KR20160131196A - Device for detecting an obstacle - Google Patents

Abstract

The present invention relates to a device for detecting an obstacle. According to an embodiment of the present invention, the device for detecting an obstacle comprises: a detection unit for detecting an object around a vehicle to generate the object detecting information; an analysis unit for analyzing the surrounding image filmed around the vehicle and sorting the type of the object to generated the object sorting information; and a control unit for setting the risk of the object by using the object detecting information received through the detection unit, for transmitting the object information with respect to the object to the analysis unit, and for setting the final risk with respect to the object by using the object sorting information received through the analysis unit.

Description

[0001] The present invention relates to a device for detecting an obstacle,

BACKGROUND OF THE INVENTION 1. Field of the Invention [0002] The present invention relates to a technology for detecting an obstacle around a vehicle, and more particularly,

In recent years, the number of vehicles equipped with an Advanced Driver Assisted System (ADAS) is increasing. These driver assistance systems include devices such as obstacle detection devices, Automatic Emergency Break (AEB), Lane Departure Warning System (LDWS) or Advanced Cruise Control (ACC) do.

Among these devices included in the driver assistance system, the obstacle detection device detects the presence of objects in the vicinity of the vehicle.

As an example, the obstacle sensing apparatus may be implemented to detect the presence or absence of nearby objects using a radar. However, although the radar has excellent detection capability of objects, there is a high probability of false alarms because there is no identification and classification function.

As another example, the obstacle detection device may be implemented to detect an object around the vehicle using a camera, and to classify and classify an object type. Specifically, the obstacle detecting device detects the presence or absence of an object around the vehicle from the image photographed through the camera, and classifies the type of the detected object such as a pedestrian or an automobile. However, in order to perform object identification and classification using a camera, a high-performance processor is required because a large amount of processing time and calculation amount are required. In addition, the sensitivity to the surrounding environment such as the weather, the time zone, and the like greatly varies depending on the surrounding environment.

Accordingly, the obstacle detection device can be implemented in the form of a fusion processor that detects objects (obstacles) around the vehicle even when the driver does not know by using the camera and the radar.

1 is a view for explaining an operation of an obstacle detecting apparatus using a camera and a radar. 1, the obstacle sensing apparatus 10 receives an object_list from a RADAR processor 12 that senses an object using a RADAR sensor 11 implemented in a vehicle, Information can be provided. The obstacle detection device 10 also receives object classification information (target_list) information from a vision processor 14 that detects and classifies objects in an image photographed by a camera sensor (vision sensor) 13, Can be provided.

The obstacle sensing device 10 detects and / or tracks an object around the vehicle by performing a fusion process on the provided object_list information and target_list information. In addition, the obstacle sensing apparatus 10 determines the danger according to the result of detection and / or tracking of an object around the vehicle, and controls the operation of the vehicle brakes or the steering apparatus in the event of a danger, Can be prevented.

An object of the present invention is to provide an apparatus for detecting an obstacle in a vehicle, which can efficiently detect the presence or the danger of an object by fusing sensed information of the surroundings of the vehicle and image information of the surroundings of the vehicle.

According to an aspect of the present invention, there is provided an obstacle detection apparatus comprising: a sensing unit for sensing an object around a vehicle to generate object sensing information; An analyzing unit that classifies the object and classifies the object and classifies the object and classifying the object and classifying the object and classifying the object; And a controller for setting a final risk level for the object using the received object classification information.

According to an embodiment of the present invention, a risk is primarily set for an object around a vehicle using perimeter sensing information, and a final risk for an object is determined based primarily on the type of object classified using the surrounding image information So that the vehicle can be controlled by using it, and it is possible to improve the detection performance and reliability of the object around the vehicle with few errors.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a view for explaining an operation of a fusion type obstacle sensing apparatus using a camera and a radar. FIG.
2 is a block diagram of an apparatus for detecting an obstacle for a vehicle according to an embodiment of the present invention.
3 is a flowchart of a method for detecting an obstacle by an obstacle detecting device for a vehicle according to an embodiment of the present invention.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention, and the manner of achieving them, will be apparent from and elucidated with reference to the embodiments described hereinafter in conjunction with the accompanying drawings. The present invention may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. And is provided to fully convey the scope of the invention to those skilled in the art, and the present invention is defined by the claims. It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In the present specification, the singular form includes plural forms unless otherwise specified in the specification. It is noted that " comprises, " or "comprising," as used herein, means the presence or absence of one or more other components, steps, operations, and / Do not exclude the addition.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the drawings, like reference numerals refer to like elements throughout. In the drawings, like reference numerals are used to denote like elements, and in the description of the present invention, In the following description, a detailed description of the present invention will be omitted.

2 is a block diagram of an apparatus for detecting an obstacle for a vehicle according to an embodiment of the present invention. 2, an obstacle sensing apparatus 100 for sensing an object in the vicinity of a vehicle includes a sensing unit 110, a photographing unit 120, an analysis unit 130, and a control unit 140. As shown in FIG.

The sensing unit 110 generates a signal for detecting an obstacle and transmits the signal to the outside of the vehicle. The sensing unit 110 receives a signal reflected by an object (target) existing in the vicinity of the vehicle, The presence or absence of an object is detected.

The sensing unit 110 senses an object around the vehicle using a sensing signal received from a sensing sensor mounted at a predetermined position outside the vehicle (e.g., a bumper of the vehicle). Preferably, at this time, the sensing unit 110 can sense an object in front of the vehicle.

The sensing unit 110 may be physically spaced from the other components of the obstacle sensing apparatus 100, and in this case, may be implemented to be connected via a vehicle network. Here, the detection sensor may be a radar sensor. Alternatively, the sensing sensor may be an ultrasonic sensor and a Liar sensor.

The sensing unit 110 generates object detection information (object_list) according to the detection result of the surrounding object. In addition, the sensing unit 110 transmits the generated object sensing information to the controller 140.

At this time, the object detection information includes position information of the object including the distance r between the vehicle and the object and the azimuth angle? Of the object. In addition, the object detection information may further include information such as reception sensitivity, mobility information, and the type of object to be estimated (e.g., a pedestrian, a car, etc.). Here, the type of the object to be estimated can be estimated using distance, angle, and sensitivity.

The photographing unit 120 includes one or more cameras for acquiring a peripheral image of the vehicle. In one embodiment, the photographing unit 120 can acquire a surrounding image through a vision sensor (camera) mounted in the front, rear, left and right rooms of the vehicle.

For example, at this time, the vision sensor may be a commonly used optical camera. At this time, the photographing unit 120 may use a camera for an Around View System (AVM) of the vehicle. In addition, the photographing unit 120 may further use a camera for a lane departure warning system (LDWS). Alternatively, the photographing unit 120 may use an infrared camera and Light Detection And Ranging (Lidar).

The analyzing unit 130 receives the surrounding image from the photographing unit 120 on a predetermined time basis, analyzes the received surrounding image through the image processing processor, and detects, identifies, and classifies the objects around the vehicle to generate object classification information do. For example, the analysis unit 130 can classify objects existing around the vehicle such as automobiles, pedestrians, median separators, and the like. The classification criterion of the object type can be preset in advance and stored in a separate memory.

The analysis unit 130 may generate object classification information by further using the object information received from the control unit 140 together with the peripheral image received from the photographing unit 120. [ This will be described in detail later.

The control unit 140 may be an electronic control unit (ECU) or a body control unit (BCM) for controlling a plurality of electronic control units.

The control unit 140 determines the position of the object using the object detection information received from the sensing unit 110 and sets a risk level for the object. The control unit 140 classifies objects by the degree of risk and generates object information for each degree of risk. The generated object information is transmitted to the analysis unit 130.

Also, the control unit 140 may set the final risk level for the object using the object classification information received from the analysis unit 130, and may output a control signal for controlling the vehicle according to the set final risk level.

Hereinafter, the operation of the analysis unit 130 and the control unit 140 according to the embodiment of the present invention will be described in detail.

First, the control unit 140 determines the position of an object using the object detection information received from the sensing unit 110, and sets a risk level for the object. At this time, the control unit 140 can set the risk level of the object by further using the mobility information of the vehicle (the vehicle) together with the object detection information.

Here, the mobility information of the vehicle may include at least one of speed information and acceleration information of the vehicle. Alternatively, the mobility information may include at least one of roll information, pitch information, and yaw information indicating changes (shakes) of the X, Y, and Z axes about the central axis of the vehicle. Also, the mobility information may include direction information of the vehicle.

In setting the risk level based on the object detection information, the control unit 140 can set a high risk of an object having a distance to the vehicle. For example, it is also possible to set a higher risk by adding a weight to an object existing on the path front of the vehicle rather than an object existing on the side of the vehicle. Further, the risk can be set high by adding a weight to an object approaching the vehicle, rather than an object moving away from the vehicle in the path of the vehicle.

In addition, the control unit 140 may classify each of a plurality of objects existing in the vicinity of the vehicle into a plurality of groups according to the risk set using the object detection information and the vehicle motion information. At this time, the control unit 140 can classify objects around the vehicle into a danger group (first group) and an interested group (second group) through Equation (1).

[Equation 1]

i = {1 to N} (i)

if (targt (i). risk> = threshold)

target (i) .group = Risky

else

target (i) .group = Interested group

Here, N is the number of objects sensed by the sensing unit 110, and the threshold value may be a predetermined value for sorting into a risk group and a target group. The function f may be a predetermined arithmetic expression to determine the risk using the object detection information_ (i) and the autoregularity information.

For example, through Equation (1), any one of the risk 0 to the risk 10 can be obtained. In addition, an object with a risk of 6 or more at a risk of 0 to 10 can be classified as a risk group. In some cases, 0 to 3 may be categorized into three groups of interest, 4 to 6 as a warning group, and 7 to 10 as a risk group. In this case, three groups can be classified using two threshold values.

Also, the control unit 140 can set the risk level of the object by further using the estimated path of the vehicle (self-vehicle) as shown in Equation (2).

&Quot; (2) "

i = {1 ~ N}, where i is the distance from the object to the target,

The control unit 140 can determine whether there is a possibility of collision within a predetermined time (for example, 1 second, 2 seconds, 5 seconds) and set a risk level when the vehicle has the same mobility.

The control unit 140 generates object information by grouping information on objects classified into the same group and transmits the object information to the analysis unit 130. For example, the control unit 140 may generate dangerous object information on objects classified as a dangerous group, object-of-interest information on objects classified as a target of interest, and transmit the generated object information to the analysis unit 130.

At this time, the control unit 140 may preferentially generate the risk group object information and transmit the information to the analysis unit 130. [

Here, regarding the position information of the object included in the object information, the position of the object in accordance with the mobility of the vehicle is corrected, and the control unit 140 can include the position information of the corrected object in the object information. The control unit 140 performs tracking of the object in relation to the position of the object and transmits the time of the sensing information of the sensed object received from the sensing unit 110 and the generated object information to the analysis unit 130 It is possible to generate the risk group object information including the estimated position information estimated based on the time difference.

Further, it may be necessary to convert the position of the object included in the object information into position coordinates that the analysis unit 130 can recognize. In this case, the control unit 140 may perform the position coordinate conversion process so that the analysis unit 130 recognizes the object, and may then transmit the object information to the analysis unit 130.

The analysis unit 130 generates object classification information using the object information (risk group object information) received from the control unit 140 through the above process. At this time, the analysis unit 130 can classify the object type by using the position information of the object included in the object information received from the control unit 140.

For example, the analyzer 130 may receive object information (risk group object information) for a plurality of high-risk groups determined to be a high-risk group in the control unit 140 and analyze the object types of the plurality of objects. The analysis unit 130 may also receive and analyze interest group object information for a plurality of objects determined to be of interest by the control unit 140 according to circumstances.

The analysis unit 130 may generate object classification information by correcting the position of the object included in the object information by using the object information and the surrounding image information received from the control unit 140. [ In addition, the analysis unit 130 can grasp the size or dimension of an object in the vicinity of the vehicle, and can include information on the size or dimension of the object in the object classification information.

The object classification information generated through the above process is transmitted to the controller 140.

The control unit 140 updates the risk level of the object using the object classification information received from the analysis unit 130 to set the final risk level.

At this time, the control unit 140 may need to convert the corrected position or dimension information of the object included in the object classification information received from the analysis unit 130 into the generally used position information.

Specifically, the control unit 140 sets the final risk level for the object using the dangerous object classification information received from the analysis unit 130. [ At this time, the final risk can be set as shown in Equation (3).

&Quot; (3) "

i = {1 to N}, the final risk = f2 (target (i)

or

i = {1 to N}, the final risk = f2 (target i), the classifier of the object, the dimension of the object,

Here, the classifier refers to the kind of object classified by the analyzing unit 130. The function f2 may be a predetermined arithmetic expression to further determine the risk by using the risk of the target (i), the classifier of the object, and the dimensions of the object.

In determining the final risk, the control unit 140 sets the final risk in consideration of the type of the object. For example, by applying a weight to a pedestrian, even if the pedestrian and the vehicle are likely to collide within the same time (e.g., 2 seconds), the pedestrian and the object classified as a vehicle can have different final risks.

The control unit 140 may output a control signal for controlling the vehicle according to the final risk set on the object. At this time, the control unit 140 outputs a driving control signal for traveling on a path without any risk of collision, while maintaining the path of the vehicle (own vehicle) or changing the path to the minimum. If the risk of collision can not be avoided, a driving control signal for driving on a path that minimizes the collision damage is output. If necessary, the control unit 140 may output the driving control signal or the warning control signal according to whether collision avoidance is possible with respect to the object having the highest collision risk among the plurality of objects for warning, that is, the object having the highest final risk.

At this time, the travel control signal may include a braking signal for controlling the braking of the vehicle and a steering signal for controlling the steering wheel of the vehicle. In addition, the warning control signal may include a warning screen output signal for outputting a warning message screen through the vehicle's cluster or AVN, and an alarm sound output signal for outputting a warning sound through a speaker or the like.

For example, when the final risk of the vehicle ahead of the vehicle exceeds a predetermined first value th1 (final risk> th1), the control unit 140 determines that avoidance of collision with an object is difficult, And outputs the driving control signal. At this time, the travel control signal may include a braking signal for controlling the braking of the vehicle and a steering signal for controlling the steering wheel of the vehicle.

If the final risk of the forward object of the vehicle exceeds a predetermined second value and is equal to or less than the first value th1 (th2 <final risk? Th1), the control unit 140 determines that collision with an object is possible And output a warning control signal for warning the driver of the presence of an object. For example, the control unit 140 may output an alarm control signal to output a warning screen through the vehicle's cluster or the AVN.

In addition, when the final risk of the vehicle ahead is less than or equal to the second value (final risk? Th2), the control unit 140 may perform tracking of the object based on the priority according to the final risk of the plurality of objects .

In this way, the control unit 140 can set a route free from the risk of collision of surrounding objects, and can perform tracking of the object based on the priority according to the final risk of a plurality of objects while maintaining the set route.

On the other hand, the control unit 140 can also set the final risk for the object classified as the interest group. For example, when the control unit 140 transmits the risk group object information to the analysis unit 130, the control unit 140 can set a timer. Thereafter, the time at which the risk group object classification information on the transmitted risk group object information is received from the analysis unit 130 is confirmed. For example, the remaining time of the timer can be obtained by comparing the time from the time when the transmission of the object information to the analysis unit 130 is started to the time when the object classification information is received from the analysis unit 130 and the set time of the timer. To this end, the set time value of the timer may be greater than the time value from the time when the object information is transmitted to the analysis unit 130 to the time when the object classification information is received from the analysis unit 130.

If the remaining time of the timer is equal to or longer than the preset time, the interested object information is transmitted to the analyzer 130. Thereafter, the control unit 140 may set the final risk for an object classified into the interest group through the final risk setting process using the object group classification information of interest received from the analysis unit 130. [

As described above, according to the embodiment of the present invention, by using the perimeter sensing information of the vehicle, it is possible to primarily set a risk level for an object around the vehicle, and secondarily to the object depending on the type of object classified using the surrounding image information The final risk for the vehicle is set and the vehicle is controlled by using it, so that it is possible to improve the detection and identification performance and reliability of objects around the vehicle with few errors.

3 is a flowchart illustrating a method for detecting an obstacle around a vehicle according to an exemplary embodiment of the present invention.

The object detection information is generated according to the object detection result of the surroundings of the vehicle detected through the sensor (S301). At this time, the sensor may be a radar sensor mounted in a predetermined position outside the vehicle (for example, a bumper of the vehicle) and communicated via a vehicle network.

The object detection information generated according to the object detection result of the surroundings of the vehicle includes the position information of the detected object. In this case, the position information includes the distance r between the vehicle and the object and the azimuth angle? Of the object. In addition, the object detection information may further include information such as reception sensitivity, mobility information, and the type of object to be estimated (e.g., a pedestrian, a car, etc.). Here, the type of the object to be estimated can be estimated using distance, angle, and sensitivity.

The risk of the object is set using the object detection information generated in step S301 (S303). At this time, the position of the object can be grasped by using the information on the motion of the vehicle (the vehicle) together with the object detection information, and the risk can be primarily set.

Here, the mobility information of the vehicle may include at least one of speed information and acceleration information of the vehicle. Alternatively, the mobility information may include at least one of roll information, pitch information, and yaw information indicating changes (shakes) of the X, Y, and Z axes about the central axis of the vehicle. Also, the mobility information may include direction information of the vehicle.

For example, through Equation (1), any one of the risk 0 to the risk 10 can be obtained. At this time, in setting the risk level based on the object detection information, the risk of an object having a distance to the vehicle can be set high. Further, the risk can be set high by adding a weight to an object existing on the front surface rather than an object existing on the side surface of the vehicle.

Each of the plurality of sensed objects is classified into a plurality of groups according to the risk (S305). At this time, the objects around the vehicle can be classified into the danger group (the first group) and the interest group (the second group). In addition, the risk of the object can be set by considering the expected path of the vehicle (the vehicle) as shown in Equation (2). For example, when the vehicle has the same motility as before, it is possible to determine whether there is a possibility of collision within a predetermined time (for example, 1 second, 2 seconds, 5 seconds), and to set the risk.

In step S305, object information for each group obtained by collecting information on the objects classified into the same group is generated (S307). For example, in step S305, risk group object information for risky objects classified as the dangerous group, interest object information on the interest group objects classified as the interest group, and the like can be respectively generated. At this time, the object information may include position information of objects.

The peripheral image information of the vehicle is acquired through the camera (S309). At this time, peripheral images can be acquired through a vision sensor (camera) mounted on the front, rear, left and right sides of the vehicle,

The object type is classified using the object information generated in step S307 and the peripheral image information acquired in step S309, and a timer is set (S311). At this time, it is possible to classify the object types of the dangerous object in priority by using the dangerous object information.

The obstacle sensing apparatus 100 analyzes surrounding image information obtained in units of a predetermined time through an image processing processor, and identifies and classifies objects around the vehicle to generate object classification information. For example, the position of a dangerous object may be corrected using the dangerous object information, and the type of the object may be classified into an automobile, a pedestrian, or a median barrier.

The classification criterion of the object type can be preset in advance and stored in a separate memory. Also, when correcting the position of an object, it may be necessary to convert the coordinates of the object position included in the dangerous object information to the position coordinates used in the image processing processor.

Further, a timer may be set at the time when the sorting operation of the risk group object is started. At this time, the set timer is used to determine whether or not to classify the object type for the object of interest in the future.

The object classification information is generated using the result of step S311 (S313). For example, the risk group object classification information can be generated by correcting the position of the object included in the risk group object information. Further, the size or dimension of an object in the vicinity of the vehicle can be grasped, and information on the size or dimension of the object can also be included in the risk group object classification information.

The final risk level for the object is set in consideration of the object classification information generated in step S313 (S315). At this time, it may be necessary to convert the corrected position or dimension information of the object included in the object classification information into position information that is generally used.

For example, a relatively high weight is applied to a pedestrian than a car through a dangerous object classification information, so that even if a pedestrian and an object classified as a car are likely to collide within the same time (for example, 2 seconds) May have different final risks.

In step S315, whether collision avoidance is possible or not is determined according to the final risk of the object set (S317). At this time, the position of the object and the distance between the vehicle and the object are checked based on the vehicle, and it is confirmed whether collision between the vehicle and the object can be avoided.

For example, when the final risk of the object having the highest risk exceeds the predetermined first value th1 (final risk> th1), it can be determined that it is difficult to avoid the collision with the object. If the final risk of the object having the highest risk exceeds the predetermined second value and is equal to or less than the first value th1 (th2 <final risk? Th1), it can be determined that avoidance of collision with the object is possible . For this purpose, a collision avoidance determination algorithm according to the position and distance of the vehicle and the object may be stored.

If it is determined in step S317 that the collision between the vehicle and the object can be avoided, a driving control signal for controlling the vehicle for collision avoidance is output (S319). At this time, the travel control signal includes a braking signal for controlling the braking of the vehicle and a steering signal for controlling the steering wheel of the vehicle so as to avoid collision with an object.

For example, when there is a sufficient distance between the vehicle and the object, collision with an object can be avoided by braking the vehicle by turning on the brake of the vehicle. Alternatively, by switching the steering wheel of the vehicle, it is possible to avoid collision with an object by turning the direction of the vehicle.

If it is determined in step S317 that the collision between the vehicle and the object is impossible, the driving control signal for controlling the vehicle in the direction of low risk of collision is output (S321). At this time, the travel control signal includes a braking signal for controlling the braking of the vehicle and a steering signal for controlling the steering wheel of the vehicle so as to avoid collision with an object.

For example, by switching the steering wheel of the vehicle, it is possible to reduce the amount of impact of collision with an object by turning the vehicle. At this time, by turning on the brake of the vehicle to reduce the speed of the vehicle, the amount of impact can be reduced.

In addition, it outputs a driving control signal of the vehicle and outputs a warning control signal to warn the driver of the risk of collision with an object. At this time, the warning control signal may include a warning screen output signal for outputting a warning message screen through a vehicle's cluster or AVN, and an alarm sound output signal for outputting a warning sound through a speaker or the like.

In step S315, a control signal for controlling the vehicle is outputted in accordance with the final risk of the object set (S317). At this time, the running control signal or the warning control signal may be outputted depending on whether collision with the object having the highest risk of collision among the plurality of objects is highest or can be avoided.

For example, when the final risk of an object having the highest risk exceeds the predetermined first value th1 (final risk> th1), it is determined that avoidance of collision with an object is difficult, Signal is output. At this time, the travel control signal may include a braking signal for controlling the braking of the vehicle and a steering signal for controlling the steering wheel of the vehicle.

If the final risk of the object having the highest risk exceeds the predetermined second value and is equal to or less than the first value th1 (th2 <final risk? Th1), it is determined that it is possible to avoid the collision with the object, It is possible to output a warning control signal for warning of the presence of an object. For example, the control unit 140 may output an alarm control signal to output a warning screen through the vehicle's cluster or the AVN.

When the sorting operation of the object type of the dangerous object is completed, the remaining time of the timer set in step S311 is checked to determine whether to classify the object type for the object of interest (S319S323). If the remaining time of the timer is equal to or longer than the preset time, the process returns to step S311 to classify the object type for the object of interest by using the interest object information. Thereafter, through steps 313 to S319S321, the final risk level can be set for the object of interest and the vehicle control signal corresponding thereto can be outputted to the object of interest similarly to the dangerous object.

As described above, according to the embodiment of the present invention, by using the perimeter sensing information of the vehicle, it is possible to primarily set a risk level for an object around the vehicle, and secondarily to the object depending on the type of object classified using the surrounding image information The final risk for the vehicle is set and the vehicle is controlled by using it, so that it is possible to improve the detection and identification performance and reliability of objects around the vehicle with few errors.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It is to be understood that the invention may be embodied in other specific forms. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. The scope of the present invention is defined by the appended claims rather than the detailed description, and all changes or modifications derived from the scope of the claims and their equivalents shall be construed as being included within the scope of the present invention.

100: Obstacle detection device
110: sensing unit 120:
130: Analyzer 140:

Claims (1)

A sensing unit that senses an object in the vicinity of the vehicle and generates object detection information;
An analysis unit for analyzing a surrounding image photographed around the vehicle and sorting the types of the objects to generate object classification information; And
Wherein the analyzing unit is configured to set the risk level of the object using the object sensing information received through the sensing unit, to transmit object information about the object to the analyzing unit, and to use the object classification information received through the analyzing unit, A control unit for setting a final risk level for the device;
And an obstacle detection device.

Images