JP2006163637A - Driving supporting device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent an accident with notification suitable to a degree of risk by judging the degree of risk of the occurrence of an accident with a walker without using a dedicated terminal. <P>SOLUTION: An image processing part 11 applies image processing to an image around a self-vehicle photographed by a self-vehicle camera 31 and images received by road-to-vehicle communication equipment 35 and inter-vehicle communication equipment 36, then an image recognizing part 12 recognizes the existence/absence of a walker. When the walker exists as a result of this recognition, a risk degree calculating part 13 uses outputs of a navigation system 20, a speed sensor 32, a radar 33 and a raindrop sensor 34 or the like to calculate the possibility (risk degree) that the self-vehicle comes into contact with the walker, and a notification means control part 14 controls a notifying means such as a loudspeaker 51 outside the vehicle, a horn 52, a headlight 533, a turning signal lamp 54 and a brake lamp 55 to carry out notification processing corresponding to the degree of risk. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a driving support device that collects information related to traveling of a vehicle and supports driving of the vehicle based on the collected information, and more particularly to a driving support device that ensures the safety of a pedestrian.

  Conventionally, prevention of vehicle accidents, particularly accidents between vehicles and pedestrians (including bicycles) has been a very important issue. Such an accident often occurs when the driver of the vehicle overlooks a pedestrian or when the pedestrian is unaware of the approach of the vehicle and does not recognize each other.

  Therefore, techniques for preventing accidents have been devised by supporting mutual recognition of the vehicle side and the pedestrian side. For example, in the traffic fault notification system disclosed in Patent Document 1, a passer such as a vehicle or a pedestrian is provided with a dedicated terminal, and the presence of each other is recognized by communication between the terminals.

JP 2002-245595 A

  However, the conventional technology described above is based on the premise that each passer has a terminal, and there is a problem in that a passerby without a terminal cannot be recognized.

  In addition, the risk of accident occurrence varies depending on the state of the vehicle, the state of the pedestrian, the surrounding situation, and the like. If a pedestrian is simply notified to a pedestrian, the pedestrian may be notified even though the pedestrian is safe, which is annoying for the pedestrian. Therefore, it has been an important issue to comprehensively determine the risk of accident occurrence and to notify the pedestrian of the presence of the vehicle.

  The present invention has been made to solve the above-described problems of the prior art and to solve the problems, and determines the risk of an accident with a pedestrian without using a dedicated terminal. It is an object of the present invention to provide a driving support device that prevents accidents through adaptive notification.

  In order to solve the above-described problems and achieve the object, the driving support apparatus according to the invention of claim 1 collects information related to the traveling of the vehicle, and performs driving that supports driving of the vehicle based on the collected information. The support device, the image recognition means for recognizing the pedestrian from the image around the vehicle acquired by the image acquisition means, and the walking based on the risk of contact between the pedestrian recognized by the image recognition means and the host vehicle. And / or notification control means for controlling notification processing for the surrounding vehicle.

  According to the first aspect of the present invention, the driving support apparatus acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of risk that the recognized pedestrian and the host vehicle are in contact with each other, and is calculated. Based on the degree of risk, the notification processing for pedestrians and surrounding vehicles is controlled.

  According to a second aspect of the present invention, there is provided the driving support apparatus according to the first aspect of the invention, wherein the notification control means uses any one of a horn, a vehicle lamp, a vehicle communication means, and a sound output means, or a warning sound. A notification process using a combination of a lighting device, an in-vehicle lamp, an in-vehicle communication unit, and an audio output unit is performed.

  According to the second aspect of the present invention, the driving support apparatus acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of risk that the recognized pedestrian and the host vehicle come into contact, and is calculated. Based on the degree of risk, a notification process for a pedestrian or a surrounding vehicle is performed by any one of or a combination of a horn, an in-vehicle lamp, an in-vehicle communication unit, and an audio output unit.

  According to a third aspect of the present invention, in the driving support device according to the second aspect of the present invention, when the notification control means executes the notification process using the alarm, the output sound volume and / or the output cycle is provided. Is changed based on the degree of risk.

  According to the invention of claim 3, the driving support device acquires an image around the vehicle, recognizes the pedestrian, recognizes a pedestrian, and calculates the degree of danger of contact between the recognized pedestrian and the own vehicle. The output volume and output cycle of the are changed according to the risk level.

  According to a fourth aspect of the present invention, in the driving support device according to the second or third aspect of the present invention, when the notification control means executes the notification process using the vehicle lamp, the lighting state and / or the lighting The period is changed based on the degree of risk.

  According to the fourth aspect of the invention, the driving support device acquires an image around the vehicle, recognizes the pedestrian, recognizes a pedestrian, and calculates the degree of risk that the recognized pedestrian contacts the own vehicle. Change the lighting state and lighting cycle according to the degree of danger.

  According to a fifth aspect of the present invention, in the driving support device according to the second, third, or fourth aspect of the invention, when the notification control means executes the notification process using the voice output means, The type and / or output volume is changed based on the risk level.

  According to the invention of claim 5, the driving support device acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of risk that the recognized pedestrian and the own vehicle come into contact with each other, and outputs the sound. Change the type of output sound and output volume according to the risk level.

  According to a sixth aspect of the present invention, in the driving support device according to any one of the first to fifth aspects, the degree of risk includes a road shape, a state of the pedestrian, and a relative relationship between the pedestrian and the host vehicle. It is calculated using distance and relative velocity.

  According to the sixth aspect of the present invention, the driving support device acquires an image around the vehicle and recognizes the pedestrian, recognizes the road shape, the state of the pedestrian, the relative distance between the pedestrian and the host vehicle, and the relative The degree of danger of contact between the pedestrian and the host vehicle is calculated using the speed, and the notification process for the pedestrian and surrounding vehicles is controlled based on the calculated degree of danger.

  According to a seventh aspect of the present invention, in the driving support device according to the sixth aspect of the present invention, the degree of risk further includes at least one of a state of surrounding vehicles, map information of the surrounding area, time information, and weather information. It is calculated using.

  According to the seventh aspect of the present invention, the driving support device acquires an image around the vehicle to recognize the pedestrian, and uses the state of the surrounding vehicle, the surrounding map information, the time information, the weather information, and the like. The degree of risk of contact between the vehicle and the host vehicle is calculated, and notification processing for pedestrians and surrounding vehicles is controlled based on the calculated risk.

  According to an eighth aspect of the present invention, there is provided a driving support apparatus according to the seventh aspect of the present invention, wherein at least one of the road shape, the pedestrian state, the relative distance and relative speed, the state of surrounding vehicles, and weather information. One of them is acquired by image processing.

  According to the eighth aspect of the present invention, the driving support device acquires an image around the vehicle and the presence or absence of the pedestrian, the state of the pedestrian, the road shape, the relative distance and the relative speed between the pedestrian and the host vehicle, the surrounding vehicle The degree of danger that the pedestrian and the host vehicle come into contact is calculated by recognizing the situation, weather, etc., and the notification process for the pedestrian and surrounding vehicles is controlled based on the calculated degree of danger.

  According to a ninth aspect of the present invention, in the driving support device according to the seventh or eighth aspect of the present invention, the risk degree calculation means is at least one of the road shape, the surrounding map information, and the time information. One is obtained from the navigation device.

  According to the ninth aspect of the present invention, the driving support device acquires an image around the vehicle and recognizes a pedestrian, and also acquires a road shape, surrounding map information, time information, and the like from the navigation device. The degree of danger of contact with the pedestrian is calculated, and the notification process for the pedestrian and surrounding vehicles is controlled based on the calculated degree of danger.

  According to a tenth aspect of the present invention, in the driving support apparatus according to any one of the seventh to ninth aspects, the state of the surrounding vehicle is acquired by inter-vehicle communication with the surrounding vehicle.

  According to the invention of claim 10, the driving support apparatus acquires an image around the vehicle to recognize the pedestrian, and acquires the state of the surrounding vehicle by inter-vehicle communication with the surrounding vehicle to acquire the state of the pedestrian and the host vehicle. The degree of danger of contact with the pedestrian is calculated, and the notification process for the pedestrian and surrounding vehicles is controlled based on the calculated degree of danger.

  Further, in the driving support apparatus according to the invention of claim 11, in any one of the inventions of claims 1 to 10, the image acquisition means acquires an image of the periphery of the vehicle from an in-vehicle imaging device mounted on the own vehicle. It is characterized by that.

  According to the eleventh aspect of the invention, the driving support device takes an image around the own vehicle, recognizes the pedestrian, and calculates the degree of risk of contact between the recognized pedestrian and the own vehicle. The notification process for pedestrians and surrounding vehicles is controlled based on the degree of danger.

  According to a twelfth aspect of the present invention, the driving support apparatus according to any one of the first to eleventh aspects is characterized in that the image acquisition means acquires an image around the vehicle by communication with the outside. To do.

  According to the twelfth aspect of the present invention, the driving support device acquires an image around the own vehicle by communicating with the outside, recognizes the image of the pedestrian, and determines the degree of danger that the recognized pedestrian and the own vehicle contact each other. The notification process for the pedestrian and surrounding vehicles is controlled based on the calculated degree of risk.

  According to the first aspect of the present invention, the driving support device acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of risk that the recognized pedestrian contacts the host vehicle, and calculates the calculated risk. Because it controls the notification process for pedestrians and surrounding vehicles based on the degree of driving, it is possible to determine the risk of accidents with pedestrians without using a dedicated terminal, and to support accidents that prevent accidents through notifications adapted to the risk level There exists an effect that an apparatus can be obtained.

  Further, according to the invention of claim 2, the driving support device acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of danger that the recognized pedestrian and the own vehicle come into contact, and is calculated. Based on the degree of danger, the alarm processing for the pedestrian and the surrounding vehicle is performed by any one of the alarming device, the in-vehicle lamp, the in-vehicle communication unit, the audio output unit, or a combination thereof, so that the pedestrian can be used without using a dedicated terminal. It is possible to obtain a driving support device that determines the risk of occurrence of an accident and executes a notification adapted to the risk using existing in-vehicle devices.

  According to the invention of claim 3, the driving support device acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of danger of contact between the recognized pedestrian and the host vehicle, and generates a warning sound. Because the output volume and output cycle of the device are changed according to the degree of danger, the risk of accident occurrence with pedestrians is judged, and the driving support that performs notifications adapted to the degree of danger by changing the output of the alarm There exists an effect that an apparatus can be obtained.

  According to a fourth aspect of the present invention, the driving support device acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of danger of contact between the recognized pedestrian and the vehicle, Since the lighting state and lighting cycle of the vehicle are changed according to the risk level, a driving support device that determines the risk level of accidents with pedestrians and performs notifications adapted to the risk level by changing the output of the on-vehicle lamp There is an effect that it can be obtained.

  Further, according to the invention of claim 5, the driving support device acquires an image around the vehicle, recognizes the image of the pedestrian, calculates the degree of danger that the recognized pedestrian and the own vehicle come into contact, and outputs the sound. Since the type of output sound and the output volume by the means are changed according to the degree of danger, the degree of danger of an accident with a pedestrian is determined, and the notification adapted to the degree of danger is executed by changing the output of the sound output means There exists an effect that a driving assistance device can be obtained.

  According to the invention of claim 6, the driving support device acquires an image around the vehicle and recognizes the pedestrian, recognizes the road shape, the state of the pedestrian, the relative distance between the pedestrian and the host vehicle, and The degree of risk of contact between the pedestrian and the vehicle is calculated using the relative speed, and the alarm processing for the pedestrian and surrounding vehicles is controlled based on the calculated risk, so the risk of accidents with pedestrians It is possible to obtain a driving support device that comprehensively determines the degree and prevents accidents by notification adapted to the degree of danger.

  According to the seventh aspect of the present invention, the driving support device acquires an image around the vehicle and recognizes the pedestrian, and walks using the state of the surrounding vehicle, map information around the vehicle, time information, weather information, and the like. The degree of risk of contact between the person and the vehicle is calculated, and the notification process for pedestrians and surrounding vehicles is controlled based on the calculated degree of risk, so the risk of accidents with pedestrians can be accurately determined. There is an effect that it is possible to obtain a driving support device that prevents an accident by a notification adapted to the degree of danger.

  According to the invention of claim 8, the driving support device acquires an image of the surroundings of the vehicle, the presence or absence of a pedestrian, the state of the pedestrian, the road shape, the relative distance and relative speed between the pedestrian and the host vehicle, the surroundings Recognize the situation of the vehicle, weather, etc., calculate the degree of danger that the pedestrian and the host vehicle contact, and control the notification process for the pedestrian and surrounding vehicles based on the calculated degree of risk. There is an effect that it is possible to obtain a driving support device that acquires various kinds of information, determines the risk of occurrence of an accident, and prevents the accident by notification adapted to the risk.

  According to the ninth aspect of the present invention, the driving support device acquires an image around the vehicle to recognize the pedestrian, and acquires the road shape, the surrounding map information, the time information, and the like from the navigation device to acquire the own vehicle. The degree of danger of contact with the pedestrian and surrounding vehicles is controlled based on the calculated degree of danger, so the risk of accident occurrence is determined in cooperation with the navigation device. It is possible to obtain a driving assistance device that prevents accidents by notification adapted to the above.

  According to the invention of claim 10, the driving support device acquires an image around the vehicle to recognize the pedestrian, recognizes a pedestrian, acquires the state of the surrounding vehicle by inter-vehicle communication with the surrounding vehicle, and self-identifies with the pedestrian. The degree of risk of contact with the vehicle is calculated, and the notification process for pedestrians and surrounding vehicles is controlled based on the calculated risk level. There is an effect that it is possible to obtain a driving assistance device that prevents accidents by notification adapted to the degree.

  According to the invention of claim 11, the driving support device captures an image around the own vehicle, recognizes the pedestrian, calculates the degree of danger that the recognized pedestrian and the own vehicle come into contact, and calculates Since the notification process for pedestrians and surrounding vehicles is controlled based on the determined risk level, the risk level of the accident occurrence is determined by the own vehicle alone, and a driving support device that prevents the accident by the notification adapted to the risk level is obtained. There is an effect that can be.

  According to the twelfth aspect of the present invention, the driving support device acquires an image around the own vehicle by communicating with the outside to recognize the pedestrian, and the degree of danger that the recognized pedestrian and the own vehicle come into contact with each other. And controls the notification process for pedestrians and surrounding vehicles based on the calculated risk level. There is an effect that it is possible to obtain a driving support device that prevents an accident by the adapted notification.

  Exemplary embodiments of a driving support apparatus according to the present invention will be described below in detail with reference to the accompanying drawings.

  FIG. 1 is an explanatory diagram illustrating a schematic configuration of a driving support apparatus according to an embodiment of the present invention. As shown in the figure, the driving support device 10 includes a navigation device 20, a host vehicle camera 31, a speed sensor 32, a radar 33, a raindrop sensor 34, a road-to-vehicle communication device 35, an inter-vehicle communication device 36, an in-vehicle notification system 40, and an outside of the vehicle. Connect to the notification system 50 or the like.

  The navigation device 20 is a device that sets and guides the travel route of the host vehicle. Specifically, the navigation device 20 acquires the current position of the host vehicle by GPS (Global Positioning System), specifies the road on which the host vehicle is traveling using the map data 21, and uses the in-vehicle notification system 40. To perform route guidance. In addition, the navigation device 20 transmits the position information of the host vehicle, the road shape, the surrounding map information, and the time information to the driving support device 10.

  The own vehicle camera 31 is a vehicle-mounted photographing means for photographing the periphery of the own vehicle, and the speed sensor 32 is a sensor that measures the traveling speed of the own vehicle. The radar 33 is a sensor that detects an object around the host vehicle by radio waves or the like, and the raindrop sensor 34 is a sensor that detects a rainfall state. Further, the road-to-vehicle communication device 35 is a communication unit that communicates with a roadside communication terminal installed on the road, and the vehicle-to-vehicle communication device 36 is a communication unit that performs communication with other vehicles.

  The in-vehicle notification system 40 includes a display 41, an in-vehicle speaker 42, and the like, and notifies the passenger and provides information. The in-vehicle notification system 40 is shared by a plurality of in-vehicle devices such as the driving support device 10, the navigation device 20, and an audio device (not shown).

  The outside notification system 50 includes an outside speaker 51, a horn 52, a headlight 53, a blinker lamp 54, a brake lamp 55, and the like, and performs notification and information provision to the vicinity of the host vehicle. Specifically, the vehicle exterior speaker 51 performs notification to the periphery of the vehicle by outputting sound to the outside of the vehicle. The horn 52 is a horn that notifies the presence of the host vehicle around the vehicle.

  The headlight 53, the blinker lamp 54, and the brake lamp 55 are on-vehicle lamps that respectively perform illumination in front of the host vehicle, notification of turning left and right, notification of deceleration, and the like. Moreover, the driving assistance apparatus 10 uses these vehicle-mounted lamps for danger notification with respect to the vehicle periphery. Furthermore, the driving assistance device 10 uses inter-vehicle communication by the inter-vehicle communication device 36 for notification to other nearby vehicles.

  The driving support device 10 includes an image processing unit 11, an image recognition unit 12, a risk level calculation unit 13, and a notification unit control unit 14 therein. The image processing unit 11 performs image processing on an image around the host vehicle and outputs the image to the image recognition unit 12. In addition, the imaging | photography result of the own vehicle camera 31 may be used for the image around the own vehicle, and the surrounding image which the road-vehicle communication apparatus 35 received from the roadside communication terminal may be used. In addition, an image captured by another vehicle can be received and used by the inter-vehicle communication device 36.

  The image recognition part 12 determines the presence or absence of a pedestrian by the image recognition with respect to this image. If the presence of a pedestrian is detected, the risk level calculation unit 13 calculates the possibility (risk level) that the pedestrian and the vehicle are in contact with each other. The notification means selection unit 14 controls the outside notification system 50 and the inter-vehicle communication device 36 based on the degree of risk, and executes notification control for pedestrians and surrounding vehicles.

  Here, the risk level calculation unit 13 calculates the risk level using various types of information indicating the state of the host vehicle and the surrounding conditions. FIG. 2 is an explanatory diagram for explaining information used by the risk calculation unit 13 for calculating the risk and a method for acquiring the information.

  In the figure, “Pedestrian presence / absence”, “Pedestrian position”, “Pedestrian state”, “Pedestrian characteristics”, “Vehicle position”, “Vehicle speed”, “Vehicle speed” “Presence / absence”, “state of surrounding vehicle”, “road shape”, “map information around”, “time”, “weather”, etc. are used as information used for risk calculation.

  Here, “presence / absence of pedestrian” can be acquired by image recognition as described above, and can also be detected by the radar 33. “Position of pedestrian” is information indicating the relative positional relationship between the vehicle and the pedestrian, and can be acquired by image recognition by the image recognition unit 12 or the radar 33.

  The “pedestrian state” is information indicating a state of a pedestrian's movement and direction, riding a bicycle, using a mobile phone, or holding an umbrella, and is recognized by the image recognition unit 12. Can be obtained by: Further, the moving direction of the pedestrian can be acquired by the radar 33.

  The “pedestrian characteristic” is information indicating a rough age of the pedestrian, for example, a child or an elderly person, and can be acquired by image recognition by the image recognition unit 12.

  Information indicating “the position of the host vehicle” can be acquired from the navigation device 20, and the “speed of the host vehicle” can be acquired from the speed sensor 33.

  “Presence / absence of surrounding vehicle” is information indicating whether or not another vehicle, for example, an oncoming vehicle or a following vehicle exists in the vicinity of the host vehicle. In addition to image recognition by the image recognition unit 12 and detection by the radar 33, the vehicle It can be obtained by inter-vehicle communication by the inter-communication device 36.

  “Neighboring vehicle state” is information indicating the moving direction, moving speed, planned travel route, and the like of other vehicles, and includes image recognition by the image recognition unit 12, inter-vehicle communication by the inter-vehicle communication device 36, and detection by the radar 33. Can be obtained by:

  Information indicating the “road shape” can be obtained by acquiring the map data 21 from the navigation device 20 or image recognition by the image recognition unit 12. Further, the position of the white line on the road, the road surface state, and the like can also be acquired by image recognition.

  “Nearby map information” is information indicating what kind of building (house, store, school, etc.) is in the vicinity of the road on which the vehicle is traveling, and acquires map data 21 from the navigation device 20. Can be obtained. Further, information indicating “time” can be obtained by acquiring time information used by the navigation device 20 in GPS.

  Information relating to “weather” such as the presence or absence of rain, the presence or absence of fog, and the like can be acquired by the raindrop sensor 34. Further, the presence or absence of rain or fog may be determined from the operation state of the wiper or the lighting state of the fog lamp.

  The risk level calculation unit 13 calculates the possibility (risk level) that the pedestrian and the host vehicle come into contact using these pieces of information, and the notification unit control unit 14 executes notification control corresponding to the risk level. FIG. 3 is a diagram illustrating a specific example of a correspondence relationship between the risk level and the notification control.

  In FIG. 3, the risk level calculation unit 13 calculates the risk level in six stages from “risk level 0” having the lowest risk to “risk level 5” having the highest risk. “Danger degree 0” corresponds to a state in which there is a pedestrian but it is safe, so the notification control is not executed.

  And in "risk degree 1", the alerting | reporting means control part 14 performs alerting | reporting with the low sound volume by the exterior speaker 51, and alerting | reporting to the other vehicle by the inter-vehicle communication part 36. Further, in the “risk level 2”, notification at a large volume by the outside speaker 51, notification by blinking the blinker lamp 54 in a long cycle, and notification to other vehicles by the inter-vehicle communication unit 36 are executed.

  Further, in the “risk level 3”, a notification by blowing the horn 52 at a long cycle, a notification by flashing the headlight 53 at a long cycle, a notification by flashing the blinker lamp 54 at a medium cycle, and a brake lamp 55 By flashing in a long cycle, and notification to other vehicles by the inter-vehicle communication unit 36 is executed.

  Similarly, in the “risk level 4”, a notification by blowing the horn 52 in a short cycle, a notification by flashing the headlight 53 in a short cycle, a notification by flashing the blinker lamp 54 in a short cycle, a brake lamp Notification by blinking 55 in a short cycle and notification to other vehicles by the inter-vehicle communication unit 36 are executed.

  Further, in the “risk level 5”, the continuous sounding of the horn 52, the continuous lighting of the headlight 53, the continuous lighting of the blinker lamp, the continuous lighting of the brake lamp, and notification to other vehicles by the inter-vehicle communication unit 36 are executed. .

  As described above, the notification means to be used corresponding to the risk level and the combination thereof are changed, and the control content of each notification means is changed corresponding to the risk level, thereby realizing notification suitable for the situation. Can do.

  Here, for the pedestrian, notification using the outside speaker 51, the horn 52, and the headlight 53 is particularly effective. In addition, notification using the horn 52, the headlight 53, the blinker lamp 54, the brake lamp 55, and the inter-vehicle communication device 36 is effective for surrounding vehicles, and the headlight 53 is particularly effective for oncoming vehicles. The notification used is effective for the following vehicle using the brake lamp.

  The correspondence relationship between the risk level and the notification control shown in FIG. 3 is merely an example, and the risk level can be set to an arbitrary number of stages. Further, the content of the corresponding notification control can be changed as appropriate. For example, it may be configured to output different types of sounds corresponding to the degree of danger in response to the notification using the speaker 51 outside the vehicle, or the output volume from the horn 52 may be changed corresponding to the degree of danger. You may comprise.

  Next, the processing operation of the driving support device 10 will be described. FIG. 4 is a flowchart for explaining the processing operation of the driving support device 10. The processing flow shown in FIG. 4 is repeatedly executed when the driving support device 10 is turned on.

  First, the driving assistance apparatus 10 acquires an image around the own vehicle from the own vehicle camera 31, the road-vehicle communication device 35, or the inter-vehicle communication device 36 (step S101). Then, after performing image processing by the image processing unit 11 on the acquired image, the image recognition unit 12 recognizes a pedestrian (step S102).

  As a result, if the presence of a pedestrian is recognized (step S103, Yes), the risk calculation unit 13 collects information around the vehicle to calculate the risk (step S104), and the notification unit control unit 14 detects the danger. Notification processing corresponding to the degree is executed (step S105).

  If the pedestrian is not aware of the danger even after performing the notification process (step S106, No), an image around the host vehicle is acquired again (step S101).

  On the other hand, when the pedestrian notices danger (step S106, Yes) or when it is determined that there is no pedestrian (step S103, No), the driving support device 10 ends the process.

  Note that whether or not the pedestrian has noticed can be determined by recognizing the pedestrian's body direction, face direction, and line-of-sight direction.

  Next, a specific example of risk calculation by the risk calculation unit 13 will be described. First, in FIG. 5, the host vehicle C10 is traveling on the road R10, and the pedestrian H10 is moving in the direction of the host vehicle from the front.

  In such a state, the risk level calculation unit 13 calculates the risk level from the relative distance and the relative speed between the host vehicle C10 and the pedestrian H10. That is, as the relative distance is smaller and the relative speed is larger, that is, when the pedestrian and the host vehicle are approaching each other, the risk value is set larger.

  Here, the value of the risk varies depending on the shape of the road. For example, as shown in FIG. 6A, when the road is divided into a roadway R20 and a sidewalk R21 and the pedestrian H20 is on the sidewalk, the relative distance and relative speed between the host vehicle C20 and the pedestrian H20 are shown in FIG. Even if it is the same as the situation shown, the degree of risk is made smaller than the situation shown in FIG.

  Next, a case where the road shape is the same as that shown in FIG. Fig. 6-2 shows the situation where pedestrians are going out on the road from the sidewalk, approaching each other due to the relative speed and relative distance between the pedestrian and the vehicle, and there is a high possibility of a collision in the future. Is detected.

  In FIGS. 5 and 6-1, there is a high possibility of passing each other while approaching each other, but in FIG. 6-2, it is not passing, and the pedestrian is moving laterally with respect to the traveling direction of the own vehicle. The possibility of a collision is extremely high. In such a case, the degree of danger is increased as compared to the cases of FIGS. 5 and 6-1, and the warning level for the pedestrian is increased.

  However, when the pedestrian's face or line of sight is facing the vehicle, a pedestrian crossing is detected on the road where the pedestrian crosses, and the traffic light for the traveling direction of the vehicle is detected red If it can be discriminated), the risk decreases. On the other hand, when the pedestrian is an elderly person or a child, or when the surrounding environment of the vehicle is bad such as nighttime or fog, the degree of risk becomes even larger than that in FIG.

  In this embodiment, when determining the degree of risk, the relative speed with the own vehicle is taken into account, but this is not limiting, and a pedestrian is recognized within a predetermined monitoring area set around the own vehicle, In addition, when the fact that the pedestrian is about to cross the road on which the vehicle is traveling is determined by the recognized pedestrian state, the risk is increased and a warning is given to the pedestrian. May be.

  The risk value also changes when the road is curved. For example, when the road R30 is curved as shown in FIG. 7, even if the relative distance and relative speed between the host vehicle C30 and the pedestrian H30 are the same as those shown in FIG. The degree becomes smaller. Since this is a curve, especially in image processing, it appears that a pedestrian approaches the destination of the host vehicle, and it is determined that there is a high possibility of a collision. When road information such as white line recognition and navigation is taken in and it is detected that a pedestrian is walking on a sidewalk in a curve, it is judged that the pedestrians are simply passing each other and the risk is reduced.

  By the way, the relative distance between the host vehicle C30 and the pedestrian H31 is larger than the relative distance with respect to the pedestrian H30, and the relative speed between the host vehicle C30 and the pedestrian H31 is smaller than the relative speed with respect to the pedestrian H30. However, since the host vehicle C30 travels along the road 30, the risk level for the pedestrian H31 is greater than the risk level for the pedestrian H30.

  Next, the danger level when there is an oncoming vehicle will be described with reference to FIG. In FIG. 8, the host vehicle C40 is traveling on the road R40, the pedestrian H40 is moving in the direction of the host vehicle from the front, and the oncoming vehicle C41 is traveling.

  For this reason, it is difficult for the host vehicle C40 to avoid the pedestrian H40, and the danger level of the pedestrian H40 is greater than when there is no oncoming vehicle. The degree of risk also changes depending on the position and speed of the oncoming vehicle.

  Therefore, in such a situation, the driving assistance device 10 ensures the safety of the pedestrian H40 by notifying the oncoming vehicle C41 in addition to notifying the pedestrian H40. As already described, it is preferable to use the horn 52, the headlight 53, the blinker lamp 54, the inter-vehicle communication device 36, etc. for the notification to the oncoming vehicle C41.

  Next, variations of image acquisition will be described. 5-8, the own vehicle can image a pedestrian directly by the own vehicle camera 31, and can recognize the presence. However, when the pedestrian is behind a building or other vehicle, the vehicle camera 31 cannot capture the pedestrian.

  For example, in the situation shown in FIG. 9, the preceding vehicle C51 is traveling in front of the host vehicle C50, and the camera mounted on the host vehicle C50 cannot capture the pedestrian H50. Thus, the host vehicle C50 recognizes the presence of the pedestrian H50 by acquiring an image captured by the preceding vehicle C51 through inter-vehicle communication and performing image recognition on the image.

  Here, in order to obtain information such as the position of the pedestrian from the image captured by the preceding vehicle C51, the position information of the preceding vehicle C51 is required. The position information of the preceding vehicle C51 may be acquired from the preceding vehicle C51 by inter-vehicle communication, or may be calculated by capturing the preceding vehicle C51 with a camera mounted on the host vehicle C50 and performing image recognition.

  In addition, in a state where the own vehicle can photograph a pedestrian and a pedestrian cannot be photographed from the succeeding vehicle, the image captured by the own vehicle is transmitted to the succeeding vehicle, thereby assisting the pedestrian recognition in the succeeding vehicle.

  In addition, although FIG. 9 demonstrated the case where the image image | photographed by the other vehicle was used, the terminal provided with the camera and the communication means was installed on the road, and the image image | photographed by this roadside terminal can also be received and utilized. .

  Next, a specific example in the case where peripheral map information is used for calculation of the risk will be described with reference to FIG. In FIG. 10, the host vehicle C60 is traveling on a road R60, and a building B60 exists around the road R60. And pedestrian group H60 exists in front of building B60.

  In this situation, for example, when the building B60 is a tavern or the like, the pedestrian group H60 may be drunk, so the risk level is calculated to a high value. Similarly, when the building B60 is a school or the like, since the child may jump out to the road side, the risk is calculated to be high.

  In this way, by estimating the state of a pedestrian according to the types of buildings around the road and calculating the risk value, it is possible to calculate the risk that is more suitable for the situation.

  Although common to the examples shown in FIGS. 5 to 10, the risk is corrected to a high value when the weather is rainy. This is because the rain drops the visibility and makes it difficult to hear the vehicle running sound, so the vehicle and the pedestrian are delayed in recognizing each other. This is to cope with the improvement of sex.

  Similarly, when it is determined that it is nighttime from time information or the like, the degree of risk is corrected to a high value. This is to cope with a decrease in the visibility and a delay in recognition of each other between the vehicle and the pedestrian, increasing the possibility of an accident.

  Furthermore, the risk level is increased even when the pedestrian is using a mobile phone, wearing an umbrella, a child, an elderly person, and the like.

  As described above, the driving support device according to the present embodiment recognizes the presence of a pedestrian from an image around the vehicle, and based on the state and characteristics of the pedestrian, the state of the own vehicle, and the surrounding situation, By calculating the possibility (risk level) of a pedestrian to contact and executing notification control according to the risk level, accidents can be prevented using existing in-vehicle devices without using a dedicated terminal. .

  In addition, by acquiring images from other vehicles and roadside communication terminals, it is possible to protect pedestrians at positions that cannot be captured from the host vehicle, and to transmit images captured by the host vehicle to nearby vehicles. By doing so, pedestrian protection in surrounding vehicles can be supported. In this example, notification is given to pedestrians and other vehicles. However, the present invention is not limited to this, and may be applied to notification to the driver of the host vehicle.

  As described above, the driving support device according to the present invention is suitable for preventing vehicle accidents, particularly for preventing contact accidents with pedestrians.

1 is a schematic configuration diagram illustrating a schematic configuration of a driving support apparatus according to an embodiment of the present invention. It is explanatory drawing explaining the information used for calculation of a risk, and its acquisition method by the risk calculation part shown in FIG. It is a figure which shows the specific example of the correspondence of a risk degree and alerting | reporting control. It is a flowchart explaining the processing operation of the driving assistance apparatus shown in FIG. It is explanatory drawing explaining the specific example of the risk calculation by the risk calculation part shown in FIG. It is explanatory drawing explaining the risk calculation when there exists a sidewalk (the 1). It is explanatory drawing explaining the risk degree calculation when there exists a sidewalk (the 2). It is explanatory drawing explaining the risk level calculation in a curve. It is explanatory drawing explaining risk degree calculation in case there exists an oncoming vehicle. It is explanatory drawing explaining the acquisition of the image using communication between vehicles. It is explanatory drawing explaining the risk degree calculation using map information.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Driving assistance apparatus 11 Image processing part 12 Image recognition part 13 Risk degree calculation part 14 Notification means control part 20 Navigation apparatus 21 Map data 31 Own vehicle camera 32 Speed sensor 33 Radar 34 Raindrop sensor 35 Road-to-vehicle communication apparatus 36 Inter-vehicle communication apparatus 40 In-car notification system 41 Display 42 In-car speaker 50 Out-of-car notification system 51 Out-of-car speaker 52 Horn 53 Headlight 54 Blinker lamp 55 Brake lamp

Claims (12)

A driving support device that collects information related to traveling of a vehicle and supports driving of the vehicle based on the collected information,
Image recognition means for recognizing a pedestrian from an image around the vehicle acquired by the image acquisition means;
Notification control means for controlling notification processing for the pedestrian and / or surrounding vehicle based on the risk of contact between the pedestrian recognized by the image recognition means and the host vehicle;
A driving support apparatus comprising:   The notification control unit executes a notification process using any one of a warning device, a vehicle lamp, a vehicle communication unit, and a voice output unit, or a combination of a warning device, a vehicle lamp, a vehicle communication unit, and a voice output unit. The driving support device according to claim 1, wherein:   The driving according to claim 2, wherein the notification control means changes an output volume and / or an output cycle based on the degree of risk when performing the notification process using the horn. Support device.   The said notification control means changes a lighting state and / or a lighting period based on the said risk, when performing a notification process using the said vehicle-mounted lamp, The Claim 2 or 3 characterized by the above-mentioned. Driving assistance device.   The said notification control means changes the kind and / or output volume of an output sound based on the said risk, when performing a notification process using the said audio | voice output means. Or the driving assistance device of 4.   The risk level is calculated by using a road shape, a state of the pedestrian, a relative distance and a relative speed between the pedestrian and the own vehicle. Driving assistance device.   The driving assistance device according to claim 6, wherein the risk level is calculated by further using at least one of a state of surrounding vehicles, surrounding map information, time information, and weather information.   The driving according to claim 7, wherein at least one of the road shape, the pedestrian state, the relative distance and speed, the state of surrounding vehicles, and weather information is acquired by image processing. Support device.   The driving support device according to claim 7 or 8, wherein at least one of the road shape, the surrounding map information, and the time information is acquired from a navigation device.   The driving support device according to any one of claims 7 to 9, wherein the state of the surrounding vehicle is acquired by inter-vehicle communication with the surrounding vehicle.   The driving support device according to any one of claims 1 to 10, wherein the image acquisition unit acquires an image around the vehicle from an in-vehicle imaging device mounted on the host vehicle.   The driving support apparatus according to any one of claims 1 to 11, wherein the image acquisition unit acquires an image around the vehicle by communication with the outside.

Images