KR101302798B1 - Method and system for protecting pedestrian using environment information - Google Patents

Abstract

One embodiment of the present invention relates to a method and system for protecting active pedestrians using environmental information, receiving environmental information through at least one sensor or a specific device to detect an external vehicle condition, and based on the detected external vehicle condition, Selecting or controlling a specific discrimination algorithm and a specific tracking algorithm, the pedestrian is determined by analyzing the input image according to the specific discrimination algorithm, and the risk of the pedestrian determined according to the specific tracking algorithm is determined.

According to an embodiment of the present invention, there is an effect of adaptively providing a pedestrian protection function such as pedestrian discrimination, pedestrian risk determination and protection processing according to the external environment.

Pedestrian protection, environmental information

Description

Active pedestrian protection method and system using environmental information {METHOD AND SYSTEM FOR PROTECTING PEDESTRIAN USING ENVIRONMENT INFORMATION}

One embodiment of the present invention relates to an active pedestrian protection method and system using environmental information. More particularly, the present invention relates to a technology for adaptively performing a parental protection function for an external environment.

The conventional active pedestrian protection system (APPS) uses sensors and cameras to identify pedestrians, determine pedestrian risk, and perform pedestrian protection accordingly.

In the conventional active pedestrian protection system, the recognition rate, which is a pedestrian discrimination capability, and a false detection rate that misjudges non-pedestrians as pedestrians may vary greatly according to an operating environment. For example, in rainy weather conditions, the driver's visibility cannot be easily secured because the driver's visibility is not easy, and the braking distance can be greatly increased when braking the vehicle to protect the pedestrians. It may not be easy for pedestrians. In addition, in a cryogenic situation where the temperature drops sharply, the pedestrian may wear thick clothes, earplugs, a hat, etc., and thus the pedestrian characteristics may be changed, and thus, the pedestrian discrimination ability (recognition rate) may be changed. If the pedestrian does not hear the horn of the vehicle due to the walking speed, the wearing of an earplug or a hat, or the ice road, the braking distance may be very long when the vehicle is braked to protect the pedestrian.

As described above, although the recognition rate or false detection rate of the active pedestrian protection system may vary greatly depending on the operating environment, the conventional active pedestrian protection system does not consider the operating environment and determines the pedestrian, the pedestrian risk determination and the pedestrian. The protection process is performed, and thus, pedestrian protection functions such as accurate pedestrian determination, pedestrian risk determination, and protection processing cannot be provided, and thus there is a problem in that pedestrian safety cannot be guaranteed.

Against this background, an object of an embodiment of the present invention is to obtain environmental information to detect an external condition of a vehicle and to provide a pedestrian protection function such as pedestrian discrimination, pedestrian risk determination, and protection processing according to the detected external condition of the vehicle to the external environment. To provide an adaptive fit.

One embodiment of the present invention, the environment sensor for receiving the environmental information through one or more sensors or a specific device to detect the external state of the vehicle; A selection unit for selecting or controlling a specific determination algorithm and a specific tracking algorithm based on the detected external state of the vehicle, from among a predetermined determination algorithm and a tracking algorithm for pedestrian determination; A pedestrian determination unit for determining a pedestrian by analyzing an input image according to the specific determination algorithm; And it provides an active pedestrian protection system using the environmental information including a pedestrian risk determination unit for determining the risk for the determined pedestrian according to the specific tracking algorithm.

In addition, an embodiment of the present invention, the environmental sensing step of detecting the external state of the vehicle by receiving environmental information through one or more sensors or a specific device; A selection step of selecting or controlling a specific determination algorithm and a specific tracking algorithm, based on the detected external state of the vehicle, from among a predetermined determination algorithm and a tracking algorithm for pedestrian determination; A pedestrian determination step of discriminating pedestrians by analyzing an input image according to the specific determination algorithm; A pedestrian risk determination step of determining a risk level for the determined pedestrian according to the specific tracking algorithm; And a pedestrian protection step of performing a pedestrian protection function by controlling a driver warning, a pedestrian warning, and an automatic braking according to the determined risk level.

As described above, according to an exemplary embodiment of the present invention, environmental information is obtained by detecting environmental information, and a pedestrian protection function such as pedestrian discrimination, pedestrian risk determination, and protection processing is performed according to the detected vehicle external condition. It is effective to provide adaptation to suit.

Hereinafter, some embodiments of the present invention will be described in detail with reference to exemplary drawings. In adding reference numerals to the components of each drawing, it should be noted that the same reference numerals are assigned to the same components as much as possible even though they are shown in different drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear.

In addition, in describing the component of this invention, terms, such as 1st, 2nd, A, B, (a), (b), can be used. These terms are intended to distinguish the constituent elements from other constituent elements, and the terms do not limit the nature, order or order of the constituent elements. When a component is described as being "connected", "coupled", or "connected" to another component, the component may be directly connected to or connected to the other component, It should be understood that an element may be "connected," "coupled," or "connected."

1 is a block diagram of an active pedestrian protection system (APPS) 100 using environmental information according to an embodiment of the present invention.

Referring to FIG. 1, the active pedestrian protection system 100 using environmental information according to an exemplary embodiment of the present disclosure may detect an environment in which an external state of a vehicle is detected based on environmental information acquired by the environmental information obtaining apparatus 110. Part 120, the selection unit 130 for selecting or controlling a specific determination algorithm and a specific tracking algorithm suitable for the external state of the vehicle, and the specific determination using the input image or information input through the input device 150, etc. Pedestrian determination unit 144 for determining the pedestrian according to the algorithm and a specific tracking algorithm, pedestrian risk determination unit 145 for determining the pedestrian risk for the determined pedestrian and the pedestrian protection function through the output device 160 Pedestrian protection unit 146 and the like.

The environment detecting unit 120 includes an environment through the environment information obtaining apparatus 110 including one or more sensors such as a raindrop sensor 111, a temperature sensor 112, or a specific device such as the GPS / MAP device 113. Receive information and detect the external condition of the vehicle.

The selector 130 selects or controls a specific determination algorithm and a specific tracking algorithm, based on the detected exterior state of the vehicle, from among a predetermined determination algorithm and tracking algorithm for the pedestrian determination. The above-mentioned classification algorithm and tracking algorithm, database and threshold value used in the algorithm execution in each algorithm mean misrecognition rate for pedestrian discrimination ability and error rate for non-pedestrians as pedestrians. These factors may affect the detection rate, and since these factors may vary greatly depending on the external state of the vehicle, the external state is sensed to select or control these elements according to the detected external state of the vehicle.

The pedestrian determining unit 144 analyzes the input image and determines pedestrians according to a specific determination algorithm selected or controlled based on an external state of the vehicle by the selecting unit 130.

The pedestrian risk determination unit 145 determines a risk level for the pedestrian determined according to a specific tracking algorithm selected or controlled based on an external state of the vehicle by the selector 130.

The above-described environment sensor 120 receives environment information including at least one of rainfall information, temperature information, and location information from the environment information acquisition device 110 including at least one sensor or a specific device. On the basis of the environmental information, it is possible to detect whether the external state of the vehicle is a rainy situation or a cryogenic situation.

The selection unit 130 classifies the detected external state of the vehicle according to the degree to which the pedestrian recognition is disturbed or the degree to which the braking distance is lengthened.

For example, when the external state of the vehicle is in a rainy situation, the driver's visibility may not be easily secured, and thus the pedestrian recognition ability of the driver may be further degraded. Also, the braking distance in rainy weather may be longer than the braking distance in rainy weather. Therefore, it is possible to classify the external state of the vehicle in a rainy state into a state in which the driver's recognition of the pedestrian is disturbed or the degree of braking distance is increased. For example, it may be classified as a phase among phases, middle and bottom.

In addition, when the exterior of the vehicle is in a cryogenic situation, the pedestrian may wear thick clothes, earplugs, or a hat, and thus the pedestrian outline may be changed, thereby greatly reducing the pedestrian discrimination ability (ie, recognition rate) based on the input image. Can be. In addition, pedestrians in cryogenic conditions tend to show a slow walking speed, pedestrians may not be able to hear the horn of the vehicle by wearing earplugs or hats, and the braking distance is very high due to ice in cryogenic conditions. Can be long. Therefore, the external condition of the vehicle, which is a cryogenic condition, may be classified into a state in which the driver's recognition of the pedestrian is disturbed or the degree of braking distance is increased. For example, it may be classified as a phase among phases, middle and bottom.

The selector 130 described above may select or control a specific determination algorithm and a specific tracking algorithm defined to correspond to the classified vehicle exterior states.

Each of the specific determination algorithm and the specific tracking algorithm selected or controlled by the selection unit 130 described above are classified and defined by performing reinforcement learning (eg, machine learning) to correspond to the detected external state of the vehicle. Reinforcement learning is performed to correspond to the external state of the vehicle, and the defined thresholds and database are used to execute the algorithm.

As described above, the selection unit 130 selects or controls a specific determination algorithm (including a suitable threshold and a database) and a specific tracking algorithm (including a suitable threshold and a database) suitable for the vehicle exterior state. The pedestrian determination unit 144 and the pedestrian risk determination unit 145 using algorithms also perform pedestrian determination and pedestrian risk determination in consideration of the vehicle external condition. Therefore, the pedestrian protection unit 146 can also perform a pedestrian protection function suitable for the external state of the vehicle.

The pedestrian protection unit 146 described above may drive a driver warning according to the degree of danger determined by the pedestrian risk determination unit 145 and the required braking deceleration amount according to the distance between the pedestrians determined by the pedestrian determination unit 144. Alarm section, pedestrian warning (Horning) section and automatic braking (ESC Braking) section, the sound command signal, the horn command signal and the braking command signal generated in the corresponding section, the vehicle included in the output device 160 A driver warning, a pedestrian warning, and an automatic braking function are performed through a HMI (Human Machine Interface) 161, a speaker 162, a horn device 163, and an electronic control unit (ESC).

In the following, the above-described guardian determination, guardian risk determination and guardian protection will be described by taking an example in rainy weather and cryogenic conditions.

First, when the environment detecting unit 120 detects that the vehicle external state is a rainy situation, the selector 130 may reduce the pedestrian discrimination ability (recognition rate) in the rainy situation, and thus, the schedule defined in the rainy situation Select or control a specific discrimination algorithm and a specific tracking algorithm with a high level of reliability.

The pedestrian discrimination unit 144 analyzes the input image according to a specific discrimination algorithm by using the database and the threshold value that have been reinforced by the pedestrian in rainy conditions to discriminate pedestrians.

The pedestrian risk determination unit 145 determines a risk level for the pedestrian determined according to a specific tracking algorithm by using a tracking model defined in a rainy situation and a determination threshold.

The pedestrian protection unit 146 may control the generation time, the number of occurrences, or the magnitude of the sound command signal, the horn command signal, and the braking command signal based on the rain. For example, in rainy conditions, drivers or pedestrians may be unable to hear sound or horns, or the braking distance may be very long. You can brake quickly.

Next, when the environment sensing unit 120 detects that the vehicle external state is a cryogenic condition below a certain temperature, the selector 130 may have a pedestrian discrimination ability (recognition rate) due to the pedestrian characteristic in the cryogenic situation. As such, a specific discrimination algorithm and a specific tracking algorithm having a high level of reliability that are defined in a cryogenic situation are selected or controlled.

The pedestrian discrimination unit 144 analyzes the input image according to a specific discrimination algorithm and discriminates pedestrians by using a database for which the pedestrians have been reinforced by the pedestrian in the cryogenic situation and the discrimination threshold.

The pedestrian risk determination unit 145 determines a risk level for the pedestrian determined according to a specific tracking algorithm by using the tracking model defined in the cryogenic situation and the determination threshold.

The pedestrian protection unit 146 may control the generation time, the number of occurrences, or the magnitude of the sound command signal, the horn command signal, and the braking command signal based on the cryogenic situation. For example, in a cryogenic situation, the pedestrian discrimination ability (recognition rate) may be reduced, the driver or pedestrian may not hear the sound or the horn, or the braking distance may be very long. You can brake quickly, generate multiple times, or brake faster.

On the other hand, the selector 130 described above, for the reaction according to the distance based on the pedestrian distance obtained through the range sensor (151), such as a laser scanner, and the detected external state of the vehicle. The maximum allowable time may be determined, and according to the determined maximum allowable time, a specific determination algorithm and a specific tracking algorithm may be selected or controlled.

On the other hand, the active pedestrian protection system 100 using the environmental information according to an embodiment of the present invention, as shown in Figure 1, through the range sensor 151, such as a laser scanner included in the input device 150 The input device according to the position of the pedestrian candidate group detecting unit 142 for detecting the pedestrian candidate group in the detected dangerous section through the danger section detecting unit 141 and the range sensor 151 detecting the danger section. An input image generating unit 143 or the like which cuts only an image of a region of interest (ROI) as an input image from the image acquired by the camera 152 included in the 150 and inputs it to the pedestrian discriminator 144. It may include.

The above-described dangerous section detecting unit 141, the pedestrian candidate group detecting unit 142, the input image generating unit 143, the pedestrian determining unit 144, the pedestrian risk determining unit 145, and the pedestrian protection unit 146 may include a controller ( 140 may be included and implemented.

2 is a view showing an operating range of the active pedestrian protection system 100 using the environmental information according to an embodiment of the present invention by way of example.

Referring to FIG. 2, an operating range of the active pedestrian protection system 100 using environmental information according to an exemplary embodiment of the present invention includes an input including a range sensor 151 such as a laser scanner and a camera 152. It may be determined by the performance of the device 150.

In the example of FIG. 2, the data-based pedestrian candidate group extraction and tracking distance from the range sensor 151, which is a laser scanner, is 80 m at maximum, and the image-based pedestrian discrimination distance from the camera 152 is about 30 m when using a 12 mm lens. Assumed

2, a vehicle 200 equipped with an active pedestrian protection system 100 using environmental information according to an exemplary embodiment of the present invention is parked in a right front within 30m of a vehicle 201, 202. When the pedestrian 220 walks or jumps out between, the active pedestrian protection system 100 using the environmental information according to an embodiment of the present invention provides a dangerous section between the spaces between the vehicles 201 and 202 parked on the right front side. Detects at 210, determines the pedestrian 220 in the detected danger section 210, determines the degree of danger, and performs a pedestrian protection function accordingly.

In addition, the operating speed of the active pedestrian protection system 100 using the environmental information according to an embodiment of the present invention, the active pedestrian protection system 100 using the environmental information according to an embodiment of the present invention It is determined by the speed. For example, the operating speed range can be 20k / h-50k / h.

3 is a diagram exemplarily illustrating a method for differentially providing pedestrian protection in an active pedestrian protection system 100 using environmental information according to an exemplary embodiment of the present invention.

3, the active pedestrian protection system 100 using the environmental information according to an embodiment of the present invention, the driver warning (Sound Alarm) according to the pedestrian risk degree, the amount of required braking deceleration according to the distance to the pedestrian ) Section, pedestrian warning (Horning) section and automatic braking (ESC Braking) section, the sound command signal, the horn command signal and the braking command signal generated in the corresponding section, the vehicle included in the output device 160 A driver warning, a pedestrian warning, and an automatic braking function are performed through a HMI (Human Machine Interface) 161, a speaker 162, a horn device 163, and an electronic control unit (ESC).

4 is a flowchart illustrating an active pedestrian protection method using environmental information according to an embodiment of the present invention.

Referring to FIG. 4, the active pedestrian protection method using environmental information according to an exemplary embodiment of the present invention receives an environmental information through an environmental information acquisition device 110 including one or more sensors or a specific device, and thus may provide a vehicle external state. A step of selecting or controlling a specific determination algorithm and a specific tracking algorithm based on the detected external state of the vehicle, from among the environmental detection step S400 for detecting a detection state and a predefined determination algorithm and a tracking algorithm for pedestrian determination. And a pedestrian determination step (S404) for determining a pedestrian by analyzing an input image according to a specific determination algorithm, a pedestrian risk determination step (S406) for determining a risk level for the pedestrian determined according to a specific tracking algorithm, and Pedestrian protection by controlling driver warnings, pedestrian warnings and automatic braking based on risk Protection step S408, and the like.

As described above, according to an exemplary embodiment of the present invention, environmental information is obtained by detecting environmental information, and a pedestrian protection function such as pedestrian discrimination, pedestrian risk determination, and protection processing is performed according to the detected vehicle external condition. It has the effect of adapting to suit.

While the present invention has been described in connection with what is presently considered to be the most practical and preferred embodiments, it is to be understood that the invention is not limited to the disclosed embodiments. In other words, within the scope of the present invention, all of the components may be selectively operated in combination with one or more. In addition, although all of the components may be implemented as one independent hardware, some or all of the components may be selectively combined to perform a part or all of the functions in one or a plurality of hardware. As shown in FIG. Codes and code segments constituting the computer program may be easily inferred by those skilled in the art. Such a computer program may be stored in a computer readable storage medium and read and executed by a computer, thereby implementing embodiments of the present invention. As the storage medium of the computer program, a magnetic recording medium, an optical recording medium, a carrier wave medium, or the like may be included.

It is also to be understood that the terms such as " comprises, "" comprising," or "having ", as used herein, mean that a component can be implanted unless specifically stated to the contrary. But should be construed as including other elements. All terms, including technical and scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined. Terms commonly used, such as predefined terms, are to be interpreted as being consistent with the contextual meaning of the related art, and are not to be construed as ideal or overly formal, unless expressly defined to the contrary.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are intended to illustrate rather than limit the scope of the present invention, and the scope of the technical idea of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

1 is a block diagram of an active pedestrian protection system using environmental information according to an embodiment of the present invention;

2 is a view showing an operating range of an active pedestrian protection system using environmental information according to an embodiment of the present invention by way of example,

3 is a diagram illustrating a method of differentially providing pedestrian protection in an active pedestrian protection system using environmental information according to an embodiment of the present invention;

4 is a flowchart illustrating an active pedestrian protection method using environmental information according to an embodiment of the present invention.

Description of the Related Art

100: Active Pedestrian Protection System (APPS)

111: rain sensor

112: temperature sensor

120: environmental sensing unit

130: selection

144: pedestrian discrimination unit

145: pedestrian risk determination unit

146: pedestrian protection

151: range sensor

152: camera

Claims (10)

An environment sensor configured to receive environmental information through at least one sensor or a specific device to detect an external state of the vehicle; A selection unit for selecting or controlling a specific determination algorithm and a specific tracking algorithm, based on the detected external state of the vehicle, from among predefined determination algorithms and tracking algorithms for pedestrian determination; A pedestrian determination unit for determining a pedestrian by analyzing an input image according to the specific determination algorithm; A pedestrian risk determination unit that determines a risk level for the determined pedestrian according to the specific tracking algorithm; And The pedestrian risk determination unit divides the driver warning section, the pedestrian warning section and the automatic braking section according to the determined risk degree and the required braking deceleration amount according to the determined distance from the pedestrian, and in this section, the sound command signal, Active pedestrian protection system using environmental information including a pedestrian protection unit for generating a horn command signal and a braking command signal. The method of claim 1, The environmental sensor, Active pedestrian protection system using environmental information, characterized in that for receiving the environmental information including at least one of rainfall information, temperature information and location information from the at least one sensor or the specific device. The method of claim 1, Wherein the selection unit comprises: Classify the detected external state of the vehicle according to the degree of obstruction of pedestrian recognition or the length of braking distance, And selecting or controlling the specific determination algorithm and the specific tracking algorithm defined to correspond to the classified vehicle exterior conditions. The method of claim 1, Each of the specific determination algorithm and the specific tracking algorithm, Reinforcement learning is performed to be classified and defined to correspond to the sensed external state of the vehicle, Active pedestrian protection system using environmental information, characterized in that the reinforcement learning is performed to correspond to the detected external state of the vehicle to use the defined threshold value and database for the algorithm execution. delete The method of claim 1, When the environmental sensor detects that the vehicle external condition is a rainy situation, The selecting unit selects or controls the specific determination algorithm and the specific tracking algorithm having a certain level or more of reliability defined in the rainy situation, The pedestrian discriminator determines a pedestrian by analyzing an input image according to the specific discrimination algorithm by using the database and the discrimination threshold that have been strengthened by the pedestrian in the rainy situation. The pedestrian risk determination unit determines a risk level for the pedestrian determined according to the specific tracking algorithm by using a tracking model defined in the rain and a determination threshold value. The pedestrian protection unit, based on the rainy weather conditions, active pedestrian protection system using environmental information, characterized in that for controlling the occurrence time, number of occurrences or magnitude of the sound command signal, the horn command signal and the braking command signal. . The method of claim 1, When the environmental sensing unit detects that the vehicle external condition is a cryogenic situation below a certain temperature, The selecting unit selects or controls the specific determination algorithm and the specific tracking algorithm having a certain level or more of reliability defined in the cryogenic situation, The pedestrian discriminator determines a pedestrian by analyzing an input image according to the specific discrimination algorithm by using a database and a discrimination threshold that are reinforced by the pedestrian in the cryogenic situation. The pedestrian risk determination unit determines a risk level for the pedestrian determined according to the specific tracking algorithm by using a tracking model defined in the cryogenic situation and a determination threshold value. The pedestrian protection unit, based on the cryogenic situation, the active pedestrian protection system using environmental information, characterized in that for controlling the occurrence time, number of occurrences or magnitude of the sound command signal, the horn command signal and the braking command signal. . The method of claim 1, Wherein the selection unit comprises: Based on the pedestrian distance obtained through the range sensor and the detected external state of the vehicle, determine the maximum allowable time for response according to the distance, Active pedestrian protection system using environmental information, characterized in that for selecting or controlling the specific determination algorithm and the specific tracking algorithm according to the determined maximum allowable time. The method of claim 1, Danger section detection unit for detecting a danger section through the range sensor; A pedestrian candidate group detection unit detecting the pedestrian candidate group in the danger section through the range sensor; And An input image generation unit which cuts only an image of a region of interest from the image acquired by the camera according to the detected position of the pedestrian candidate group as the input image and inputs it to the pedestrian determination unit Active pedestrian protection system using environmental information, characterized in that it further comprises. An environmental sensing step of sensing an external state of the vehicle by receiving environmental information through at least one sensor or a specific device; A selection step of selecting or controlling a specific determination algorithm and a specific tracking algorithm, based on the detected external state of the vehicle, from among a predetermined determination algorithm and a tracking algorithm for pedestrian determination; A pedestrian determination step of discriminating pedestrians by analyzing an input image according to the specific determination algorithm; A pedestrian risk determination step of determining a risk level for the determined pedestrian according to the specific tracking algorithm; And Including the pedestrian protection step of performing a pedestrian protection function by controlling the driver warning, pedestrian warning and automatic braking according to the determined risk, The pedestrian protection step may be divided into a driver warning section, a pedestrian warning section and an automatic braking section according to the determined risk level and the required braking deceleration amount according to the determined distance from the pedestrian. An active pedestrian protection method using environmental information, characterized in that generated by controlling the horn command signal and the braking command signal.

Images