JP2002144991A - Obstacle sensing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an obstacle sensing device of simple constitution capable of avoiding a touch with obstacle by sensing any obstacle beside a vehicle. SOLUTION: The obstacle sensing device includes two sensors, and when the first sensor 2 has sensed any obstacle 23 ahead of the vehicle 1, the moving time Δ of the vehicle 1 required until the obstacle 23 appears in the second range 22 beside the vehicle as the range which is covered with the second sensor 3 is presumed, and when the presumed time has elapsed, the second sensor is actuated. Judgement is made whether the object sensed by the second sensor is the same as the obstacle which the first sensor sensed, and judgement about existence or none of obstacle is passed on the basis of the given result, and if yes, another judgement is made whether the obstacle 23 is approaching the vehicle 1, and if yes, an alarm is issued.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an obstacle detecting device for detecting an obstacle on the side of a vehicle.

[0002]

2. Description of the Related Art Conventionally, as an obstacle detecting device for a vehicle, one described in Japanese Patent Application Laid-Open No. 2000-123298 is known. This conventional obstacle detection device uses CC
The presence of obstacles is detected by the D camera, and the same range as the imaging range of the CCD camera is searched by the ultrasonic sensor to detect the presence of the obstacle and the distance to the obstacle. The object and the obstacle detected by the ultrasonic sensor are determined in an integrated manner, and the presence of the obstacle and its position are detected.

[0003]

However, in the conventional obstacle detection device, when an obstacle existing on the rear side of the vehicle is to be detected, for example, a predetermined range on the rear side of the vehicle is imaged. Camera and an ultrasonic sensor that emits ultrasonic waves to a predetermined range on the rear side of the vehicle, so that the camera needs high-speed processing capability in image processing, that is, high computational capability, However, there is a problem that a high-performance arithmetic device is required, and the cost of the device is increased.

For example, when a garage is put forward into the parking lot 100 on a course shown by an arrow A in FIG. Obstacles 101 like pillars,
If the vehicle 102 is present, the side of the vehicle may be rubbed against the obstacle 101 or 102 due to an inner wheel difference. Is desirable. But,
In such a situation, if the conventional obstacle detection device is configured to generate a warning, there is a problem that the cost becomes too high as described above.

The present invention has been made to solve such a technical problem, and an object of the present invention is to provide an obstacle detecting device capable of detecting a lateral obstacle at low cost.

[0006]

According to the first aspect of the present invention, there is provided an obstacle detection apparatus comprising: a first sensor for detecting an object in a first range in front of a vehicle; and an object in a second range on a side of the vehicle. A second sensor for detecting, a time estimating unit for estimating a moving time of the vehicle required until the object appears in the second range when the object is detected by the first sensor; After the sensor detects the object, the second sensor is operated at the time estimated by the time estimating means, and the coincidence between the object detected by the second sensor and the object detected by the first sensor is determined. And an obstacle judging means for judging the presence or absence of an obstacle based on the judgment result of the coincidence judging means.

According to a second aspect of the present invention, in the obstacle detecting device according to the first aspect, when the coincidence determining means determines the coincidence, the object is detected based on information detected by the second sensor. An approach determining means for determining whether the object is approaching the vehicle, and an alarm means for outputting an alarm when the approach determining means determines that the object is approaching the vehicle.

According to a third aspect of the present invention, in the obstacle detecting device of the second aspect, the approach determining means determines that the area of a portion corresponding to the object detected by the second sensor is increasing. It is characterized in that it is determined that the object is approaching the vehicle.

According to a fourth aspect of the present invention, in the obstacle detecting device of the first to third aspects, the first sensor is a laser radar and the second sensor is a camera. .

According to a fifth aspect of the present invention, in the obstacle detecting device according to the fourth aspect, the coincidence determining means determines a luminance distribution of a detection signal of the second sensor by detecting a luminance distribution of the detection signal of the object by the first sensor. It is characterized by comparing with the luminance distribution and judging the coincidence.

[0011]

According to the first aspect of the present invention, when an obstacle in front of the vehicle is detected by the first sensor, the obstacle is located in the second range on the side of the vehicle, which is the detection range of the second sensor. The travel time of the vehicle required to appear is estimated, and when the estimated time has elapsed, the second sensor is operated. Then, the matching between the object detected by the second sensor and the obstacle detected by the first sensor is determined, and the presence or absence of the obstacle is determined based on the determination result.

Thus, the presence of an obstacle on the side of the vehicle can be determined.
Apparently, detection can be performed by both the first sensor and the second sensor, and there is no need to make an integrated judgment by the first sensor and the second sensor at the same time. There is no need to use high-speed, high-performance computing devices,
The cost of the device can be reduced.

According to the second aspect of the present invention, when the coincidence judging means judges the coincidence, it is judged whether or not the obstacle is approaching the vehicle from the detection information of the second sensor, and the obstacle is approaching the vehicle. An alarm is output when it is determined that there is.

Thus, when an obstacle on the side of the vehicle is relatively close to the vehicle, an alarm can be output to the driver so as to avoid contact with the obstacle.

According to the third aspect of the present invention, the approach judging means comprises a second
When it is determined that the obstacle is approaching the vehicle when the area of the portion corresponding to the obstacle detected by the sensor is increasing, it is possible to accurately detect the approach of the obstacle to the vehicle.

According to the fourth aspect of the present invention, a laser radar, which is usually mounted as an inter-vehicle distance measuring device, is used by using a laser radar for the first sensor and a camera for the second sensor. It can also be used for obstacle detection, and the total cost can be reduced when an obstacle detection device is also mounted on a vehicle equipped with an inter-vehicle distance measurement device.

According to the fifth aspect of the present invention, the coincidence judging means judges the coincidence by comparing the luminance distribution of the detection signal from the second sensor with the luminance distribution of the detection signal of the object from the first sensor. The presence of an obstacle on the side of the vehicle can be accurately detected.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 shows a device configuration according to an embodiment of the present invention. The obstacle detection device according to the present embodiment includes a laser radar 2 that detects the presence or absence of an object existing in front of and in front of the vehicle 1 and a distance to the object, and a CCD camera that captures an image of the side of the vehicle 1. , A vehicle speed sensor 4 for detecting a vehicle speed, a speaker 5 for warning of approaching an obstacle, and a controller 6 for controlling each device and executing necessary arithmetic processing.

As shown in FIG. 2, the laser radar 2 periodically emits a laser beam to the front side of the vehicle 1 and
When there is a target inside, it receives the reflected light that is reflected back to the target and measures the distance to the target based on the time difference from the laser beam emission timing to the reflected light reception timing. If no reflected light is received, it is determined that there is no target.

The camera 3 is incorporated in the door mirror 7 of the vehicle 1 or at the same place as the door mirror 7, and has a visual field 22 for detecting an obstacle in the rear of the vehicle. This camera 3
Under the control of the controller 6, when the laser radar 2 detects the presence of an obstacle on the front side of the vehicle, the same obstacle will
It operates at the estimated time delay until it appears in the
It captures the situation on the side of the vehicle and outputs the video to the controller 6.

The controller 6 executes a program necessary for operating the obstacle detecting device,
A ROM that stores programs and other fixed information, a RAM that temporarily stores information, and an I / O processing unit that transmits and receives signals to and from each external device are provided. To execute repeatedly.

Next, the operation of the obstacle detecting device according to the above embodiment will be described with reference to FIGS. Controller 6
Periodically repeats the calculation processing of the flowchart of FIG. In step S05, it is determined whether or not the laser radar 2 has detected the presence of the object 23 in the visual field 21 on the front side. If the presence of the object 23 has been detected, a delay time Δ until the object 23 appears in the visual field 22 of the camera 3 with the movement of the vehicle 1 is estimated in step S10.

This operation is as follows. As shown in FIG. 4, first, a vehicle coordinate system is set. The x-axis is the front-back direction of the vehicle 1, and the y-axis is the width direction of the vehicle 1. Then, the front right end of the vehicle 1 is set as the origin of the vehicle coordinate system. Hereinafter, in order to simplify the description, it is assumed that the vehicle 1 moves one-dimensionally in the-(minus) direction of the x-axis.

Here, the speed of the vehicle is assumed to be v. this is,
From the speed pulse p obtained from the vehicle speed sensor 4 of the vehicle 1,
Multiplied by constants such as tire radius and reduction ratio, it is obtained from Equation 1.

[0025]

## EQU1 ## Now, the front right end of the vehicle 1 is at the origin, a laser beam is emitted from this point by the laser radar 2, and an obstacle 23 exists in the detection range 21 in FIG. Assume that the position is calculated from the reception intensity of the reflected signal at each scanning angle of the laser radar 2 and the obstacle 23 is equivalently detected at (x0, y0) in the vehicle coordinate system. In the following equation, the position of the obstacle 23 in the vehicle coordinate system at the time t is represented by (x0 (t), y0 (t)).
Notation.

It should be noted that the detection of the presence of the obstacle 23 by the laser radar 2 alone does not output an alarm because the obstacle 23 does not exist on the path of the vehicle 1.

Next, the vehicle 1 has a velocity v parallel to the -x axis direction.
For δt hours. At this time, in the vehicle coordinate system,
The obstacle 23 is observed at the position (x0 + v · δt, y0). This position is described again as (x1 (t + Δt), y0 (t + Δt)).

At this time, if the obstacle 23 is a fixed object that does not move with respect to the ground, it is possible to estimate when the movement of the vehicle 1 will appear in the field of view 22 of the camera 3.

Now, the end 22 of the field of view 22 of the camera 3
Assuming that A is represented as y = ax + b by the vehicle coordinate system, the coordinates (x1 (t + δt), y0 (x0 (t), y0 (t)) when the obstacle 23 enters the visual field range 22. t + δt)), and this x1 is expressed by the following equation (2).

[0030]

(Equation 2) Now, since the speed of the vehicle 1 is v, from the time t, the visual field range 22 of the camera 3 after Δ time given by the following equation (3).
It is presumed that an obstacle 23 'appears in the area.

[0031]

(Equation 3) Therefore, after calculating the Δ time in step S10, the camera 3 is operated to image the visual field range 22. Then, in steps S15 and S16, a change in luminance at the end 22A of the visual field 22 of the camera 3 after Δ time is observed, and luminance information Is (x, y) and Ic (x, y) described below are obtained. Extract.

As shown in FIG. 5A, luminance information Is (x, y) in a predetermined range of the end 22A of the visual field 22 is extracted (step S15). On the other hand, luminance information Ic (x, y) is similarly extracted from a plurality of pixels at a predetermined position in the + x direction in the observation field 22 of the camera 3 different from the above-described predetermined portion (step S).
20). Then, in step S25, the invariant variance Vr (Is (x, y) for the luminance information Is (x, y) and Ic (x, y) is obtained.
y)) and Vr (Ic (x, y)), and their ratios.

[0033]

## EQU4 ## Vr (Is (x, y)) / Vr (Ic (x, y)) In the next step S30, Fr is calculated using the ratio of equation (4).
Perform a test to determine if a significant difference has appeared. If there is a significant difference, it is considered that the obstacle 23 has appeared, and in the present embodiment, a process of step S35 is further performed. However, if the F-test determines that there is no significant difference, the obstacle 2
Regarding that 3 is gone, return.

Here, the F test is as follows. In each figure of FIG. 5, the left side shows the graph of the luminance information, and the darker the right, the darker the object (that is, the detection of an object different from the background). A hatched rectangular portion 24 at the center indicates the end 22A of the viewing range of the camera 3. The left side shows the visual field range 22 of the camera 3.

FIG. 5A shows an image of the obstacle 23 'captured by the camera 3 at a time Δ time after the time t at which the laser radar 2 detects the obstacle 23 described above. As described above, when the obstacle 23 'enters the end portion 22A of the visual field range, the portion corresponding to the image of the obstacle 23' has a luminance different from that of the background, and the luminance distribution Is (x, y) is the same. As shown on the left side of the figure. In step S20, the viewing range 22
Similarly, the luminance information Ic (x,
y).

Then, in step S25, the luminance information Is (x,
y), the invariant variance Vr (Is (x, y)) for Ic (x, y), Vr (Ic (x,
y)) is obtained, and an F test is executed in the subsequent step S30.

The F test is a method for judging whether or not the distribution of a signal is different from the average distribution.
Basically, the variance V1 of the conventional signal (the amount of deviation from the average value) is determined, and if a signal is input to check whether the signal is different from the conventional signal, the variance V2 is calculated, In this method, the variance ratio V1 / V2 is obtained and compared with a mathematically determined value in a numerical table, and the difference between the signals is determined based on whether the variation of the data is the same or different.

Thus, in step S30, FIG.
Using the image signal of the camera 3 after the time Δ shown in (a) until the time (Δ + α) shown in FIG. 5C for detecting the opposite end of the same obstacle 23 ′, F
A test is performed to determine whether there is a significant difference from the background-only image.

As a result of the F test, there is a significant difference.
If it is determined that the obstacle 23 'is detected on the side of, the average luminance value and the standard deviation of the luminance information Is (x, y) are obtained in the subsequent step S35. In the following step S40, the movement of the vehicle causes a new in-field end 2 in the observation field 22.
If the luminance of 2A is within the range of the average luminance ± standard deviation, it is recognized that the obstacle 23 'has been continuously detected (step S).
45) If it is out of the range, it is determined that the opposite end of the obstacle 23 'has been reached (step S50).

As shown in FIG. 5B, while the same obstacle 23 ′ detected after Δ time is continuously detected, the measurement is repeated, but as shown in FIG. If the luminance of the inner end 22A is out of the range of the average luminance ± standard deviation, it is determined that the opposite end of the obstacle 23 'has been detected, and the clustering process is executed (step S5).
5). In this clustering process, as shown in FIG. 5D, the contour of the obstacle 23 '(different from the contour of the actual object) is clustered by luminance.

After this clustering, the same obstacle 2
3 'is continuously tracked in the visual field 22, and it is determined whether or not the clustered area 25 is enlarged (steps S60 and S65). And the clustered area 25
Is expanded, the obstacle 23 'is approaching the vehicle, and a warning is output from the speaker 5 as a possibility of contact (step S70).

As described above, according to the present embodiment,
The obstacle 23 is detected by the laser radar 2 as the first sensor.
Is detected, the time Δ until the same obstacle 23 ′ appears in the field of view 22 of the camera 3, which is the second sensor, is estimated in consideration of the speed of the vehicle. ′ Is detected, it is further determined whether or not the image of the obstacle 23 ′ captured by the camera 3 has a tendency to expand, and when it is determined that the image has a tendency to expand, an alarm is output as a possibility of contact. Therefore, while the first sensor and the second sensor each independently detect obstacles, apparently two sensors cooperate to monitor obstacles and avoid contact. Alarm can be issued, and since the first and second sensors are operated independently, complicated arithmetic processing is not required.
There is no need to mount a high-speed, high-performance arithmetic unit in the controller 6, and the cost can be kept low.

In the above embodiment, not only the detection of the obstacle 23 'after Δ time, but also the clustering processing of the image of the obstacle 23' is performed, and it is judged whether or not the obstacle 23 'is approaching, and an alarm is issued. However, in order to simplify the apparatus, if it is determined that there is a significant difference in step S30 in the flowchart of FIG. 3, a warning is issued assuming that the obstacle 23 'is approaching the side of the vehicle. Is also good.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram showing the configuration of an embodiment of the present invention.

FIG. 2 is a plan view showing a visual field range of each of a laser radar and a camera in the embodiment.

FIG. 3 is a flowchart of an obstacle detection process according to the embodiment.

FIG. 4 is an explanatory diagram of an obstacle detection operation according to the embodiment.

FIG. 5 is an explanatory diagram showing a relationship between a luminance distribution obtained from an image signal of a camera and a position of an obstacle in the embodiment, and a clustering result.

FIG. 6 is a layout of a parking lot where the present invention is effective.

[Explanation of symbols]

 Reference Signs List 1 vehicle 2 laser radar 3 camera 4 vehicle speed sensor 5 speaker 6 controller

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G01S 17/93 G06T 1/00 330Z G06T 1/00 330 7/20 100 7/20 100 G08B 21/00 U G08B 21/00 G08G 1/16 C G08G 1/16 G01S 17/88 A F-term (reference) 5B057 AA16 BA01 BA02 CA08 CA12 CA16 CB08 CB12 CB16 CC01 CH20 DA07 DB02 DB09 DC04 DC22 DC32 5C086 AA53 BA22 CA21 CA25 CB28 CA30 CB36 CB40 DA08 DA40 EA45 FA02 5H180 AA01 BB15 CC03 CC04 CC14 LL02 LL07 LL17 5J084 AA01 AA05 AC02 AD01 AD05 BA34 CA66 CA67 DA10 EA22 EA29 5L096 AA06 BA04 CA02 CA27 DA03 FA33 FA59 GA08 HA02 JA11

Claims (5)

[Claims] 1. A first sensor for detecting an object in a first range in front of a vehicle, a second sensor for detecting an object in a second range on the side of the vehicle, and an object is detected by the first sensor. A time estimating means for estimating a moving time of the vehicle required until the object appears in the second range when the object is detected, and the time estimating means estimating the object after the first sensor detects the object. Activating said second sensor at a time;
Coincidence determining means for determining the coincidence between the object detected by the second sensor and the object detected by the first sensor; and determining the presence or absence of an obstacle based on the result of the determination by the coincidence determining means. An obstacle detection device comprising: an obstacle determination unit. 2. An approach judging means for judging whether or not the object is approaching a vehicle based on information detected by the second sensor when the coincidence judging means judges the coincidence; The obstacle detection device according to claim 1, wherein the approach determination unit includes an alarm unit that outputs an alarm when it is determined that the object is approaching the vehicle. 3. The method according to claim 2, wherein the approach determining means determines that the object is approaching the vehicle when an area of a portion corresponding to the object detected by the second sensor is increasing. The obstacle detection device according to claim 2. 4. The obstacle detection device according to claim 1, wherein the first sensor is a laser radar, and the second sensor is a camera. 5. The coincidence determining means compares the luminance distribution of the detection signal from the second sensor with the luminance distribution of the detection signal of an object from the first sensor to determine coincidence. The obstacle detection device according to claim 4.