Notification device and notification method in autonomous vehicle

Abstract

The present application provides a kind of notification device and notification method in automatic driving vehicle.Can accurately determine its driving condition in vehicle side, and can be executed to the person in front of the notification suitable for this condition.The notification device in one mode of the present application is equipped in the vehicle in automatic driving, to the person in front of the person in front of the existence of the above vehicle, comprising: shooting mechanism, the person in front of the above is photographed;Inferential mechanism, at least the characteristics of the person in front of the above photographed by the shooting mechanism are inferred;Determination mechanism, based on the inference result of the inferential mechanism, determines the notification mode for the person in front of the above;And notification control mechanism, based on the notification mode determined by the determination mechanism, makes the notification mechanism that the existence of the above vehicle is notified to the person in front of the above work.

Description

TECHNICAL FIELD

[0001] The present application relates to a notification technology suitable for a vehicle, for example, and more particularly to a notification device and a notification method that are mounted on a vehicle in autonomous driving and notify the presence of the vehicle to the outside. BACKGROUND

[0002] In modern society, as a mobile mechanism, a car is indispensable, and various vehicles travel on a road in daily life. On a road where vehicles pass, there are cases where a pedestrian, a cyclist, or the like who is a forwarder exists ahead of the traveling direction of the vehicle, and it is necessary to alert the forwarder to the presence of the vehicle.

[0003] In particular, in recent years, electric vehicles and hybrid vehicles that mount an electric motor have increased, and it can be said that the necessity of such alerting is increasing.

[0004] In this regard, for example, a warning device that can notify a pedestrian who is hidden in an obstruction and cannot be seen of the approach of the vehicle without causing a sound that becomes noise to spread to the surroundings is disclosed in Patent Literature 1.

[0005] PRIOR ART DOCUMENTS

[0006] PATENT LITERATURE

[0007] Patent Literature 1: Japanese Patent Application Laid-Open No. 2017-43197 SUMMARY

[0008] PROBLEMS TO BE SOLVED BY THE INVENTION

[0009] However, not limited to each of the above patent literatures, it cannot be said that the market demand is appropriately satisfied with the current technology, and the following problems exist.

[0010] That is, in the above-described Patent Literature 1, the fact that the frequency of the warning sound is changed according to the vehicle speed further improves the degree of alerting, but as described below, it is necessary to say that the notification method is only sound is lacking in flexibility.

[0011] For example, the situation on the road where vehicles pass sometimes changes, and therefore, depending on the environment, the appropriate notification method is not uniform or different.

[0012] In particular, in autonomous driving that can be expected in the future, since the opportunity to actively operate the vehicle according to the driver's intention decreases, it is required that the vehicle side accurately determines its situation and performs a notification method appropriate for the situation.

[0013] The present application has completed the above problems in view of one example, and aims to provide a notification device and a notification method that accurately determine the traveling situation of the vehicle on the vehicle side and can perform a notification appropriate for the situation to a forwarder.

[0014] Technical Solution

[0015] To solve the above problems, a notification device (1) in one embodiment of the present application is mounted on a vehicle traveling autonomously, notifies the presence of the vehicle to a preceding person who is preceding, and includes: a photographing mechanism that photographs the preceding person; an inference mechanism that infers at least a characteristic of the preceding person photographed by the photographing mechanism; a determination mechanism that determines a notification method for the preceding person based on an inference result of the inference mechanism; and a notification control mechanism that causes a notification mechanism that notifies the preceding person of the presence of the vehicle to operate based on the notification method determined by the determination mechanism.

[0016] Note that in the notification device described in (1) above, (2) it is preferable that the inference mechanism detect the height or the movement of the preceding person based on the photographing information of the photographing mechanism, infer whether the preceding person is any one of the elderly, adults other than the elderly, children, and the disabled with high likelihood, and the notification control mechanism change the notification method of the notification mechanism according to the category of the preceding person.

[0017] In addition, in the notification device described in (2) above, (3) it is preferable that the notification control mechanism cause a sound emitting mechanism to emit a sound as the notification mechanism in the case where the preceding person is any one of the elderly, the adults, and the children, and change the frequency and / or the amplitude of the sound based on the characteristic of the preceding person.

[0018] In addition, in the notification device described in any one of (1) to (3) above, (4) it is preferable that the inference mechanism infer the characteristic of an equipment held by the preceding person based on the photographing information of the photographing mechanism, and the notification control mechanism change the notification method of the notification mechanism according to the equipment of the preceding person.

[0019] In addition, in the notification device described in any one of (1) to (4) above, (5) it is preferable that the inference mechanism identify the surrounding environment in which the vehicle is traveling, and the notification control mechanism change the notification method of the notification mechanism according to the surrounding environment.

[0020] In addition, in the notification device described in (5) above, (6) it is preferable that the inference mechanism infer the state of the surrounding environment based on at least one of illumination information, air temperature information, wind speed information, and information on the presence or absence of precipitation from a sensor mounted on the vehicle, and the notification control mechanism change the notification method of the notification mechanism according to the state of the surrounding environment.

[0021] In addition, in the notification device according to any one of (1) to (6) described above, (7) it is preferable that the estimation mechanism estimate a region in which the preceding person is highly likely to exist in a large number based on map information, and the notification control mechanism cause the notification mechanism to operate regardless of the presence or absence of the preceding person when the vehicle is traveling in the region.

[0022] In addition, in the notification device according to any one of (1) to (7) described above, (8) it is preferable that the estimation mechanism estimate a region in which the preceding person is highly likely to exist in a large number based on map information, and the notification control mechanism change the notification method of the notification mechanism based on the time of passage when the vehicle is traveling in the region.

[0023] In addition, in the notification device according to any one of (1) to (8) described above, (9) it is preferable that, in a case where it is determined based on the captured information of the capturing mechanism that the number of the preceding persons is large, the notification control mechanism cause the notification mechanism to operate until the foremost preceding person among the large number of preceding persons is not captured by the capturing mechanism.

[0024] In addition, in the notification device according to any one of (1) to (9) described above, (10) it is preferable that the estimation mechanism further estimate the evacuation space in the presence or absence of the preceding person, and the notification control mechanism cause the notification mechanism to operate in a case where the evacuation space in the presence of the preceding person is present.

[0025] In addition, in the notification device according to (10) described above, (11) the estimation mechanism estimates the evacuation space based on at least either the captured information of the capturing mechanism or map information.

[0026] In addition, in the notification device according to any one of (1) to (11) described above, (12) it is preferable that the notification control mechanism determine whether or not the preceding person has reacted to the notification of the notification mechanism based on the captured information of the capturing mechanism, and cause the notification mechanism to operate in a notification method different from the notification method of the last time in a case where it is determined that the preceding person has not reacted to the notification.

[0027] In addition, in order to solve the above problem, a notification method according to an embodiment of the present application for a preceding preceding person by the presence of a vehicle in automatic travel includes: a capturing step of capturing the preceding person; an estimation step of estimating at least a characteristic of the preceding person captured in the capturing step; a notification method determination step of determining a notification method for the preceding person based on at least the estimation result in the estimation step; and a notification step of performing notification of the presence of the vehicle to the preceding person based on the notification method determined in the notification method determination step.

[0028] Effects of Invention

[0029] According to the present application, it is possible to accurately determine the running situation of the vehicle side, and to perform the notification suitable for the situation to the oncoming person. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 is a schematic diagram showing the configuration of the notification device.

[0031] Figure 2 is a flowchart explaining the notification method of the first embodiment.

[0032] Figure 3 is a flowchart showing the details of the feature detection in the notification method of the first embodiment.

[0033] Figure 4 is a flowchart showing the details of the feature detection by other methods in the notification method of the first embodiment.

[0034] Figure 5 is a flowchart showing the details of the notification mode determination in the notification method of the first embodiment.

[0035] Figure 6 is a flowchart showing the details of the notification control in the notification method of the first embodiment.

[0036] Figure 7 is a schematic diagram for explaining the details of the retreat determination of the oncoming person on the road.

[0037] Figure 8 is a flowchart explaining the notification method of the second embodiment.

[0038] Figure 9 is a flowchart showing the details of the area situation determination (one) in the notification method of the second embodiment.

[0039] Figure 10 is a flowchart showing the details of the area situation determination (two) in the notification method of the second embodiment.

[0040] Figure 11 is a flowchart showing the details of the notification mode determination in the notification method of the second embodiment.

[0041] Figure 12 is a flowchart explaining the notification method of the third embodiment.

[0042] Figure 13 is a flowchart explaining the modification of the notification method of the third embodiment.

[0043] SYMBOL EXPLANATION

[0044] 100: notification device

[0045] 10: Filming equipment (filming mechanism)

[0046] 20: Inference Department (Inference Mechanism)

[0047] 30: Notification method to be determined by the department (or the organization).

[0048] 40: Notification device (notification mechanism)

[0049] 50: Notify the Control Department (Notify the Control Organization)

[0050] 60: Sensor

[0051] 70: Communication device

[0052] 80: HUD device

[0053] 90: Navigation device Detailed Implementation

[0054] The preferred embodiments for carrying out the present invention will be described below. Furthermore, for configurations other than those detailed below, publicly known technologies incorporating the aforementioned patent documents may be appropriately added.

[0055] Implementation Method 1

[0056] <Notification Device 100>

[0057] First, refer to Figure 1 The configuration of the notification device 100 in a preferred embodiment of the present invention will be described.

[0058] As shown in the figure, the notification device 100 of this embodiment is mounted on a vehicle V that is driving automatically, and has the function of notifying a traveler P that is moving forward relative to the vehicle V of the presence of the vehicle.

[0059] Here, "vehicle V" can be exemplified by 4-wheeled vehicles such as SUVs, cars, and buses.

[0060] Furthermore, the vehicle V is particularly preferably a hybrid electric vehicle or an electric vehicle equipped with a battery and an electric motor.

[0061] Furthermore, the vehicle V in this embodiment is configured to have a known automatic driving function as described above; however, the automatic driving function is not essential and can be omitted appropriately. In this case, for example, by equipping the vehicle V driven by an elderly person with the notification device 100 of this embodiment, safety can be further improved.

[0062] "The person moving forward, P" refers to a person moving forward relative to the vehicle V. Examples of such persons include pedestrians, people riding bicycles or other light vehicles. In this embodiment, the person moving forward, P can also include animals, such as dogs, cats, or other animals that may be on the road besides people.

[0063] In addition, "moving forward relative to vehicle V" refers to the state of being in front of the vehicle from the side of the vehicle, relative to the direction of travel. For example, on a straight road, it is the case that the vehicle is in front of the vehicle in the direction of travel and enters the predetermined route, or the case that the vehicle is to the side (lateral) of the vehicle V and moves in the same direction as the vehicle V and may enter the predetermined route.

[0064] More specifically, the notification device 100 is configured to include a shooting device 10, an inference unit 20, a notification method determination unit 30, a notification device 40, and a notification control unit 50. Furthermore, by Figure 1 It is understood that the notification device 100 is preferably configured to also include a sensor 60, a communication device 70, a HUD device 80, and a navigation device 90.

[0065] The shooting device 10 is the "shooting mechanism" in this embodiment, which has the function of shooting the aforementioned advance vehicle P. As a specific example of such a shooting device 10, a known vehicle-mounted camera, such as a CCD camera capable of shooting outside the vehicle, can be used. In addition, there are no particular restrictions on the number of shooting devices 10 or their placement, as long as the aforementioned function can be performed. For example, it can be configured to carry multiple CCD cameras to acquire more shooting information.

[0066] The inference unit 20 is the "inference mechanism" in this embodiment, which has the function of inferring at least the characteristics of the pioneer P captured by the shooting device 10, which is the shooting mechanism.

[0067] As a characteristic of such a pioneer P, it can be exemplified that one is most likely to infer whether the pioneer P is an elderly person, an adult other than an elderly person, a child, or a person with a disability, or whether there are visual / hearing difficulties (details are described later).

[0068] Furthermore, the inference unit 20 may be configured to have at least one of the following functions: inferring the presence or absence of retreat space SP for the advancer P; inferring the characteristics of the equipment (decorations) held by the advancer P; or inferring, based on map information, areas where the likelihood of a large number of advancers is high.

[0069] In this embodiment, the inference unit 20 is included in the control device C mounted on the vehicle V. The components of the control device C may be composed of a known circuit board (hardware), a central processing unit such as a CPU, and a program (software) for making it function.

[0070] The notification method determination unit 30 is the "determination mechanism" in this embodiment, and has the function of determining the notification method for the predecessor P based on the inference result obtained by the inference unit 20, which is the inference mechanism. In addition, the notification method determination unit 30 in this embodiment also has the function of identifying the surrounding environment based on various information from the sensor 60 described later.

[0071] As will be described later, in the notification device 100 of this embodiment, as a notification method, a total of four notification methods can be selected: sound emitted by the vehicle exterior speaker 41, vibration performed by the ultrasonic device 42, light emitted by the light-emitting device 43, and wind delivered by the air-blowing device 44.

[0072] Therefore, the notification method determination unit 30 has the function of determining whether to select one of these notification methods, or to combine one of these notification methods.

[0073] In this embodiment, the notification method determination unit 30 is also included in the control device C mounted on the vehicle V, just like the inference unit 20 described above.

[0074] As described above, the notification device 40 is configured to include at least one of the following: an external speaker 41, an ultrasonic device 42, a light-emitting device 43, and an air-blowing device 44. However, this embodiment is not limited to this example; for instance, the air-blowing device can be equipped with a temperature control function to deliver cold or warm air. Furthermore, as the ultrasonic device 42, a known ultrasonic generating device with ultrasonic tactile function utilizing an ultrasonic transducer can be used, for example. Additionally, as the light-emitting device 43, a known LED or organic EL device can be used.

[0075] The notification control unit 50 is the "notification control mechanism" in this embodiment. It has the function of notifying the traveler P of the presence of the vehicle V based on the notification method determined by the notification method determination unit 30, which is the aforementioned determination mechanism.

[0076] Furthermore, the notification control unit 50 of this embodiment preferably has at least one of the following functions, which are described in detail below: the function of different notification methods of the notification mechanism 40 according to the type of the forerunner P; the function of different notification methods of the notification mechanism 40 according to the equipment of the forerunner P; the function of different notification methods of the notification mechanism 40 according to the surrounding environment; and the function of operating the notification mechanism 40 only when there is a retreat space SP for the forerunner P.

[0077] In this embodiment, the notification control unit 50 is also included in the control device C mounted on the vehicle V, just like the inference unit 20 and the notification method determination unit 30 described above.

[0078] Sensor 60 is mounted on vehicle V and has the function of detecting various information. In this embodiment, as an example, vehicle V is equipped with a known laser detection device 61, a rain sensor 62, an illuminance sensor 63, an external temperature sensor 64, and a road surface sensor 65. It should be noted that in this embodiment, an example of the above sensors being mounted on vehicle V is shown, but any of these may be omitted depending on functional requirements.

[0079] The laser detection device 61 may exemplify the use of a known LiDAR (Light Detection and Ranging, Laser Imaging Detection and Ranging) system that employs a pulsed laser beam capable of detecting objects outside the vehicle V. In object detection using this LiDAR system, the aforementioned beam is illuminated around the vehicle V, reflected light from each object is received, and each object is detected based on the obtained data.

[0080] The communication device 70 may be a known in-vehicle communication device that has the function of communicating with the outside of the vehicle V, and can receive various information such as traffic congestion information and road traffic information through the communication device 70.

[0081] HUD device 80 is a known head-up display device that displays information directly in the human field of vision, showing real and virtual images on the windshield, rear window, or other glass surfaces of a vehicle V. It should be noted that while HUD device 80 is used in this embodiment, it can be replaced by, for example, a large LCD device or LED display device mounted on the dashboard, or a wearable device such as a known HMD (Head Mounted Display) device.

[0082] The navigation device 90 is a known device that has the function of retrieving a route from the current location to a destination based on map information stored in a memory (not shown). Therefore, the navigation device 90 of this embodiment can obtain the current location of the vehicle V via GPS or the like. It should be noted that the map information can be stored by the navigation device 90 or configured to be downloadable via a communication device 70 and a network N.

[0083] Furthermore, the network N in this embodiment can be exemplified by, for example, the well-known Internet.

[0084] <One of the notification methods>

[0085] Next, we will also refer to the appropriate references Figures 2-7 The notification method executed by the control device C in this embodiment will be described. It should be noted that the notification method of this embodiment is particularly preferred for use, for example, when a vehicle V in autonomous driving reminds the aforementioned preceding traveler P of the vehicle's presence.

[0086] That is, the notification method of this embodiment is a notification method for notifying a forward traveler P of the existence of an automatically driving vehicle, including: a shooting step, shooting the forward traveler P; an inference step, inferring at least the characteristics of the forward traveler P captured in the shooting step; a notification method determination step, determining a notification method for the forward traveler P based on the inference result in the inference step; and a notification step, notifying the forward traveler P of the existence of the vehicle V based on the notification method determined in the notification method determination step.

[0087] The following describes each process in detail with reference to the accompanying drawings.

[0088] In the notification method of this embodiment, firstly, it is determined whether an automatic driving (autonomous driving) function is being performed in the vehicle V (S100). Then, if autonomous driving is not being performed, the control device C monitors the vehicle until the autonomous driving function is performed.

[0089] On the other hand, when autonomous driving begins, the control device C determines whether a preceding traveler P exists on the predetermined travel path of the vehicle V (S200). More specifically, in this embodiment, the camera device 10 is used to capture an image of the front of the vehicle V, and the control device C detects the presence of the preceding traveler P based on the captured image information.

[0090] Alternatively, the control device C can use the laser detection device 61 to detect whether there is a traveler P ahead of the vehicle V, instead of the above situation. Or, the control device C can determine whether there is a traveler P ahead of the vehicle V on the predetermined travel route based on the information detected by both the imaging device 10 and the laser detection device 61.

[0091] Furthermore, as a predetermined route, when a driving route is set via the navigation device 90, the map information of the predetermined route can be used to set either the front or the side of the vehicle V as the predetermined route when no driving route is set.

[0092] If it is determined that a preceding traveler P exists on the aforementioned predetermined route, the control device C then performs feature detection on the identified preceding traveler P (S300). This will be explained later. Figure 3 and Figure 4 The specific details of feature detection are described.

[0093] In the subsequent notification method determination, a notification method suitable for the detected predecessor P is determined based on the characteristics of the predecessor P (S400). According to the subsequently determined notification method, the notification control mechanism 50 is used to drive the notification mechanism 40 (S500).

[0094] Repeat steps S100 to S500 until vehicle V stops (S600).

[0095] [Feature Detection (Human Identification)]

[0096] Next, refer to Figure 3 The feature detection related to human identification (body feature identification) in the notification method of this embodiment will be described in detail. As described below, in the feature detection of this embodiment, the inference unit 20 of the control device C detects the height or movement of the person in front P based on the shooting information of the shooting device 10.

[0097] That is, firstly, in the control device C, as a shooting process, it is determined whether image (image) information has been acquired in the shooting device 10 (S301). As described above, if the presence of the predecessor P can be detected using the shooting device 10, the image information at this time can be used directly. On the other hand, if the shooting device 10 is not used in detecting the presence of the predecessor P, the control device C controls the shooting device 10 to acquire the image information of the predecessor P.

[0098] Upon acquiring the image information, the control device C performs image analysis (S302). It should be noted that in this image analysis, known moving object detection algorithms such as background subtraction and frame subtraction can be used, enabling the detection of various objects, such as the person P, in the resulting image information. Furthermore, by combining known person extraction algorithms, such as those described in Japanese Patent Application Publication No. 2015-69460, it is also possible to determine whether a moving object within the image is a person.

[0099] Therefore, as an inference process, the control device C (inference unit 20) can identify the person P in the image if the analysis result indicates that the person is a human being (S303a). It should be noted that if no human being is present in the obtained image, it is determined that the presence or absence of the person P cannot be determined from the vehicle V (Mode E) (S311). It should be noted that in the case of Mode E, the notification method determination described later can, for example, be based on information about the surrounding environment.

[0100] In S303a described above, if human identification is possible based on the obtained image information, the inference unit 20 determines whether the person P can visually recognize the vehicle V (S304). More specifically, the inference unit 20 determines from the obtained image information whether the person P is holding a white cane. If the person P is holding a white cane, it is determined that the person is visually difficult to identify (no in S304); if the person P is not holding a white cane, it is inferred that the person P is visually identifiable (yes in S304). It should be noted that the criteria for determining whether the person P is visually identifiable are not limited to the above, and other known features such as being led by a guide dog can be extracted for determination.

[0101] Next, assuming the aforementioned forerunner P can visually identify vehicle V, the inference unit 20 then infers whether vehicle V can be identified audibly (S305). More specifically, the inference unit 20 determines from the obtained image information whether the forerunner P is wearing headphones, for example. If headphones are worn, it is determined that auditory recognition is difficult (no in S305), and if headphones are not worn, it is inferred that recognition is possible (yes in S305). It should be noted that such a criterion related to whether auditory recognition is possible is not limited to the above. For example, if other known feature quantities such as whether hearing aids are worn can be extracted, then such feature quantities can also be used for determination.

[0102] By proceeding through S305 above, it is determined whether the forerunner P parsed in the image information can be identified by either visual or auditory means (Mode A), or whether the vehicle V can be identified visually but is difficult to identify aurally (Mode B).

[0103] On the other hand, even if it is difficult to identify vehicle V visually (not in S304), the same as above, it is then determined whether vehicle V can be identified by hearing (S308). Then, by going through S308, it is determined whether the forerunner P parsed from the image information is difficult to identify visually but can be identified by hearing (mode C), or whether it is difficult to identify vehicle V by both vision and hearing (mode D).

[0104] It should be noted that in Mode A, since the predecessor P can recognize both sound and light, various notification methods can be used. On the other hand, since either sound or light can be recognized in Modes B and C, this condition is included in the notification method determination described later. Furthermore, since sound and light are difficult to recognize in Mode D, notification methods other than sound and light, such as ultrasound or airflow, are given priority.

[0105] It should be noted that the identification of people from image information is not limited to the methods described above. For example, by using it in conjunction with the laser detection device 61, it is also possible to obtain distance information from the person ahead P, detect the height of the person ahead P, and infer whether they are an adult or a child. In addition, although it is difficult to determine, for example, by identifying the hair color of the person ahead P based on the image information, it is also possible to infer whether the person ahead P is an elderly person.

[0106] Feature detection (object identification)

[0107] Next, refer to Figure 4 The feature detection related to object identification in the notification method of this embodiment will be described in detail.

[0108] As described below, in the feature detection of this embodiment, the inference unit 20 of the control device C detects the features of the equipment (decorations) held by the predecessor P based on the shooting information of the shooting device 10.

[0109] That is, firstly, in the control device C, it is determined, as described above, whether image (image) information has been acquired in the shooting device 10 (S301). Furthermore, as described above, if the image information has been acquired, the control device C performs image analysis (S302). It should be noted that, during image analysis, known algorithms such as Region CNN (Convolutional Neural Networks) utilizing deep learning can be used, thereby enabling the determination of people and equipment contained in the image information. In addition to the above, known equipment detection algorithms, such as those exemplified in Japanese Patent Application Publication No. 2018-84890, can also be used during image analysis.

[0110] Therefore, the control device C (inference unit 20) can identify the equipment (decorations) carried by the forerunner P from the image based on the analysis result (object identification) (S303b). It should be noted that if the aforementioned equipment is not present in the obtained image, it is determined that the equipment of the forerunner P cannot be determined from the vehicle V (mode E) (S311). It should be noted that, similarly to the above, in the case of mode E, in the notification method determination described later, the notification method can be determined, for example, based on information about the surrounding environment.

[0111] Next, if the inference unit 20 can identify the equipment of the aforementioned pioneer P, it then determines whether the pioneer P is equipped with (carries) for example, a white cane (S303c). Then, if the pioneer P is not carrying a white cane (no in S303c), the inference unit 20 then determines whether the pioneer P is wearing, for example, headphones or a hearing aid (hereinafter, these devices that have characteristics that make it difficult to perform hearing are referred to as "headphones, etc.") (S303d).

[0112] Then, in S303d, if the forerunner P is not wearing headphones or the like, the inference unit 20 infers that the vehicle V can be identified by both sight and hearing, and determines it as Mode A (S306). On the other hand, if the forerunner P is wearing headphones or the like in S303d, the inference unit 20 infers that it is difficult to identify the vehicle V by hearing, and determines it as Mode B (S307).

[0113] In addition, if the forerunner P is carrying a white cane (which is true in S303c), the inference unit 20 then determines whether the forerunner P is wearing, for example, headphones (S303e).

[0114] Then, in S303e, if the forerunner P is not wearing headphones, the inference unit 20 infers that the vehicle V can be identified by hearing and determines it to be the above-mentioned mode C (S309).

[0115] On the other hand, in S303e, when the forerunner P is wearing headphones, the inference unit 20 infers that it is difficult to identify the vehicle V through both vision and hearing, and determines it to be the above-mentioned pattern D (S310).

[0116] It should be noted that the above-mentioned method of identification is just one example. For instance, when determining whether a vehicle V can be identified visually, in addition to detecting someone carrying a white cane, one can also detect known equipment such as wearing blindfolds that indicate some kind of visual impairment. Furthermore, even when determining whether a vehicle V can be identified by hearing, one can detect other known equipment such as earmuffs that impair hearing.

[0117] It should be noted that the inference unit 20 of this embodiment can use at least one of the above-described human identification and object identification to infer that the predecessor P is an elderly person, an adult other than an elderly person, a child, or the most likely candidate among the disabled persons.

[0118] At this time, as one method of determining the notification method described later, for example, if the traveler P is any of the elderly, adults, and children, the notification control unit 50 may also cause the external speaker 41 to emit a sound as the notification mechanism 40. At this time, the notification control unit 50 may also change the frequency and / or amplitude of the sound based on the characteristics of the traveler P, for example, increasing the frequency and amplitude in the case of the elderly, and on the other hand, increasing only the frequency in the case of the adult, etc.

[0119] [Notification Method Determination]

[0120] Next, refer to Figure 5 The notification method determination in this embodiment will be described in detail. As described below, in the notification method determination of this embodiment, the notification method determination unit 30 of the control device C determines the surrounding environment of the vehicle V based on at least one of the following: illuminance information, temperature information, wind speed information, and precipitation information from the sensor 60 mounted on the vehicle V.

[0121] That is, firstly, the control device C determines whether the notification method is the first time (S401). More specifically, the number of notifications for the predecessor P is determined based on whether the predecessor P parsed from the image information is the same as the previous one.

[0122] Then, if the notification method is not the first time, since the previous notification to the predecessor P has already been made, a decision is made on whether to add a new notification method or increase its intensity (S402). It should be noted that, as a new notification method, for example, if the previous notification method was an audio notification made through the external speaker 41, this time it can be replaced by, for example, a light notification made through the light-emitting device 43 or an additional notification.

[0123] On the other hand, if the notification method is determined to be the first time (in S401), the control device C acquires various information from the sensor 60 (S403). More specifically, the control device C acquires information about whether there is precipitation from the rain sensor 62 mounted on the vehicle V, illuminance information from the illuminance sensor 63, temperature information from the external temperature sensor 64, and information about the road surface from the road surface sensor 65.

[0124] In addition, the control device C can also obtain wind speed information from a wind speed sensor (not shown) mounted on the vehicle V.

[0125] Thus, in this embodiment, various types of information are obtained from each sensor. However, as information from sensor 60, at least one of the following can also be obtained: illuminance information, temperature information, wind speed information, and information on whether there is precipitation.

[0126] Next, the notification mode determination unit 30, which obtains information from the sensor 60, identifies the surrounding environment of the vehicle V based on the information obtained from the sensor 60, and determines the notification mode corresponding to the surrounding environment (S404).

[0127] As an example, examples of notification patterns based on illumination and wind speed information are shown in Tables 1 and 2, which serve as information about the surrounding environment.

[0128] Table 1

[0129]

[0130] Table 2

[0131]

[0132] According to these tables, for example, if the illuminance is less than A lux and the wind speed is less than B m / s, it is in a normal state where there is no obstruction to the notification, and the notification mode α in Table 2 is selected. In this notification mode α, when the notification is made for the first time, the sound emitted by the external speaker 41 is selected, and when the notification is made for the second time, in addition to the sound, the light emitted by the light-emitting device 43 is added.

[0133] It should be noted that in any of the notification modes α to σ, the intensity increases after the 4th notification, but as n increases, the maximum intensity of each device also reaches its upper limit. Furthermore, the specific values ​​of A, which serves as the reference for illuminance, and B, which serves as the reference for wind speed, can be set arbitrarily according to their respective regions and seasons.

[0134] In addition, the examples of illumination and wind speed shown above are sensor information obtained from the surrounding environment, but are not limited to these examples. For example, the notification mode can also be determined based on other parameters such as whether there is precipitation or not, temperature information, etc.

[0135] After determining the notification mode in S404, the notification mode determination unit 30 of the control device C determines the notification strength of the notification device 40 (S405). When determining the notification strength, the notification mode determination unit 30 determines its initial value based on the sensor information obtained from the sensor 60.

[0136] As an example, Table 3 shows an example of determining the initial value of the volume of the external speaker 41 based on the amount of precipitation detected by the rain sensor 62, and Table 4 shows an example of determining the initial value of the volume of the external speaker 41 based on the wind speed detected by the wind speed sensor mentioned above.

[0137] Table 3

[0138] Precipitation [mm] Volume [dB] 0 to less than 1 60 1 to less than 2 63 2 to less than 5 66 5 to less than 10 70 10 to less than 20 75 20 to less than 50 80 50 to less than 80 85 80～ 90

[0139] Table 4

[0140] Wind speed [m / s] Volume [dB] 0 to less than 3 60 3 to less than 5 65 5 to less than 8 70 8 to less than 10 75 10 to less than 16 80 16～ 90

[0141] These tables show that, for example, when the precipitation is less than 1 mm, or when the wind speed is less than 3 m / s, the initial value of the volume of the external speaker 41 is set to, for example, 60 decibels.

[0142] It should be noted that, in addition to the aforementioned precipitation and wind speed, the initial value of the volume of the external speaker 41 can also be set based on other parameters, such as the temperature information obtained from the external temperature sensor 64. Furthermore, the initial values ​​of the ultrasonic waves of the ultrasonic device 42, the initial values ​​of the light emission quantity / intensity of the light-emitting device 43, or the initial values ​​of the airflow of the air supply device 44 can also be set based on the aforementioned parameters.

[0143] [Notification Control]

[0144] Next, refer to Figure 6 The notification control in the notification method of this embodiment will be described in detail. In the notification control of this embodiment, the notification control unit 50 of the control device C controls the notification device 40 by determining the notification method based on the notification method determination.

[0145] First, the notification control unit 50 determines whether this is the first time the notification control has been performed. As a criterion for this determination, similar to the criterion in the notification method determination described above, it can be based on whether the predecessor P parsed from the image information is the same as the previous one.

[0146] Then, in the case of initial notification control (yes in S501), the notification control unit 50 activates the notification mechanism (notification device 40) based on the notification mode and intensity determined by the notification method determination (S502). On the other hand, in the case of subsequent notification control (no in S501), the notification control unit 50 determines whether the pedestrian P has noticed the vehicle V (S503). More specifically, based on image information from the imaging device 10, the notification control unit 50 detects that the pedestrian P has turned from forward to backward along the direction of travel using known face detection algorithms, and infers that the pedestrian P has noticed the vehicle V.

[0147] Then, if it is determined that the preceding traveler P did not notice the vehicle V (no in S503), the notification mechanism (notification device 40) is activated again based on the notification mode and intensity determined by the notification method determination (S502). For example, in the case where the notification mode is "mode α", when n=2, notification is given by light in addition to sound.

[0148] In this way, the notification control unit 50 can, for example, determine whether the pioneer P has responded to the notification from the notification device 40 based on the shooting information from the shooting device 10. If it is determined that the pioneer P has not responded to the notification, the notification device 40 is activated to make the notification method different from the previous notification method.

[0149] Then, if it is determined that the foreground traveler P has noticed the vehicle V (in case S503), the notification control unit 50 determines whether there is a retreat space SP for the foreground traveler P under this condition (S504). If it is determined in S504 that there is a retreat space SP, the notification control unit 50 restarts the notification mechanism (notification device 40) based on the notification mode and intensity determined in the notification mode determination (S502). On the other hand, if it is determined that there is no retreat space SP, the notification control unit 50 determines that notification is not required and stops the operation of the notification mechanism (notification device 40) (S505).

[0150] Here, refer to Figure 7 The determination of the retreat space SP mentioned above will be explained.

[0151] As shown in the figure, the notification control unit 50 determines the type of area to the side (left side if left-hand traffic is allowed) of the preceding traveler P in front of vehicle V. This area type determination can be made by analyzing image information acquired from the imaging device 10, or by analyzing map information.

[0152] As an example, Table 5 illustrates the determination of whether or not to withdraw, which is carried out by the notification control unit 50.

[0153] Table 5

[0154] Area content Distance until area Can retreat Sidewalk No limit ○ Wall 0.8 m or more ○ Wall 0.8 m or less ×

[0155] From Table 5 and Figure 7 As can be seen from (a), if the image information acquired by, for example, the imaging device 10 is analyzed and it is determined that there is a sidewalk to the side of the pedestrian P walking on the road RD, the control unit 50 is notified to determine that the pedestrian P can retreat. At this time, the distance d1 between the pedestrian P and the sidewalk (retreat space SP) is unlimited, and it is determined that the pedestrian can retreat regardless of the size of the distance.

[0156] It should be noted that in this example, the notification control unit 50 determines whether there is a backoff space SP based on image information, but it can also determine whether there is a backoff space SP based on map information, or it can determine whether there is a backoff space SP based on both image information and map information. Alternatively, the notification control unit 50 can use the laser detection device 61 to determine whether there is a backoff space SP, or it can determine whether there is a backoff space SP based on both the detection result of the laser detection device 61 and the image capture result (image information) of the imaging device 10.

[0157] On the other hand, as shown in Table 5 and Figure 7 (b) shows that when there is no sidewalk on the side of the foreground traveler P but an obstacle X exists (for example, a wall as an obstacle X as shown in the figure), the control unit 50 is notified to determine whether the distance d2 to the obstacle X is within a predetermined distance. As an example, in this embodiment, if the distance d2 from the foreground traveler P to the wall as an obstacle X is 0.8m or more, it is determined that the foreground traveler P can retreat. It should be noted that in this embodiment, "0.8m" is used as a reference for whether retreat is possible, but it is not limited to this value and can be set appropriately.

[0158] According to the notification device and notification method in the first embodiment described above, the driving status of vehicle V can be accurately determined, and a notification appropriate to the driving status can be executed for the preceding traveler P.

[0159] In particular, the control device C (notification control unit 50) in this embodiment can change the notification method of the notification device 40 according to the type of the pioneer P. Furthermore, the control device C (notification control unit 50) can change the notification method of the notification device 40 according to the equipment of the pioneer P.

[0160] Furthermore, the control device C (notification control unit 50) can also change the notification method of the notification device 40 according to the surrounding environment. In addition, when it is determined that there is a retreat space SP for the advance traveler P, the control device C (notification control unit 50) activates the notification device 40.

[0161] It should be noted that, as described above, in this embodiment, a notification mode is determined in the notification mode determination unit 30 based on the feature detection of the predecessor P, but the method is not limited to this. For example, as described above, if the feature detection of the predecessor P is not performed, or if the feature detection is omitted, notification control for the predecessor P can be performed based on the notification mode determined by the notification mode determination unit 30 based on information obtained from the surrounding environment.

[0162] Implementation Method 2

[0163] Next, refer to Figure 1 andFigures 8-11 The notification device and notification method in the second embodiment will be described. It should be noted that in this embodiment, reference numerals are used to denote components that are identical to those described in the first embodiment, and their descriptions are appropriately omitted.

[0164] As can be understood from these figures, the notification device in this embodiment is a notification device that notifies a preceding vehicle V of the presence of the vehicle V in an autonomous driving vehicle V. It includes: an area information acquisition mechanism that acquires information about the area of ​​the predetermined travel route of the vehicle V; a laser detection mechanism that detects whether a preceding vehicle exists in the predetermined travel route; and a notification control mechanism that activates the notification device in the area based on the information of the area and the detection result of the laser detection mechanism.

[0165] Furthermore, the notification method of this embodiment is a notification method for notifying a preceding traveler P of the presence of an automatically driving vehicle V, including: a region information acquisition step, acquiring information about the region of a predetermined travel route traveled by vehicle V; a laser detection step, detecting whether a preceding traveler exists on the predetermined travel route; and a notification step, notifying the region based on the region information and the detection result of the laser detection step.

[0166] The following is a detailed description of each process with reference to the accompanying drawings.

[0167] In the notification method of this embodiment, similar to the first embodiment, it is first determined whether the automatic driving (autonomous driving) function is being performed in the vehicle V (S100).

[0168] On the other hand, when driving via autonomous driving begins, the control device C determines the area's condition (S350) and determines a notification method based on the area's condition (S450). Then, the control device C performs notification control according to the notification method determined by the notification method determination (S550).

[0169] Repeat steps S100 to S550 until vehicle V stops (S600).

[0170] [Regional Status Assessment (Part 1)]

[0171] Next, refer to Figure 9 The area status determination (one of the methods in this embodiment) of the notification method will be described in detail.

[0172] First, as a regional information acquisition process, the control device C determines whether map information has been acquired (S351). If map information has been acquired, it determines whether a warning area is included based on the regional information of the predetermined travel route of vehicle V (S352).

[0173] Here, in this embodiment, the "warning area" refers to a region where the likelihood of a large number of pioneers P existing is high. Examples include areas around train stations, schools, busy streets such as commercial districts, parks, or areas with frequent accident reports. Furthermore, this embodiment not only infers the likelihood of pioneers P existing but also introduces the concept of a "warning time period," thus considering real-world time periods when inferring the presence of pioneers P. It should be noted that the "warning time period" in this embodiment refers to a time period during which the likelihood of pioneers P appearing in that region is conventionally or statistically higher than in other time periods.

[0174] Table 6 illustrates such combinations of alert zones and alert periods.

[0175] Table 6

[0176] Warning area Warning time period Station surroundings Tram operation time + 1 h School surroundings School time period Busy street Shop business hours + 1 h Area with many accident information 24 hours

[0177] Therefore, as Figure 9 As shown, if the predetermined route of vehicle V includes, for example, "the vicinity of a station (e.g., within a radius of several hundred meters from the station)" (yes in S352), then in S352, the inference unit 20 of the control device C determines that a warning area exists. On the other hand, if the area shown in Table 6 does not exist in the predetermined route (no in S352), then the inference unit 20 determines that notification via the notification device 40 is not required (S354).

[0178] If, in step S352 above, it is determined that a warning area exists on the predetermined travel route, the inference unit 20 of the control device C then determines whether the time period during which vehicle V passes through the warning area is included in the warning time period (S353). Then, if it is determined that the time period during which vehicle V passes through the warning area is not the warning time period, the inference unit 20, similarly as above, determines that notification via the notification device 40 is not required (S354).

[0179] It should be noted that in this embodiment, a "warning period" is also added to the "warning area". However, the "number of people" detected by the laser detection device 61 described later can also be added to determine whether notification is required.

[0180] More specifically, for example, within the aforementioned warning area, a case where the average number of people detected by the laser detection device 61 is 3 or more is defined as "more," and a case where the number of people is less than 3 is defined as "less." It should be noted that "3 people" as the benchmark value for the average number of people detected is just an example, and other values ​​can be appropriately set.

[0181] Then, as shown in Table 7 below, if there is a warning area in the predetermined route and a large number of advance travelers P are actually detected in that area, the inference unit 20 of the control device C can issue a notification even if it is not a warning period.

[0182] Table 7

[0183]

[0184] Furthermore, although Table 7 includes "warning time periods" corresponding to the type of area, these warning time periods can be set uniformly, independent of the characteristics of the area, as shown in Table 8. It should be noted that in Table 8, the baseline for the average number of people tested in each of the aforementioned specified ranges is set at 5.

[0185] Table 8

[0186]

[0187] Then, if it is determined in S353 above that vehicle V passes through the warning area during the warning period, the control device C then uses the laser detection device 61 to detect whether a pedestrian P actually exists in the warning area as a laser detection process (S356). It should be noted that, at this time, it is preferable that the control device C uses the laser detection device 61 to detect whether a pedestrian P exists in at least one of the front or side of vehicle V.

[0188] Then, if the laser detection device 61 does not detect the presence of the preceding traveler P (no in S356), the inference unit 20 of the control device C determines that even if the vehicle V passes through the warning area, no notification is required (S354). On the other hand, if the laser detection device 61 detects the actual presence of the preceding traveler P (yes in S356), the inference unit 20 of the control device C determines that the warning area is a warning area where attention to the preceding traveler P is required (S357).

[0189] [Regional Situation Assessment (Part Two)]

[0190] Next, refer to Figure 10 The second part of the notification method of this embodiment will be described in detail. It should be noted that the parts of the second part of the notification method that are the same as those in the first part of the notification method (S351 to S356) are omitted from the description.

[0191] That is, in the determination of the situation in this area (Part Two), the characteristic is that the laser detection device 61 is used to detect whether there are multiple predecessors P.

[0192] That is, since the laser detection device 61 of this embodiment uses a LiDAR system, it is able to detect the number of people P ahead and their distance. Therefore, as a laser detection process, if multiple people P are detected by the laser detection device 61 (yes in S357), the inference unit 20 of the control device C determines that the warning area is the area requiring the most attention to multiple people P ahead (S359). On the other hand, if multiple people P are not detected by the laser detection device 61 (no in S357), the inference unit 20 of the control device C determines that the warning area is the area requiring attention to a single person P ahead (S358).

[0193] [Notification Method Determination]

[0194] Next, refer to Figure 11 The notification method determination in the notification method of this embodiment will be described in detail.

[0195] That is, the notification method determination unit 30 of the control device C determines whether the warning area on the predetermined travel route is a region of attention (S451). Then, if the warning area is not a region of attention (no in S451), the notification method determination unit 30 determines that the "region of most attention" is not required to notify, and therefore performs feature detection on multiple forwarders P (S452).

[0196] It should be noted that S452 can be omitted if the above-mentioned area condition determination (part two) is not applied.

[0197] The feature detection in S452 above can be applied, for example, using Figure 3 and Figure 4 The feature detection is explained. By passing through S452, the results of identifying the person and the object of the predecessor P are obtained, so the notification mode determination unit 30 can also determine the notification mode based on the features of the predecessor P.

[0198] If the area is determined to be an area of ​​attention in S451 or after feature detection in S452, the process moves to S453, where the notification method determination unit 30 determines the number of notification method determinations.

[0199] Because the content after S453 is related to Figure 5 The notification method determination (S400) of the first embodiment shown is the same, so its description is omitted here to avoid repetition.

[0200] [Notification Control]

[0201] In S550, as a notification process, notification control is performed based on the notification method determined by the notification method determination described above. It should be noted that, since the notification control in this embodiment is related to... Figure 6 The notification control (S500) in the first embodiment shown is the same, so its description is omitted here to avoid repetition.

[0202] According to the notification device and notification method in the second embodiment described above, the driving status of vehicle V can be accurately determined, and a notification appropriate to the driving status can be executed to the preceding vehicle P.

[0203] In particular, the control device C in this embodiment can determine the area (warning area) where there are multiple forerunners P based on map information, and select the notification method of the notification device 40 based on the condition of the area when the vehicle V is driving (warning time period, whether there are actual forerunners P, or the surrounding environment inferred by the sensor 60).

[0204] Thus, for example, if the vehicle V's planned route passes a school and it passes during school hours, the surrounding environment can be inferred from the various sensors 60 mentioned above. For example, a notification can be sent via the light-emitting device 43 in the case of low illumination, or a notification can be sent via the air supply device 44 if a hearing-impaired traveler P is present.

[0205] In addition, for example, when vehicle V passes through the restricted area during the aforementioned school hours, in addition to only providing sound notifications outside of school hours, a notification method combining light can be selected, so that the notification method of the notification device 40 can be different based on the time when vehicle V passes through the restricted area.

[0206] It should be noted that in the second embodiment described above, in addition to detecting information about the area requiring vigilance, the presence of the actual preceding traveler P is also detected by the laser detection device 61, but this is not limited to this example. For example, the inference unit 20 of the control device C can also infer areas with a high probability of multiple preceding travelers P by determining the vigilance area based on the map information described above, and notify the control unit 50 to activate the notification device 40 when the vehicle V is traveling in that area, regardless of whether there is a preceding traveler P present.

[0207] In addition, the notification control unit 50 of the control device C can determine the distance between itself and the preceding traveler P based on the detection result of the laser detection device 61, and change the notification method of the notification device 40 according to the determined distance between itself and the preceding traveler P (for example, changing the intensity of the notification according to the distance, or switching from sound to light or other notification devices according to the distance).

[0208] Third Implementation Method

[0209] Next, refer toFigure 12 and Figure 13 The notification device and notification method in the third embodiment will be described. In this embodiment, the control device C has the following characteristics: when there is a preceding traveler P on the predetermined travel route, it first determines whether there is a retreat space SP for the preceding traveler P, and if there is a retreat space SP, it determines the notification method. It should be noted that in this embodiment, for configurations that are the same as those described in the first and second embodiments, the same reference numerals are used and their descriptions are appropriately omitted.

[0210] That is, in the notification method of this embodiment, similar to the first embodiment, firstly, it is determined whether the automatic driving (autonomous driving) function is being performed in the vehicle V (S100), and then, it is detected whether there is a forerunner P in the predetermined travel route of the vehicle V (e.g., in front, to the side, or on the predetermined route set by the navigation device 90) (S200).

[0211] Then, if the presence of the forerunner P is detected in S200 above, the inference unit 20 of the control device C performs a retreat space determination (S375) to determine whether the forerunner P has a retreat space SP. It should be noted that this retreat space determination can be applied, for example, using... Figure 7 The method of S504 described in the first embodiment. Therefore, it is possible to determine whether the vehicle V can safely overtake the preceding traveler P.

[0212] Then, if the inference unit 20 determines that there is a retreat space SP, the notification method determination unit 30 of the control device C performs a notification method determination for the advance vehicle P (S450). It should be noted that this notification method determination can be applied, for example, using... Figure 11 The notification method determination in S450 described in the second embodiment. Furthermore, as the notification method determination in this embodiment, for example, using... Figure 5 The notification method determination in S400 described in the first embodiment is used instead of the notification method determination in S450.

[0213] Then, the notification control unit 50 of the control device C executes control (S550) to notify the advance traveler P using the notification device 40, according to the notification method determined by the notification method determination unit 30. It should be noted that this notification control can be applied, for example, using... Figure 6 The notification control of S500 described in the first embodiment is omitted here, as the backoff space determination has already been performed in this embodiment. Figure 6 The S504 process in the process.

[0214] According to the notification device and notification method in the third embodiment described above, the driving status of vehicle V can be accurately determined, and a notification appropriate to the driving status can be executed for the preceding traveler P.

[0215] In particular, since the control device C in this embodiment notifies the forerunner P after determining whether there is a retreat space SP when the forerunner P exists on the predetermined travel route, it suppresses the unnecessary notification to the forerunner P when there is no retreat space SP.

[0216] It should be noted that the above-described embodiments are preferred examples of the present invention. New structures and controls can be achieved by appropriately combining the elements of the embodiments without departing from the spirit of the invention. Hereinafter, variations applicable to these embodiments will be described.

[0217] <Variation Example>

[0218] Figure 13 This refers to a notification method in a variation of the third embodiment.

[0219] As shown in the figure, the control device C can determine whether the forerunner P has retreated (S380) after the retreat space determination (S375), and then perform notification method determination and notification control. At this time, in the above-mentioned S380, the inference unit 20 of the control device C can, for example, detect whether the forerunner P has entered the retreat space SP based on the image information acquired by the imaging device 10.

[0220] Then, if it is detected that the forerunner P has entered the retreat space SP (in S380), the control device C will not subsequently perform a notification to the forerunner P. On the other hand, if it is determined that even after a predetermined time (e.g., 1 minute) has elapsed, the control device C performs the notification method determination described above (S400).

[0221] Additionally, as shown in the figure, after the notification control for the foreground vehicle P (S500), the control device C can also determine whether the foreground vehicle P has retreated to the retreat space SP according to the notification (S380). Then, if it is detected that the foreground vehicle P has entered the retreat space SP (yes in S380), it is considered that the notification for the foreground vehicle P has ended, and the control device C then determines whether the vehicle V has stopped (S600).

[0222] The preferred embodiments and modifications of the present invention have been described in detail above with reference to the accompanying drawings, but the present invention is not limited to the above examples. Those skilled in the art will recognize that, within the scope of the technical concept described in the claims, they will attempt to further modify these embodiments and modifications, and these modifications naturally fall within the scope of the present invention.

[0223] For example, in various embodiments and variations, if the number of people identified as forwarders P by the shooting device 10 and the laser detection device 61 is multiple, the control device C may cause the notification device 40 to issue a notification until at least the forwarder in the front row among the multiple forwarders P is not photographed by the shooting device 10 (i.e., until it is not identified by the control device C).

[0224] Alternatively, the configurations of the third embodiment and its variations described above can be appropriately combined into the first embodiment and the second embodiment, respectively.

Claims

1. A notification device, characterized in that, A vehicle equipped with autonomous driving capabilities notifies a pedestrian of its presence of the vehicle's presence, including: The filming agency that filmed the aforementioned pioneers; An inference mechanism that infers at least the characteristics of the pioneer captured by the imaging mechanism; The determining agency, based on the inference results of the inference agency, determines the notification method for the pioneer; and The notification control agency, based on the notification method determined by the determining agency, activates the notification mechanism that notifies the preceding traveler of the presence of the vehicle. The inference mechanism infers the characteristics of the equipment maintained by the advance party based on the shooting information from the shooting mechanism. The notification control mechanism changes its notification method based on the equipment of the advance party.

2. The notification device according to claim 1, characterized in that, The inference mechanism detects the height or movement of the person in front based on the shooting information from the shooting mechanism, and infers whether the person in front is most likely to be an elderly person, an adult other than an elderly person, a child, or a disabled person. The notification control mechanism changes the notification method based on the category of the pioneer.

3. The notification device according to claim 2, characterized in that, In the case that the forerunner is any one of the elderly person, the adult, and the child, the notification control mechanism causes the sound-emitting mechanism to emit a sound as the notification mechanism, and changes the frequency and / or amplitude of the sound based on the characteristics of the forerunner.

4. The notification device according to any one of claims 1 to 3, characterized in that, The inference mechanism assesses the surrounding environment of the vehicle while it is in motion. The notification control mechanism changes the notification method based on the surrounding environment.

5. The notification device according to claim 4, characterized in that, The inference mechanism infers the conditions in the surrounding environment based on at least one of the following: illuminance information, temperature information, wind speed information, and precipitation information from sensors mounted on the vehicle. The notification control mechanism changes the notification method based on the surrounding environment.

6. The notification device according to any one of claims 1 to 3, characterized in that, The inference mechanism infers areas where the number of pioneers is high based on map information. The notification control mechanism activates the notification mechanism when the vehicle is traveling in the area, regardless of whether there is a preceding traveler.

7. The notification device according to any one of claims 1 to 3, characterized in that, The inference mechanism infers areas where the number of pioneers is high based on map information. The notification control mechanism changes the notification method based on the time the vehicle passes through the area.

8. The notification device according to any one of claims 1 to 3, characterized in that, If, based on the shooting information from the shooting agency, it is determined that there are multiple people among the trailers, the notification control agency activates the notification mechanism until at least the trailer in the front row among the multiple trailers is not photographed by the shooting agency.

9. The notification device according to any one of claims 1 to 3, characterized in that, The notification control mechanism determines whether the advance party has responded to the notification from the notification mechanism based on the shooting information from the shooting mechanism. If it is determined that the advance party has not responded to the notification, the notification mechanism is activated using a notification method different from the previous notification method.

10. The notification device according to claim 1, characterized in that, The inference mechanism also infers whether there is any retreat space for the aforementioned advance traveler. The notification control mechanism activates only when there is retreat space for the aforementioned advancers.

11. The notification device according to claim 10, characterized in that, The inference mechanism detects the height or movement of the person in front based on the shooting information from the shooting mechanism, and infers whether the person in front is most likely to be an elderly person, an adult other than an elderly person, a child, or a disabled person. The notification control mechanism changes the notification method based on the category of the pioneer.

12. The notification device according to claim 11, characterized in that, In the case that the forerunner is any one of the elderly person, the adult, and the child, the notification control mechanism causes the sound-emitting mechanism to emit a sound as the notification mechanism, and changes the frequency and / or amplitude of the sound based on the characteristics of the forerunner.

13. The notification device according to any one of claims 10 to 12, characterized in that, The inference mechanism assesses the surrounding environment of the vehicle while it is in motion. The notification control mechanism changes the notification method based on the surrounding environment.

14. The notification device according to claim 13, characterized in that, The inference mechanism infers the conditions in the surrounding environment based on at least one of the following: illuminance information, temperature information, wind speed information, and precipitation information from sensors mounted on the vehicle. The notification control mechanism changes the notification method based on the surrounding environment.

15. The notification device according to any one of claims 10 to 12, characterized in that, The inference mechanism infers areas where the number of pioneers is high based on map information. The notification control mechanism activates the notification mechanism when the vehicle is traveling in the area, regardless of whether there is a preceding traveler.

16. The notification device according to any one of claims 10 to 12, characterized in that, The inference mechanism infers areas where the number of pioneers is high based on map information. The notification control mechanism changes the notification method based on the time the vehicle passes through the area.

17. The notification device according to any one of claims 10 to 12, characterized in that, If, based on the shooting information from the shooting agency, it is determined that there are multiple people among the trailers, the notification control agency activates the notification mechanism until at least the trailer in the front row among the multiple trailers is not photographed by the shooting agency.

18. The notification device according to claim 10, characterized in that, The inference mechanism infers whether or not there is a retreat space based on at least one of the shooting information from the shooting mechanism and the map information.

19. The notification device according to any one of claims 10 to 12, characterized in that, The notification control mechanism determines whether the advance party has responded to the notification from the notification mechanism based on the shooting information from the shooting mechanism. If it is determined that the advance party has not responded to the notification, the notification mechanism is activated using a notification method different from the previous notification method.

20. A notification method, characterized in that, Notifying the preceding driver of the presence of the autonomously driving vehicle, including: The filming process involves filming the aforementioned pioneers; The inference process, at least inferring the characteristics of the pioneers captured in the shooting process; The notification method determination process, based at least on the inference results in the inference process, determines the notification method for the predecessor; and The notification process, based on the notification method determined in the notification method determination process, involves notifying the traveler of the presence of the vehicle. In the inference process, the characteristics of the equipment maintained by the explorer are inferred based on the shooting information from the shooting process. In the notification method determination process, the notification method of the notification process is changed according to the equipment of the pioneer.

21. The notification method according to claim 20, characterized in that, In the inference process, it is also inferred whether there is retreat space for the preceding traveler. The notification to inform the preceding traveler of the vehicle's presence shall be made only if there is sufficient retreat space for the preceding traveler.

Images