JP2000251200A - Obstacle detector for vehicle - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent or reduce the situation of erroneously judging that a transversal moving speed in front is dangerous. SOLUTION: The transversal moving speed in a direction that a transversally moving object in front (a pedestrian H in front for instance) is to advance to the traveling path 10 of a present vehicle 1, the position and size of the object are detected by utilizing a radar. At least the transversal moving speed, the position and the size are collated with the setting conditions of a danger coefficient set beforehand as parameters, the danger coefficients (a)-(d) for one detected transversally moving object are set and a danger degree Di for the one transversally moving object is calculated by the calculation formula of a×b×c×d. When the danger degree Di is larger than a prescribed value Tw, an alarm device is operated. The prescribed value Tw is corrected to a small value at nighttime and at the time of raining when visibility is deteriorated and an alarm is easily issued. The danger degree Di decided at each sampling timing is accumulated for plural times and the accumulated danger degree can be set as a final danger degree.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

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

[0002]

2. Description of the Related Art Various methods have been proposed for detecting an obstacle in front of a vehicle, that is, an object in front of the vehicle, for safe driving. In particular, it has been proposed to detect a pedestrian. In Japanese Patent Application Laid-Open No. 10-105891, the lateral movement speed of a forward object in the traveling direction of the own vehicle is detected, and based on the detected lateral movement speed, the forward object enters the traveling path of the own vehicle. A proposal has been made to determine the possibility of occurrence, that is, the presence or absence of danger. Further, Japanese Patent Application Laid-Open No. 8-313632 proposes an apparatus that makes a danger determination based on a relative positional relationship between a host vehicle and a pedestrian ahead.

[0003]

By the way, a forward object is detected by, for example, a scanning type laser radar. However, a large number of forward objects may be detected. In this case, a determination is made as to whether or not each of the horizontally moving objects considered to be highly dangerous among the many forward objects detected is dangerous. Actually, it is a situation that does not require danger judgment, such as a stationary object or a vehicle ahead in front of which the driver of the host vehicle is already sufficiently aware. However, to accurately identify whether the object is actually a dangerous lateral moving object or a non-hazardous lateral moving object, including the accurate identification of whether the object is a lateral moving object, as described above. As described in the gazette, there is a limit in simply judging only the lateral movement speed and only the relative distance as parameters. In particular, it is desired to accurately detect a pedestrian who has a high possibility of entering a traveling path of the vehicle as a dangerous laterally moving object. Must be set to perform the danger determination as much as possible, and the possibility of erroneously determining that a non-dangerous horizontally moving object is dangerous is extremely high, that is, the noise of the danger determination increases. Will be.

The present invention has been made in view of the above circumstances, and has as its object not to be particularly dangerous so that it can be more accurately determined whether or not a horizontally moving object existing ahead is dangerous. It is an object of the present invention to provide a vehicle obstacle detection device capable of preventing or reducing a situation in which a laterally moving object is erroneously determined to be dangerous.

[0005]

Means for Solving the Problems To achieve the above object, the present invention provides the following as a solution. That is, as described in claim 1 of the claims, an object detecting means for detecting a laterally moving object on the traveling path of the vehicle, a lateral moving speed of the laterally moving object detected by the object detecting means, An object information acquisition unit that acquires a position and a size, and a risk determination unit that determines a danger related to the traveling of the own vehicle based on the lateral movement speed, the position, and the size acquired by the object information acquisition unit, It is made to have. A preferable mode based on the above solution is as described in claim 2 of the claims.

[0006]

According to the first aspect of the present invention, it is determined whether or not the forward moving object is a dangerous object in consideration of the speed, position and size of the moving object. To determine whether or not the object is really dangerous, especially considering the size of the laterally moving object, and erroneously determining that a horizontally moving object of a size that should be excluded from the risk judgment target is a dangerous object This is preferable for preventing or reducing the situation of making a judgment. According to the second aspect, when there is a laterally moving object having a degree of risk equal to or more than a predetermined value, the driver is alerted by a warning, which is extremely preferable in terms of safety.

According to the third aspect, a small laterally moving object that is easily overlooked by the driver is positioned as having a higher degree of danger, which is preferable in terms of safety. According to the fourth aspect, the final danger level is determined by accumulating a predetermined number of danger levels, so that the obtained danger level is determined as compared with the case where the danger level is determined only once. Can be made more accurate. According to the fifth aspect, the risk level is determined in consideration of the visibility of the driver, so that the risk level can be determined in accordance with the actual driving situation. According to claim 6, it is preferable in terms of safety in rainy weather when visibility deteriorates. According to claim 7, it is preferable from the viewpoint of safety at nighttime when visibility is deteriorated.

According to the eighth aspect, the risk is determined in consideration of a traveling system road in the near future from the current position of the host vehicle, so that the vehicle exists in a direction in which the host vehicle is unlikely to run.
In other words, this is preferable for preventing a situation in which a horizontally moving object having no danger or an extremely low risk is erroneously determined to have a high risk. According to the ninth aspect, the diverged portion of the traveling path is a place where it is clearly separated whether or not the own vehicle is a future traveling system road to be traveled from now on, and a pedestrian who is strongly desired to make a danger determination exists. However, the effect corresponding to claim 8 can be sufficiently exhibited in such a branched portion. According to the tenth aspect, it is possible to clearly recognize the future traveling route that the driver of the own vehicle is about to travel using the turn signal, and to sufficiently exhibit the effect corresponding to the eighth aspect. According to the eleventh aspect, the direction in which the own vehicle is about to travel, that is, the travel path can be clearly recognized by the operation state of the turn signal, and therefore, the predetermined travel path in which the own vehicle is not likely to travel in the forward travel path is provided. It is possible to prevent a situation in which a horizontally moving object existing near the traveling road and having no danger is erroneously determined to have a high risk.

[0009]

FIG. 1 shows a vehicle 1 as a host vehicle, which is equipped with a radar (laser radar in the embodiment) 2 for detecting an object in front and a traveling line (in front of the vehicle). A camera (CCD camera in the embodiment) 3 for recognizing a white line) is provided. The vehicle 1 also includes a navigation device 4 for receiving road information from outside the vehicle and the position of the own vehicle 1, as shown in FIG.
Alternatively or additionally, a road-to-vehicle communication device (VICS) 5 for obtaining road information is provided.
In FIG. 2, U is a controller constituted by using a microcomputer. The controller U receives signals from the devices 2 to 5 and a vehicle speed sensor 6.
The vehicle speed signal of the own vehicle 1 is input. In addition, the controller U outputs signals to various alarm devices 11 to 13 for alerting the driver of the vehicle 1.
The alarm device 11 is activated when a dangerous front object is detected, and issues an alarm in a visual display format such as a lamp. The alarm device 12 is activated when an object with high danger is detected, and issues an alarm in the form of auditory or sound display using a speaker or the like. The warning device 13 visually displays the distance to a dangerous front object.

The outline of the control by the controller U will be described with reference to FIGS. First, in FIG. 3 showing a straight lane of one lane in one way (a total of two lanes including an oncoming lane), a left traveling road on which the host vehicle 1 is traveling is indicated by reference numeral 10A, and a right traveling road on the right is shown. Reference numeral 10B denotes a center line that separates the two, and reference numeral 10C denotes a center line. Among the boundary lines between the traveling path 10 and the sidewalk, the left boundary line is indicated by reference numeral 10L, and the right boundary line is indicated by reference numeral 10R. FIG. 3 shows a state where the pedestrian H is about to enter the traveling path 10A of the host vehicle 1 from the left side (over the left boundary line 10L) at a normal walking speed. On a straight road, the radar 2 detects a lateral movement speed (indicated as VX in FIG. 3) in a direction perpendicular to the traveling direction of the vehicle 1.

A forward object having a lateral movement speed (a lateral movement speed in a direction to enter the traveling path 10) is grasped as a laterally moving object, and the danger of the laterally moving object to the own vehicle 1 is at least laterally. The moving speed and its position and size are determined as parameters. Specific examples for determining the degree of risk are shown in FIGS. 4 to 7 show the case where the first to fourth risk factors a to d are determined by using the size of the object and the like as parameters, and the value obtained by multiplying each of the risk factors a to d is one. It is designed to indicate the degree of risk for a laterally moving object, more specifically,
In the embodiment, the degree of danger is “maximum”, “large”,
There are five types of “medium”, “small”, and “extremely small”, and these five types of danger are finally determined by danger factors a (in the case of FIG. 4), b
(In the case of FIG. 5), c (in the case of FIG. 6) or d (in the case of FIG. 7), it is converted into a numerical value. That is, when the risk factors a to d are represented by numerical values, for example, the large is “3”,
“2” can be set for medium, “1” for small, and “0” for minimal.

FIG. 4 shows parameters of the width and length of the horizontally moving object.
The first risk coefficient a is determined as the data, for example, when the width of the horizontally moving object is smaller than a predetermined value and the length of the horizontally moving object is smaller than a predetermined value, the risk is “small”. Then, the risk factor a corresponding to FIG. 4 is set to “1”. Similarly, from FIG. 5, the risk factor b is set using the lateral moving speed of the horizontally moving object as a parameter, and the risk in the vertical direction (the distance from the host vehicle 1 to the horizontally moving object) is set as a parameter from FIG. The coefficient c is set, and from FIG.
The danger coefficient d is set using the lateral distance to the vehicle and the state of the lateral movement of the laterally moving object from the host vehicle 1 (whether the vehicle is moving away from or approaching the traveling road) as parameters. The settings shown in FIGS. 4 to 7 are created and stored in advance.
By multiplying each of the risk factors a to d obtained by collating with FIG.
The risk for one laterally moving object is calculated. That is, the degree of danger = a × b × c × d.

The risk obtained as described above is compared with a predetermined value (judgment threshold value). When the risk is larger than the predetermined value, it is determined that the object is a dangerous laterally moving object. The devices 11 and 12 are operated, and the distance to the horizontally moving object is displayed on the distance display device 13. Further, the predetermined value is corrected (changed) in accordance with the visibility of the driver, so that a warning is more easily generated when the visibility is poor than when the visibility is good. For example, at night or during rainy weather, the predetermined value is corrected to a small value. It should be noted that two types of first predetermined value and second predetermined value (> first predetermined value) are set, and when the risk is larger than the second predetermined value which is a large value, the degree of danger is determined to be extremely high and the alarm device is set. 1
2 is activated to give a strong alert by sound, and when the degree of danger is larger than the first predetermined value, which is a small value, but smaller than the second predetermined value, only the alarm device 11 is operated to give gentle attention. You can also evoke.

4 to 7 will be described in more detail. First, FIG. 4 shows the setting of the risk coefficient based on the size of the horizontally moving object. Basically, the risk is set to be the highest when the width and length of the horizontally moving object are medium. Otherwise, the degree of risk is set as small. The specific size setting when the width and the length are medium are assumed to be pedestrians. Although different from FIG. 4, it is also possible to set so that the risk increases as the size of the horizontally moving object decreases.

FIG. 5 is set such that the lower the vertical movement speed, the higher the risk. The slow speed in FIG. 5 is set assuming the walking speed of the pedestrian. FIG.
Is set such that the closer the vertical distance, the higher the risk. In FIG. 7, the risk is set to be higher as the distance from the traveling road is shorter, the risk is set to be higher as the lateral movement speed in the direction away from the traveling road is smaller, and the direction approaching the traveling road is set. Is set so that the risk increases as the lateral movement speed increases.

Next, the above-described control by the controller U will be described with reference to the flowcharts of FIGS. 8 to 10. In the following description, Q indicates a step. First,
7, signals from the sensors 6 to 8 are input (acquisition of vehicle information), signals from the devices 4 and 5 are input (acquisition of road information) in Q2, and signals from the camera 3 in Q3. A signal is input (travel path estimation). Q
At 4, the forward object is detected based on the signal from the radar 2. Thereafter, in Q5, object identification including the detected lateral movement speed of the forward object is performed, and in Q6, danger determination control including alarm control is performed.

The details of Q6 are shown in FIG. First, in Q11, after setting of a determination threshold value Tw according to visibility, which will be described later, i, which is a code for identifying a plurality of horizontally moving objects, is initialized to zero. In Q13, for the i-th laterally moving object obj (i), the degree of risk Di (i) is calculated based on FIGS. 4 to 7 (as described above, the risk coefficient a × b × c × d is calculated). Calculation) is performed.
In Q14, it is determined whether or not the calculated degree of risk Di (i) is larger than a predetermined value Tw. In the determination of Q14, Y
In the case of ES, the above-described alarm is issued in Q15.

After Q15 or in the case of NO in the determination of Q14, after counting up i by one in Q16, in Q17, it is determined whether or not i is smaller than the total number of detected laterally moving objects. Is determined. This Q
If the determination in Step 17 is YES, the process returns to Q13 again (determination classification in Q14 for all detected horizontally moving objects). If the determination in Q17 is NO, the danger determination and the control termination of the alarm accompanying this determination are returned.

FIG. 10 shows the details of Q11.
First, in Q21, it is determined whether or not it is currently rainy. This determination is made by, for example, checking whether or not the wiper switch is turned on. This Q21
Is YES, the predetermined value Tw is set in Q22.
Is corrected by multiplying the basic Tw by a correction coefficient Kw. The correction coefficient Kw is set as 0 <Kw <1, and the predetermined value Tw after the correction in Q22 is set to a small value. It becomes correction.

After Q22 or when NO in the determination of Q21, it is determined in Q23 whether or not the current time is nighttime. For example, this determination is performed by checking whether or not the headlight switch is turned on. This is done by: If the determination in Q23 is YES, in Q24, the predetermined value Tw is corrected by multiplying by the correction coefficient Kn. The correction coefficient Kn is set as 0 <Kn <1, and the correction is performed in a direction in which an alarm is easily generated. Q2
If the determination in 3 is NO, the return is made.

FIG. 11 shows a modification of FIG. In FIG. 11, the risk of one horizontally moving object is accumulated a predetermined number of times (a plurality of times, for example, four times), and the accumulated value is compared with a predetermined value Tw. 9 differs from the case of FIG. 9 in the portions of Q24 and Q25 for accumulating the degree of risk. That is, Q2
In 4, the current risk Di (i) calculated in Q23 is newly added to the risk Di (i) calculated up to the previous time,
The accumulated value Ds (i), which is the value when the number of additions reaches the predetermined number, is compared with the predetermined value Tw in Q25. The predetermined value Tw in Q25 is set to a value larger than that in FIG. 9 in consideration of the number of times of accumulation. The predetermined value Tw in Q25 can be set to the same size as that in FIG. 9 by arithmetically averaging the accumulated value Ds (i) in Q25. In addition, when the risk is accumulated, a process of setting the weight of the new risk higher than that of the old risk can be performed. For example, when four times are accumulated, the degree of reflection of the current risk is 40%, the degree of reflection of the previous risk is 30%, the degree of reflection of the second risk is 20%, and the degree of reflection is three times Can be weighted such that the degree of reflection of the degree of risk is set to 10%.

FIG. 12 and FIG. 13 show examples in which a specific laterally moving object among the detected laterally moving objects is excluded from danger judgments according to the traveling path of the own vehicle in the near future. In the flowchart of FIG. 12, the road condition ahead of the host vehicle 1 due to navigation or the like and the host vehicle 1
The specific laterally moving object to be excluded is selected in consideration of the operation state of the turn signal. First, FIG.
In Q2 of No. 2, the distance L to the intersection K, which is a branch portion of the traveling road as shown in FIG. 13, and the distance R (i) to the laterally moving object detected from the own vehicle are determined. . In Q32, it is determined whether or not R (i) is larger than L. If the determination in Q32 is NO, Q
At 38, the horizontally moving object is subjected to danger determination.
That is, a horizontally moving object before the intersection K is included in the danger determination target (for example, the pedestrian H1 in FIG. 13 is a danger determination target).

When the determination in Q32 is YES, for example, a pedestrian H2 is present farther than the intersection K as shown in FIG. At this time, in Q33, it is determined whether or not the right turn signal is operated.
If the determination in Q33 is YES, in Q34,
It is determined whether or not the host vehicle is present in the rightmost lane. When the determination in Q34 is YES, it is when it is clearly confirmed that the own vehicle 1 does not travel to the position of the pedestrian H2 beyond the intersection K, so that Q3 is used.
In 5, the laterally moving object is excluded from the target of the risk determination (the pedestrian H2 in FIG. 13 is excluded from the target of the risk determination).

If NO in Q33, or if Q3
If the determination in Step 4 is NO, it is determined in Q36 whether or not the left turn signal is activated. This Q
If the determination in 36 is YES, in Q37, it is determined whether or not the vehicle 1 is present in the leftmost traveling lane. If the determination in Q37 is YES, the process shifts to Q35 and is excluded from the target of danger determination. When the determination in Q36 is NO, or when the determination in Q37 is NO, the process shifts to Q38. The determination of Q34 and Q37 is a clearer confirmation that the host vehicle 1 surely turns right or left at the intersection K.
The determination of Q37 can be eliminated.

FIG. 14 also shows a case where a specific laterally moving object among the detected laterally moving objects is excluded from the danger judgment. Byway 1 in the direction of turning diagonally left in front of own vehicle 1
FIG. 13 shows a case where a large intersection K as shown in FIG. The pedestrian crossing the side road 10X is H
The pedestrian that is indicated by 3 and that exists immediately before the intersection K is indicated by H4. Since both the pedestrians H3 and H4 are close to the own vehicle 1, the pedestrians H3 and H4 are basically laterally moving objects that are subject to danger determination. However, just before the branching portion K2 to the side road 10X, the host vehicle 1 is operating the left turn signal.
There is no possibility of collision with the pedestrian H4 while traveling at 0X. For this reason, when it is confirmed (estimated) that the traveling of the side road 10X is driven by the operation of the left turn signal, only the pedestrian H4 is excluded from the target of the danger determination.

FIGS. 15 and 16 show a method for improving the S / N ratio of the light receiving signal of the scanning laser radar 2 for improving the light receiving intensity at the current scanning position sufficiently. Has become. First, in FIG. 15, the laser light emitted from the light emitting element 21 passes through the lens 22, is reflected by the scanning mirror 23, and is projected toward the front of the host vehicle 1. The rotating position of the scanning mirror 23 can be changed by a motor 24 with an encoder, and the laser drive is controlled by a motor drive control unit 25 to scan the laser beam. The laser light emitted forward of the host vehicle 1 is reflected by a forward object and passes through a light receiving lens 26 and enters a light receiving unit (photodiode) 27.

Immediately before the light receiving section 27, as shown in FIG. 16, a slit plate 28 is disposed so as to be displaceable. The slit plate 28 has a slit 28a that extends in the vertical direction and is screwed to a screw rod 30 that is driven to rotate by a motor 29 with an encoder. In response to the rotation of the screw rod 30, the slit plate 28, that is, the slit 28a is displaced in the left-right direction. The drive of the motor 29 is controlled by a motor drive circuit 32. The synchronization signal generation circuit 31 controls the drive circuits 25 and 32 so that the emission timing of the laser light and the rotational positions of the motors 24 and 29 are synchronized, and the slit is set at a position corresponding to the scanning direction of the laser light. Control is performed so that the slit 28a of the plate 28 is located. That is, the light receiving lens 26 and the light receiving unit 27 are set so as to be able to receive light over the entire detection range to be scanned. However, since only the laser beam passing through the slit 28a is actually incident on the light receiving unit 27, That is, since only the reflected laser light from the scanning direction is incident on the light receiving section 27, the S / N ratio is improved.

FIGS. 17 and 18 show that the intensity of the laser beam is changed in the vertical direction so that the noise reflected from the road surface is reduced and the object ahead, especially a pedestrian, can be clearly detected. That is, when the laser light projected forward of the host vehicle 1 hits a strongly reflective object on the road surface, for example, a pedestrian crossing mark drawn on the road surface, a forward object present farther than the mark, particularly a walking object, is walked. It becomes difficult to detect a person. For this reason, as shown in FIG. 17, for example, a filter or the like is used so that the intensity of the middle portion in the vertical direction of the laser beam is higher than that of the periphery thereof in consideration of the height of the pedestrian from the road surface. It has been set. FIG.
In the example of the above, the intensity of the laser light is changed in four steps from a to d in the up-down direction, and the intensity of the laser light is set in the order of a, b, c, and d from the highest to the lowest. . As a result, the intensity of the laser light in the vertical direction becomes as shown in FIG. 18, and it is possible to reliably detect a pedestrian having a predetermined height from the road surface.

FIGS. 19 and 20 show modified examples of FIGS. 17 and 18, in which the light receiving sensitivity of the laser beam is changed in the vertical direction so that the intermediate portion in the vertical direction has the highest sensitivity. It is. That is, the light receiving sensitivity is changed in four steps of a to d in the vertical direction using, for example, a filter, and the light receiving sensitivity is sequentially changed from a higher to a lower, a, b, and
They are set in the order of c and d. The change in the light emission intensity in the vertical direction shown in FIGS. 17 and 18 and the change in the light reception sensitivity in the vertical direction shown in FIGS. 19 and 20 can be performed together.

Although the embodiment has been described above, the detection of a forward object, that is, a laterally moving object, can be performed by not only the laser radar but also the laser radar.
It can be used as appropriate such as a millimeter wave radar or an ultrasonic radar. Each step (step group) shown in the flowchart or various members such as a sensor and a switch can be expressed by adding a name of a means to a higher-level expression of its function. Further, the function of each step (step group) shown in the flowchart can be expressed as a functional unit (control unit) configured inside the controller U. The purpose of the present invention is
It is implied to provide not only what is explicitly stated but also what is substantially preferred or expressed as an advantage. Further, the present invention can be expressed as a control method.

[Brief description of the drawings]

FIG. 1 is a perspective view showing an example of a vehicle to which the present invention is applied.

FIG. 2 is a block diagram showing a control system of the present invention.

FIG. 3 is a simplified explanatory diagram showing a state where a pedestrian is about to enter a straight running path.

FIG. 4 is a diagram showing a setting example of a risk factor.

FIG. 5 is a diagram showing an example of setting a risk factor.

FIG. 6 is a diagram showing an example of setting a risk coefficient.

FIG. 7 is a diagram showing a setting example of a risk factor.

FIG. 8 is a flowchart showing a control example in which a risk degree is determined using the risk coefficients shown in FIGS. 4 to 7;

FIG. 9 is a flowchart showing a control example in which a risk degree is determined by using the risk coefficients shown in FIGS. 4 to 7;

FIG. 10 is a flowchart showing a control example in which a risk degree is determined using the risk coefficients shown in FIGS. 4 to 7;

FIG. 11 is a flowchart showing a control example in which the risk level is determined using the cumulative value of the risk levels.

FIG. 12 is a flowchart showing a control example for excluding a specific laterally moving object from objects of danger judgment.

FIG. 13 is a simplified explanatory view of excluding a specific laterally moving object from targets of danger determination.

FIG. 14 is a simplified explanatory diagram of excluding a specific horizontally moving object from targets of danger determination.

FIG. 15 is a main part system diagram showing an example of improving the S / N ratio of a received light signal using a slit plate.

FIG. 16 is a detailed perspective view of a main part of a slit plate portion shown in FIG. 15;

FIG. 17 is a simplified explanatory diagram showing a method for detecting a forward object by reducing the noise from the road surface by changing the emission intensity of the laser light in the vertical direction.

18 is a characteristic diagram showing the relationship between the vertical position and the light emission intensity obtained in the case of FIG. 17;

FIG. 19 is a simplified explanatory diagram showing a method of detecting a forward object by reducing the noise from the road surface by changing the light receiving sensitivity in the vertical direction.

20 is a characteristic diagram showing the relationship between the height position in the vertical direction and the light receiving sensitivity obtained in the case of FIG. 19;

[Explanation of symbols]

 1: own vehicle 2: radar (for detecting an object ahead) 3: camera (for detecting a traveling path) 4: navigation (for detecting road information) 5: VICS (for detecting road information-road-vehicle communication) 10: Traveling path U: Controller H (H1 to H4): Pedestrian (front object) K: Intersection (branch) K2: Branch

Claims (11)

[Claims] 1. An object detecting means for detecting a laterally moving object on a traveling path of a vehicle, and an object information acquiring means for acquiring a lateral moving speed, a position and a size of the laterally moving object detected by the object detecting means. And a danger determining unit that determines a danger related to the traveling of the host vehicle based on the lateral movement speed, the position, and the size acquired by the object information acquiring unit. Obstacle detection device. 2. The vehicle fault according to claim 1, further comprising: an alarm control unit that issues an alarm when the danger determination unit determines that the danger is equal to or more than a predetermined value. Object detection device. 3. The vehicle obstacle detecting device according to claim 1, wherein the risk determining means determines that the smaller the size of the laterally moving object, the greater the risk. . 4. The method according to claim 1, wherein the risk determining means is set to make a final risk determination based on the accumulation of a predetermined number of risks. Obstacle detection device for vehicles. 5. The visibility information detecting means according to claim 3, wherein said visibility information detecting means detects information relating to the visibility of a driver, and said visibility is low when said visibility information detected by said visibility information detecting means is poor. An obstacle detection device for a vehicle, further comprising: a correction unit that makes it easier for the alarm control unit to issue an alarm than when it is good. 6. The apparatus according to claim 2, further comprising a correction unit that corrects the predetermined value to be smaller in rainy weather than in fine weather, so that the warning is easily performed. Obstacle detection device for vehicles. 7. The apparatus according to claim 2, further comprising a correction unit that corrects the predetermined value to be smaller at nighttime than at daytime so that the warning is easily performed. Obstacle detection device for vehicles. 8. The driving system according to claim 1, further comprising: a future traveling route detecting means for detecting a traveling route in the near future to travel from a current position of the own vehicle. An obstacle detecting device for a vehicle, wherein the degree determining means determines the degree of risk in consideration of a future traveling system detected by the future traveling system detecting means. 9. A direction different from the direction in which the host vehicle is going to travel when the future driving system road detecting means detects that a branch portion of the host vehicle is present in front of the host vehicle. An obstacle detection device for a vehicle, wherein a laterally moving object is excluded from a target of danger determination. 10. A vehicle according to claim 8, wherein said future traveling system detecting means detects an operation state of a turn signal of the own vehicle, and a laterally moving object in a direction different from a direction indicated by the turn signal of the own vehicle is provided. An obstacle detection device for a vehicle, which is excluded from a target of danger determination. 11. A vehicle according to claim 1, wherein said vehicle is not likely to travel on the basis of an operation state of said turn signal of said vehicle and a traveling system road condition close to a current position of said vehicle. An obstacle detection device for a vehicle, wherein a traveling route is determined, and a laterally moving object existing near the determined predetermined traveling route is excluded from a target of danger determination.