movement estimation route of the host vehicle is calculated.
Furthermore, a vehicle periphery monitoring device according to the present invention for achieving the above object acquires a driving state of the host vehicle from a detection unit that calculates a movement estimation area of the target vehicle from a camera video signal and a sensor signal. Whether or not the own vehicle state recognition unit calculates the movement estimation route of the own vehicle from the driving state of the own vehicle and whether or not the movement estimation route of the own vehicle enters the movement estimation area of the target vehicle after a predetermined time. A risk determination unit for determining, and a warning unit that outputs a warning signal when the risk determination unit determines that the vehicle is approaching, and the detection unit detects the temperature in the target vehicle. When this happens, the movement estimation route of the host vehicle is calculated.

In the present invention, the movement estimation area of the target vehicle is calculated from the camera video signal, the driving state of the host vehicle is acquired from the sensor signal, and the movement estimation route of the host vehicle is calculated from the driving state of the host vehicle. In addition, it is determined whether or not the movement estimation route of the host vehicle enters the movement estimation area of the target vehicle after a certain time, and when it is determined that the target vehicle enters, a warning signal is output, and the target vehicle has vibration. The vehicle periphery monitoring method for calculating the movement estimation route of the own vehicle when detecting that the target vehicle has wiped, or detecting the temperature in the target vehicle. Is provided.
In the present invention, the movement estimation area of the target vehicle is calculated from the camera video signal, the driving state of the host vehicle is acquired from the sensor signal, and the movement estimation route of the host vehicle is calculated from the driving state of the host vehicle. At the same time, it is determined whether or not the movement estimation route of the own vehicle enters the movement estimation area of the target vehicle after a predetermined time. When the vehicle is detected, a vehicle periphery monitoring method is provided that calculates the movement estimation route of the host vehicle.
Furthermore, in the present invention, the movement estimation area of the target vehicle is calculated from the camera video signal, the driving state of the host vehicle is acquired from the sensor signal, and the movement estimation route of the host vehicle is calculated from the driving state of the host vehicle. And determining whether or not the movement estimation route of the own vehicle enters the movement estimation area of the target vehicle after a predetermined time, and outputs a warning signal and detects the temperature in the target vehicle when it is determined to enter. When this is done, there is provided a vehicle periphery monitoring method for calculating the movement estimation route of the host vehicle.

Then, based on the information on the danger area of the target vehicle 30 and the estimated travel route of the host vehicle 20, the danger determination unit 14 is estimated that the host vehicle 20 enters the danger area of the target vehicle 30 after a certain time, that is, enters. (Step S6), and when it is determined that the vehicle will enter after a certain time, the warning unit 15 is notified.
For example, the movement distances x (t) and y (t) that travel after a certain time from the vehicle speed and the yaw rate can be estimated by the following equations.

In order to achieve the above object, a vehicle periphery monitoring device according to the present invention includes a detection unit that calculates a movement estimation area of a target vehicle from a camera video signal, and a host vehicle that acquires a driving state of the host vehicle from a sensor signal. A state recognition unit calculates a movement estimation route of the host vehicle from the driving state of the host vehicle, and determines whether the movement estimation route of the host vehicle enters the movement estimation area of the target vehicle after a predetermined time. A risk determination unit, and a warning unit that outputs a warning signal when the risk determination unit determines that the vehicle is entering, and the risk determination unit has the detection unit having vibration of the target vehicle. When it is detected that the target vehicle has performed a wiper operation, or when the temperature in the target vehicle is detected, the movement estimation route of the host vehicle is calculated.
In addition, a vehicle periphery monitoring apparatus according to the present invention for achieving the above object acquires a driving state of the host vehicle from a detection unit that calculates a movement estimation area of the target vehicle from a camera video signal and a sensor signal. Whether or not the own vehicle state recognition unit calculates the movement estimation route of the own vehicle from the driving state of the own vehicle and whether or not the movement estimation route of the own vehicle enters the movement estimation area of the target vehicle after a predetermined time. A risk determination unit for determining, and a warning unit that outputs a warning signal when the risk determination unit determines that the vehicle is approaching. The risk determination unit includes a wiper operation of the target vehicle. When it is detected that the vehicle has traveled, the movement estimation route of the host vehicle is calculated.

In the present invention, the movement estimation area of the target vehicle is calculated from the camera video signal, the driving state of the host vehicle is acquired from the sensor signal, and the movement estimation route of the host vehicle is calculated from the driving state of the host vehicle. In addition, it is determined whether or not the movement estimation route of the host vehicle enters the movement estimation area of the target vehicle after a certain time, and when it is determined that the target vehicle enters, a warning signal is output, and the target vehicle has vibration. The vehicle periphery monitoring method for calculating the movement estimation route of the own vehicle when detecting that the target vehicle has wiped, or detecting the temperature in the target vehicle. Is provided.
In the present invention, the movement estimation area of the target vehicle is calculated from the camera video signal, the driving state of the host vehicle is acquired from the sensor signal, and the movement estimation route of the host vehicle is calculated from the driving state of the host vehicle. At the same time, it is determined whether or not the movement estimation route of the host vehicle enters the movement estimation area of the target vehicle after a predetermined time. When it is determined that the target vehicle enters, a warning signal is output and the wiper operation of the target vehicle is performed. When the vehicle is detected, a vehicle periphery monitoring method is provided that calculates the movement estimation route of the host vehicle.

Claims (18)

A detection unit that calculates a movement estimation area of the target vehicle from the camera video signal;
A host vehicle state recognition unit that calculates a movement estimation route of the host vehicle from the sensor signal;
A risk determination unit that determines whether or not the movement estimation route of the host vehicle enters the movement estimation area of the target vehicle after a predetermined time;
A vehicle periphery monitoring device comprising: a warning unit that outputs a warning signal when the danger determination unit determines that the vehicle enters. The vehicle periphery monitoring device according to claim 1, wherein the detection unit calculates a movement area of the host vehicle when a person in the target vehicle is specified. The vehicle periphery monitoring device according to claim 1, wherein the detection unit calculates a movement area of the host vehicle when a face orientation of a person in the target vehicle is specified. The vehicle periphery monitoring device according to claim 1, wherein the detection unit calculates a movement area of the host vehicle when detecting that the vibration of the target vehicle is present. The vehicle periphery monitoring device according to claim 1, wherein the detection unit calculates a movement area of the host vehicle when detecting that the target vehicle has a wiper operation. The vehicle periphery monitoring device according to claim 1, wherein the detection unit calculates a movement area of the host vehicle when detecting that a light of the target vehicle is turned on. The vehicle periphery monitoring device according to claim 1, wherein the detection unit calculates a movement area of the host vehicle when detecting a temperature in the target vehicle. The vehicle periphery monitoring device according to claim 1, wherein the warning unit warns the driver with a warning sound or a warning video. The vehicle periphery monitoring device according to claim 1, wherein the risk determination unit performs determination based on a risk level set by quantifying the predetermined time. The vehicle periphery monitoring apparatus according to claim 1, wherein the danger determination unit is activated only during automatic driving. The vehicle periphery monitoring apparatus according to claim 1, further comprising a control signal generation unit that outputs an automatic brake signal to a vehicle control unit that reduces the speed of the host vehicle when the warning signal is output. The vehicle periphery monitoring device according to claim 1, further comprising a driver monitoring device that determines whether or not the driver of the host vehicle pays attention to the target vehicle. The vehicle periphery monitoring device according to claim 1, wherein the sensor signal is an output signal from a speed sensor and a yaw rate sensor. It further includes an acoustic processing unit that determines whether or not the target vehicle exists based on ambient sound from a microphone and gives the target vehicle to the risk determination unit, and the risk determination unit is configured on the condition that the ambient sound is detected. The vehicle periphery monitoring device according to claim 1, wherein the determination is performed. The vehicle periphery monitoring device according to claim 1, wherein the danger determination unit performs the determination only when the host vehicle is located in a parking lot based on current location information of the host vehicle from a car navigation system. Calculate the movement estimation area of the target vehicle from the camera video signal,
From the sensor signal, calculate the movement estimation route of the vehicle,
Determining whether the movement estimation route of the subject vehicle enters the movement estimation area of the target vehicle after a predetermined time;
A vehicle periphery monitoring method that outputs a warning signal when it is determined that the vehicle is approaching.   A vehicle periphery monitoring program for causing a computer to execute the vehicle periphery monitoring method according to claim 16.   A computer-readable storage medium storing the vehicle periphery monitoring program according to claim 17.

Images