KR102467227B1 - Method and apparatus for preventing collisions during parking using ultra wideband radar - Google Patents

Abstract

The present invention relates to a parking collision avoidance method and apparatus using an ultra-wideband radar, and a parking collision avoidance device mounted on a vehicle according to an embodiment of the present invention includes first to Nth radars transmitting ultra-wideband radar signals and the above When ultra-wideband radar signals are transmitted, a reception signal processing unit that processes the received radar signal, an obstacle detection unit that detects an obstacle based on the processed radar signal, and a distance calculation that calculates the separation distance from the vehicle to the detected obstacle and a boundary shape mapping unit generating an obstacle map including the boundary shape of the detected obstacle based on the processed radar signal and a boundary shape display unit outputting an obstacle map including the boundary shape.

Description

Parking collision avoidance method and apparatus using ultra-wideband radar

The present invention relates to parking collision avoidance, and more particularly, by utilizing an ultra-wideband (UWB) radar having high distance resolution (several cm) to provide driver convenience and safety through highly reliable recognition of surrounding situations when parking a vehicle. Parking collision using ultra-wideband radar, which improves safety and visually checks the location and boundary shape of dangerous obstacles behind the vehicle when parking through obstacle detection and mapping close to the vehicle using multiple ultra-wideband radars It relates to prevention methods and devices.

Vehicles such as automobiles are required to function as various convenience means to provide users with a more stable and comfortable driving state beyond their function as means of transportation.

Accordingly, various electronic products and sensors are being reflected in automobiles.

In the case of the ultrasonic sensor used as an existing parking assist device, the detection distance was short, and the performance deteriorated according to changes in the external environment such as temperature, humidity, and rain.

In addition, in the case of a parking assistance system using a camera sensor, there is a disadvantage in that distortion occurs due to image reconstruction and a significant performance degradation occurs in situations such as night, backlight, and rainy weather, and only the image captured by the camera is displayed on the display. There were limits to being

In the case of Korean Registered Patent No. 10-1406211 (AVM image providing device and method for vehicles, Hyundai Autron), when an image is generated using a plurality of cameras, in order to prevent obstacle recognition failure caused by distortion of the image junction, A method of generating divided images using a top view image and a panoramic image of a corner portion of a vehicle has been disclosed. However, since the present invention is based only on video images, it is vulnerable to ambient noise and external environments such as nighttime, backlight, and rain, and it is difficult to perform its function when a failure occurs in one of a plurality of cameras.

In addition, in the case of Korea Patent Publication No. 10-2011-0061885 (obstacle position warning device for automatic parking assistance system, Hyundai Motor Company), only the alarm sound according to the distance is transmitted to alleviate the driver's difficulty in determining the exact location of the obstacle A method has been disclosed in which a warning sound is output from a speaker corresponding to a direction in which a corresponding obstacle exists by partitioning the surrounding area of the vehicle into several areas. However, it is difficult to determine the exact location and distance of an obstacle through only the warning sound, and there is a possibility that corresponding danger information may not be recognized when there is noise around.

As such, there is a need for a driver's convenience technology capable of alleviating the anxiety of inexperienced drivers and elderly drivers when parking and reversing, and active research is being conducted.

In addition, there is a demand for a parking assistance system capable of intuitively providing information about obstacles close to the vehicle with high distance resolution without performance degradation due to changes in the external environment.

In the case of ultra-wideband (UWB) radar, it is possible to measure precise distance with excellent time resolution, has strong immunity to narrowband noise, and has the advantage of being able to detect obstacles at a distance of up to 10 m.

Therefore, a safer and more convenient parking assistance system can be provided by utilizing the ultra-wideband radar.

The present invention has been devised to solve the problems of the prior art described above, and an object of the present invention is to provide a parking collision avoidance method and apparatus using an ultra-wideband radar.

Another object of the present invention is a parking collision avoidance method using ultra-wideband radar capable of preventing collisions that may occur during parking in advance by providing precise obstacle detection and warning without performance degradation by the external environment using ultra-wideband radar. and to provide an apparatus.

Another object of the present invention is to provide a parking collision avoidance method and apparatus using ultra-wideband radar capable of improving driver convenience during parking by visually displaying location information and boundary shapes of dangerous obstacles close to the vehicle.

The technical problems to be achieved in the present invention are not limited to the above-mentioned technical problems, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

The present invention provides a parking collision avoidance method and apparatus using ultra-wideband radar.

A parking collision avoidance device mounted on a vehicle according to an embodiment of the present invention includes first to Nth radars that transmit ultra-wideband radar signals and a reception signal processing unit that processes radar signals received when the ultra-wideband radar signals are transmitted. An obstacle detecting unit detecting an obstacle based on the processed radar signal, a distance calculating unit calculating a separation distance from the vehicle to the detected obstacle, and a boundary shape of the detected obstacle based on the processed radar signal It may include a boundary shape mapping unit generating an included obstacle map and a boundary shape display unit outputting an obstacle map including the boundary shape.

Here, the received signal processing unit includes a band pass filtering unit for removing signal components other than the frequency band corresponding to the transmitted ultra-wideband radar signal from the received radar signal, and the transmitted ultra-wideband radar signal from the received radar signal. may include a ground noise removal unit that removes signal components reflected from the ground and a clutter removal unit that removes unnecessary signal components other than the transmitted ultra-wideband radar signal using a moving average method.

In addition, the obstacle detecting unit may determine that an obstacle exists within a predetermined radar recognition distance when the intensity of the processed radar signal exceeds a predetermined threshold value.

In addition, the parking collision avoidance device further includes a warning alarm unit outputting a warning alarm, and the distance calculating unit may control the warning alarm unit to output the warning alarm when the separation distance is within a predetermined warning distance threshold. .

Here, the warning alarm may be output through at least one of a speaker, a vibration element, and a lamp provided in the vehicle.

In addition, the boundary shape mapping unit generates an obstacle map by applying cross-correlation between a map generator for generating first to Nth obstacle maps corresponding to the first to Nth radars and the first to Nth obstacle maps. and a filtering unit for applying 2D Gaussian filtering to the obstacle map to which the cross-correlation is applied, and a correction unit for applying an interpolation method to the filtered obstacle map.

In addition, the cross-correlator may generate the cross-correlated obstacle map by cross-correlating obstacle maps corresponding to physically adjacent radars among the first to N-th radars and adding the cross-correlation.

Here, the cross-correlated obstacle map may include boundary information of the detected obstacle.

According to another embodiment of the present invention, a parking collision avoidance method in a vehicle equipped with first to Nth radars that transmit ultra-wideband radar signals includes the steps of processing a radar signal received when the ultra-wideband radar signal is transmitted The step of detecting an obstacle based on the processed radar signal, the step of calculating a separation distance from the vehicle to the detected obstacle, and the obstacle map including the boundary shape of the detected obstacle based on the processed radar signal. It may include generating and outputting an obstacle map including the boundary shape.

Here, the processing of the radar signal includes removing signal components other than the frequency band corresponding to the transmitted ultra-wideband radar signal from the received radar signal, and the transmitted ultra-wideband radar signal from the received radar signal. It may include removing signal components reflected from the ground and removing unnecessary signal components other than the transmitted ultra-wideband radar signal using a moving average method.

In addition, when the strength of the processed radar signal exceeds a predetermined threshold value, it may be determined that an obstacle exists within a predetermined radar recognition distance.

The method may further include outputting a warning alarm, and if the separation distance is within a predetermined warning distance threshold, the warning alarm may be output.

Here, the warning alarm may be output through at least one of a speaker, a vibration element, and a lamp provided in the vehicle.

In addition, the generating of the obstacle map may include generating the first to Nth obstacle maps corresponding to the first to Nth radars, and applying cross-correlation between the first to Nth obstacle maps to obtain an obstacle map. It may include generating, applying 2D Gaussian filtering to the obstacle map to which the cross-correlation is applied, and applying an interpolation method to the filtered obstacle map.

Here, the cross-correlated obstacle map may be generated by cross-correlating and adding obstacle maps corresponding to physically adjacent radars among the first to N-th radars.

Here, the cross-correlated obstacle map may include boundary information of the detected obstacle.

Another embodiment of the present invention may provide a program recorded on a recording medium for realizing the parking collision avoidance method and a recording medium on which the program is recorded.

The above aspects of the present invention are only some of the preferred embodiments of the present invention, and various embodiments in which the technical features of the present invention are reflected are detailed descriptions of the present invention to be detailed below by those skilled in the art. It can be derived and understood based on.

The effects of the method and apparatus according to the present invention are described as follows.

The present invention has the advantage of providing a parking collision avoidance method and apparatus using ultra-wideband radar.

In addition, the present invention utilizes ultra-wideband radar to provide precise obstacle detection and warning without performance degradation by the external environment, thereby preventing collisions that may occur during parking in advance. A parking collision avoidance method using ultra-wideband radar and There are advantages to providing the device.

In addition, the present invention has the advantage of providing a parking collision avoidance method and apparatus using ultra-wideband radar capable of improving driver convenience during parking by visually displaying location information and boundary shapes of dangerous obstacles close to the vehicle.

Advantages and effects obtainable from the present invention are not limited to the above-mentioned advantages and effects, and other effects not mentioned will become clear to those skilled in the art from the description below. You will be able to understand.

The accompanying drawings are provided to aid understanding of the present invention, and provide embodiments of the present invention together with a detailed description. However, the technical features of the present invention are not limited to specific drawings, and features disclosed in each drawing may be combined with each other to form a new embodiment.
1 is a block diagram for explaining the structure of a parking collision avoidance device according to an embodiment of the present invention.
Figure 2 is a block diagram for explaining the configuration of the received signal processing unit of Figure 1 according to an embodiment of the present invention.
3 is a diagram for explaining the operation of an obstacle detecting unit according to an embodiment of the present invention.
4 is a diagram for explaining an operation of a distance calculation unit according to an embodiment of the present invention.
5 is a diagram for explaining a method of generating an obstacle map according to an embodiment of the present invention.
6 is a diagram for explaining a procedure for generating an obstacle map using cross-correlation according to an embodiment of the present invention.
7 is a diagram for explaining advantages of generating an obstacle map using cross-correlation according to an embodiment of the present invention.
8 is a diagram for explaining a method of filtering an obstacle map generated according to an embodiment of the present invention.
9 is a diagram for explaining a method for shaping an obstacle boundary using an interpolation method according to an embodiment of the present invention.
10 is a block diagram for explaining the configuration of the boundary shape mapping unit of FIG. 1 described above.

Hereinafter, an apparatus and various methods to which embodiments of the present invention are applied will be described in more detail with reference to the drawings. The suffixes "module" and "unit" for components used in the following description are given or used together in consideration of ease of writing the specification, and do not have meanings or roles that are distinct from each other by themselves.

In the above, even though all the components constituting the embodiment of the present invention have been described as being combined or operated as one, the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all of the components may be selectively combined with one or more to operate. In addition, although all of the components may be implemented as a single independent piece of hardware, some or all of the components are selectively combined to perform some or all of the combined functions in one or a plurality of hardware. It may be implemented as a computer program having. Codes and code segments constituting the computer program may be easily inferred by a person skilled in the art. Such a computer program may implement an embodiment of the present invention by being stored in a computer readable storage medium, read and executed by a computer. A storage medium of a computer program may include a magnetic recording medium, an optical recording medium, and the like.

In addition, terms such as "include", "comprise" or "have" described above mean that the corresponding component may be inherent unless otherwise stated, excluding other components. It should be construed as being able to further include other components. All terms, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs, unless defined otherwise. Commonly used terms, such as terms defined in a dictionary, should be interpreted as consistent with the meaning in the context of the related art, and unless explicitly defined in the present invention, they are not interpreted in an ideal or excessively formal meaning.

Also, terms such as first, second, A, B, (a), and (b) may be used in describing the components of the present invention. These terms are only used to distinguish the component from other components, and the nature, order, or order of the corresponding component is not limited by the term. When an element is described as being “connected,” “coupled to,” or “connected” to another element, that element is or may be directly connected to the other element, but there is another element between the elements. It will be understood that elements may be “connected”, “coupled” or “connected”.

1 is a block diagram for explaining the structure of a parking collision avoidance device according to an embodiment of the present invention.

Referring to FIG. 1, the parking collision avoidance apparatus 100 includes first to Nth radars 110, a received signal processing unit 120, an obstacle detection unit 130, a distance calculating unit 140, a boundary shape mapping unit ( 150), a boundary shape display unit 160, and a warning alarm unit 170. Here, N may be a natural number equal to or greater than 2.

Note that the first to Nth radars 110 may be mounted on one side of the rear of the vehicle, but this is only one embodiment, and may be mounted on the side, front, etc. as well as the rear according to the configuration of the vehicle. Should be.

Each radar is an ultra wideband (UWB) radar and may include a transmitter 111 and a receiver 112.

A UWB radar is a radar that uses a very wide frequency band of several gigahertz (GHz) bands in the base band, and uses very short pulses of several nanoseconds or several pico seconds.

Therefore, UWB radar can provide high-resolution signals capable of accurate distance and location measurement, and has very low spectral power, such as noise in existing wireless communication systems, so it is suitable for communication systems such as mobile communication networks, broadcasting networks, and satellite networks. mutual interference can be minimized.

The transmitter 111 may transmit a radar signal in a predetermined UWB frequency band with a predetermined strength. Here, the strength of the radar signal may be determined based on a maximum obstacle detection distance required by the corresponding vehicle.

The receiving unit 112 may receive the radar signal and transmit it to the signal processing unit 120 .

The received signal processing unit 120 may remove unnecessary signal components and noise components included in radar signals received from the first to Nth radars 110 . A signal processed by the reception signal processing unit 120 may be provided to the obstacle detection unit 130 .

The received signal processing unit 120 according to an embodiment performs a bandpass filtering function to remove signal components other than the frequency band corresponding to the transmitted radar signal, and transmits an incoming signal after the transmitted radar signal hits the ground and is reflected. It may be configured to include at least one of a ground noise removal function to remove and a clutter removal function to remove unnecessary signal components other than the target by using a moving average method.

The obstacle detection unit 130 may determine whether or not an obstacle exists within a predetermined radar recognition distance based on a signal received from the reception signal processing unit 120 .

For example, the obstacle detection unit 130 may compare the strength of the signal received from the reception signal processing unit 120 with a predetermined threshold value considering characteristics of engine vibration and road vibration of the vehicle. As a result of the comparison, if the intensity of the received signal exceeds the corresponding threshold value, the obstacle detecting unit 130 may determine that an obstacle exists within the corresponding radar recognition distance.

The obstacle detecting unit 130 may provide an obstacle detection result to the distance calculating unit 140 through a predetermined control signal.

When an obstacle is detected by the obstacle detecting unit 130, the distance calculator 140 may calculate a separation distance from the vehicle to the detected obstacle. Here, the separation distance may be calculated by measuring the degree of attenuation of the transmitted radar signal. In addition, the measured separation distance may be corrected and tracked using a Kalman filter. Here, the Kalman filter is an optimal mathematical calculation process that allows an accurate position or distance to be predicted after a certain period of time by analyzing existing measurement values mixed with noise through the least squares method.

The distance calculation unit 140 sends a predetermined control signal to the warning alarm unit 170 when the calculated separation distance is within a predefined alarm distance threshold so that the warning alarm unit 170 outputs a predetermined warning alarm message. The alarm unit 170 may be controlled. The warning alarm message may be output through a speaker, a vibration element, a lamp, etc., but is not limited thereto.

Also, the warning alarm message may be different according to the calculated separation distance. For example, as the detected obstacle approaches the vehicle, the intensity and repetition period of the warning sound output through the speaker may increase. As another example, the intensity of vibration of the vibration element provided on one side of the driver's seat may be controlled to increase as the detected obstacle approaches the vehicle.

When it is determined that the detected obstacle is located within a predetermined warning distance threshold, the distance calculating unit 140 may transmit a predetermined control signal instructing boundary shape mapping to the boundary shape mapping unit 150 .

The boundary shape mapping unit 150 may map the location and boundary shape of the obstacle to a 2D obstacle map when the obstacle is detected within the warning distance. Here, the 2D obstacle map may be generated by a radar imaging technique based on intensity information of radar signals received through the first through Nth radars 110 and the received signal processing unit 120 .

For example, the obstacle map may have a two-dimensional lattice structure including a plurality of pixels, and each pixel has a boundary shape of an obstacle calculated based on information on the strength of a signal received through each radar corresponding to a corresponding position. information may be included.

After the 2D obstacle map is generated, the boundary shape mapping unit 150 may map the boundary shape of the corresponding obstacle to the closest distance between the vehicle and the detected obstacle.

In addition, the boundary shape mapping unit 150 may determine the degree of danger based on the minimum separation distance from the obstacle, and determine the color of the mapped boundary shape according to the determined degree of danger.

For example, if the risk level is “low”, the color of the boundary shape may be green, and if the risk level is “high”, the color of the boundary shape may be displayed in red, but this is only one embodiment, and those skilled in the art Depending on the design, the detailed definition of the degree of risk and the corresponding color mapping method may be different, and may be implemented to enable user setting through a predetermined user interface.

That is, the boundary shape mapping unit 150 according to the present invention displays the color of the line representing the boundary shape of the obstacle differently according to the degree of danger determined based on the distance from the vehicle to the obstacle boundary, thereby providing a driver's visual perception of the obstacle. may help understanding.

The boundary shape display unit 160 may display the boundary shape of the obstacle on the display based on the boundary shape mapping result of the detected obstacle by the boundary shape mapping unit 150 .

For example, the boundary shape display unit 160 may be implemented through a screen of an AVN (Audio Video Navigation) terminal provided in a vehicle or a cluster screen, but is not limited thereto, and a separate display means-eg For example, a transparent display mounted on a windshield of a vehicle, a hologram display displayed on one side of a windshield of a vehicle, and the like may be used.

The boundary shape mapping unit 150 according to an embodiment includes information about the boundary of an obstacle by using cross correlation between the first to Nth obstacle maps generated in response to the first to Nth radar signals. Obstacle maps can be created.

Here, the first to Nth radar signals may be generated through the first to Nth radars 110 and the received signal processing unit 120 .

In addition, the boundary shape mapping unit 150 according to the embodiment of the present invention may apply a 2D Gaussian filter mainly used for photo and video recognition to the created obstacle map.

When a 2D Gaussian filter is applied, not only can the obstacle map image be filtered smoothly, but also the signal to noise ratio (SNR) of the obstacle map image can be improved, so that noise can be removed and spatial resolution loss can be minimized. have.

In addition, the boundary shape mapping unit 150 according to another embodiment of the present invention identifies the location where the corresponding obstacle is located closest to the vehicle on the obstacle map where cross-correlation and Gaussian filtering are completed, and identifies the identified location as a target point can be determined by

The boundary shape mapping unit 150 may more accurately estimate the boundary shape of the obstacle by performing interpolation based on the determined target point.

Figure 2 is a block diagram for explaining the configuration of the received signal processing unit of Figure 1 according to an embodiment of the present invention.

Referring to FIG. 2 , the received signal processing unit 120 includes a signal receiving unit 210, a bandpass filtering unit 220, a ground noise removing unit 230, a clutter removing unit 240 and a signal transmitting unit 250. can be configured to include

The signal receiver 210 may receive a raw signal corresponding to the radar signal transmitted through the transmitter 111 from the receiver 112 of the first to Nth radars 110 .

The bandpass filtering unit 220 may remove signal components other than the frequency band corresponding to the transmitted radar signal from the raw signal.

The ground noise removing unit 230 may remove a signal component that is reflected when the transmitted radar signal hits the ground and is reflected from the low signal.

The clutter remover 240 may remove unnecessary signal components other than the target signal - that is, the target signal - by using a moving average method.

The signal transmission unit 250 includes the obstacle detection unit 130, the distance calculation unit 140, and the boundary shape mapping unit 150 of FIG. ) can be provided to at least one of them.

3 is a diagram for explaining the operation of an obstacle detecting unit according to an embodiment of the present invention.

Referring to FIG. 3 , reference numeral 310 shows a change in intensity of a received radar signal measured by the obstacle detecting unit 130 when an obstacle does not exist within a radar recognition distance of the corresponding vehicle.

As shown in reference numeral 310, when there is no obstacle within the radar recognition distance, the strength of the received radar signal for the mode sample index is less than the threshold value.

Reference numeral 320 shows a change in strength of a received radar signal measured by the obstacle detecting unit 130 when an obstacle exists within a radar recognition distance of the corresponding vehicle.

As shown in reference numeral 310, if an obstacle exists within a radar recognition distance, the intensity of the received radar signal for some sample indices is greater than a threshold value, as shown in reference 321.

4 is a diagram for explaining an operation of a distance calculation unit according to an embodiment of the present invention.

Referring to FIG. 4 , when receiving a predetermined control signal indicating that an obstacle has been detected from the obstacle detecting unit 130, the distance calculating unit 140 processes the first to Nth radar signals processed by the received signal processing unit 120. A separation distance from the vehicle to the detected obstacle may be calculated based on the strength of .

If, as shown in reference number 410, it is confirmed that the distance corresponding to the received radar signal is within the predetermined warning distance, the distance calculation unit 140 sets the warning alarm unit 170 to output a predetermined warning alarm. You can control it.

5 is a diagram for explaining a method of generating an obstacle map according to an embodiment of the present invention.

A 2D obstacle map for mapping obstacle locations according to an embodiment of the present invention may be generated based on radar signal strength information received from a plurality of radars.

Referring to FIG. 5 , an obstacle map I 510 according to an embodiment may have a 2D grid structure having N×M pixels. Here, N and M may be natural numbers greater than 2.

In the parking collision avoidance apparatus 100 according to the present invention, an obstacle map may be generated for each radar provided in a vehicle. That is, when K UWB radars for parking collision avoidance are provided in a vehicle, K obstacle maps corresponding to each radar may be generated. Thereafter, cross-correlation is performed on obstacle maps of physically adjacent radars to generate an obstacle map I 510 including information about the boundary of obstacles.

As an example, assume that the first to Kth radars 520 are provided in a vehicle. In this case, the distance DN(xm, yn) from the Nth radar 521 to the corresponding first pixel 511 corresponding to the first pixel 511 - that is, the pixel (xm, yn) - is expressed by Equation 1 below: As such, it can be calculated based on the strength of the signal received through the Nth radar.

<Equation 1>:

X coordinate of

Radar /

Y coordinate of

Radarhere,

X coordinate of

Radar /

Y coordinate of

Radar

The obstacle map Sn(xm, yn) corresponding to one N-th radar for the first pixel 511 is the strength RN of the signal received through the N-th radar and the above Equation 1, as shown in Equation 2 below. It can be generated based on the product of the distance DN(xm, yn) calculated by

<Formula 2>:

과 N-1번째 레이더에 의해 생성된 장애물 앱

을 상호 상관하여 생성될 수 있다.The obstacle app I(xm, yn) is the obstacle app of the Nth radar generated by Equation 2, as shown in Equation 3 below.

and the obstacle app generated by the N-1th radar

It can be generated by cross-correlating.

<Formula 3>:

The obstacle position estimation method using the conventional radar imaging technique is a method of simply adding N obstacle maps generated from each radar in pixel units. In particular, there was a problem that the resolution for the cross-range was inferior.

However, since the parking collision avoidance method according to the present invention can generate an obstacle map including boundary shape information for detected obstacles by cross-correlating N obstacle maps, the resolution for the cross-range is There are high advantages.

6 is a diagram for explaining a procedure for generating an obstacle map using cross-correlation according to an embodiment of the present invention.

Referring to FIG. 6, the parking collision avoidance apparatus 100 according to the present invention measures the strength of signals received through the first to Nth radars 610 mounted on the vehicle, respectively, and based on the strength of the measured signals. Thus, first to Nth obstacle maps 620 corresponding to respective radars may be generated.

Subsequently, as shown in reference numeral 630, the parking collision avoidance apparatus 100 cross-correlates obstacle maps corresponding to two physically adjacent radars, and adds up the cross-correlation values to obtain boundary information corresponding to the detected obstacle. An embedded obstacle map 640 can be created.

7 is a diagram for explaining advantages of generating an obstacle map using cross-correlation according to an embodiment of the present invention.

In detail, FIG. 7 shows test results performed in an experimental environment in which four UWB radars are mounted on a vehicle and a wall exists behind the vehicle within a radar recognition distance, as shown at reference numeral 710 .

Reference numeral 720 shows an obstacle map generated by simple addition of first to fourth obstacle maps generated corresponding to respective radars without cross-correlation in the experimental environment of reference 710 above. On the other hand, reference numeral 730 shows an obstacle map generated by cross-correlating the first to fourth obstacle maps generated corresponding to each radar in the experimental environment of reference numeral 710 .

Comparing reference numbers 720 and 730, the obstacle map of reference number 730 has a higher cross-range resolution than the obstacle map of reference number 720. In addition, the obstacle map of reference numeral 730 has the advantage of more clearly displaying the position and boundary shape of the obstacle than the obstacle map of reference numeral 720 .

8 is a diagram for explaining a method of filtering an obstacle map generated according to an embodiment of the present invention.

Referring to FIG. 8 , reference numeral 810 shows an obstacle map generated through cross-correlation of first through Nth obstacle maps.

The parking collision avoidance device 100 according to an embodiment of the present invention generates an obstacle map 830 having a more smoothly filtered image by performing a convolution operation on the 2D Gaussian filter 820 and the obstacle map 810. can The filtered obstacle map 830 has an improved Signal to Noise Ratio (SNR), so noise is removed and spatial resolution loss can be minimized.

9 is a diagram for explaining a method for shaping an obstacle boundary using an interpolation method according to an embodiment of the present invention.

Referring to FIG. 9 , the parking collision avoidance apparatus 100 according to an embodiment of the present invention may apply an interpolation method to more clearly define an obstacle boundary shape included in an obstacle map generated by applying cross-correlation and Gaussian filtering. have.

Reference numeral 911 shows an obstacle map generated in a state in which a wall surface is disposed to the rear of the vehicle, as shown by reference numeral 910 . Reference numeral 912 shows an obstacle map after interpolation is applied to the obstacle map of reference numeral 911.

Comparing drawing number 911 and drawing number 912, in drawing number 911, there is a part where the boundary shape of the obstacle is broken, but in the drawing number 912 to which the interpolation method is applied, not only is there no break in the boundary shape of the obstacle, but also the border of the obstacle has the advantage of being more clearly distinguished than the drawing number 911.

10 is a block diagram for explaining the configuration of the boundary shape mapping unit of FIG. 1 described above.

Referring to FIG. 10 , the boundary shape mapping unit 150 may include a map generator 1010, a cross-correlation unit 1020, a filtering unit 1030, and a correction unit 1040.

The map generator 1010 may generate first to Nth obstacle maps respectively corresponding to the first to Nth radars based on the first to Nth radar received signals processed by the received signal processing unit 120. .

The cross-correlator 1020 may cross-correlate obstacle maps corresponding to two physically adjacent radars and add up the cross-correlation values to generate an obstacle map including boundary information of detected obstacles.

The filtering unit 1030 may generate an obstacle map having a more smoothly filtered image by applying a 2D Gaussian filter to the obstacle map generated by the cross-correlation unit 1020 . Here, the signal-to-noise ratio (SNR) of the filtered obstacle map is improved so that noise is removed and spatial resolution loss can be minimized.

The correction unit 1040 may apply an interpolation method to the filtered obstacle map to generate an obstacle map including a clear obstacle boundary shape without interruption.

It is apparent to those skilled in the art that the present invention can be embodied in other specific forms without departing from the spirit and essential characteristics of the present invention.

Accordingly, the above detailed description should not be construed as limiting in all respects and should be considered illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention.

Claims (18)

A device mounted on a vehicle to prevent collision during parking,
First through N-th radars transmitting ultra-wideband radar signals;
a reception signal processor processing a radar signal received when the ultra-wideband radar signal is transmitted;
an obstacle detector detecting an obstacle based on the processed radar signal;
a distance calculation unit that calculates a separation distance from the vehicle to the detected obstacle;
Based on the processed radar signal, an obstacle map is generated for each of the first to Nth radars, and cross-correlation between the obstacle maps generated for each of the first to Nth radars is applied to obtain a boundary shape of the detected obstacle. a boundary shape mapping unit generating an included obstacle map; and
Boundary shape display unit outputting an obstacle map including the boundary shape
Including, parking anti-collision device. According to claim 1,
The received signal processing unit
a band pass filtering unit which removes signal components other than a frequency band corresponding to the transmitted ultra-wideband radar signal from the received radar signal;
a ground noise canceling unit configured to remove a signal component that is reflected from the received radar signal after the ultra-wideband radar signal transmitted hits the ground and is reflected; and
A clutter removal unit that removes unnecessary signal components other than the transmitted ultra-wideband radar signal using a moving average method
Including, parking anti-collision device. According to claim 1,
Wherein the obstacle detection unit determines that an obstacle exists within a predetermined radar recognition distance when the intensity of the processed radar signal exceeds a predetermined threshold value. According to claim 3,
Further comprising a warning alarm unit for outputting a warning alarm,
The distance calculation unit controls the warning alarm unit to output the warning alarm when the separation distance is within a predetermined warning distance threshold value,
Parking anti-collision device. According to claim 4,
The warning alarm is output through at least one of a speaker, a vibration element, and a lamp provided in the vehicle. According to claim 1,
The boundary shape mapping unit
a map generator for generating an obstacle map for each of the first to Nth radars;
a cross-correlation unit generating an obstacle map including the boundary shape by applying cross-correlation between obstacle maps generated for each of the first to N-th radars;
a filtering unit that applies 2D Gaussian filtering to the obstacle map including the boundary shape; and
A correction unit for applying an interpolation method to the obstacle map to which the 2D Gaussian filtering is applied
Including, parking anti-collision device. According to claim 6,
The cross-correlation unit cross-correlates obstacle maps corresponding to two physically adjacent radars among the first to N-th radars, and then adds the cross-correlated values to generate an obstacle map including the boundary shape. . According to claim 7,
The obstacle map including the boundary shape includes boundary information of the detected obstacle, the parking collision avoidance device. A method for preventing a collision during parking in a vehicle equipped with first to Nth radars transmitting ultra-wideband radar signals,
processing a radar signal received when the ultra-wideband radar signal is transmitted;
detecting an obstacle based on the processed radar signal;
calculating a separation distance from the vehicle to the detected obstacle;
Based on the processed radar signal, an obstacle map is generated for each of the first to Nth radars, and cross-correlation between the obstacle maps generated for each of the first to Nth radars is applied to obtain a boundary shape of the detected obstacle. generating an embedded obstacle map; and
Outputting an obstacle map including the boundary shape
Including, parking collision avoidance method. According to claim 9,
Processing the radar signal
removing signal components other than a frequency band corresponding to the transmitted ultra-wideband radar signal from the received radar signal;
removing an incoming signal component from the received radar signal after the ultra-wideband radar signal transmitted hits the ground and is reflected; and
Removing unnecessary signal components other than the transmitted ultra-wideband radar signal using a moving average method
Including, parking collision avoidance method. According to claim 9,
Parking collision avoidance method, wherein it is determined that an obstacle exists within a predetermined radar recognition distance when the intensity of the processed radar signal exceeds a predetermined threshold. According to claim 11,
Further comprising the step of outputting a warning alarm,
If the separation distance is within a predetermined warning distance threshold, the warning alarm is output. According to claim 12,
The warning alarm is output through at least one of a speaker, a vibration element, and a lamp provided in the vehicle. According to claim 9,
Creating the obstacle map
applying 2D Gaussian filtering to the obstacle map including the boundary shape; and
Applying an interpolation method to the filtered obstacle map
Including, parking collision avoidance method. According to claim 14,
After cross-correlating obstacle maps corresponding to two physically adjacent radars among the first to N-th radars, an obstacle map including the boundary shape is generated by adding the cross-correlated values. According to claim 15,
The obstacle map including the boundary shape includes boundary information of the detected obstacle. A program recorded on a recording medium realizing the method according to any one of claims 9 to 16. A recording medium on which the program according to claim 17 is recorded.

Images