Driving assistance system and driving assistance method

Abstract

The system determines the threat level of a traffic participant that is a threat, taking into account the blind spot of the driver of the vehicle, depending on the positional relationship between the vehicle and other traffic participants such as other vehicles located in a specified area, and issues an alarm to the driver. [Solution] The driving assistance system comprises an external server communicatively connected to a first vehicle and a second vehicle located in a predetermined area. The driving assistance system acquires first position information and driver's line of sight information of the first vehicle and second position information of the second vehicle, generates a traffic situation map by superimposing the first position information, line of sight information, and second position information on a digital map showing the periphery of the predetermined area, calculates blind spot areas that are not visible from the first vehicle or the second vehicle on the traffic situation map, determines the threat level of a threatening object as seen by the first vehicle or the second vehicle, and decides a method of alerting the driver based on the threat level.

Description

[Technical Field]

[0001] The present invention relates to a driving assistance system and a driving assistance method. [Background technology]

[0002] There are known driving assistance devices that provide information and warnings to vehicle drivers according to traffic conditions. In particular, at intersections, multiple traffic participants simultaneously exist in different positions and move in different directions. Therefore, the traffic conditions at the intersection as seen by the vehicle driver change in real time.

[0003] Patent Document 1 discloses a driving assistance device that assists a vehicle in making a right turn according to traffic conditions. The driving assistance device predicts the vehicle situation at a right-turn intersection based on the traffic conditions around the intersection where the vehicle is to make a right turn, obtained through vehicle-to-vehicle communication and road-to-vehicle communication, calculates the timing at which the vehicle can turn right for each right-turn possible condition, and instructs the driver accordingly. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-265832 Summary of the Invention [Problem to be solved by the invention]

[0005] At intersections, multiple traffic participants (e.g., vehicles, motorcycles, bicycles, pedestrians, etc.) simultaneously exist and stop or move according to traffic signals. Therefore, collisions and contact accidents between vehicles and pedestrians may occur around intersections. Various technologies have been developed to prevent collisions and contact accidents between traffic participants and ensure traffic safety. For example, vehicles are equipped with onboard sensors (e.g., onboard cameras, positioning devices, millimeter-wave radar, etc.) to acquire image information about the vehicle's surroundings, its position, and the distance between the vehicle and preceding or following vehicles. Technologies have also been developed to detect driver information, such as the driver's line of sight, head movement, and posture, using driver monitoring systems (DMS). Furthermore, technologies for three-dimensionally measuring the situation around intersections have also been developed. For example, LiDAR (Laser Imaging Detection and Ranging) has been developed, which detects the surrounding situation by emitting laser light and analyzing the radio waves reflected from objects. Combining LiDAR with an onboard camera makes it possible to measure the shape of objects in real time with centimeter-level accuracy. In addition to other vehicles, pedestrians, and obstacles, objects (structures) such as buildings, traffic lights and traffic signs installed at intersections, guardrails, block walls, etc. Obstacles and objects on the road are collectively referred to as "targets."

[0006] On the other hand, vehicle information obtained through V2X (Vehicle to X) communication technologies, such as vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I), can be used to detect the vehicle's surroundings, such as the presence or absence of oncoming vehicles, pedestrians, traffic conditions, and weather conditions. Vehicle information can also be obtained by communicating with a server, such as a data center, via a base station using mobile communication technologies (e.g., 4G, 5G, WiFi, etc.). Vehicle information can be useful for detecting traffic participants in blind spots (i.e., areas blind to the driver) that cannot be detected by onboard sensors. Collisions and contacts between vehicles at intersections and junctions where multiple roads intersect, as well as collisions and contacts between right-turning vehicles and straight-moving vehicles or pedestrians, are problematic. Examples of vehicle information obtained through the above-mentioned communication technologies include information detected by the vehicle itself, information detected by other vehicles, information provided by pedestrians, and information from traffic systems such as surveillance cameras. Vehicle information can be used to help prevent collisions and contact accidents between the vehicle and other traffic participants who may be in the blind spot of the vehicle, such as other vehicles or pedestrians in areas that are difficult for the driver of the vehicle to see or recognize.

[0007] Under Japanese traffic regulations, vehicles must drive on the left side of the road, and on two-lane roads, sidewalks are provided on the left side of each lane. When a vehicle makes a left turn at an intersection, the driver must be aware of bicycles, pedestrians on the sidewalk, and pedestrians in crosswalks that may be on the left side of the vehicle. When a vehicle makes a right turn at an intersection, the driver may need to stop within the intersection to allow oncoming vehicles (vehicles traveling straight) to pass. Drivers must also pay attention to pedestrians on the sidewalk just before the crosswalk, as well as pedestrians on the crosswalk before the crosswalk. Furthermore, there may be oncoming vehicles waiting for the green light to turn on the road before the vehicle turns right. However, landmarks on the road before the vehicle turn right are hidden in the driver's blind spot and cannot be seen. In this case, it may be necessary to consider using a two-dimensional bird's-eye view of multiple vehicles, pedestrians, landmarks, etc. near the intersection. The situation at an actual intersection is a 3D image that changes from moment to moment, and the driver's field of vision is blocked by roadside buildings, roadside trees, landmarks, etc., so there are limits to the range that a driver can see when turning right. In other words, there are areas that are difficult for the driver to see due to the driver's blind spot, blind spots at the intersection, and visibility of the road.

[0008] In driving assistance technologies using distance measurement sensors such as on-board sensors (for example, on-board cameras, LIDAR, etc.), such as the driving assistance device of Patent Document 1, when an oncoming vehicle enters the detection range of the distance measurement sensor, control intervention is performed such as issuing a warning display or alarm to the driver of the vehicle, or inhibiting the vehicle from starting.In addition, V2X communication technology can be used to obtain information on other traffic participants and objects in the blind spot of the vehicle driver, such as other vehicles, pedestrians, obstacles, and traffic signs, and provide this information to the driver.

[0009] However, the above-mentioned driving assistance technologies have a problem in that they do not provide high accuracy. For example, when a vehicle attempts to turn right at an intersection and an oncoming vehicle is traveling straight in the oncoming lane, the above-mentioned driving assistance technology provides driving assistance by taking into account the positions and direction of the vehicle and the oncoming vehicle without specifying the lane in which the oncoming vehicle is traveling. This driving assistance technology can provide sufficient driving assistance when there is only one oncoming vehicle when the vehicle makes a right turn. However, when there are multiple traffic participants around the intersection, it is difficult to accurately determine which traffic participants, such as oncoming vehicles and pedestrians on the sidewalk or crosswalk, are likely to pose a threat to the driver (threats that may pose a risk of collision or contact accident). For this reason, in a right-turn use case, for example, when a right-turning vehicle and a straight-moving vehicle in the oncoming lane simultaneously enter an intersection, it is necessary to provide highly accurate driving assistance, such as issuing a warning at an appropriate time based on information about the vehicle attempting to turn right and information about other traffic participants (e.g., vehicles traveling straight, pedestrians, etc.).

[0010] In order to solve the above-mentioned problems, an embodiment of the present invention aims to provide a driving assistance system and a driving assistance method that can determine the threat level (risk of collision or contact accident) regarding the possibility of a collision or contact between the vehicle and other traffic participants, and perform driving assistance at an appropriate time, such as issuing an alarm to the driver depending on the threat level. [Means for solving the problem]

[0011] A first aspect of the present invention is a driving assistance system including a first vehicle that intends to turn around in a predetermined area and move in a first movement direction, a second vehicle that intends to pass through the predetermined area in a second movement direction, and an external server communicatively connected to the first and second vehicles. The driving assistance system includes an information acquisition unit that acquires first position information indicating the position of the first vehicle, line-of-sight information indicating the line-of-sight direction of a driver of the first vehicle, and second position information indicating the position of the second vehicle, a traffic condition map generation unit that generates a traffic condition map by superimposing the first position information and line-of-sight information related to the first vehicle and the second position information related to the second vehicle on a digital map that indicates the periphery of the predetermined area, a threat level determination unit that calculates a first blind spot area not visible from the first vehicle on the traffic condition map to determine a first threat level posed by a first threat as seen from the first vehicle, and calculates a second blind spot area not visible from the second vehicle to determine a second threat level posed by a second threat as seen from the second vehicle, and a notification determination unit that determines a first notification method for the driver of the first vehicle in accordance with the first threat level and a second notification method for the driver of the second vehicle in accordance with the second threat level.

[0012] A second aspect of the present invention is a driving assistance method that communicates with a first vehicle that is about to turn around in a predetermined area and move in a first movement direction, and a second vehicle that is about to pass through the predetermined area in a second movement direction, obtains first position information that indicates the position of the first vehicle, line-of-sight information that indicates the line-of-sight direction of the driver of the first vehicle, and second position information that indicates the position of the second vehicle, generates a traffic situation map by superimposing the first position information and line-of-sight information related to the first vehicle and the second position information related to the second vehicle on a digital map that indicates the predetermined area, calculates a first blind spot area that is not visible from the first vehicle in the traffic situation map, and determines a first threat level posed by a first threat object as seen from the first vehicle, calculates a second blind spot area that is not visible from the second vehicle, and determines a second threat level posed by a second threat object as seen from the second vehicle, determines a first notification method for the driver of the first vehicle in accordance with the first threat level, and determines a second notification method for the driver of the second vehicle in accordance with the second threat level. [Effects of the Invention]

[0013] According to an embodiment of the present invention, it is possible to determine the threat level (risk of a collision or contact accident) regarding the possibility of a collision or contact between the vehicle and other traffic participants, and to provide driving assistance at an appropriate time, such as issuing an alarm to the driver depending on the threat level. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a schematic diagram showing an overhead view of the positional relationship of a plurality of traffic participants present around an intersection to which a driving assistance system according to an embodiment of the present invention is applied. [Figure 2] 1 is a block diagram showing a schematic configuration of a driving assistance system according to an embodiment of the present invention; [Figure 3] 1 is a block diagram showing a specific configuration of a driving assistance system according to an embodiment of the present invention. [Figure 4] 4 is a flowchart showing a procedure for determining a threat level applied to a driving assistance method according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0015] A driving assistance system and a driving assistance method according to an embodiment of the present invention will be described with reference to the accompanying drawings.

[0016] This embodiment aims to prevent a collision between a right-turning vehicle and a straight-moving vehicle at an intersection. In this embodiment, vehicle information acquired through V2X communication and high-precision digital map information, such as a 3D digital map, are taken into consideration, taking into account the presence of other traffic participants, such as other vehicles and pedestrians. This embodiment is specialized for a right-turn use case in which a right-turning vehicle communicates with a straight-moving vehicle using V2X communication at an intersection, but is not limited to this. To prevent a possible collision between a right-turning vehicle and a straight-moving vehicle in a right-turn use case, this embodiment formulates operational logic that takes into consideration a high-precision digital map, i.e., logic for providing information to the driver and issuing a warning under predetermined conditions.

[0017] Traditionally, 2D maps have been digitized, but what drivers actually see is a constantly changing 3D image. External object recognition technology using onboard cameras, LIDAR, and other onboard sensors has improved the accuracy of digital maps by providing higher resolution and faster (real-time) measurement methods for measuring road shapes and the terrain around the vehicle. For example, it is possible to use multiple 3D point clouds as the source data for a map, extracting only the point cloud related to the center lines of road lanes and combining it with the white lines that separate the lanes. Such 3D digital maps are called high-precision digital maps (or HD (High Density) maps) to distinguish them from the digital maps installed in conventional navigation systems.

[0018] In a right-turn use case, a host vehicle attempting to turn right at an intersection, i.e., a right-turning vehicle, needs to pass a vehicle traveling straight in the oncoming lane. However, in the oncoming lane of the intersection, not only the vehicle traveling straight but also other right-turning vehicles may be waiting for an opportunity to turn right. The driver of the host vehicle's view is blocked by the body of the other right-turning vehicle in the oncoming lane, making it impossible to see the presence or absence of a vehicle traveling straight in the distance. In other words, the vehicle traveling straight in the oncoming lane may be blocked by a "blind spot" created by the other right-turning vehicle in the oncoming lane and may not be visible to the driver of the host vehicle. In such an intersection situation, traffic safety is improved by notifying the driver of the presence of a vehicle traveling straight in the oncoming lane that is in the blind spot of the driver of the host vehicle. Alternatively, when a vehicle traveling straight in the oncoming lane is approaching an intersection, traffic safety is improved by issuing a stop instruction when the driver starts the host vehicle (i.e., starts to turn right). In this embodiment, driving assistance such as a notification to the driver of the host vehicle, brake control, or start suppression is performed. In this embodiment, to detect other vehicles in the oncoming lane, position information of the vehicle and other vehicles is superimposed on a high-precision digital map. Furthermore, in this embodiment, the threat level (i.e., the threat level indicating the risk of collision or contact between the vehicle and other traffic participants) of other traffic participants (e.g., oncoming vehicles, pedestrians, etc.) that are the subject of a warning or other alert is determined. In other words, this embodiment selects the level of alert based on the threat level and issues the alert, thereby achieving collision avoidance according to the threat level.

[0019] Next, we will explain blind spots that are difficult for the driver of the vehicle to see. Blind spots of a vehicle based on the driver include physical blind spots such as pillars at the four corners of the vehicle, driving blind spots (blind spots on the left, right, and rear sides of the vehicle) that the driver may not pay attention to even if they consciously focus their attention while driving the vehicle, and blind spots outside the field of view of the vehicle's front camera.

[0020] It is difficult for the driver to directly see physical blind spots, but if vehicle-to-vehicle (or road-to-vehicle) communication with other vehicles is performed and image information as seen from the other vehicle can be acquired, targets in the blind spot can be recognized using an in-vehicle monitor, etc. For driving blind spots, the driver can recognize targets in the blind spot by changing their position and looking at the side mirrors or rearview mirror. For blind spots outside the camera's field of view, the driver can recognize targets in the blind spot by combining the vehicle's front camera with a driver monitoring system.

[0021] Furthermore, three types of blind spots may be defined: "road blind spots" such as shadows of other vehicles or buildings; blind spots to which the driver should pay attention, such as on the left, right, or rear of the vehicle; "driver's attention blind spots" such as blind spots caused by the driver paying too much attention to targets that are likely to attract the driver's attention; and "vehicle blind spots" such as blind spots due to the vehicle's structure. For example, a high-precision digital map can be used to provide an overview of the area around an intersection, and vehicle information (such as location information and movement information) acquired from the vehicle's own vehicle and other vehicles can be superimposed on the high-precision digital map on an external server, providing the driver with a traffic condition map showing the traffic conditions around the intersection. This allows the driver to check targets in blind spots that are difficult for the driver to see based on the traffic condition map provided by the external server.

[0022] Next, an example of a use case of a right turn in a traffic situation near an intersection will be described with reference to FIG. 1. FIG. 1 is a schematic diagram showing an overhead view of the relative positions of multiple traffic participants present near an intersection to which a driving assistance system 1 according to this embodiment is applied. Two roads X and Y intersect at an intersection IS (Intersection). Both roads X and Y are two-lane roads with one lane on each side. In FIG. 1, the intersection IS is defined as a circular area indicated by a dotted line, but the scope of the intersection IS is not limited to this. Furthermore, blind spot areas S (Sa, Sb, etc.) occur at the intersection IS depending on the traffic situation. For ease of explanation, one of the two lanes constituting each of roads X and Y is designated by the symbol "a" and the other by the symbol "b." In other words, road X consists of lanes Xa and Xb, and lane Xb is an oncoming lane when viewed from lane Xa. In addition, at intersection IS where two lanes X and Y intersect, there are multiple crosswalks (two crosswalks for each of roads X and Y). Note that we will not discuss the traffic lights installed at intersection IS.

[0023] In this embodiment, a traffic environment is described in which two-lane roads X and Y, each with one lane on each side, intersect at an intersection IS as shown in FIG. 1 . However, the present invention can also be applied to a four-lane road with two lanes on each side. To determine a right-turn use case, for example, on road X with one lane on each side, it is determined whether the driver of vehicle A operates a turn signal to indicate an intention to turn right. On a road with two lanes on each side, a right-turn lane is provided in addition to the lanes for going straight and turning left. Therefore, to determine a right-turn use case, for example, it is determined whether the driver indicates an intention to turn right by moving vehicle A into the right-turn lane. In this way, in this embodiment, the operation logic of the right-turn use case is activated when the driver of vehicle A indicates an intention to turn right at intersection IS.

[0024] Around the intersection IS shown in FIG. 1, there are multiple traffic participants: vehicles A, B, C, and D, and a pedestrian P. The line of sight of the drivers of vehicles A, B, C, and D is indicated by arrows AD, BD, CD, and DD. Specifically, vehicle A, which is attempting to turn right along lane Xa in the direction of travel Ta, is located on one side of the intersection IS, while vehicle B, which is attempting to travel straight along lane Xb in the direction of travel Tb, is located on the other side of the intersection IS. On one side of the intersection IS, building BL1 is located to the right of lane Xa in which vehicle A is located, at a distance from adjacent lane Xb. Therefore, the driver of vehicle A's view is obstructed by building BL1, creating a blind spot area Sa. In other words, even if the driver of vehicle A looks to the right to view the range from line of sight AD1 to AD2, his or her line of sight will be obstructed by building BL1. On the other side of the intersection IS, building BL2 is located at a distance from the right side of lane Xb in which vehicle B is located (or from the left side as seen from the driver of vehicle B). As a result, the driver of vehicle B's view is blocked by building BL2, creating a blind spot area Sb. Note that the blind spot areas Sa and Sb are shown schematically and differ from blind spots that occur in an actual intersection environment. In addition, in Figure 1, the blind spot areas Sa and Sb overlap on the road Y side.

[0025] Vehicle A on lane Xa of road X is turning right at intersection IS in the direction of travel Ta and moving onto lane Ya of road Y. Vehicle B on lane Xb is going straight through intersection IS in the direction of travel Tb. In addition, other vehicles C and D are parked in tandem on lane Yb of road Y, which intersects with road X, waiting for the traffic light on lane Yb to change from red to green. Vehicle C is partially obscured by blind spots Sa and Sb, but is visible to the drivers of vehicles A and B if they pay close attention. On the other hand, vehicle D is in blind spots Sa and Sb and is therefore invisible to the drivers of vehicles A and B.

[0026] Pedestrian P is on the sidewalk near building BL1 on the right side of lane Xb, moving in direction D and attempting to cross the crosswalk on road Y at intersection IS. In Figure 1, pedestrian P is located within the gaze direction AD1 to AD2 of the driver of vehicle A when he directs his attention to the right. In other words, the driver of vehicle A in lane Xa can visually confirm pedestrian P by directing his gaze to the right. Meanwhile, the driver of vehicle B in lane Xb can visually confirm pedestrian P by looking toward intersection IS. Furthermore, pedestrian P is in a position that makes it difficult for the driver of vehicle C, which is stopped at the stop line just before the crosswalk on road Y in lane Yb, to see pedestrian P. However, if the passenger in the front passenger seat of vehicle C directs his attention to the left, he can see pedestrian P. Furthermore, the driver of vehicle C can also visually confirm pedestrian P by directing his attention to the left.

[0027] As described above, the threat level of a threat object (i.e., an object that poses a potential risk of causing a collision accident) to each traffic participant varies depending on the relative positions of vehicles A to D near intersection IS and the moving direction D of pedestrian P. Possible threat levels include, for example, the urgency of a need to alert the vehicle driver or to intervene in vehicle control. In this embodiment, the position information of vehicles A and B and the driver's line of sight information are superimposed on a high-precision digital map recorded on an external server and stored as a traffic situation map, and calculations are performed using predetermined logic to determine the threat level to each traffic participant.

[0028] In this embodiment, three threat levels are set for threats that are other traffic participants as seen from the vehicle (for example, vehicle B and pedestrian P as seen from vehicle A) according to the position information of each traffic participant (vehicles A, B, etc.) and the gaze information of the drivers of vehicles A and B. The relative positional relationship between vehicles at the intersection IS can be grasped by storing the vehicle information of the traffic participants in an external server via V2X communication. The driver's gaze information is measured by a driver monitoring system (DMS), and the gaze information is included in the vehicle information and transmitted to the external server. This makes it possible to grasp the gaze directions of the drivers of vehicles A and B facing each other at the intersection IS on a traffic situation map formed by superimposing the position information and gaze information of the traffic participants on a high-precision digital map.

[0029] Threat Level 1: The driver is looking at the threat and can easily avoid a collision (or contact). Threat Level 2: If the driver can change their line of sight and see the threat, they can avoid a collision (or a collision). Threat Level 3: The threat is in the blind spot area S, so the driver cannot see the threat even if he or she changes his or her line of sight.

[0030] The above-mentioned threat level is not a fixed value, but changes from moment to moment depending on the traffic conditions at the intersection IS. In this embodiment, the threat level of the threat target is determined, and information and an alarm are issued to the driver according to the threat level. For example, the threat level is determined based on the line of sight AD of the driver of vehicle A (a vehicle turning right) and the line of sight BD of the driver of vehicle B (a vehicle going straight), and if the threat level is threat level 2 or threat level 3, information regarding vehicle B is provided to the driver of vehicle A and an alarm is issued.

[0031] Furthermore, if the visibility at the intersection IS is good and the threat level of vehicle B (a vehicle going straight) is determined to be 1 from the perspective of vehicle A (a vehicle turning right), and the threat level of vehicle A (a vehicle going straight) is determined to be 1 from the perspective of vehicle B (a vehicle going straight), the effect (collision avoidance effect) of issuing an alert to the drivers of vehicles A and B is small. For this reason, an alert is issued to the driver of a vehicle that encounters a threat object with a relatively high threat level, that is, a threat object that is estimated to be 2 or 3, higher than 1. This allows the driver to become aware of the presence of the threat object early and take action to avoid the collision, thereby reducing the risk of collision. Actions to avoid the collision include the driver slowing down or stopping the vehicle or changing steering direction.

[0032] With reference to FIG. 1 , the threat level determination method of this embodiment will be described in detail. Vehicle B is visible from vehicle A, and a collision can be avoided, so vehicle B is determined to be at threat level 1. Regarding vehicle C as seen from vehicle A, if the driver of vehicle A can turn his gaze to the right and see vehicle C, the possibility of a collision can be avoided. Therefore, vehicle C is determined to be at threat level 2 from the perspective of the driver of vehicle A. Regarding pedestrian P as seen from vehicle A, if the driver of vehicle A can turn his gaze to the right and see pedestrian P, the possibility of a collision (or contact) can be avoided, so pedestrian P is determined to be at threat level 2. In other words, if the driver of vehicle A changes his gaze direction AD to the right and pays attention to gaze direction AD1 to AD2 to see pedestrian P, the threat level of pedestrian P will decrease from threat level 2 to threat level 1. Vehicle D is present in blind spot area Sa as seen from vehicle A. Even if the driver of vehicle A were to look to the right, he or she would be able to see pedestrian P, but vehicle D is completely inside blind spot area Sa, so the driver of vehicle A would not be able to see vehicle D. In other words, the threat level of vehicle D from vehicle A's perspective would be determined to be threat level 3.

[0033] On the other hand, if the driver of vehicle B keeps his eyes fixed on the road ahead, he can see vehicle A and pedestrian P through intersection IS. Therefore, the threat level of vehicle A and pedestrian P, which are threats from the driver of vehicle B's perspective, is determined to be threat level 1. Furthermore, for vehicle C, which is partially obstructed by blind spot area Sb, the driver of vehicle B can see vehicle C if he changes his line of sight BD and looks left through intersection IS. Therefore, the threat level of vehicle C from the perspective of vehicle B is determined to be threat level 2. However, as vehicle D is completely inside blind spot area Sb, it is located behind vehicle C, so the driver of vehicle B cannot see vehicle D even if he changes his line of sight BD. Therefore, the threat level of vehicle D from the perspective of vehicle B is determined to be threat level 3.

[0034] As described above, vehicle D is determined to be a threat level 3 from the perspective of vehicles A and B, and, if necessary, vehicles A and B are provided with information or an alarm that vehicle D is present behind vehicle C. However, it is also assumed that vehicle D is connected to an external server for communication, and vehicle information about vehicle D is reflected on a high-precision digital map. In this case, although the drivers of vehicles A and B cannot directly see vehicle D, if information is provided based on the high-precision digital map, the drivers of vehicles A and B can confirm that vehicle D is present behind vehicle C.

[0035] The traffic condition map of the intersection IS (or the digital map showing the traffic conditions at the intersection IS), which is obtained by superimposing vehicle information on a high-precision digital map stored on an external server, is updated continuously in response to changes in the location or direction of movement of traffic participants, and therefore the threat level determination results for threat targets are also updated continuously at short intervals. For this reason, a traffic participant, for example, can specify a specific area (e.g., the area around the intersection IS) on the high-precision digital map and make an inquiry to sequentially download (or browse) the traffic conditions for the specific area. This allows a traffic participant (e.g., vehicle A) to avoid the possibility of a collision accident while checking the conditions of other traffic participants (e.g., vehicles B and C, pedestrian P, etc.) in real time. In this way, this embodiment can provide driving assistance that allows vehicles A and B to operate smoothly while paying attention to pedestrian P in a use case where vehicles turn right at the intersection IS.

[0036] FIG. 2 is a block diagram showing an example of the configuration of a driving assistance system 1 according to this embodiment. In the driving assistance system 1, a host vehicle 10 (e.g., vehicle A) and another traffic participant 20 are connected to a data server 30 via V2X communication. Examples of the other traffic participants 20 include terminal devices with communication capabilities carried by other vehicles B and C and pedestrian P, and a surveillance camera installed at an intersection IS. The surveillance camera can acquire images showing the traffic conditions at the intersection IS. The surveillance camera can also identify the positions of vehicles A, B, and C approaching the intersection IS, or the position of pedestrian P (or the terminal device carried by pedestrian P) through image processing or image analysis. The data server 30 is an external server connected to the Internet, etc., and may be a cloud server installed in a remote location or a roadside device or edge server installed near a road network. A high-precision digital map M is stored in the data server 30. In this embodiment, a specific area around the intersection IS is focused on from all the digital data in the high-precision digital map M.

[0037] The vehicle 10 and other traffic participants 20 (particularly other vehicles) are equipped with a global positioning system (GPS), which allows them to acquire navigation position information. The driving assistance system 1 according to this embodiment is equipped with a global navigation satellite system (GNSS) 40 to acquire detailed position information as viewed from a communication satellite in real time. The vehicle 10 and other traffic participants 20 query the GNSS 40 to acquire detailed position information.

[0038] The data server 30 is also linked to a plurality of road cameras 50 installed around the roads. The road cameras 50 may also include surveillance cameras installed at intersections IS. Therefore, images of a plurality of vehicles traveling on the roads are captured in real time by the road cameras 50 and transmitted to the data server 30 as traffic information. The data server 30 receives images from the road cameras 50 and reflects the ever-changing traffic conditions (such as the operating status of a plurality of vehicles) on the high-precision digital map M.

[0039] Next, the functions of the driving assistance system 1 will be described. The subject vehicle 10 (for example, vehicle A) queries the GNSS 40 to obtain first position information a. Other traffic participants 20 (for example, vehicle B) also query the GNSS 40 to obtain second position information b. Meanwhile, the data server 30 obtains traffic information c (i.e., traffic conditions at an intersection IS where vehicles A to D, pedestrian P, etc. are present in close proximity) based on images captured by a road camera 50, such as the positions and movement directions of multiple vehicles and pedestrians. The first position information a, second position information b, and traffic information c are obtained from external sensors, and constitute peripheral parts of the subject vehicle 10, other traffic participants 20, and data server 30, which are the core of the driving assistance system 1.

[0040] The host vehicle 10 generates first vehicle information d including first position information a, the turn signal activation state, the driver's line of sight information and head direction (driver information detected by a driver monitoring system), the driving situation and vehicle body information of the host vehicle 10, and transmits it to the data server 30. In this embodiment, the turn signal activation state is treated as a direction indication indicating the driver's intention to go straight, turn right, or turn left. In FIG. 1, the driver of vehicle A activates the turn signal to indicate his / her intention to turn right. On the two-lane road X and Y shown in FIG. 1, the driver's intention regarding the direction indication he / she arbitrarily decides is determined by whether or not the turn signal is activated. However, on a four-lane road, the driver's intention may also be determined by whether or not the turn signal is activated. Similarly, other traffic participants 20 also generate second vehicle information e consisting of second position information b, the operation status of the turn signal, the driver's line of sight information and head direction (driver information detected by the driver monitoring system), and the driving status and vehicle information of other traffic participants 20 (other vehicles), and transmit it to the data server 30.

[0041] If the other traffic participant 20 is a pedestrian P, the pedestrian P can see the traffic lights at the intersection IS, vehicles A, B, C, etc. In this embodiment, a terminal device such as a smartphone is assumed as the means by which the pedestrian P communicates with the driving assistance system 1. In this case, the position of the terminal device of the pedestrian P may be acquired as the second position information b acquired from the GNSS 40. Alternatively, position information acquired by the terminal device of the pedestrian P in cooperation with another satellite communication system may be used. Furthermore, the second vehicle information e may include pedestrian information consisting of the position information of the terminal device and the moving direction of the pedestrian P (or the orientation of the head of the pedestrian P) detected by the terminal device.

[0042] The data server 30 stores a map of the road network in a specified jurisdiction of Japan (for example, the Kanto area) and a 3D map showing the positions and heights of buildings as a high-precision digital map M. The data server 30 updates the high-precision digital map M in accordance with traffic information c obtained from the road cameras 50. The data server 30 also acquires first vehicle information d from the vehicle 10 and second vehicle information e from other traffic participants 20. Using a specified algorithm, the data server 30 reflects the positions of vehicles A to D and pedestrian P at intersections IS in real time on the high-precision digital map M to generate a traffic situation map.

[0043] The data server 30 then determines the threat level of the threat object using a predetermined operational logic based on the traffic condition map. In this embodiment, threat levels are set from 1 to 3 as threat levels, but the threat level varies depending on the threat object as seen by the traffic participants, and the threat level is updated depending on changes in the driver's line of sight and traffic conditions. If the subject vehicle 10 is vehicle A, vehicle B and pedestrian P in the driver's field of view are threat objects, and vehicle C may also be a threat object when turning right. If the other traffic participant 20 is vehicle B, vehicle A and pedestrian P in the driver's field of view are threat objects, and vehicle C may also be a threat object depending on the status of the traffic light at the intersection IS. Therefore, the data server 30 determines the threat level for each traffic participant, notifies the subject vehicle 10 of a first threat level f related to the first threat object, and notifies the other traffic participants 20 of a second threat level g related to the second threat object. Note that the first threat level f and the second threat level g may be the same threat level or different threat levels.

[0044] Next, a specific configuration of the driving assistance system 1 according to this embodiment will be described with reference to the block diagram of Fig. 3. In Fig. 3, in addition to the host vehicle 100 (e.g., vehicle A) corresponding to the host vehicle 10 shown in Fig. 2, other vehicles 210 (e.g., vehicle B) as other traffic participants 20, a terminal device 220 carried by a pedestrian P, and a monitoring device 230 installed at an intersection IS are assumed. Also, an external server 300 is provided corresponding to the data server 30 shown in Fig. 2. The host vehicle 100, the other vehicles 210, the terminal device 220, the monitoring device 230, and the external server 300 are communicatively connected, and the host vehicle 100 and the other vehicles 210 can exchange information with each other through V2V communication.

[0045] The host vehicle 100 includes sensors 101 (e.g., an on-board camera, a lidar, a driver monitoring system, etc.), a vehicle information generation unit 102, a vehicle information transmission unit 103, a notification determination unit 104, and a notification unit 105. The other vehicle 210 has the same configuration as the host vehicle 100. The vehicle information generation unit 102 generates first vehicle information based on first position information from the GNSS 40 and driver information (e.g., driver's line of sight information) detected by the sensors 101. The vehicle information transmission unit 103 of the host vehicle 100 transmits the first vehicle information to the external server 300. Similarly, the other vehicle 210 generates second vehicle information based on second position information from the GNSS and driver information, etc., and transmits the second vehicle information to the external server 300. The external server 300 stores a high-precision digital map M.

[0046] The terminal device 220 carried by the pedestrian P can measure its own position (referred to as third position information) using a GPS function. The terminal device 220 can also detect the behavior of the pedestrian P. For example, the terminal device 220 can perform facial recognition on the pedestrian P to detect the direction of the head of the pedestrian P, or detect the direction of movement of the pedestrian P. Therefore, the terminal device 220 generates pedestrian information based on the third position information of the terminal device 220 and the behavior of the pedestrian P, instead of the first and second vehicle information, and transmits the pedestrian information to the external server 300.

[0047] The monitoring device 230 is installed at the intersection IS together with traffic lights, and includes a monitoring camera 231 and a speaker 232. The position of the monitoring device 230 (referred to as fourth position information) is set in advance, and the position of each of the vehicle 100, other vehicle 210, and pedestrian P can be detected by analyzing images captured by the monitoring camera 231. That is, the monitoring device 230 generates monitoring information indicating the current positions of the vehicle 100, other vehicle 210, and pedestrian P at the intersection IS, and transmits the information to the external server 300. The speaker 232 can transmit predetermined messages by voice to multiple traffic participants present around the intersection IS. If the pedestrian P is not carrying the terminal device 220, the speaker 232 of the monitoring device 230 can transmit a warning or the like from the external server 300 to the pedestrian P by voice.

[0048] The external server 300 includes an information acquisition unit 301, a traffic condition map generation unit 302, a threat level determination unit 303, and a threat level transmission unit 304. The information acquisition unit 301 acquires traffic information from the road camera 50 shown in FIG. 2 . The information acquisition unit 301 also acquires first vehicle information from the host vehicle 100, second vehicle information from another vehicle 210, pedestrian information from the terminal device 220 held by the pedestrian P, and monitoring information from the monitoring device 230. The traffic condition map generation unit 302 reflects the first vehicle information, second vehicle information, pedestrian information, and monitoring information on the high-precision digital map M. As a result, the traffic condition map generation unit 302 generates a traffic condition map that projects the positional relationship between the host vehicle 100 and the other vehicle 210, as well as the positions of the terminal device 220 and the monitoring device 230, onto a three-dimensional map of a predetermined area such as an intersection IS. The threat level determination unit 303 determines the threat level of a threat object from the perspective of a predetermined traffic participant using a predetermined determination logic according to the relative positions of multiple traffic participants (e.g., the subject vehicle 100, the other vehicle 210, the mobile terminal 220 held by the pedestrian P, etc.) on the traffic situation map. The threat level transmission unit 304 transmits the respective threat levels to the subject vehicle 100, the other vehicle 210, the terminal device 220, and the monitoring device 230. Note that if the monitoring device 230 does not detect a threat object but has detected the pedestrian P, the threat level of the threat object from the pedestrian P is transmitted from the external server 300.

[0049] In the host vehicle 100, the notification determination unit 104 determines the notification method according to the threat level notified from the external server 300. The notification unit 105 activates the speaker of the host vehicle 100 in accordance with the notification method to notify the driver of an alarm using a predetermined sound. In this case, the volume may be changed according to the threat level. Alternatively, the notification unit 105 may display a warning such as a predetermined image or a predetermined message on the navigation device (display) of the host vehicle 100.

[0050] Although the notification determination unit 104 that determines the notification method for the host vehicle 100 is provided within the host vehicle 100, the function of the notification determination unit 104 may be provided in the external server 300. Similarly, the function of determining the notification method for the other vehicle 210 may also be provided in the external server 300. In this case, the notification method for the host vehicle 100 (or the other vehicle 210) is determined based on the threat level determined by the threat level determination unit 303 in the external server 300. Furthermore, the threat level transmission unit 304 of the external server 300 functions as a notification method transmission unit that transmits the notification method determined as described above to the notification unit 105 of the host vehicle 100.

[0051] In FIG. 3 , the processes for generating a traffic condition map based on the high-precision digital map M and determining the threat level of a threat object are implemented on the external server 300. However, these functions may be distributed between the external server 300 and an on-board computer installed in the vehicle 100 or the other vehicle 210. Because the high-precision digital map M requires a large amount of information, it must be implemented on the external server 300. Furthermore, because generating a traffic condition map requires a large amount of computing power, it must be implemented on the external server 300. However, the determination of the threat level of other traffic participants, taking into account the driver's blind spots, may be implemented on the on-board computer. For example, the on-board computer can query the external server 300 to retrieve a traffic condition map around the intersection IS and grasp the positional relationship between the blind spot area as seen from the driver of the vehicle and other traffic participants in a simple diagram (e.g., a two-dimensional diagram as shown in FIG. 1 ). In other words, if the on-board computer can obtain a simplified version of the traffic condition map (or a reduced version focusing on the intersection IS) with a relatively small amount of information from the external server 300, it can determine the threat level of a threat object even with the on-board computer's relatively limited computing power.

[0052] Next, a method of determining the threat level by the data server 30 (or the threat level determination unit 303 of the external server 300) will be described with reference to the flowchart in FIG. 4. The data server 30 (or the external server 300) comprises a processor and a storage device, and the storage device stores a program related to the threat level determination process. The processor may be a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). The processor executes the threat level determination process and calculates the threat level for each of a plurality of conditions.

[0053] When there are multiple traffic participants who are threats, the control flow is executed in parallel for each traffic participant to calculate the threat level for each traffic participant. At the intersection IS in Figure 1, when focusing on vehicles A and B (host vehicle 100, other vehicle 210) and pedestrian P (terminal device 220), the threat level is calculated for six possible combinations.

[0054] (1) The threat level to vehicle B from vehicle A's perspective. (2) The threat level to pedestrian P from vehicle A's perspective. (3) The threat level to vehicle A from vehicle B's perspective. (4) The threat level to pedestrian P from vehicle B's perspective. (5) The threat level of vehicle A from the perspective of pedestrian P. (6) The threat level of vehicle B from the perspective of pedestrian P.

[0055] The flowchart in Fig. 4 includes an input process IN and steps S10 and S20 that describe a plurality of conditions. The flowchart in Fig. 4 is described with a focus on the data server 30 in the driving assistance system 1 shown in Fig. 2, and classifies other traffic participants 20 as seen by the driver of the vehicle 10 into threat levels 1 to 3. If the other traffic participants 20 are not in the driver's blind spot, and the driver visually recognizes the other traffic participants 20, the threat level is determined to be 1, and if the driver does not visually recognize the other traffic participants 20, the threat level is determined to be 2. Furthermore, if the other traffic participants 20 are in the driver's blind spot, the threat level is determined to be 3. Here, the threshold is set as "Threat Level 1 < Threat Level 2 < Threat Level 3", but this is not limiting.

[0056] First, the data server 30 inputs various sensing data (IN). The sensing data inputted include first position information a of the subject vehicle 10, second position information b of the other traffic participants 20, and the line of sight direction and head orientation of the driver of the subject vehicle 10 (or another vehicle that is another traffic participant 20).

[0057] When another traffic participant 20 that is a threat to the host vehicle 10 is present in the blind spot area S, the threat level of the other traffic participant 20 is determined to be threat level 3 (step S10). When the other traffic participant 20 that is a threat to the host vehicle 10 is not present in the blind spot area S but the driver of the host vehicle 10 is looking at the other traffic participant 20, the threat level of the other traffic participant 20 is determined to be threat level 1 (step S20, YES). On the other hand, when the driver of the host vehicle 10 is not looking at the other traffic participant 20, the threat level of the other traffic participant 20 is determined to be threat level 2 (step S20, NO). In this way, even if the other traffic participant 20 is not in the blind spot area S, different threat levels are calculated depending on the line of sight of the driver of the host vehicle 10.

[0058] Consider the positional relationship between vehicles A and B and pedestrian P in FIG. 1. Just before intersection IS, the drivers of vehicles A and B have their gaze directions AD and BD aligned with their respective straight-ahead directions. Therefore, vehicle B is visible to the driver of vehicle A, and vehicle B is therefore determined to be at threat level 1 (step S20, YES). Because the gaze direction AD of the driver of vehicle A differs from the position of pedestrian P, pedestrian P is determined to be at threat level 2 (step S20, NO). However, if the driver of vehicle A visually recognizes pedestrian P, pedestrian P is determined to be at threat level 1 (step S20, YES). In this case, depending on the status of the traffic lights at intersection IS, when vehicle A turns right along its travel direction Ta and approaches a pedestrian crossing, there is a possibility that vehicle A may come into contact with or collide with pedestrian P walking on the pedestrian crossing. Therefore, for threats determined to be at threat level 3, vehicle A issues a strong warning to the driver or performs control intervention, such as braking.

[0059] On the other hand, vehicle A and pedestrian P could be threats from vehicle B, but because vehicle A and pedestrian P attempting to turn right are visible within the field of view of the driver of vehicle B in the line of sight BD, vehicle A and pedestrian P are determined to be at threat level 1 from vehicle B's perspective (step S10 → step S20). On the other hand, vehicle C, which is waiting for a traffic light to change from red to green in lane Yb of road Y, could also be a threat from vehicle B depending on the traffic light's status. Because vehicle C is not in vehicle B's blind spot Sb, vehicle C is determined to be at threat level 1 (step S20). However, depending on the congestion at intersection IS, it may take vehicle B time to pass through intersection IS, causing the traffic light located along road X to change from green to red. In this case, there is a risk that the traffic light located along road Y will change from red to green, causing vehicle C, which is waiting in lane Yb, to depart. Therefore, it is possible that vehicle C, which is a threat to vehicle B, may change from threat level 1 to threat level 2 or threat level 3.

[0060] Furthermore, vehicles A, B, and C may be threats from the perspective of pedestrian P. When pedestrian P crosses the crosswalk on road Y in the direction of movement D, vehicle B traveling straight along lane Xb is visible to pedestrian P, and is therefore determined to be at threat level 1 (step S20). However, with regard to vehicle A turning right at intersection IS, pedestrian P may not notice the presence of vehicle A until vehicle A approaches the crosswalk that pedestrian P is crossing. For this reason, vehicle A is determined to be at threat level 2 or 3 from the perspective of pedestrian P. Since the driver of vehicle A is normally obligated to pay attention to pedestrians on the crosswalk, an alarm is issued to vehicle A regarding the presence of pedestrian P.

[0061] 3, if pedestrian P has terminal device 220, the threat level is transmitted from external server 300 and an alarm is issued to terminal device 220. If pedestrian P does not have terminal device 220, monitoring device 230 detects a proximity state between vehicle A, which has turned right at an intersection and is approaching a crosswalk on road Y, and pedestrian P, who is moving across the crosswalk, and external server 300 notifies monitoring device 230 of the threat level based on the monitoring information. If, as viewed from monitoring device 230, vehicle A and pedestrian P are in a mutual threat level 2 or threat level 3 state, an alarm is issued to vehicle A and an audio warning is given to pedestrian P via speaker 232 of monitoring device 230.

[0062] Next, modifications and effects of the driving assistance system and driving assistance method (particularly, the threat level determination method and the notification method such as an alarm) according to the present invention will be described. Here, the description will be made with reference to the block diagram of FIG. 3, but the functions shown in FIG. 3 are illustrative and not limiting. For example, the function of the notification determination unit 104 installed in the host vehicle 100 may be installed in the external server 300.

[0063] The driving assistance system 1 includes a host vehicle 100 that attempts to turn around in a predetermined area such as an intersection and move in a first movement direction, another vehicle 210 that attempts to pass through the predetermined area in a second movement direction, and an external server 300 that is communicatively connected to the host vehicle 100 and the other vehicle 210. The driving assistance system 1 generates a traffic situation map by superimposing first position information of the host vehicle 100, driver's line of sight information, and second position information of the other vehicle on a digital map showing the periphery of the predetermined area. In the traffic situation map, a first blind spot area that is not visible from the host vehicle 100 is calculated to determine a first threat level posed by a first threat object as seen from the host vehicle 100, and a second blind spot area that is not visible from the other vehicle 210 is calculated to determine a second threat level posed by a second threat object as seen from the other vehicle 210. Thereafter, a first notification method for the driver of the host vehicle 100 is determined according to the first threat level, and a second notification method for the driver of the other vehicle 210 is determined according to the second threat level.

[0064] In this way, an appropriate notification method can be implemented for the driver of the vehicle 100 at an appropriate timing according to the threat level of the threatening object determined in consideration of the blind spot as seen from the driver of the vehicle 100. This makes it possible to avoid the risk of collision (or contact) between the vehicle 100 and other traffic participants such as the other vehicle 210.

[0065] A first threat level and a second threat level higher than the first threat level may be set as the first threat level of the host vehicle 100 vis-à-vis the first threat and the second threat level of the other vehicle 210 vis-à-vis the second threat. In this case, when the first threat level or the second threat level is determined to be the second threat level, a notification method for the driver of the host vehicle 100 or the other vehicle 210 may be determined. For example, the first threat level may be determined to be the second threat level when the first threat level of the host vehicle 100 is located in a first blind spot area, or the second threat level may be determined to be the second threat level when the second threat level of the other vehicle 210 is located in a second blind spot area. This allows an alarm or the like to be constantly notified to the driver of the host vehicle 100 or the other vehicle 210 regardless of the level of the first threat level of the first threat level or the second threat level of the second threat level, thereby avoiding a situation in which the driver feels annoyed and does not pay attention to a serious alarm.

[0066] Assume that the predetermined area is an intersection, and the host vehicle 100 intends to turn right at the intersection while another vehicle 210 traveling in the oncoming lane continues straight through the intersection. In this case, when the driver of the host vehicle 100 indicates an intention to turn right at the intersection, a first threat level may be determined for the other vehicle 210 that is the first threat based on the line of sight of the driver of the host vehicle 100 on the traffic situation map. The process of determining the first threat level of the other vehicle 210 that is the first threat based on the traffic situation map imposes a large computational processing load. Therefore, by executing a predetermined operational logic to determine the first threat level when the driver of the host vehicle 100 indicates an intention to turn right at the intersection, a situation in which a large computational processing load is constantly imposed on the external server 300 can be avoided.

[0067] In the above, the second threat level of the host vehicle 100 that is the second threat target may be determined according to the line of sight direction of the driver of the other vehicle 210 on the traffic situation map.

[0068] It is also assumed that the specified area is an intersection, there is a pedestrian near the intersection about to cross a crosswalk, and the terminal device 220 carried by the pedestrian is communicatively connected to the external server 300. In this case, the driving assistance system 1 may determine pedestrian information indicating the position and movement direction of the pedestrian detected by the terminal device 220 on a traffic condition map. Furthermore, when the driver of the vehicle indicates an intention to turn right at an intersection, the first threat level of the pedestrian who is the first threat may be determined based on the line of sight of the driver of the vehicle 100 and the line of sight of the pedestrian on the traffic condition map. By executing a predetermined operational logic to determine the first threat level when the driver of the vehicle 100 indicates an intention to turn right at an intersection, it is possible to avoid a situation in which a large computational load is constantly placed on the external server 300.

[0069] In the above, when the pedestrian's own vehicle 100 or another vehicle 210 is determined as a third threat target from the pedestrian's perspective, a third threat level may be determined based on the pedestrian's movement direction or line of sight direction on the traffic situation map, and the third threat level may be notified to the pedestrian's terminal device 220. Note that the pedestrian's movement direction can be determined using a position / direction sensor built into a terminal device such as a smartphone, or a traffic monitoring camera installed around the intersection. Similarly, the pedestrian's line of sight direction (or head direction) can be determined using a traffic monitoring camera installed around the intersection.

[0070] In the above, a monitoring device 230 may be installed at an intersection, and the relative positions of the host vehicle 100, the other vehicle 210, and the pedestrian may be identified from video captured by the monitoring camera 231 of the monitoring device 230. The monitoring device 230 generates monitoring information indicating the traffic conditions around the intersection. The monitoring information is reflected in a traffic condition map. For example, the monitoring device 230 can track the position of a pedestrian at an intersection even if the pedestrian is not carrying a mobile terminal 220. When a pedestrian is walking on a crosswalk, the host vehicle 100 or the other vehicle 210 may appear to the pedestrian as a third threat that may pose a risk of contact or collision. In this case, a third threat level of the third threat from the pedestrian's perspective may be determined based on the monitoring information from the monitoring device 230, and a warning may be issued to the pedestrian from a speaker 232 mounted on the monitoring device 230. This ensures the safety of pedestrians at intersections even if the pedestrian is not carrying a terminal device 220.

[0071] By notifying not only the subject vehicle 100 and the other vehicle 210 but also other traffic participants such as pedestrians of the threat level determined with reference to the traffic situation map, it is possible to encourage both the subject vehicle 100, the other vehicle 210, and the pedestrians to take action to avoid the risk of collision. In particular, when other traffic participants such as pedestrians are present in the blind spots of the drivers of the subject vehicle 100 and the other vehicle 210, it is possible to notify both of them of the potential risk of collision, thereby effectively avoiding the risk of collision.

[0072] Although several embodiments of the present invention have been described, these embodiments are presented as examples and are not intended to limit the scope of the invention. These novel embodiments can be embodied in various other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their modifications are included within the scope and spirit of the invention, and are also included in the inventions and their equivalents as defined in the claims. [Explanation of symbols]

[0073] 10 Vehicle 20 Other traffic participants 30 Data Servers 40 GNSS 50 Road Camera 100 Vehicle 101 Sensors 102 Vehicle information generation unit 103 Vehicle information transmission unit 104 Notification Decision Unit 105 Information Department 210 Other vehicles 220 Terminal Equipment 230 Monitoring equipment 300 External Server 301 Information Acquisition Department 302 Traffic condition map generation unit 303 Threat Assessment Unit 304 Threat Level Transmission Unit A, B, C, D vehicles BL1 and BL2 buildings IS Intersection M High-precision digital map P Pedestrian S, Sa, Sb blind spot area X, Y road Xa, Xb, Ya, Yb lanes

Claims

1. A driving assistance system including a first vehicle that attempts to change direction in a predetermined area and move in a first movement direction, a second vehicle that attempts to pass through the predetermined area in a second movement direction, and an external server that is communicatively connected to the first vehicle and the second vehicle, an information acquisition unit that acquires first position information indicating a position of the first vehicle, line-of-sight information indicating a line-of-sight direction of a driver of the first vehicle, and second position information indicating a position of the second vehicle; a traffic condition map generation unit that generates a traffic condition map by superimposing the first position information and the line of sight information related to the first vehicle and the second position information related to the second vehicle on a digital map showing the periphery of the predetermined area; a threat level determination unit that calculates a first blind spot area that is not visible from the first vehicle in the traffic condition map to determine a first threat level posed by a first threat as seen from the first vehicle, and that calculates a second blind spot area that is not visible from the second vehicle to determine a second threat level posed by a second threat as seen from the second vehicle; a notification determination unit that determines a first notification method for the driver of the first vehicle in accordance with the first threat level, and that determines a second notification method for the driver of the second vehicle in accordance with the second threat level.

2. 2. The driving assistance system according to claim 1, wherein a first threat level and a second threat level higher than the first threat level are set for the first threat level and the second threat level, respectively, and when the threat level determination unit determines that the first threat level is the second threat level, the notification determination unit determines the first notification method for the driver of the first vehicle, and when the threat level determination unit determines that the second threat level is the second threat level, the notification determination unit determines the second notification method for the driver of the second vehicle.

3. 3. The driving assistance system according to claim 2, wherein the threat level determination unit determines the first threat level to be the second threat level when the first threat is located in the first blind spot area of ​​the first vehicle, and determines the second threat level to be the second threat level when the second threat is located in the second blind spot area of ​​the second vehicle.

4. 2. The driving assistance system according to claim 1, wherein when the specified area is an intersection and the driver of the first vehicle indicates an intention to turn at the intersection, the threat level determination unit determines the first threat level for the second vehicle, which is the first threat as seen from the first vehicle, based on the line of sight direction of the driver of the first vehicle on the traffic situation map.

5. The driving assistance system according to claim 4, characterized in that the threat level determination unit determines the second threat level for the first vehicle, which is the second threat as seen from the second vehicle, based on the line of sight direction of the driver of the second vehicle on the traffic situation map.

6. 2. The driving assistance system of claim 1, wherein when the specified area is an intersection and a pedestrian having a terminal device that is communicatively connected to the external server is present in the vicinity of the intersection, the information acquisition unit acquires pedestrian information indicating the position and movement direction of the pedestrian measured by the terminal device, the traffic condition map generation unit generates the traffic condition map by reflecting the pedestrian information, and when a driver of the first vehicle indicates an intention to turn at the intersection, the threat level determination unit determines the first threat level for the pedestrian who is the first threat as seen from the first vehicle, based on the line of sight of the driver of the first vehicle in the traffic condition map and the pedestrian information.

7. 7. The driving assistance system according to claim 6, wherein the threat level determination unit determines a third threat level for the first vehicle, which is a third threat from the pedestrian's perspective, based on the line of sight direction of the driver of the first vehicle and the pedestrian information on the traffic condition map, and notifies the terminal device of the pedestrian of the third threat level.

8. 8. The driving assistance system according to claim 7, wherein a first threat level and a second threat level higher than the first threat level are set for each of the first threat level, the second threat level, and the third threat level, and when the pedestrian information from the terminal device indicates that the pedestrian is located in the first blind spot area of ​​the first vehicle, the threat level determination unit determines the third threat level to be the second threat level, and the notification determination unit determines a third notification method and notifies the terminal device.

9. 2. The driving assistance system according to claim 1, wherein, when the specified area is an intersection and a monitoring device that generates monitoring information by identifying the relative positions of the first vehicle, the second vehicle, and a pedestrian from video is installed at the intersection, the information acquisition unit acquires the monitoring information, the traffic situation map generation unit generates the traffic situation map by reflecting the monitoring information, the threat level determination unit determines a third threat level for the first vehicle or the second vehicle, which is a third threat from the pedestrian's perspective, and the monitoring device determines a third notification method for the pedestrian based on the third threat level.

10. communicating with a first vehicle that intends to turn around in a predetermined area and move in a first movement direction and a second vehicle that intends to pass through the predetermined area in a second movement direction; acquiring first position information indicating a position of the first vehicle, line-of-sight information indicating a line-of-sight direction of a driver of the first vehicle, and second position information indicating a position of the second vehicle; generating a traffic situation map by superimposing the first position information and the line-of-sight information related to the first vehicle and the second position information related to the second vehicle on a digital map showing the predetermined area; calculating a first blind spot area that is not visible from the first vehicle in the traffic situation map to determine a first threat level of a first threat object as seen from the first vehicle; calculating a second blind spot area that is not visible from the second vehicle to determine a second threat level of a second threat object as seen from the second vehicle; determining a first notification method for a driver of the first vehicle according to the first threat level; A driving assistance method, comprising determining a second notification method for the driver of the second vehicle in accordance with the second threat level.

Images