JPH1068777A - On-vehicle preceding car detecting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect a preceding car existing in a current lane accurately. SOLUTION: A detection signal from a scanning laser radar device 10 is supplied to ECU 12. An image from a camera 24 is processed by a computer 26 to detect form of a road and road form data such as curvature of a curve is supplied to the ECU 12. The ECU 12 decides whether a detected preceding car is in presence in a current line or not based on the data and when the preceding car exists in the current line, respective actuators 20, 22 are controlled so as to follow up the preceding car.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicular preceding vehicle detecting device, and more particularly to a vehicular preceding vehicle detecting device for surely catching a preceding vehicle existing in its own lane.

[0002]

2. Description of the Related Art Conventionally, a preceding vehicle detecting device for a vehicle for detecting an inter-vehicle distance to a preceding vehicle, a relative speed, and the like for the purpose of performing automatic following travel or the like has been known. As such a preceding vehicle detection device, a laser radar device has been conventionally used.However, when the preceding vehicle and the own vehicle do not exist on the same straight road, for example, on a curved road, a laser radar is used. If the direction of the laser beam emitted from the radar device is fixed, the preceding vehicle cannot be detected. Therefore, conventionally, a scanning type laser radar device which scans a laser beam within a predetermined range or a steer type laser radar device which detects a curved track by some means and turns the beam in accordance with the curve direction has been proposed. I have. For example, in a vehicle radar sensor disclosed in Japanese Patent Application Laid-Open No. 51-30490, a radiator of the radar sensor is linked with a steering wheel of a vehicle, and a radar beam radiated from the radiator in accordance with a steering angle of the steering wheel. It changes the direction.

[0003]

However, in such a configuration in which the laser beam direction of the laser radar device is changed in conjunction with the steering wheel, the own vehicle is still in spite of the fact that the preceding vehicle is traveling on a curved road. If the vehicle is traveling on a straight road before the curve, it cannot be handled. That is, in such a case, the driver of the own vehicle has not yet operated the steering wheel in accordance with the curved road, and the radiator of the laser radar device linked to the steering wheel still detects the front of the vehicle. There was a problem that it could not be detected.

In particular, when such a laser radar device is incorporated in a system for detecting a preceding vehicle and automatically following the preceding vehicle in accordance with the inter-vehicle distance or relative speed with the preceding vehicle, such a lost target is used for the system. There is a problem that greatly affects the reliability of the device.

The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to detect a preceding vehicle present in its own lane irrespective of the shape of a curved road or the like, and to provide an automatic following system and the like. It is an object of the present invention to provide a vehicle preceding vehicle detection device capable of improving reliability.

[0006]

According to a first aspect of the present invention, there is provided a radar apparatus for detecting a predetermined range in front of a vehicle, a photographing means for photographing a front area of the vehicle, and a preceding vehicle inspection by the radar apparatus. Determining means for determining the presence or absence of the preceding vehicle in the own lane based on an output result and a road shape acquired from the photographing means; By taking into account the road shape obtained by the photographing means, it is possible to improve the accuracy of determining whether or not the preceding vehicle is present in the own lane.

In a second aspect based on the first aspect, the road shape is a curvature of the road. Thus, it is possible to reliably determine whether or not the preceding vehicle is in the curved lane even on a curved road.

In a third aspect based on the first aspect, the vehicle further comprises steering angle detecting means for detecting a steering angle of the own vehicle, wherein the determining means changes the preceding vehicle from a detection state to a detection impossible state. When the steering angle detecting means detects a substantially no steering state, it is determined whether or not the preceding vehicle exists in the own lane according to the road shape.
Thus, even if the preceding vehicle is not traveling on a straight road and the own vehicle is traveling on a straight road, it is possible to reliably determine whether or not the preceding vehicle exists in the own lane.

[0009]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of this embodiment. A scanning laser radar device 10 is installed at the front of a vehicle. As is well known, a scanning laser radar device scans a narrowly focused laser beam in a horizontal plane at fixed angles, and determines the distance to the object from the time it takes for the laser beam to be reflected and warped off the object after being transmitted. The distance L to the object is calculated for each scan angle θ.

The distance data and the angle data from the scanning laser radar device 10 are input to an ECU 12 which is a control computer.

On the other hand, a vehicle speed sensor 1 for detecting the speed of the vehicle
4. A steering sensor 16 for detecting a steering angle of the own vehicle and a system operation switch 18 for turning ON / OFF the device are provided at predetermined positions of the vehicle, respectively.
Detection signals from the sensors are also input to the ECU 12. It should be noted that a sensor for detecting the number of rotations of the transmission gear can be used as the vehicle speed sensor 14, and a slit plate provided on the steering shaft and a photo interrupter are combined as the steering sensor 16 so that the rotation speed of the slit plate is increased. A sensor for detecting a light blocking state of the photo interrupter can be used.

The ECU 12 has an input port 12a for inputting a detection signal from each sensor, and a vehicle obstacle recognition unit 12 for extracting an obstacle existing in front of the vehicle from the input detection signal.
b, a target vehicle identification unit 12c that further extracts a target preceding vehicle from the extracted obstacles, a calculation unit 12d that calculates an inter-vehicle distance and a relative speed with the identified preceding vehicle,
It includes an output port 12e for outputting a processing signal to the outside and a memory 12f for storing data, and outputs a control signal from the output port 12e in accordance with the following distance or relative speed with respect to a preceding vehicle. ,
The brake actuator 22 is operated.

Further, in addition to the scanning laser radar device 10, a camera 24 for detecting a front running track is installed at the front of the vehicle, and image data from the camera 24 is supplied to an image processing computer 26. The image processing computer 26 detects a road shape ahead (for example, a curve radius R on a curved road) based on the input image and outputs the detected road shape to the ECU 12. The ECU 12 determines whether or not the preceding vehicle detected based on the preceding vehicle detection result of the laser radar device 10 and the road shape data from the image processing computer 26 is present in the own lane. To control the running of the vehicle.

Hereinafter, the processing performed by the ECU 12 will be described in detail with reference to FIG.

FIG. 2 shows a processing flowchart of the ECU 12. First, in S101, it is determined whether or not the system operation switch 18 is turned on. This S
If YES is determined in 101, the next S102
Then, the set vehicle speed V0 set by the driver is input. The set vehicle speed V0 is input via the system operation switch 18. The set vehicle speed V0 gives the maximum speed during the following running, and the following running is performed at a vehicle speed lower than V0.

After the set vehicle speed V0 is input, the distance data L and the angle data θ from the scanning laser radar device 10 are input via the input port 12a (S10).
3). Further, at S104, the vehicle speed V from the vehicle speed sensor 14 is obtained.
Enter Then, in the next S105, recognition of an obstacle existing in front of the vehicle is performed by the vehicle obstacle recognition unit 12b.

That is, the scanning laser radar device 1
When 0 detects an obstacle at a distance L from the angle θn to θn + m, the size of the obstacle is L as shown in FIG.
(Θn + m−θn) = Lmθ.

After recognizing the vehicle or obstacle in S105, the process proceeds to the next S106, where the steering sensor 16
Is input, and it is determined whether or not the vehicle recognized using the steering angle θs input in S107 is the own lane target traveling on the same lane as the own vehicle. Whether or not the recognized vehicle is the own lane target is determined by the curve radius R, which is determined by the steering angle θs. Of course, the relationship between the curve radius R and θs changes depending on the vehicle speed, and is generally expressed by R = P · θs · V (P is a constant). Then, it is determined whether or not the preceding vehicle detected by the laser radar device 10 is on the same lane as the own vehicle based on the curve radius R (S108).
Since the curve radius R can be obtained by processing an image obtained by the camera 24 as described later, the value of the curve radius R obtained from the image of the camera 24 can be used instead of the steering angle θs. It is.

If it is determined in step S108 that the vehicle has its own lane target, the flow shifts to step S109 to perform an inter-vehicle control calculation for performing the following travel. In the present embodiment, the following distance control calculation for the following running is performed as follows. That is, first, the safe inter-vehicle distance RS is calculated from the relative speed VR calculated in S105 and the vehicle speed V input in S104. The safe inter-vehicle distance RS is defined as an inter-vehicle distance at which the vehicle can safely stop without collision with the preceding vehicle, and is generally a function of the own vehicle speed V and the relative speed VR, the own vehicle acceleration α, and the preceding vehicle acceleration β. = F1 (V, VR, α, β). For example, RS = V · t + V ² / 2α− (V + VR) ² / 2β. Here, t is the idle running time, α is the time derivative of the vehicle speed V, and β is the time derivative of V + VR. The safe inter-vehicle distance RS thus obtained is a safe inter-vehicle distance within a range in which only the throttle is controlled. In contrast,
If the inter-vehicle distance from the preceding vehicle is suddenly reduced due to an interruption from another lane or an interruption such as passing, there is an inter-vehicle distance in which braking control is also required, as such throttle control alone is not enough. The dangerous inter-vehicle distance Rd requiring the brake control is represented by Rd = f2 (V, VR, α ', β). Here, α ′ is the maximum acceleration when the throttle is fully closed and the engine brake is operated. Then, the safe inter-vehicle distance RS and the dangerous inter-vehicle distance Rd are calculated in S109, the throttle and the brake are controlled such that the inter-vehicle distance becomes equal to or longer than RS, Rd, and the vehicle follows the preceding vehicle (S110).

As described above, when it is determined in S108 that the own lane target is present, the vehicle can follow, but when there is no own lane target, the process proceeds to S111. In S111, it is determined whether or not the own lane target exists in the previous processing. If the own lane target does not exist in the previous time, there is no need to perform the following travel, and the set vehicle speed V input in S102 is not required.
Speed control calculation is performed so as to be 0 (S112), and throttle or brake control is performed (S110). On the other hand, if it is determined in S111 that the own lane target was present in the previous control, the preceding vehicle, which is the own lane target, has changed lanes or is running in the same lane. It is conceivable that only the preceding vehicle enters the curved lane and cannot be recognized as the own lane target.

In step S113, it is determined whether or not the steering angle θs is substantially 0 degree.
When it is determined to be YES, it is determined that only the preceding vehicle is traveling on the curved lane, and that the own vehicle is still traveling on the lane just before the curved lane, and the process proceeds to the following steps from S114. If NO is determined in S113, it is determined that the preceding vehicle has changed lanes.
The routine proceeds to 112, where vehicle speed control is performed so as to attain the set vehicle speed V0.

As described above, the steps after S114 correspond to the processing when the preceding vehicle is traveling on a curved road and the own vehicle is traveling on a straight road immediately before the curved road. In the flow, a curve radius R of a curved lane is detected based on image data from the lane detection camera 24 shown in FIG. 1, and whether or not a preceding vehicle is present in the own lane according to the curve radius R is determined. Is determined.

That is, first, in step S114, the image processing computer 26 uses the curve radius R based on the input image data.
Is calculated. The image processing computer 26 includes an input port 26a, a CPU 26b, an output port 26c, and a memory 2.
6d, the CPU 26b processes the image data captured through the input port 26a,
The curve radius R is supplied from c to the input port of the ECU 12. Here, there are several methods for detecting the curve radius R from the image data. In the present embodiment, two white line portions for displaying lanes are extracted from the obtained image, and the white line portion 2 is extracted. Calculate the tangent at the point,
The curve radius R is calculated by calculating the difference between these inclinations. That is, when the tangents at the two white line portions are y = a1 x + b1 y = a2 x + b2, the slope difference Δa is given by the following equation.

Δa = | 1 / a1 / a2 | This inclination difference Δa is generated because the white line representing the lane is curved according to the curve radius R. The difference between the inclination difference Δa and the curve radius R Memory 26
By storing the data in d and comparing the detected inclination difference Δa with the stored data, it is possible to calculate the curve radius R of the curved road existing in front of the vehicle.

After the curve radius R is calculated and input to the ECU 12, it is determined in S115 whether or not the calculated curve radius R is almost infinite. Here, when it is determined that the curve radius R is almost infinite,
It is determined that the own vehicle will travel in a straight line in the future, and S11
Move to 2. On the other hand, if the curve radius R is a finite value, the laser beam direction of the laser radar device 10 is changed in S116 according to the curve radius R, and the own lane target is identified again.

Hereinafter, the identification processing performed in S116 will be described in detail.

Let L be the distance between the vehicle and the preceding vehicle before losing sight of the preceding vehicle, VR be the relative speed at that time, and Lc be the distance to the start of curve running.

If the vehicle traveling direction is x and the vehicle width direction is y, the relationship between x and y at the curve radius R is expressed as y = R- (R ² -x ² ) ^0.5 . Accordingly, in FIG. 3, the deviation amount of the preceding vehicle deviating from the straight line after the calculation processing time t seconds is calculated from this equation as y = R− [R ² − {(V + VR) t + (L−Lc)}
² ] ^0.5 . That is, it can be seen that the deviation amount y is a function of the curve radius R, the vehicle speed V, the relative speed VR, and the inter-vehicle distance L. Here, (V + VR) t is the distance that the preceding vehicle travels during the calculation processing time t.

Therefore, by correcting the center of the scan width of the laser radar device 10 in accordance with the deviation amount, the preceding vehicle can be captured again. Considering that the normal lane width is 3.5 m, the scan width may be set to 3.5 m + w (0 <w <1).

Then, the own lane target is re-identified by changing the scanning direction of the laser radar device 10 in S116, and if it is determined in the next S117 that the own lane target has been captured again, the preceding vehicle is detected. Determines that the vehicle is traveling on the same lane as the own vehicle even on the curved road, and executes the following distance control calculation for following traveling in S109 described above. If the preceding vehicle is not in the own lane, speed calculation is performed in S112 so as to reach the set vehicle speed V0.

As described above, in the present embodiment, the shape of the road (such as the curvature of a curved road) is detected from the forward image obtained by the camera 24 as the photographing means, and the preceding vehicle can detect its own shape based on the road shape. Since it is determined whether or not the vehicle is present in the lane, the preceding vehicle traveling in the own lane can be reliably recognized regardless of the road shape, and appropriate traveling control according to the recognition result can be performed. Become.

[0033]

As described above, according to the preceding vehicle detecting device for a vehicle according to the present invention, the preceding vehicle existing in the own lane can be reliably detected regardless of the road shape such as a curved road, and the automatic following can be performed. The reliability of the system and the like can be improved.

[Brief description of the drawings]

FIG. 1 is a configuration block diagram of an embodiment.

FIG. 2 is a control flowchart of the embodiment.

FIG. 3 is an explanatory diagram of a curve deviation amount during a curved road.

FIG. 4 is a diagram illustrating detection of a laser radar device.

[Explanation of symbols]

Reference Signs List 10 laser radar device, 12 ECU, 16 steering sensor, 20 throttle actuator, 22
Brake actuator, 24 cameras, 26 Image processing computer.

Continued on the front page (51) Int.Cl. ⁶ Identification number Agency reference number FI Technical indication location G01S 13/93 G01S 17/87 17/87 G08G 1/16 D G08G 1/16 G01S 13/93 Z

Claims (3)

[Claims] A radar device for detecting a predetermined range in front of the vehicle; a photographing means for photographing the front of the vehicle; a preceding vehicle detected by the radar device and a road shape acquired from the photographing means. A determination device for determining the presence or absence of the vehicle in the own lane. 2. The preceding vehicle detecting device for a vehicle according to claim 1, wherein the road shape is a curvature of the road. 3. The vehicle further comprises a steering angle detecting means for detecting a steering angle of the own vehicle, wherein the determining means changes the detection state of the preceding vehicle from a detection state to an undetectable state, and the steering angle detection means performs almost no steering. The vehicle preceding vehicle detection device according to claim 1, wherein when a state is detected, the presence / absence of the preceding vehicle in the own lane is determined according to the road shape.