JP2008027309A - Collision determination system and collision determination method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a collision determination system and a collision determination method for hardly receiving the influence of the calculation error of the detection position of an obstacle, and for precisely determining collision with the obstacle whose separation from a risk level based on the point of view of a driver is small. <P>SOLUTION: Image data imaged by a plurality of imaging apparatuses which image the periphery of a vehicle are acquired, and the presence of the obstacle in an image is detected, and the relative position data of the obstacle to the vehicle are time sequentially detected, so that the possibility of collision can be determined. The peripheral region of the vehicle is divided into a plurality of regions on the basis of the relative position to the vehicle, and a parameter showing the scale of the possibility of collision is stored in each of the divided regions. The corresponding region is extracted on the basis of the detected relative position data, and the parameter corresponding to the extracted region is read, so that the possibility of collision is determined. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention determines the possibility of a collision with an obstacle based on image data captured by a plurality of imaging devices that capture the outside of the vehicle according to the relative position of the detected obstacle with the vehicle. The present invention relates to a collision determination system and a collision determination method.

  An imaging device that captures the surroundings is mounted on a vehicle such as an automobile, and the possibility of a collision with the host vehicle is detected by recognizing the presence of an obstacle such as a pedestrian or bicycle in front of the vehicle and detecting the movement of the obstacle. A collision determination system for determining the above has been developed.

  For example, from an image captured by an imaging device, an area where an obstacle, for example, a human being is present, is extracted by pattern matching with a predetermined reference pattern, and the distance to the area and a relative movement vector with respect to the vehicle are calculated by stereo vision By doing so, the possibility of collision with an obstacle is determined (see Non-Patent Document 1).

And in order to improve the calculation accuracy of the possibility of collision, for example, in Patent Document 1, the actual relative position with the obstacle included in the image data acquired by the imaging means is calculated in time series, and the calculated actual relative By approximating the position linearly, a relative movement vector of the obstacle with the host vehicle is calculated, and the possibility of collision with the host vehicle is determined.
JP 2001-006096 A "Honda R & D Technical Review" Vol.13 No.1, April 2001

  The movement vector in the above-described conventional collision determination system is easily affected by an error in distance measurement. A calculation error is also included when the position of an obstacle whose distance has been measured by calculating parallax is converted into three-dimensional coordinates. Therefore, when the presence of an obstacle such as a pedestrian is detected in the course of the host vehicle, it is difficult to increase the reliability of the movement vector, and it is difficult to increase the collision determination accuracy. .

  In addition, since the movement vector is calculated based on the past movement trajectory of an obstacle such as a pedestrian, sudden movement such as sudden jump-out cannot be predicted. Therefore, it is difficult to detect movement in a short time such as several frames, and the reliability of collision determination is low.

  Furthermore, the result of determining the possibility of a collision based on the movement vector is different from the sensory risk determination result based on the viewpoint of the driver of the vehicle. The problem that is large is also left behind.

  The present invention has been made in view of such circumstances, and is not easily affected by the calculation error of the detected position of the obstacle, and makes a collision determination with the obstacle with little deviation from the risk based on the driver's viewpoint. It is an object of the present invention to provide a collision determination system and a collision determination method that can be performed with high accuracy.

  In order to achieve the above object, a collision determination system according to a first aspect of the present invention provides a plurality of imaging devices that capture the periphery of a vehicle, acquires image data captured by the plurality of imaging devices, and detects the presence of an obstacle in the image. Obstacle detection means to detect, and when detecting an obstacle by the obstacle detection means, a relative position data for detecting relative position data of the obstacle with respect to the vehicle in time series from image data acquired by the plurality of imaging devices In a collision determination system that includes a position detection unit and determines the possibility of a collision with a detected obstacle, a vehicle peripheral region is divided into a plurality of regions based on a relative position with respect to the vehicle. Means for storing a parameter indicating the possibility of collision for each area; means for extracting a corresponding area based on the detected relative position data; and a parameter corresponding to the extracted area. And means for reading over data, characterized in that based on the read parameters are to be determined the possibility of a collision.

  According to a second aspect of the present invention, the collision determination system according to the first aspect includes a means for detecting the speed of the vehicle and a means for deforming the shape of the divided area according to the detected speed of the vehicle. It is characterized by.

  Further, the collision determination system according to the third aspect of the present invention is the first or second aspect of the invention, wherein the means for judging whether or not the vehicle is turning, and if it is judged that the means is turning, the turning direction and turning It is characterized by comprising means for calculating a radius and means for deforming the shape of the divided area in accordance with the calculated turning direction and turning radius.

  According to a fourth aspect of the present invention, there is provided the collision determination system according to any one of the first to third aspects, wherein the parameters corresponding to the extracted area are read out in time series. A means for storing in a time series and a means for calculating an index value indicating the possibility of a collision according to a time-series change of a parameter are determined, and the possibility of a collision is determined based on the calculated index value. It is made to do so.

  Moreover, the collision determination method according to the fifth aspect of the present invention acquires image data captured by a plurality of imaging devices that capture the periphery of the vehicle, detects the presence of an obstacle in the image, and detects an obstacle. In the collision determination method for detecting the relative position data of the obstacle with respect to the vehicle in time series from the image data acquired by the plurality of imaging devices and determining the possibility of the collision, the vehicle is based on the relative position with respect to the vehicle. The surrounding area is divided into a plurality of areas, parameters indicating the possibility of collision are stored for each divided area, and the corresponding area is extracted and extracted based on the detected relative position data. It is characterized in that a parameter corresponding to the selected area is read out and the possibility of collision is determined based on the read out parameter.

  In the first invention and the fifth invention, image data captured by a plurality of imaging devices that capture the periphery of the vehicle is acquired to detect the presence of an obstacle in the image. When an obstacle is detected, relative position data of the obstacle with respect to the vehicle is detected in time series from image data acquired by a plurality of imaging devices, and the possibility of a collision is determined. Based on the relative position with respect to the vehicle, the peripheral region of the vehicle is divided into a plurality of regions, and a parameter indicating the possibility of a collision is stored for each of the divided regions. A corresponding region is extracted based on the detected relative position data, a parameter corresponding to the extracted region is read, and the possibility of collision is determined based on the read parameter. The possibility of collision, that is, the degree of danger, can be determined according to the area corresponding to the position where the obstacle actually exists, and even if there is an error in the distance measurement, the result of the danger degree judgment has an effect. It is hard to receive. In addition, even if there is a sudden movement such as a sudden jump of an obstacle such as a pedestrian, the degree of danger can be determined according to whether or not it exists in a predetermined area. It is possible to maintain a high reliability. Furthermore, by storing a parameter indicating the possibility of collision for each divided area in accordance with the sensory risk level based on the driver's viewpoint, the risk level actually felt by the driver Can be kept to a minimum, and a collision determination that matches the driver's feeling can be performed.

  In the second invention, the speed of the vehicle is detected, and the shape of the divided area is deformed according to the detected speed of the vehicle. By doing in this way, the risk determination of collision with an obstacle can be dynamically changed according to the traveling speed of the vehicle, and the position where the possibility of collision is low when the vehicle speed is low Even if the obstacle is detected at, it is possible to determine that the possibility of a collision is high when the speed of the vehicle increases.

  In the third invention, it is determined whether or not the vehicle is turning. If it is determined that the vehicle is turning, the turning direction and the turning radius are calculated, and the shape of the divided area is deformed. In this way, it is possible to dynamically change the risk determination of a collision with an obstacle depending on whether or not the vehicle is turning. For example, when the vehicle is traveling straight, a collision is possible. Even if the obstacle is detected at a position with a low probability, it is possible to determine that the possibility of a collision is high when the vehicle turns. Further, by using it together with the second invention, it becomes possible to perform a more accurate collision determination according to the traveling state of the vehicle.

  In the fourth invention, the read parameters are stored in time series, and an index value indicating the possibility of collision is calculated according to the time series changes of the parameters. The possibility of a collision is determined based on the calculated index value. By doing in this way, it is possible to prevent an erroneous collision possibility determination due to an estimation error of the distance to the obstacle, the traveling state of the vehicle, etc., and an appropriate risk determination according to the movement of the obstacle. Can be performed.

  According to the present invention, the possibility of collision, that is, the degree of danger can be determined according to the area corresponding to the position where the obstacle is actually present, and the degree of danger even if there is an error in distance measurement. The result of the judgment is not easily affected. In addition, even if there is a sudden movement such as a sudden jump of an obstacle such as a pedestrian, the degree of danger can be determined according to whether or not it exists in a predetermined area. It is possible to maintain a high reliability. Furthermore, by storing a parameter indicating the possibility of collision for each divided area in accordance with the sensory risk level based on the driver's viewpoint, the risk level actually felt by the driver Can be kept to a minimum, and a collision determination that matches the driver's feeling can be performed.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following embodiments, an example will be described in which the distance to an obstacle existing in front of a vehicle is calculated based on an image captured by a far-infrared imaging device during traveling.

(Embodiment 1)
FIG. 1 is a schematic diagram showing a configuration of a collision determination system according to Embodiment 1 of the present invention. In the first embodiment, far-infrared imaging devices 1 and 1 that capture surrounding images while traveling are juxtaposed in a front grill near the center in front of the vehicle. The far-infrared imaging devices 1 and 1 are imaging devices using infrared light having a wavelength of 7 to 14 micrometers. Image data captured by the far-infrared imaging devices 1 and 1 is transmitted to a determination device 3 connected via a video cable 7 compatible with an analog video system such as NTSC or a digital video system.

  The determination device 3 is connected to the far infrared imaging devices 1 and 1 and the display device 4 having an operation unit via a cable 8 corresponding to a video system such as NTSC, VGA, DVI, etc. An output device such as an alarm device 5 that emits an audible warning by a sound effect or the like is connected via an in-vehicle LAN cable 6 compliant with CAN.

  FIG. 2 is a block diagram showing the configuration of the far-infrared imaging device 1 of the collision determination system according to Embodiment 1 of the present invention. The image pickup unit 11 includes image pickup elements that convert optical signals into electric signals in a matrix. As an imaging device for infrared light, an infrared sensor such as a vanadium oxide bolometer type using a micromachining technique or a BST (Barium-Strontium-Titanium) pyroelectric type sensor is used. The image capturing unit 11 reads an infrared light image around the vehicle as a luminance signal, and transmits the read luminance signal to the signal processing unit 12.

  The signal processing unit 12 is an LSI, converts the luminance signal received from the image capturing unit 11 into a digital signal, performs processing for correcting variations in the image sensor, correction processing for defective elements, gain control processing, and the like, and performs image data processing. Is stored in the image memory 13. Needless to say, it is not essential to temporarily store the image data in the image memory 13, and the image data may be transmitted directly to the determination device 3 via the video output unit 14.

  The video output unit 14 is an LSI, and outputs video data to the determination device 3 via a video cable 7 compatible with an analog video system such as NTSC or a digital video system.

  FIG. 3 is a block diagram showing a configuration of the determination device 3 of the obstacle position calculation system according to Embodiment 1 of the present invention. The video input unit 31 a inputs video signals from the far infrared imaging devices 1 and 1. The video input unit 31a stores the image data input from the far-infrared imaging devices 1 and 1 in the image memory 32 in synchronization with each frame. The video output unit 31 b outputs image data to the display device 4 such as a liquid crystal display via the cable 8, and the communication interface unit 31 c is connected to the alarm device 5 such as a buzzer and a speaker via the in-vehicle LAN cable 6. In contrast, an output signal such as a synthesized sound is transmitted.

  The image memory 32 is an SRAM, flash memory, SDRAM or the like, and stores image data input from the far-infrared imaging devices 1 and 1 via the video input unit 31a. The RAM 331 stores data generated during the arithmetic processing and the estimated time-series position data of the obstacle. Further, the RAM 331 stores area information obtained by dividing an area around the vehicle into a plurality of areas in association with a risk parameter indicating the risk of collision.

  The LSI 33 that performs image processing reads out the image data stored in the image memory 32 in units of frames, and specifies a target region for which parallax is calculated based on the image data acquired from the far-infrared imaging devices 1 and 1. . The LSI 33 calculates the parallax of the left and right image data acquired from the far-infrared imaging devices 1 and 1 for the target region, and estimates the distance and direction to the target based on the principle of triangulation.

  Based on the estimated position data, the LSI 33 determines which of the surrounding areas of the vehicle it belongs to, and extracts the risk parameter stored in the RAM 331 in association with the belonging area. It is possible to notify the driver of appropriate information by changing the content of the warning information transmitted to the outside according to the extracted risk parameter. That is, information indicating that the danger is imminent when the risk parameter is larger than a predetermined value, and simple warning information when the risk parameter is less than the predetermined value, can be used properly.

  Detailed processing in the LSI 33 will be described below. FIG. 4 is a flowchart showing a procedure of risk extraction processing of the LSI 33 of the determination device 3 of the collision determination system according to the first embodiment of the present invention.

  The LSI 33 reads the image data stored in the image memory 32 (step S401), and divides the read image data into target areas of a predetermined size (step S402). The method for setting the target area to be divided is not particularly limited. For example, you may set as a square area | region which consists of m * n (m and n are natural numbers) pixels. In the first embodiment, the target area is set as a square area composed of 3 × 3 pixels.

  The LSI 33 calculates a correlation value in units of images for each set target area (step S403). For example, with the target region set based on the image captured by the left far infrared imaging device 1 as a reference, the same shape and the same size as the target region in the image captured by the right far infrared imaging device 1 A correlation value with the region group is calculated. The region group having the same shape and the same size as the target region means a region group in the case of parallax = 1, 2,..., N (n is a natural number), for example.

  When the LSI 33 calculates the correlation value R for each target area in image units, the correlation value R is calculated based on (Equation 1).

  In (Expression 1), N is the total number of pixels in the region to be matched, k is an integer of 0 ≦ k ≦ (N−1), and Fk is based on an image captured by the left far-infrared imaging device 1. Gk is the pixel value of the kth pixel in the set target area, and Gk is k in the area group having the same shape and the same size as the target area in the image captured by the right far-infrared imaging device 1. The pixel values of the second pixel are shown.

  The LSI 33 extracts the parallax having the maximum calculated correlation value (step S404). That is, there is a high possibility that an obstacle exists at the parallax with the maximum calculated correlation value.

  The LSI 33 specifies the target area having the maximum extracted correlation value as an obstacle candidate area (step S405). The LSI 33 estimates the actual position of the representative point of the identified obstacle candidate area (step S406). For example, the center of gravity of the obstacle candidate area is set as the representative point.

  Specifically, the LSI 33 determines the vehicle front center point and the detected obstacle based on the parallax of the obstacle candidate area corresponding to the images captured by the left and right far-infrared imaging devices 1 and 1. And the angle from the vehicle front center point indicating the direction of the obstacle. The angle indicating the direction is calculated as a positive angle on the left side and a negative angle on the right side with respect to the traveling direction of the vehicle.

  The LSI 33 converts the relative position of the obstacle from the front of the vehicle into a three-dimensional coordinate value based on the calculated distance to the obstacle and the direction in which the obstacle exists (step S407). The coordinate system for calculating the three-dimensional coordinate value is a coordinate system in which the center of the front surface of the vehicle is the origin, the forward direction is the X-axis positive direction, the right direction is the Y-axis positive direction, and the upper direction is the Z-axis positive direction.

  The LSI 33 inquires of the area information stored in the RAM 331 based on the converted three-dimensional coordinate value, and extracts a risk parameter indicating the risk of collision (step S408). FIG. 5 is a diagram schematically showing the contents of the area information stored in the RAM 331.

  As shown in FIG. 5, the front area of the vehicle is divided into a plurality of areas in a predetermined area. In the example of FIG. 5, it is divided into three in the X-axis direction according to the distance ahead. That is, it is divided into three regions: a region 51 from the front end of the vehicle to X1 (m), a region 52 from X1 (m) to X2 (m), and a region 53 from X2 (m).

  A region 51 from the front end of the vehicle to X1 (m) indicates a region at a close distance that is difficult to avoid even when the distance is estimated by stereo viewing and the warning information is notified. In the area 51, the situation does not change so much and the accuracy of the detected obstacle is high.

  A region 52 from X1 (m) to X2 (m) indicates a region where an appropriate collision determination can be performed. In the region 52, there is a relatively low possibility that the situation will change, and it is relatively easy to confirm that the detected object is an obstacle.

  A region 53 equal to or greater than X2 (m) indicates a region in which it is difficult to determine whether or not the detected object is an obstacle. In the area 53, the situation is highly likely to change, and even if it is detected as an obstacle, it cannot be guaranteed what the detected object is and which should be warned as an obstacle. .

  On the other hand, in the Y-axis direction, the vehicle is divided into four regions based on the vehicle width, the legal safety width, the width obtained by adding an integral multiple of the vehicle width, and the like. That is, the region 61 within the vehicle width range, the legal safety width with respect to the vehicle width, for example, the region 62 within the range of 1.5 m from the vehicle width, the region 63 obtained by adding the vehicle width to the region 62, and the region 63 The area is divided into the areas 64 to which the minutes are added.

  Then, an area 71 which is a risk parameter “3” indicating that the degree of risk is highest according to an area where the three areas in the X-axis direction and the four areas in the Y-axis direction intersect with each other is represented by the risk parameter “2”. A certain region 72, a region 73 having a risk parameter “1”, and a region 74 having a risk parameter “0” are set and stored in the RAM 331, respectively. The LSI 33 inquires in which area the converted three-dimensional coordinate value exists and extracts the risk parameter of the corresponding area.

  An arithmetic error occurs when converting to a three-dimensional coordinate value. Therefore, by displaying the area stored in the RAM 331 on the screen and specifying the area in association with the position of the detected obstacle, the risk parameter can be extracted with higher accuracy.

  The LSI 33 sends out warning information based on the extracted risk parameter (step S409). For example, when the risk parameter shown in FIG. 5 is “3”, a signal indicating that the degree of urgency for avoiding a collision with an obstacle is high is output to the display device 4 or the alarm device 5. The display device 4 outputs a warning display such as outputting a display indicating that the degree of urgency for avoiding a collision with an obstacle is high or blinking the warning display. The warning device 5 notifies the driver that the degree of urgency for avoiding a collision with an obstacle is high by a buzzer sounding or the like. Depending on the risk parameter, the warning display is changed, and the sounding of the alarm device 5 is changed. By doing so, the driver can determine the risk of collision with the detected obstacle.

  The first embodiment described above is effective only when the vehicle is traveling straight at a constant speed, and cannot be applied as it is when the vehicle speed changes, the vehicle is turning, or the like. . Therefore, the shape (coordinate value) of the divided area stored in the RAM 331 is changed according to the speed of the vehicle and / or whether the vehicle is turning. The speed of the vehicle is detected based on a detection value of a speed sensor that detects the speed of the vehicle. Whether or not the vehicle is turning is detected based on detection values of sensors that detect the steering angle of the steering wheel, the rotation angle of the steering shaft, and the like.

  FIG. 6 is a diagram illustrating changes in the division positions X1 and X2 in the X-axis direction of the divided area stored in the RAM 331 in accordance with changes in the vehicle speed. In the example of FIG. 6, the boundary value X1 is divided into three regions: a region 51 from the front end of the vehicle to X1 (m), a region 52 from X1 (m) to X2 (m), and a region 53 from X2 (m), X2 is changed according to the speed of the vehicle.

  In the example of FIG. 6, X1 is a constant value 15 (m) until the vehicle speed V (km / h) is 30 (km / h). X1 gradually increases after exceeding 30 (km / h), and is set to be 25 (m) at 40 (km / h). X2 is the vehicle speed V (km / h) is 0 (km / h), that is, 30 (m) when the vehicle is stopped, 55 (m) at a speed of 30 (km / h), and 40 (km / h) It is set to be 70 (m). Note that the change of the division positions X1 and X2 in the X-axis direction of the region is not limited to the example of FIG. 6, and any method may be used as long as it is changed according to the speed.

  FIG. 7 is a schematic diagram showing a change state of the area stored in the RAM 331 when the vehicle is turning right at the turning radius R. As shown in FIG. 7, a region 71 that is a risk parameter “3” indicating that the risk is the highest, a region 72 that is a risk parameter “2”, and a region 73 that is a risk parameter “1”, respectively. It is deformed. The determination of the risk of collision is performed by extracting a risk parameter of an area where the detected three-dimensional coordinate value of the obstacle exists based on the deformed area. By doing in this way, it becomes possible to determine the danger of collision with an obstacle dynamically according to the speed of the vehicle and / or whether the vehicle is turning.

  As described above, according to the first embodiment, the possibility of collision, that is, the degree of danger can be determined according to the area corresponding to the position where the obstacle actually exists, and there is an error in distance measurement. Even in such a case, the result of the risk determination is not easily affected. In addition, even if there is a sudden movement such as a sudden jump of an obstacle such as a pedestrian, the degree of danger can be determined according to whether or not it exists in a predetermined area. It is possible to maintain a high reliability. Furthermore, by storing a parameter indicating the possibility of collision for each divided area in accordance with the sensory risk level based on the driver's viewpoint, the risk level actually felt by the driver Can be kept to a minimum, and a collision determination that matches the driver's feeling can be performed.

(Embodiment 2)
Hereinafter, a collision determination system according to Embodiment 2 of the present invention will be described in detail with reference to the drawings. Since the configuration of the collision determination system according to the second embodiment, the configuration of the far-infrared imaging device 1, and the configuration of the determination device 3 are the same as those of the first embodiment, detailed description is omitted by attaching the same reference numerals. To do. The second embodiment is different from the first embodiment in that the risk parameter is extracted based on the detected movement transition of the obstacle.

  That is, the LSI 33 that performs image processing reads out the image data stored in the image memory 32 in units of frames, and specifies a target region for which parallax is calculated based on the image data acquired from the far-infrared imaging devices 1 and 1. To do. The LSI 33 calculates the parallax of the left and right image data acquired from the far-infrared imaging devices 1 and 1 for the target region, and estimates the distance and direction to the target based on the principle of triangulation.

  The RAM 331 stores the past relative position data of the detected obstacle and the area to which the relative position belongs in time series. Based on the estimated position data, the LSI 33 determines which of the surrounding areas of the vehicle it belongs to, and identifies past movement transitions. The LSI 33 extracts the risk parameter stored in the RAM 331 based on the specified movement transition. It is possible to notify the driver of appropriate information by changing the content of the warning information transmitted to the outside according to the extracted risk parameter. That is, information indicating that the danger is imminent when the risk parameter is larger than a predetermined value, and simple warning information when the risk parameter is less than the predetermined value, can be used properly.

  Detailed processing in the LSI 33 will be described below. FIG. 8 is a flowchart showing the procedure of risk extraction processing of the LSI 33 of the determination device 3 of the collision determination system according to Embodiment 2 of the present invention.

  The LSI 33 reads the image data stored in the image memory 32 (step S801), and divides the read image data into target areas of a predetermined size (step S802). The method for setting the target area to be divided is not particularly limited. For example, you may set as a square area | region which consists of m * n (m and n are natural numbers) pixels. In the first embodiment, the target area is set as a square area composed of 3 × 3 pixels.

  The LSI 33 calculates a correlation value in units of images for each set target area (step S803). For example, with the target region set based on the image captured by the left far infrared imaging device 1 as a reference, the same shape and the same size as the target region in the image captured by the right far infrared imaging device 1 A correlation value with the region group is calculated. The region group having the same shape and the same size as the target region means a region group in the case of parallax = 1, 2,..., N (n is a natural number), for example.

  The LSI 33 extracts the parallax having the maximum correlation value calculated for each parallax (step S804). That is, there is a high possibility that an obstacle exists with the parallax having the maximum calculated correlation value.

  The LSI 33 specifies the target area having the maximum correlation value as an obstacle candidate area (step S805). The LSI 33 estimates the actual position of the representative point of the specified obstacle candidate area (step S806). For example, the center of gravity of the obstacle candidate area is set as the representative point.

  Specifically, the LSI 33 determines the vehicle front center point and the detected obstacle based on the parallax of the obstacle candidate area corresponding to the images captured by the left and right far-infrared imaging devices 1 and 1. And the angle from the vehicle front center point indicating the direction of the obstacle. The angle indicating the direction is calculated as a positive angle on the left side and a negative angle on the right side with respect to the traveling direction of the vehicle.

  The LSI 33 converts the relative position of the obstacle from the front of the vehicle into a three-dimensional coordinate value based on the calculated distance to the obstacle and the direction in which the obstacle exists (step S807). The coordinate system for calculating the three-dimensional coordinate value is a coordinate system in which the center of the front surface of the vehicle is the origin, the forward direction is the X-axis positive direction, the right direction is the Y-axis positive direction, and the upper direction is the Z-axis positive direction.

  The LSI 33 reads the past three-dimensional coordinate value of the detected obstacle from the RAM 331, and specifies the movement transition (step S808). FIG. 9 is a diagram schematically showing the contents of the area information stored in the RAM 331, as in FIG.

  In the example of FIG. 9, like FIG. 5, it is divided into three in the X-axis direction according to the distance in front of the vehicle. A region from the front end of the vehicle to X1 (m) is a region N, a region from X1 (m) to X2 (m) is a region M, and a region from X2 (m) is X.

  On the other hand, in the Y-axis direction, the vehicle is divided into four regions based on the vehicle width, the legal safety width, the width obtained by adding an integral multiple of the vehicle width, and the like. The area within the range of the vehicle width is the area C0, the legal safety width with respect to the vehicle width, for example, the area from the vehicle width to the range of 1.5 m is outside the areas L1, R1, L1, R1, and the vehicle width A region obtained by adding the vehicle width to the regions L2, R2, and the regions L2, R2 is defined as regions L3, R3.

  Then, an area 71 which is a risk parameter “3” indicating that the degree of risk is highest according to an area where the three areas in the X-axis direction and the four areas in the Y-axis direction intersect with each other is represented by the risk parameter “2”. A certain region 72, a region 73 having a risk parameter “1”, and a region 74 having a risk parameter “0” are set. In the RAM 331, the transition of the area to which the current position of the detected obstacle, the position detected once before, the position detected twice, and the position detected twice before belong as area coordinates in the X-axis direction and the Y-axis direction. I remember it. The RAM 331 stores a risk parameter in association with the detected movement transition of the obstacle.

  FIG. 10 is a diagram illustrating an example of the risk parameter associated with the movement transition stored in the RAM 331. The position of the detected obstacle is stored in the area coordinates of FIG. 9, for example, (M, C0) is an area from X1 (m) to X2 (m) in the X-axis direction, and the Y-axis An area within the range of the vehicle width is shown in the direction. The LSI 33 extracts the risk parameter of the area corresponding to the specified movement transition (step S809).

  The LSI 33 sends out warning information based on the extracted risk parameter (step S810). For example, when the risk parameter shown in FIG. 10 is “5”, a signal indicating that the degree of urgency for avoiding a collision with an obstacle is high is output to the display device 4 or the alarm device 5. The display device 4 outputs a warning display such as outputting a display indicating that the degree of urgency for avoiding a collision with an obstacle is high or blinking the warning display. The warning device 5 notifies the driver that the degree of urgency for avoiding a collision with an obstacle is high by a buzzer sounding or the like. Depending on the risk parameter, the warning display is changed, and the sounding of the alarm device 5 is changed. By doing so, the driver can determine the risk of collision with the detected obstacle.

  As in the first embodiment, the shape (coordinate values) of the divided areas stored in the RAM 331 is changed according to the speed of the vehicle and / or whether the vehicle is turning or not. Therefore, it is possible to determine the risk of collision with an obstacle.

  As described above, according to the second embodiment, it is possible to prevent an erroneous collision possibility determination due to the estimation error of the distance to the obstacle, the traveling state of the vehicle, and the like according to the movement of the obstacle. Appropriate risk determination can be performed.

  In addition to the first and second embodiments described above, by using various sensors that detect the surrounding environment where the vehicle travels, by changing the setting of the area and the setting of the risk parameter, collision determination corresponding to various situations can be performed. It can be carried out. For example, map information is acquired via GPS, etc., based on information such as whether the road on which it is traveling is an urban area, a suburb, or an accident-prone area, etc. Change the parameter setting.

  Specifically, when it is determined that the vehicle is traveling in an urban area, the risk parameter is increased in order to increase the detection sensitivity of a pedestrian or the like. In addition, when performing collision determination based on movement transition, it is possible to alleviate the driver's annoyance due to frequent warning notifications by suppressing the risk parameter to a low level until entering the dangerous area.

  On the other hand, if it is determined that the vehicle is traveling on a highway at a high speed and with a small speed fluctuation range, the probability of an obstacle being present is lower than when driving on a normal road, so an alarm will be given as soon as an obstacle is detected. All risk parameters are set to their maximum values so that information is transmitted. By doing in this way, it becomes possible to notify a driver | operator early at the time of detecting an obstacle, and generation | occurrence | production of an accident can be avoided beforehand.

  Also, region information associated with a risk parameter indicating the risk of collision by estimating the road width based on map information acquired via GPS or the like, or a white line position detected by a white line detection process, etc. You can also change the classification. In general, roads with wide roads often run with high speed and little speed fluctuation. Therefore, as in the case where it is determined that the vehicle is traveling on an expressway or the like, all the risk parameters are set to the maximum value so that warning information is transmitted as soon as an obstacle is detected. By doing in this way, it becomes possible to notify a driver | operator early at the time of detecting an obstacle, and generation | occurrence | production of an accident can be avoided beforehand.

  In addition, based on the detection value of the raindrop sensor on the front window, information indicating the operation status of the wiper, information indicating the lighting status of the fog lamp, information on the weather acquired through a communication system with the outside of the vehicle such as a mobile phone, etc. The region setting and the risk parameter setting may be changed. For example, during rainy weather, snowfall, etc., the field of view is narrow and the stopping distance is extended. Therefore, the risk parameter of the proximity region is increased, and the region fluctuation range with respect to the vehicle speed is increased. By doing in this way, the collision determination with an obstacle can be performed more safely, and it is effective also from the viewpoint of ensuring safety.

  In addition, a sensor that detects biological information such as the driver's pulse and the surface temperature of the face may be provided, and warning information may be transmitted earlier by grasping the driver's arousal state. Of course, since the experience level of vehicle driving is different for each driver, it goes without saying that the setting of the above-described region and the setting of the risk parameter may be varied for each driver.

  In the above-described embodiment, the LSI 33 of the determination device 3 performs the above-described control. However, a separate control device may be provided, or a control device of another device may also be used.

It is a schematic diagram which shows the structure of the collision determination system which concerns on Embodiment 1 of this invention. It is a block diagram which shows the structure of the far-infrared imaging device of the collision determination system which concerns on Embodiment 1 of this invention. It is a block diagram which shows the structure of the determination apparatus of the obstacle position calculation system which concerns on Embodiment 1 of this invention. It is a flowchart which shows the procedure of the risk extraction process of LSI of the determination apparatus of the collision determination system which concerns on Embodiment 1 of this invention. It is a figure which shows typically the content of the area | region information memorize | stored in RAM. It is a figure which shows the change of the division position of the X-axis direction of the divided | segmented area | region memorize | stored in RAM according to the change of the speed of a vehicle. It is a schematic diagram which shows the change state of the area | region memorize | stored in RAM when the vehicle is turning right with the turning radius R. It is a flowchart which shows the procedure of the risk extraction process of LSI of the determination apparatus of the collision determination system which concerns on Embodiment 2 of this invention. It is a figure which shows typically the content of the area | region information memorize | stored in RAM. It is a figure which shows an example of the risk parameter matched with the movement transition memorize | stored in RAM.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Far-infrared imaging device 3 Judgment device 4 Display device 5 Alarm device 31a Image | video input part 31b Image | video output part 31c Image | video output part 32 Image memory 33 LSI
331 RAM

Claims (5)

A plurality of imaging devices for imaging the periphery of the vehicle;
An obstacle detection unit that acquires image data captured by the plurality of imaging devices and detects the presence of an obstacle in the image, and when the obstacle detection unit detects an obstacle, the plurality of imaging devices acquire Relative position detection means for detecting relative position data of the obstacle with respect to the vehicle from the image data in time series,
In a collision determination system that determines the possibility of a collision with a detected obstacle,
Means for dividing a peripheral area of the vehicle into a plurality of areas based on a relative position with respect to the vehicle, and storing a parameter indicating a possibility of a collision for each divided area;
Means for extracting a corresponding region based on the detected relative position data;
Means for reading out the parameter corresponding to the extracted area,
A collision determination system characterized by determining the possibility of collision based on the read parameters. Means for detecting the speed of the vehicle;
The collision determination system according to claim 1, further comprising: means for deforming the shape of the divided area according to the detected vehicle speed. Means for determining whether the vehicle is turning,
Means for calculating a turning direction and a turning radius when it is determined that the means is turning;
3. The collision determination system according to claim 1, further comprising: a unit that deforms a shape of the divided area according to the calculated turning direction and turning radius. Means for reading out the parameters corresponding to the extracted areas in time series, and storing the read out parameters in time series;
Means for calculating an index value indicating the possibility of a collision according to a time-series change of the parameter,
The collision determination system according to any one of claims 1 to 3, wherein the possibility of collision is determined based on the calculated index value. Image data captured by a plurality of imaging devices that capture the periphery of the vehicle is acquired to detect the presence of an obstacle in the image. When an obstacle is detected, the image data acquired by the plurality of imaging devices In the collision determination method for detecting the relative position data of the obstacle with respect to the vehicle in time series and determining the possibility of the collision,
Based on the relative position with respect to the vehicle, the surrounding area of the vehicle is divided into a plurality of areas, and a parameter indicating the possibility of a collision is stored for each divided area,
Extract the corresponding region based on the detected relative position data,
Read the parameters corresponding to the extracted area,
A collision determination method, characterized by determining the possibility of a collision based on a read parameter.

Images