Driving assistance method, driving assistance program, and driving assistance system

Abstract

To provide a driving support method, a driving support program, and a driving route support system that enable preventive vehicle actions to avoid risks in advance.SOLUTION: A driving support method generates route information representing a travel route from a current location of an own vehicle to a destination location, generates object information including information on attributes of a first obstacle, information on a travel lane in which the first obstacle was detected, and information on a position in the direction of travel in which the first obstacle was detected when the first obstacle is detected in an image of the road taken by traffic information acquisition means located facing the road, generates risk area information by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the travel lane and the location information, and updates the route information by selecting other travel lanes in the route information so as to bypass the portion included in the risk area information.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to a driving support method, a driving support program, and a driving support system.

Background Art

[0002] Patent Document 1 discloses a technique for classifying objects around a host vehicle according to attributes and states, calculating a potential risk map by applying potential risks corresponding to the classification, and performing control such as avoidance of the object on the host vehicle based on the calculated potential risk map.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the technique according to Patent Document 1, since control such as avoidance is performed after detecting surrounding objects, preventive vehicle actions for avoiding risks in advance cannot be taken.

[0005] An object of the present invention is to provide a driving support method, a driving support program, and a driving route support system capable of preventive vehicle actions for avoiding risks in advance.

Means for Solving the Problems

[0006] The driving assistance method according to the present invention generates route information that represents the driving route from the vehicle's current location to the destination, and includes information on multiple roads connecting the current location to the destination in series, and information on one driving lane selected based on the driving route from among multiple driving lanes arranged in parallel on each road. When a first obstacle is detected in an image of the road taken by a traffic information acquisition means positioned facing the road, object information is generated that includes information on the attributes of the first obstacle, information on the driving lane in which the first obstacle was detected, and position information in the direction of travel in which the first obstacle was detected. Risk area information is generated by associating a predetermined risk value corresponding to the attributes of the first obstacle with the driving lane information and position information. The route information is then updated by selecting another driving lane to bypass the portion included in the risk area information in the route information. When two first obstacles are detected side-by-side in the same lane along the direction of travel, a predetermined risk value is associated with the positional information from the position of the preceding first obstacle to the position of the succeeding first obstacle in the direction of travel in the object information, thereby generating risk area information. . [Effects of the Invention]

[0007] According to the present invention, the driving route can be updated to switch driving lanes before and after an obstacle before reaching a road where an obstacle exists, thereby enabling preventative vehicle behavior that avoids risks in advance. [Brief explanation of the drawing]

[0008] [Figure 1] Figure 1 is a block diagram of the driving support system of the first embodiment. [Figure 2] Figure 2 shows a simplified control flow of the driving assistance method of the present invention. [Figure 3] Figure 3 shows the control flow of the driving support system of the first embodiment, via the vehicle, traffic information acquisition device, and server. [Figure 4] Figure 4 is a control flow showing the details of the risk area information generation process and the learning database update process shown in Figure 3. [Figure 5]Figure 5 shows a case where a driving route is set to turn left from road (R1) and enter road (R2), in which lane (L1) (left lane) is initially selected on road (R1) and lane (L1) (left lane) is initially selected on road (R2), where Figure 5(a) shows the case where there are no first obstacles on road (R1), Figure 5(b) shows the case where there are two first obstacles on road (R1) and one first obstacle on road (R2), and the vehicle detects all first obstacles and performs a detour maneuver to avoid them, and Figure 5(c) shows the case where two first obstacles are detected on road (R1) via the traffic information acquisition unit and one first obstacle is detected on road (R2) via the traffic information acquisition unit, and the vehicle performs a detour maneuver to avoid the first obstacles. [Figure 6] Figure 6 shows examples of object information and route information generated according to the travel route. Figure 6(a) shows object information and route information when there is no first obstacle, as shown in Figure 5(a). Figure 6(b) shows object information and route information when the first obstacle is detected on road (R1) and road (R2) via the traffic information acquisition device, as shown in Figure 5(c). Figure 6(c) shows the state in which risk area information is generated by associating a risk value with the object information. [Figure 7] Figure 7 shows a scenario where a driving route is set to turn left from road (R1) and enter road (R2), resulting in lane (L1) (left lane) being initially selected on road (R1) and lane (L1) (left lane) being initially selected on road (R2). Figure 7(a) shows a scenario where the learning database has prior information indicating the presence of a third obstacle behind lane (L1) on road (R1), and when the vehicle detects a second obstacle in lane (L1) preceding the third obstacle, it switches lanes to avoid the second and third obstacles. Figure 7(b) shows a state where the information (location) of the second obstacle and the information (location) of the third obstacle are added to the object information. Figure 7(c) shows a state where risk area information is generated by associating a risk value with the object information. [Figure 8]Figure 8 shows the range of route information updates when the vehicle travels along a route in the order of road (R1), road (R2), and road (R3). Figure 8(a) shows the case where object information is generated up to the entrance of road (R2) when the vehicle is traveling on road (R1), and Figure 8(b) shows the case where object information is generated up to the entrance of road (R3) just before the vehicle enters road (R2). [Figure 9] Figure 9 is a block diagram of the driving support system of the second embodiment. [Figure 10] Figure 10 shows the control flow of the driving support system of the second embodiment, via the vehicle, traffic information acquisition unit, and server. [Figure 11] Figure 11 is a block diagram of the driving support system according to the third embodiment. [Figure 12] Figure 12 shows the control flow of the driving support system of the third embodiment, via the own vehicle, traffic information acquisition unit, server, and other vehicles. [Figure 13] Figure 13 is a control flow showing the details of the risk area information generation process in the third embodiment. [Figure 14] Figure 14 shows a scenario where a driving route is set to turn left from road (R1) and enter road (R2), resulting in lane (L1) (left lane) being initially selected on road (R1) and lane (L1) (left lane) being initially selected on road (R2). Figure 14(a) shows the case where another vehicle detects a fourth obstacle in the latter part of lane (L1) on road (R1), and the vehicle itself detects a second obstacle in lane (L1) prior to the fourth obstacle, and the vehicle switches lanes to avoid the second and fourth obstacles. Figure 14(b) shows the state where the information (position) of the second obstacle and the information (position) of the fourth obstacle are added to the object information. Figure 14(c) shows the state where risk area information is generated by associating risk values ​​with object information. [Modes for carrying out the invention]

[0009] Embodiments of the present invention will be described below with reference to the attached drawings.

[0010] [First Embodiment] FIG. 1 is a block diagram of a driving support system according to the first embodiment. The driving support system according to the first embodiment has a configuration for realizing the driving support method of the present invention as shown below.

[0011] The driving support system is installed in a traffic information acquisition unit 1 facing the road, a host vehicle 2, and a server 3.

[0012] Here, the "first obstacle", "second obstacle", "third obstacle", and the "fourth obstacle" described later used in this embodiment (the first embodiment, the second embodiment, and the third embodiment) refer to all things that can be an obstacle to driving for the host vehicle 2 such as a vehicle stopped on the road. The "first obstacle" is an obstacle detected by the server 3 via the traffic information acquisition unit 1, and the "second obstacle" is an obstacle detected by the host vehicle 2. The "third obstacle" is an obstacle that has been detected by the host vehicle 2 or another vehicle 4 (FIG. 11) in the past outside the detection range of the traffic information acquisition unit 1 and recorded as information. The "fourth obstacle" is an obstacle detected by another vehicle 4.

[0013] The traffic information acquisition unit 1 includes a first camera 11 and a first external communication unit 12. The first camera 11 photographs the road (especially a highway) and transmits image data (a still image photographed at a predetermined time interval or a moving image photographed constantly) to the server 3. Further, the first camera 11 can add attached information including the position and orientation of the first camera 11 to the image data. The traffic information acquisition unit 1 is arranged so as to cover most of the road on the driving route set by the driver, but there are also places where it is not arranged.

[0014] The first external communication unit 12 performs two-way communication with the server 3 or with the host vehicle 2. As the two-way communication, for example, known technologies such as a dedicated short-range communication system (DSRC) and LTE (Long Term Evolution) can be applied. When route information is input from the second external communication unit 203 of the host vehicle 2, the first external communication unit 12 transmits this to the server 3. Also, when object information is input from the server 3, the first external communication unit 12 transmits this toward the second external communication unit 203 of the host vehicle 2.

[0015] The host vehicle 2 is a vehicle that is driven by an internal combustion engine and / or an electric motor as a driving source. The host vehicle 2 also includes a navigation device that sets a travel route from the current location to the destination and guides the driver to the road to travel and the lane in the road, and further includes an automatic driving device (vehicle control unit 211) that automatically drives to the destination. When a camera or the like mounted on the host vehicle 2 detects an obstacle in the travel destination, the automatic driving device can execute control to avoid the obstacle for the host vehicle 2.

[0016] In the first embodiment, the host vehicle 2 includes an HD map 201, a route generation unit 202, a second external communication unit 203, a data analysis unit 204, a risk area information generation unit 205, a learning database 206, a second camera 207, a second object recognition unit 208, a second object information calculation unit 209, a lane planning unit 210, and a vehicle control unit 211.

[0017] The HD map 201 is a high-resolution map that covers the roads displayed by general-purpose maps (e.g., Google Maps (registered trademark)) such as highways and general roads, and has identification information of a plurality of travel lanes constituting the road and information on the area (position in the width direction, position in the traveling direction) occupied by each travel lane.

[0018] The route generation unit 202, for example, when the driver specifies a destination, refers to the HD map 201 and generates route information that connects the current location of the vehicle 2 to the destination. The route information includes information about the roads that the vehicle 2 will travel on from the current location to the destination and information about the lane the vehicle 2 will be traveling in on those roads.

[0019] Here, it is assumed that the road to which the present invention applies has at least two lanes in each direction. In the route information (driving route), for example, if turning left at the next intersection (junction), the left lane is selected preferentially, and if turning right at the next intersection (junction), the right lane is selected preferentially.

[0020] The second external communication unit 203 performs bidirectional communication with the first external communication unit 12 of the traffic information acquisition unit 1. The second external communication unit 203 transmits route information to the first external communication unit 12 of the traffic information acquisition unit 1, and object information is transmitted from the first external communication unit 12 as described later.

[0021] The data analysis unit 204 converts object information transmitted from the traffic information acquisition unit 1 into a format usable by the system within the vehicle 2.

[0022] The second camera 207 is positioned, for example, to have a field of view of the area in front of the vehicle 2, and captures still images or videos of the road ahead (driving lane).

[0023] The second object recognition unit 208 determines the attributes (e.g., passenger car, bus, truck, motorcycle, etc.) and the size (and orientation) of the object (second obstacle) that is stopped on the road (lane).

[0024] The second object information calculation unit 209 generates information about the second obstacle (part of the object information described later), including information about the attributes and size of the object (second obstacle) captured by the camera, information about the road and lane where the second obstacle was detected, and position information of the second obstacle in the direction of travel where the lane was detected, based on the path information.

[0025] The learning database 206 stores information on third-party obstacles that have been detected in the past. This information includes the attributes of the third-party obstacle, its size, information on the road and lane where the third-party obstacle was detected, and the position of the detected third-party obstacle along the direction of travel on the lane.

[0026] The learning database 206 stores risk values ​​(the probability of vehicle 2 encountering an obstacle) corresponding to the attributes of the first obstacle, second obstacle, third obstacle (and the fourth obstacle described later). For example, the risk value for the first obstacle, second obstacle, and fourth obstacle is 1. Also, since the third obstacle is an obstacle that has been detected in the past, there is a possibility that vehicle 2 will not encounter it. Therefore, the risk value for the third obstacle is set as n / N, for example, where N is the number of times vehicle 2 has passed through a road where the third obstacle was detected, and n is the number of times vehicle 2 has detected (encountered) the third obstacle.

[0027] Furthermore, the learning database 206 stores supplementary information. This supplementary information includes, for example, the location of bus stops, the location of entrances and exits to facilities, the location of traffic lights, and other information about locations where vehicles temporarily stop and their attributes.

[0028] When object information is input, the risk area information generation unit 205 refers to the learning database 206 to determine whether there is information on a third obstacle related to the object information, that is, information on past obstacles on the driving route that are outside the acquisition range of the traffic information acquisition unit 1, and if there is, it extracts the information on the third obstacle.

[0029] Furthermore, if there is information about a second obstacle and information about a third obstacle related to the object information, the risk area information generation unit 205 adds this information to the object information.

[0030] The risk area information generation unit 205 then refers to the learning database 206 to extract the risk values ​​for the first obstacle, the second obstacle, and the third obstacle, and associates each risk value with the information of each obstacle in the object information. As a result, risk area information 5 is generated in the object information. Details of risk area information 5 will be described later.

[0031] The lane planning unit 210 selects the driving lane that the vehicle 2 will travel in until it reaches its destination, based on the route information, and also selects the driving lane with the lowest probability (risk value) of the vehicle 2 encountering an obstacle. Furthermore, when the vehicle 2 is entering the next road via an intersection (junction) from the road it is currently traveling on, the lane planning unit 210 prioritizes selecting the leftmost driving lane if turning left at the intersection, and the rightmost driving lane if turning right. The lane planning unit 210 also updates the route information by switching driving lanes based on the risk area information 5 in order to bypass obstacles. The updating of route information will be described later.

[0032] The vehicle control unit 211 performs automatic driving of its own vehicle 2 based on route information updated by the lane planning unit 210, but the automatic driving is canceled, for example, when the driver takes action on the steering wheel.

[0033] Server 3 comprises a first object recognition unit 31, a first object information calculation unit 32, and a traffic information acquisition unit location database 33.

[0034] The first object recognition unit 31 refers to the traffic information acquisition unit location database 33 and analyzes the image data captured by the first camera 11 of the traffic information acquisition unit 1, which is positioned facing the road in the route information, to detect the presence or absence of an object (first obstacle) that is stopped on the road (driving lane).

[0035] The first object recognition unit 31 determines, when the first obstacle is stationary in a driving lane in the route information, the attributes of the first obstacle stationary on the road (lane), the position of the first obstacle in the direction of travel (e.g., the center position), and the size of the first obstacle.

[0036] When the first object information calculation unit 32 receives information on the attributes of the first obstacle, as well as information on the position and size of the first obstacle in the direction of travel, from the first object recognition unit 31, it generates information on the road where the first obstacle was detected, information on the lane in which the first obstacle was detected, and information on the position in the direction of travel where the first obstacle was detected, from the image data.

[0037] The first object information calculation unit 32 then generates object information based on the route information, including information on the time the first obstacle was photographed, information on the attributes of the first obstacle, information on the size of the first obstacle, information on the road where the first obstacle is located, information on the lane in which the first obstacle is located, and information on the position of the first obstacle in the direction of travel within the lane. The object information is generated in real time according to the shooting frequency of the first camera 11 and transmitted to the second external communication unit 203.

[0038] [Simplified control flow] Figure 2 shows a simplified control flow of the driving assistance method of the present invention. A simplified explanation of the driving assistance method of the present invention is as follows.

[0039] In step S101, the route generation unit 202 generates route information based on the driving route set by the driver.

[0040] In step S102, the first object information calculation unit 32 generates object information if there is a first obstacle in the driving lane related to the route information, and the risk area information generation unit 205 adds information about the second obstacle and the third obstacle to the object information if there are a second obstacle and a third obstacle in the driving lane related to the route information.

[0041] In step S103, the risk area information generation unit 205 associates risk values ​​corresponding to obstacles (first obstacle, second obstacle, third obstacle) with the object information and generates risk area information 5 in the object information.

[0042] In step S104, the lane planning unit 210 updates the route information by selecting other driving lanes to bypass the portion of the route information included in the risk area information 5.

[0043] By controlling the vehicle as described above, the driving path can be updated to switch lanes before and after an obstacle before reaching the road where the obstacle is located, enabling proactive vehicle behavior that avoids risks in advance.

[0044] [Control flow of the first embodiment] Figure 3 shows the control flow of the driving support system of the first embodiment via the vehicle 2, traffic information acquisition device, and server 3. Figure 4 shows the control flow detailing the risk area information generation process and the learning DB update process of Figure 3. The driving support system of the first embodiment is executed according to the program installed on the storage media of the traffic information acquisition unit 1, the vehicle 2, and server 3.

[0045] In step S201, the route information generation unit of the vehicle 2 generates route information based on the driving route set by the driver.

[0046] In step S202, the second external communication unit 203 of the vehicle 2 transmits route information to the first external communication unit 12 of the traffic information acquisition unit 1.

[0047] In step S203, the second camera 207 of vehicle 2 monitors (photographs) the road and driving lane in front of vehicle 2.

[0048] In step S204, the second object recognition unit 208 of the vehicle 2 detects the second obstacle if it is stationary in the image data captured by the second camera 207 and identifies its attributes. The second object recognition unit 208 determines that the second obstacle is stationary, for example, if the vehicle speed of the vehicle 2 matches the estimated approach speed of the vehicle 2 to the second obstacle from the image data.

[0049] In step S205, the second object recognition unit 208 estimates the position (e.g., center position) and size of the second obstacle in the direction of travel.

[0050] In step S206, the second object information calculation unit 209 generates information about the second obstacle based on the attributes, position, and size determined by the second object recognition unit 208, including information about the attributes and size of the second obstacle, information about the road and lane on which the second obstacle is located, and information about the position of the second obstacle in the direction of travel on the lane.

[0051] In step S207, the first external communication unit 12 of the traffic information acquisition unit 1 receives route information.

[0052] In step S208, the first camera 11 of the traffic information acquisition unit 1 monitors (photographs) the road and the driving lane.

[0053] In step S209, the first object recognition unit 31 of the server 3 selects image data captured by the first camera 11 of the traffic information acquisition unit 1, which is positioned facing the road in the route information.

[0054] In step S210, the first object recognition unit 31 analyzes the image data to detect the presence or absence of an object (first obstacle) stopped on the road (driving lane), and if the first obstacle is detected on the driving lane in the route information, it determines the attributes of the first obstacle.

[0055] In step S211, the first object recognition unit 31 determines the position and size (and orientation) of the first obstacle in the direction of travel.

[0056] In step S212, the first object information calculation unit 32 generates object information relating to the first obstacle as described above, and transmits the object information to the external communication unit of the vehicle 2 via the first external communication unit 12 of the traffic information acquisition unit 1.

[0057] In step S213, the second external communication unit 203 of the vehicle 2 receives object information transmitted from the traffic information acquisition unit 1. The data analysis unit 204 converts the object information transmitted from the traffic information acquisition unit 1 into a format usable by the system within the vehicle 2.

[0058] In step S214, the risk area information generation unit 205 refers to the learning database 206 to extract information about third obstacles related to object information (path information).

[0059] In step S215 (steps S2151-S2156 in Figure 4), the risk area information generation unit 205 adds the information of the second obstacle and the information of the third obstacle to the object information. The risk area information generation unit 205 then extracts the risk values ​​of the first obstacle, the second obstacle, and the third obstacle from the learning database 206 and associates each risk value with the information of each obstacle in the object information to generate object information that includes the risk area information 5.

[0060] As shown in Figure 4, in step S2151, the risk area information generation unit 205 adds the information of the second obstacle and the information of the third obstacle to the object information which includes the information of the first obstacle. If the second and third obstacles are not detected, there is no change in the content of the object information before and after the addition of the information of the second and third obstacles.

[0061] In step S2152, the risk area information generation unit 205 extracts the estimated stay time of the first obstacle at a predetermined location from the learning database 206 based on the attributes of the first obstacle and the location of the first obstacle. For example, if the attribute of the first obstacle is "bus" and the stopping position of the first obstacle is "bus stop", the estimated stay time of the first obstacle is set to a predetermined time (e.g., 5 minutes). Similarly, if the attribute of the first obstacle is "passenger car" and the stopping position of the first obstacle is "parking lot entrance", the estimated stay time of the first obstacle is set to a predetermined time (e.g., 5 minutes). If it cannot be determined that the first obstacle is temporarily stopped as described above, the estimated stay time of the first obstacle is set to a time longer than the time it takes for the vehicle 2 to reach its destination (e.g., 24 hours).

[0062] In step S2153, the risk area information generation unit 205 calculates the time it will take for the vehicle 2 to encounter the first obstacle based on the location information of the first obstacle, the actual location information of the vehicle 2 (GPS, etc.), and the vehicle speed of the vehicle 2.

[0063] In step S2154, the risk area information generation unit 205 refers to the learning database 206 and associates the object information with the risk values ​​(=1) corresponding to the first obstacle (the first obstacle whose estimated stay time is equal to or greater than the arrival time), the risk values ​​(=0) corresponding to the first obstacle (the first obstacle whose estimated stay time is less than the arrival time), the risk values ​​(=1) corresponding to the second obstacle, and the risk values ​​(=n / N) corresponding to the third obstacle.

[0064] In step S2155, the risk area information generation unit 205 refers to the object information (route information) and determines that it has detected multiple obstacles (first obstacle (estimated stay time ≥ arrival time), second obstacle, third obstacle) lined up in the direction of travel on the same travel lane. If the determination is YES, it proceeds to step S2156; otherwise, it proceeds to step S216 (step S2161).

[0065] In step S2156, the risk area information generation unit 205 generates risk area information 5 in the object information by associating a predetermined risk value with the position information relating to the obstacles from the foremost to the last obstacle on the same lane in the object information.

[0066] As shown in Figure 3, in step S216, the risk area information generation unit 205 updates the learning database 206 based on the object information.

[0067] As shown in Figure 4, in step S2161, the risk area information section 5 determines whether the second obstacle matches the third obstacle. If YES, it proceeds to step S2162; otherwise, it proceeds to step S2163. Here, if the attributes and location of the second obstacle match those of the third obstacle, it is determined that the second obstacle matches the third obstacle. If the attributes or location do not match, it is determined that the second obstacle does not match the third obstacle.

[0068] In step S2162, the risk area information generation unit 205 determines that the vehicle 2 has detected the third obstacle as the second obstacle on the road it is traveling on for the (N+1)th time, updates the risk value of the third obstacle to (n+1) / (N+1), and overwrites the learning database 206.

[0069] In step S2163, the risk area information generation unit 205 determines that the vehicle 2 did not detect the third obstacle recorded in the learning database 206 on the road it travels on for the (N+1)th time, and updates the risk value of the third obstacle to (n) / (N+1) and overwrites the learning database 206.

[0070] In step S2164, the risk area information generation unit 205 determines that the vehicle 2 has detected a new third obstacle and generates information about the new third obstacle (attributes of the third(2) obstacle, the road on which the third(2) obstacle was detected, the lane in which the third(2) obstacle was detected, and the position of the third(2) obstacle in the direction of travel) and stores it in the learning database 206, as well as the risk value of the new third obstacle (n / N=1 / 1) and stores it in the learning database 206.

[0071] As shown in Figure 3, in step S217, the lane planning unit 210 updates the route information by switching the driving lane based on the risk area information 5 in order to bypass each obstacle.

[0072] In step S218, the lane planning unit 210 (or vehicle control unit 211) determines whether the vehicle 2 has reached its destination. If the answer is YES, the automated driving operation ends; otherwise, the system proceeds to step S202.

[0073] [Update route information] Figure 5 shows a case where a driving route is set to turn left from road (R1) and enter road (R2), in which lane (L1) (left lane) is initially selected on road (R1) and lane (L1) (left lane) is initially selected on road (R2), where Figure 5(a) shows the case where there are no first obstacles on road (R1), Figure 5(b) shows the case where there are two first obstacles on road (R1) and one first obstacle on road (R2), and the vehicle 2 detects all first obstacles and performs a detour operation of the driving lane to avoid the first obstacles, and Figure 5(c) shows the case where two first obstacles are detected on road (R1) via the traffic information acquisition unit 1 and one first obstacle is detected on road (R2) via the traffic information acquisition unit 1, and the vehicle 2 performs a detour operation of the driving lane to avoid the first obstacles.

[0074] As shown in Figure 5, consider a case where vehicle 2 travels from its current location to its destination according to a travel route (route information), and vehicle 2 travels on road (R1), turns left at the intersection at the end of road (R1), and enters road (R2). Here, roads (R1) and (R2) have two lanes on each side (L1 and L2), for a total of four lanes, with lane (L1) on the left and lane (L2) on the right.

[0075] In this case, as shown in Figure 5(a), the driving path (route information) selects the driving lane considering a left turn, so lane (L1) is initially selected on road (R1) and lane (L1) is also initially selected on road (R2). Therefore, if there are no first obstacles on road (R1) and road (R2), the vehicle control unit 211 performs automatic driving in the order of lane (L1) on road (R1) and lane (L1) on road (R2) according to the route information.

[0076] On the other hand, consider the case shown in Figure 5(b) where two first obstacles (A) and (B) are located side by side in the direction of travel in lane (L1) of road (R1), and where there is a first obstacle (C) in lane (L1) of road (R2).

[0077] In this case, according to the prior art such as Patent Document 1, when the vehicle 2 detects the first first obstacle (A) in lane (L1) of road (R1), it moves to lane (L2) at a position ahead of the first obstacle (A) in lane (L1), passes alongside the first obstacle (A), and then returns to lane (L1). When the vehicle 2 detects the next first obstacle (B) in lane (L1) of road (R1), it moves to lane (L2) at a position ahead of the first obstacle (B) in lane (L1), passes alongside the first obstacle (B), and then returns to lane (L1). Then, when Vehicle 2 turns left and enters lane (L1) of road (R2), and Vehicle 2 detects the first obstacle (C) in lane (L1) of road (R2), it moves to lane (L2) at a position ahead of the first obstacle (C) in lane (L1), passes by the first obstacle (C), and then returns to lane (L1).

[0078] Thus, in the conventional technology, since vehicle 2 detects the first obstacles (A), (B), and (C) in sequence, it is necessary to perform a detour maneuver in the driving lane each time a first obstacle is detected. In Figure 5(b), the detour maneuver in the driving lane is performed three times.

[0079] Furthermore, when vehicle 2 is traveling ahead of the first obstacle (A), the first obstacle (A) creates a blind spot, making it difficult for vehicle 2 to detect the first obstacle (b). Similarly, when vehicle 2 is traveling on road (R1), it is difficult to detect the first obstacle (C) which is stopped on road (R2).

[0080] On the other hand, in the first embodiment, as shown in Figure 5(c), before the vehicle 2 enters the road (R1) and the road (R2), the traffic information acquisition unit 1 positioned facing the road (R1) detects the first obstacles (A) and (B), and the traffic information acquisition unit 1 positioned facing the road (R2) detects the first obstacle (C). In this embodiment, for example, lane changes are made in a way that minimizes the number of detours (switches) of the driving lane.

[0081] As shown in Figure 5(c), in the first embodiment, a risk area 51 is formed including the first obstacle (A) and the first obstacle (B) detected in the same lane, and the vehicle 2 performs a lane detour to avoid this risk area 51. Furthermore, if it is predicted that the first obstacle (C) will depart before the vehicle 2 encounters the first obstacle (C) which is stopped on the road (R2), that is, if the time it takes for the vehicle 2 to encounter the first obstacle (C) is longer than the estimated time the first obstacle (C) will stay in a predetermined position, the vehicle 2 can set a driving path that it continues to take without recognizing the first obstacle (C) as a risk area 51. Of course, if the first obstacle (C) remains stopped contrary to the prediction, the vehicle 2 can also change lanes to avoid the first obstacle (C). Therefore, in the first embodiment, compared to the conventional technology, the number of lane detours can be reduced to one (or two), enabling preventative vehicle behavior that avoids risks in advance.

[0082] Figure 6 shows examples of object information and route information generated according to the travel route. Figure 6(a) shows object information and route information when there is no first obstacle, as shown in Figure 5(a). Figure 6(b) shows object information and route information when the first obstacle is detected on road (R1) and road (R2) via the traffic information acquisition device, as shown in Figure 5(c). Figure 6(c) shows the state in which risk area information 5 is generated by associating a risk value with the object information. In Figure 5, etc., route information and object information are displayed together.

[0083] Route information (Rn, D1, D2) is data that indicates which lane to select for the road (R1, R2, R3...Rn) and the driving lanes (L1, L2) that make up the road connecting the current location of vehicle 2 to the destination, with a predetermined resolution (e.g., 10 [cm]) in the direction of travel (direction from left edge to right edge in the diagram). Here, for example, (D1, D2) = (1, 0) indicates that lane (L1) is selected as the driving lane for vehicle 2, and (D1, D2) = (0, 1) indicates that lane (L2) is selected as the driving lane for vehicle 2.

[0084] Object information (Rn,Ob1,Ob2) is data that indicates which lane (L1,L2) has a first obstacle for each predetermined resolution (e.g., 10 [cm]) in the direction of travel, along the road (R1, R2, R3...Rn) and the lanes (L1,L2) that make up the road, connecting the current location of vehicle 2 to the destination. Here, for example, (Ob1,Ob2)=(1,0) indicates that there is an obstacle in lane (L1) and no first obstacle in lane (L2), and (Ob1,Ob2)=(0,1) indicates that there is no first obstacle in lane (L1) and there is a first obstacle in lane (L2).

[0085] For example, if the length of the first obstacle (and other obstacles) in the direction of travel is 2 [m] and the resolution of the object information in the direction of travel is 10 [cm], then in the object information (e.g., Ob1), there will be 20 instances of the number "1" indicating the presence of the first obstacle. However, for the sake of simplicity, in Figure 5 and other diagrams, the length of the first obstacle is not considered, and the presence of one first obstacle is represented by a single "1".

[0086] Figure 6(a) shows a state where, as shown in Figure 5(a), a lane (L1) is initially selected on road (R1) and a lane (L1) is initially selected on road (R2), and there are no first obstacles on road (R1) and road (R2). In this case, object information (Ob1,Ob2)=(0,0) on road (R1) and road (R2), and does not interfere with the path information (D1,D2). Therefore, the path information is not updated, and the path information (D1,D2)=(1,0) on road (R1) and road (R2).

[0087] Figure 6(b) shows the case where two first obstacles (A) and (B) are detected in lane (L1) of road (R1), as shown in Figure 5(c), and a first obstacle (C) is detected in lane (L1) of road (R2). In other words, with respect to object information, there are two locations on road (R1) where Ob1=1, and one location on road (R2) where Ob1=1.

[0088] Therefore, when the path information is updated based on the object information, for example, up to the position two positions before the first "Ob=1" on road (R1), (D1,D2)=(1,0), then from the position one position before the first "1" on road (R1) to the position one position after the first "Ob=1" on road (R1), (D1,D2)=(0,1), and then it returns to (D1,D2)=(1,0) at the position two positions after the first "Ob=1" on road (R1). Similar changes occur with respect to the path information at the next "Ob=1" on road (R1) and at the position of the "Ob=1" on road (R2).

[0089] In Figure 6(c), a risk value (Risk) is associated with object information to generate risk area information 5, and the path information is updated based on the risk area information 5. As described above, two first obstacles (A) and (B) are detected in the lane (L1) of the road (R1). Therefore, in the direction of travel on the road (R1), the object information and path information from the first "Ob=1" to the next "Ob=1" are associated with a risk value (=1).

[0090] Furthermore, since the first obstacle (C) is detected on the road (R2), a predetermined risk value is associated with the position on the road (R2) where "Ob=1". However, as described above, if the time it takes for vehicle 2 to reach the first obstacle detected on the road (R2) is longer than the estimated time the first obstacle stays at the predetermined position, a risk value (=0) can be associated with the position on the road (R2) where "Ob=1".

[0091] Therefore, when the route information is updated based on the risk area information 5, for example, up to the position two positions before the first "Ob=1" (A) on road (R1), (D1,D2)=(1,0), then from the position one position before the first "1" on road (R1) to the position one position after the next "Ob=1" (B) on road (R1), (D1,D2)=(0,1), and then it returns to (D1,D2)=(1,0) at the position two positions after the next "Ob=1" (B) on road (R1). Then, on road (R2), (D1,D2)=(1,0).

[0092] Figure 7 shows a scenario where a driving route is set to turn left from road (R1) and enter road (R2), resulting in lane (L1) (left lane) being initially selected on road (R1) and lane (L1) (left lane) being initially selected on road (R2). Figure 7(a) shows a scenario where the learning database 206 has prior information indicating the presence of a third obstacle behind lane (L1) on road (R1), and when the vehicle 2 detects a second obstacle in lane (L1) preceding the third obstacle, it switches lanes to avoid the second and third obstacles. Figure 7(b) shows the state where the information (location) of the second obstacle and the information (location) of the third obstacle are added to the object information. Figure 7(c) shows the state where a risk value is associated with the object information to generate risk area information 5.

[0093] Figure 7(a) shows that the first obstacle is not detected on the road (R2). On the other hand, on the road (R1), vehicle 2 detects the second obstacle (D) while driving behind vehicle (L1), but the third obstacle (E) is in the blind spot of the second obstacle (D), and vehicle 2 is unable to detect the third obstacle (E).

[0094] However, in the first embodiment, a risk area 51 is formed in the range including the second obstacle (D) and the third obstacle (E), and the vehicle 2 can perform a lane detour to avoid the risk area 51.

[0095] As shown in Figure 7(b), the first part of the road (R1) object information that is "Ob=1" (D) is detected by vehicle 2 as a third obstacle. The next part of the road (R1) object information that is "Ob=1" (E) is extracted from the learning database 206 as a second obstacle. Therefore, when the path information is updated based on this object information, for example, up to the position two positions before the first "Ob=1" (D) on road (R1), (D1,D2)=(1,0), then from the position one position before the first "Ob=1" (D) on road (R1) to the position one position after the first "Ob=1" (D) on road (R1) (D1,D2)=(0,1), and then it returns to (D1,D2)=(1,0) at the position two positions after the first "Ob=1" (D) on road (R1). Then, a similar change occurs around "Ob=1" (E) on the road (R1), and thereafter (D1,D2)=(1,0).

[0096] In Figure 7(c), risk area information 5 is generated by associating a risk value (Risk) with object information, and route information is updated based on risk area information 5. As described above, a second obstacle (D) is detected in the lane (L1) of the road (R1), and information on the third obstacle (E) is stored in advance. Therefore, a risk value (=1) is associated with the first "Ob=1" (D) position in the object information, and a risk value (=n / N) is associated with the "Ob=1" (E) position. Then, risk area information 5 is generated by associating risk values ​​(=n / N, or 1) from the position after the first "Ob=1" (D) position in the object information to the position just before the next "Ob=1" (E) position.

[0097] Therefore, when the route information is updated based on the risk area information 5, for example, (D1,D2)=(1,0) up to the position two positions before the first "Ob=1" (D) on road (R1), then (D1,D2)=(0,1) from the position one position before the first "Ob=1" (D) on road (R1) to the position one position after "Ob=1" (E) on road (R1), and then (D1,D2)=(1,0) returns to the position two positions after "Ob=1" (E) on road (R1), and thereafter (D1,D2)=(1,0) is maintained on road (R2). And (D1,D2)=(1,0) is also maintained on road (R2).

[0098] In the first embodiment, if any two of the first, second, or third obstacles (the fourth obstacle described later) are detected to be aligned in the direction of travel on the same lane, risk area information 5 is generated within the range encompassing the two obstacles, and the vehicle 2 performs a detour maneuver in the driving lane to avoid the risk area 51.

[0099] [Scope of route information update] Figure 8 shows the range of route information updates when vehicle 2 travels along the road (R1), road (R2), and road (R3) in that order. Figure 8(a) shows the case where object information (risk area constant method) is generated up to the entrance of road (R2) when vehicle 2 is traveling on road (R1), and Figure 8(b) shows the case where object information (risk area information 5) is generated up to the entrance of road (R3) just before vehicle 2 enters road (R2).

[0100] As an example other than Figure 8, it is also preferable to repeatedly perform the following control: when the vehicle 2 reaches a position where it is a predetermined distance (e.g., 2 km) before the end of the road (R1), object information (risk area information 5) is generated for the road (R2), and when it reaches a position where it is a predetermined distance before the end of the road (R2), object information (risk area information 5) is generated for the road (R3).

[0101] In the first embodiment, it is possible to simultaneously update all route information from the current location of the vehicle 2 to the destination. However, as described above, it is also possible to obtain object information related to the next road after the currently traveling road and update the route information for the next road. This reduces the overall communication volume of the system and stabilizes the entire system.

[0102] [Second Embodiment] Figure 9 is a block diagram of the driving support system of the second embodiment. Figure 10 shows the control flow of the driving support system of the second embodiment via the vehicle 2, traffic information acquisition unit 1, and server 3.

[0103] In the driving support system of the second embodiment, the traffic information acquisition unit 1 is composed only of the first camera 11, and the first external communication unit 12 of the first embodiment is incorporated into the server 3. Therefore, the vehicle 2 communicates directly with the server 3. The control flow is the same as in the first embodiment, except that steps S207 and S212 are executed in the server 3.

[0104] [Third Embodiment] Figure 11 is a block diagram of the driving support system of the third embodiment. The driving support system of the third embodiment differs from the second embodiment in that the learning database 206 and the risk area information generation unit 205 are incorporated into the server 3, and furthermore, the server 3 can communicate not only with its own vehicle 2 but also with other vehicles 4.

[0105] In the third embodiment, the vehicle 2 transmits route information it generates to other vehicles 4 via the server 3, and when other vehicles 4 detect an obstacle (fourth obstacle) in the lane related to the route information, the system transmits information about the fourth obstacle to the server 3.

[0106] The third camera 41, third object recognition unit 42, third object information calculation unit 43, and third external communication unit 44 mounted on the other vehicle 4 are the same as those mounted on the vehicle 2. Although not shown in the diagram, the other vehicle 4 also has an HD map 201, a route generation unit 202, a lane planning unit 210, and a vehicle control unit 211, and, like the vehicle 2, generates route information and object information (risk area information 5) to enable autonomous driving.

[0107] Figure 12 shows the control flow of the driving support system of the third embodiment, via the user vehicle 2, traffic information acquisition unit 1, server 3, and other vehicles 4. Figure 13 shows the control flow detailing the risk area information 5 generation process of the third embodiment. With respect to the control flow of the third embodiment, flows common to the second embodiment (first embodiment) are denoted by the same reference numerals, and explanations are omitted unless necessary.

[0108] Steps S201 and S203 to S206 are the same as in the second embodiment.

[0109] In place of step S202 of the second embodiment, in step S202A of the third embodiment, the second external communication unit 203 of the vehicle 2 transmits route information and second obstacle information to the first external communication unit 12 of the server 3.

[0110] In step S207A of the third embodiment, instead of step S207 of the second embodiment, the first external communication unit 12 receives route information and second obstacle information. The second obstacle information is used in step S215A, which will be described later.

[0111] In step S207B, the first external communication unit 12 transmits route information to the other vehicle 4.

[0112] Steps S208 to S211 are the same as in the second embodiment.

[0113] In step S212A of the third embodiment, instead of step S212 of the second embodiment, the first object information calculation unit 32 generates object information relating to the first obstacle.

[0114] Following step S207A, in step S214, the risk area information generation unit 205 refers to the learning database 206 to extract information about third obstacles related to object information (path information).

[0115] In step S301, the third camera 41 of the other vehicle 4 monitors (photographs) the road and driving lane in front of the other vehicle 4.

[0116] In step S302, the third external communication unit 44 of the other vehicle 4 receives route information from the first external communication unit 12 of the server 3 (step S207B).

[0117] In step S303, the third object recognition unit 42 of the other vehicle 4 detects the fourth obstacle and identifies its attributes if the other vehicle 4 is traveling on a road related to the route information and the fourth obstacle exists in the image data captured by the third camera 41 of the other vehicle 4. If the other vehicle 4 is not traveling on a road related to the route information, steps S303 and subsequent steps S304 to S307 are not performed.

[0118] In step S304, the third object recognition unit 42 estimates the attributes and size of the fourth obstacle.

[0119] In step S205, the third object information calculation unit 43 generates information about the fourth obstacle based on the attributes, position, and size determined by the third object recognition unit 42, including information about the attributes and size of the fourth obstacle, information about the road and lane on which the fourth obstacle is located, and information about the position of the fourth obstacle in the direction of travel on the lane, based on the route information.

[0120] In step S306, the third external communication unit 44 transmits information about the fourth obstacle to the first external communication unit 12 of the server 3.

[0121] In step S307, the first external communication unit 12 receives information about the fourth obstacle.

[0122] In step S215A, the risk area information generation unit 205 adds the information of the second obstacle, the third obstacle, and the fourth obstacle to the object information (information of the first obstacle), and generates object information that includes the risk area information 5 by referring to the risk value of each obstacle from the learning database 206.

[0123] Step S215A is the same as step S215 in the second (first) embodiment, Step S216 is the same as in the second (first) embodiment.

[0124] In step S216A, the first external communication unit 12 transmits object information including the risk area information 5 to the second external communication unit 203 of the vehicle 2.

[0125] In step S216B, the second external communication unit 203 receives object information including risk area information 5 from the first external communication unit 12.

[0126] Steps S217 and S218 are the same as in the second (first) embodiment, but if the determination in step S218 is NO, the process proceeds to step S202A.

[0127] Figure 14 shows a case where a driving route is set to turn left from road (R1) and enter road (R2), in which lane (L1) (left lane) is initially selected on road (R1) and lane (L1) (left lane) is initially selected on road (R2), and Figure 14(a) shows a case where another vehicle 4 detects a fourth obstacle in the latter part of lane (L1) on road (R1), and the own vehicle 2 detects a second obstacle in lane (L1) prior to the fourth obstacle, and the vehicle switches lanes to avoid the second and fourth obstacles, Figure 14(b) shows object information generated by associating the information (location) of the second obstacle and the information (location) of the fourth obstacle with the route information, and Figure 14(c) shows the state in which the risk area information generation unit 205 generates risk area information 5 by associating risk values ​​based on the object information and the learning database 206.

[0128] Figure 14(a) shows a scenario similar to Figure 7(a), where the second obstacle in Figure 7(a) becomes the fourth obstacle (F) in Figure 14(a), and the fourth obstacle (F) is detected by another vehicle 4 traveling ahead of the vehicle 2. In Figure 14(a), the other vehicle 4 is traveling in lane (L1), but it could also be traveling in lane (L2). In either case, the fourth obstacle (F) cannot be detected from the perspective of the vehicle 2 because the second obstacle (D) is in a blind spot. However, in the third embodiment, risk area information 5 can be generated to include the second obstacle (D) detected by the vehicle 2 and the fourth obstacle (F) detected by the other vehicle 4, and the route information can be updated so that the vehicle 2 bypasses the risk area 51.

[0129] As shown in Figure 14(b), the first part of the road (R1) object information that reads "Ob=1" (D) is detected by vehicle 2 as a third obstacle. The next part of the road (R1) object information that reads "Ob=1" (F) is detected by other vehicle 4 as a fourth obstacle. Therefore, updating the route information based on this object information results in route information similar to that in Figure 7(b).

[0130] In Figure 14(c), risk area information 5 is generated by associating a risk value (Risk) with object information, and route information is updated based on risk area information 5. As described above, the second obstacle (D) is detected before the lane (L1) of the road (R1), and the fourth obstacle (F) is detected after it. Therefore, the risk value (=1) of the second obstacle is associated with the position where the first "Ob=1" (D) occurs in the object information, and the risk value (=1) of the fourth obstacle is associated with the position where the next "Ob=1" (F) occurs. Then, risk area information 5 is generated by associating the risk value (=1) from the position after the first "Ob=1" (D) position in the object information to the position just before the next "Ob=1" (F) position. The route information updated based on this risk area information 5 is the same as in Figure 7(c).

[0131] [Effects of this embodiment] According to the driving support method of this embodiment, route information is generated that represents the driving route from the current location of the vehicle 2 to the destination, and includes information on multiple roads connecting the current location to the destination in series, and information on one driving lane selected based on the driving route from among multiple driving lanes arranged in parallel on each road. When a first obstacle is detected in an image of the road taken by a traffic information acquisition means (traffic information acquisition unit 1) positioned facing the road, object information is generated that includes information on the attributes of the first obstacle, information on the driving lane in which the first obstacle was detected, and position information in the direction of travel in which the first obstacle was detected. Risk area information 5 is generated by associating a predetermined risk value corresponding to the attributes of the first obstacle with the driving lane information and position information, and the route information is updated by selecting another driving lane in the route information to bypass the part included in the risk area information 5.

[0132] By using the method described above, the driving route can be updated to switch lanes before and after an obstacle before reaching the road where the obstacle is located, enabling proactive vehicle behavior that avoids risks in advance.

[0133] In this embodiment, when two first obstacles are detected side by side in the same lane along the direction of travel, a predetermined risk value is associated with the positional information from the position of the preceding first obstacle to the position of the succeeding first obstacle in the direction of travel in the object information, thereby generating risk area information 5. This makes it possible to avoid both first obstacles at once, reducing the number of avoidance maneuvers and alleviating the psychological burden on the driver.

[0134] In this embodiment, when the vehicle 2 detects a second obstacle while traveling at a position ahead of the detection range of the first obstacle of the traffic information acquisition means (traffic information acquisition unit 1), information about the second obstacle, including information about the lane in which the second obstacle was detected and the position information in the direction of travel where the second obstacle was detected, is added to the object information. Furthermore, when the second obstacle and the first obstacle are detected in the same lane, a predetermined risk value is associated with the position information from the position of the second obstacle to the position of the first obstacle in the object information to generate risk area information 5. This makes it possible to avoid both the second and first obstacles together, reducing the number of avoidance maneuvers and alleviating the psychological burden on the driver.

[0135] In this embodiment, information on a third obstacle detected on the driving path at a position outside the detection range of the first obstacle of the traffic information acquisition means (traffic information acquisition unit 1), and when the vehicle 2 detects a second obstacle while traveling at a position ahead of the detection range of the first obstacle of the traffic information acquisition means (traffic information acquisition unit 1), information on the second obstacle, including information on the driving lane where the second obstacle was detected and position information in the direction of travel where the second obstacle was detected, and information on the third obstacle are added to the object information. Furthermore, when the second and third obstacles are detected in the same driving lane, a predetermined risk value is associated with the position information from the position of the second obstacle to the position of the third obstacle in the object information to generate risk area information 5. This makes it possible to avoid both the second and third obstacles together, reducing the number of avoidance maneuvers and alleviating the psychological burden on the driver.

[0136] In this embodiment, when another vehicle 4, different from the own vehicle 2, detects a fourth obstacle at a position on the driving path that is ahead of the detection range of the first obstacle of the traffic information acquisition means (traffic information acquisition unit 1), and the own vehicle 2 detects a second obstacle before the fourth obstacle while outside the detection range, information on the second obstacle, including information on the driving lane where the second obstacle was detected and the position information in the direction of travel where the second obstacle was detected, and information on the fourth obstacle, including information on the driving lane where the fourth obstacle was detected and the position information in the direction of travel where the fourth obstacle was detected, are added to the object information. Furthermore, when the second and fourth obstacles are detected in the same driving lane, a predetermined risk value is associated with the position information from the position of the second obstacle to the position of the fourth obstacle in the object information to generate risk area information 5. This makes it possible to avoid both the second and fourth obstacles together, reducing the number of avoidance maneuvers and alleviating the psychological burden on the driver.

[0137] In this embodiment, when the fourth obstacle and the first obstacle are detected in the same driving lane, a predetermined risk value is associated with the positional information from the position of the fourth obstacle to the position of the first obstacle in the object information to generate risk area information 5. This makes it possible to avoid the second obstacle, the fourth obstacle, and the first obstacle all at once, reducing the number of avoidance maneuvers and alleviating the psychological burden on the driver.

[0138] In this embodiment, risk area information 5 is generated for the road that the vehicle 2 will enter after the road it is currently traveling on. This reduces the overall communication volume of the system and stabilizes the entire system.

[0139] In this embodiment, the estimated dwell time of the first obstacle at a predetermined location is set based on the attributes and location of the first obstacle, and among the predetermined risk values ​​associated with the information of the first obstacle, the predetermined risk value associated with the information of the first obstacle where the time until the vehicle 2 encounters the first obstacle is set to zero if the time until the encounter is longer than the estimated dwell time. This makes it possible to avoid unnecessary evasive maneuvers by excluding first obstacles that are unlikely to be encountered from the risk area information 5.

[0140] In this embodiment, the vehicle 2 is driven automatically according to the driving path set by the above-described driving assistance method. This allows the driving path to be updated so that the vehicle switches driving lanes before and after an obstacle before reaching a road where an obstacle exists, thereby realizing automated driving with preventative vehicle behavior that avoids risks in advance.

[0141] According to the driving support program of this embodiment, the program performs the following steps: generating route information that represents the driving route from the current location of the vehicle 2 to the destination, and includes information on multiple roads connecting the current location to the destination in series, and information on one driving lane selected based on the driving route from among multiple driving lanes arranged in parallel on each road; generating object information that includes information on the attributes of the first obstacle, information on the driving lane in which the first obstacle was detected, and position information in the direction of travel in which the first obstacle was detected, when a first obstacle is detected in an image of the road taken by a traffic information acquisition means (traffic information acquisition unit 1) positioned facing the road; generating risk area information 5 by associating a predetermined risk value corresponding to the attributes of the first obstacle with the driving lane information and position information; and updating the route information by selecting another driving lane in the route information to bypass the part included in the risk area information 5. With the above configuration, the driving route can be updated to switch driving lanes before and after the obstacle to bypass it before reaching the road where the obstacle exists, resulting in a program that realizes preventive vehicle behavior that avoids risks in advance.

[0142] According to this embodiment of the driving support system, the system includes: a route information generation means (route generation unit 202) that generates route information representing the driving route of the vehicle 2 from its current location to its destination, including information on multiple roads connecting the current location to the destination in series, and information on one driving lane selected based on the driving route from among multiple driving lanes arranged in parallel on each road; a traffic information acquisition means (traffic information acquisition unit 1) positioned facing the road and capturing images of the road; an object information generation means (first object information calculation unit 32) that, when a first obstacle is detected in the image, generates object information including information on the attributes of the first obstacle, information on the driving lane where the first obstacle was detected, and position information in the direction of travel where the first obstacle was detected; a risk area information generation means (risk area information generation unit 205) that generates risk area information 5 by associating a predetermined risk value corresponding to the attributes of the first obstacle with the driving lane information and position information; and a route information update means (vehicle control unit 211) that updates the route information by selecting another driving lane to bypass the portion included in the risk area information 5 in the route information. According to the above configuration, the driving route can be updated to switch lanes before and after an obstacle before reaching the road where the obstacle is located, thereby realizing a system that enables proactive vehicle behavior to avoid risks in advance.

[0143] Although embodiments of the present invention have been described above, these embodiments represent only a part of the application examples of the present invention, and are not intended to limit the technical scope of the present invention to the specific configurations of the above embodiments. Furthermore, the above embodiments can be combined as appropriate. [Explanation of symbols]

[0144] 1 Traffic Information Acquisition Unit, 5 Risk Area Information, 32 First Object Information Generation Unit, 202 Route Generation Unit, 205 Risk Area Information Generation Unit, 210 Lane Planning Unit

Claims

1. The system generates route information that represents the driving route from the vehicle's current location to the destination, and includes information on multiple roads connecting the current location to the destination in series, as well as information on one of the multiple driving lanes arranged in parallel on each road, selected based on the driving route. When a first obstacle is detected in an image of the road taken by a traffic information acquisition means positioned facing the road, object information is generated that includes information on the attributes of the first obstacle, information on the lane in which the first obstacle was detected, and position information in the direction of travel where the first obstacle was detected. Risk area information is generated by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the driving lane and the position information. The route information is updated by selecting another lane to bypass the portion included in the risk area information in the aforementioned route information. A driving support method that generates risk area information by associating a predetermined risk value with the position information in the object information, from the position of the preceding first obstacle to the position of the succeeding first obstacle in the direction of travel, when two of the first obstacles are detected side by side in the same driving lane along the direction of travel.

2. To generate route information that represents the driving route of the vehicle from its current location to its destination, and includes information on multiple roads connecting the current location to the destination in series, and information on one of the driving lanes selected from among multiple driving lanes arranged in parallel on each road based on the driving route, When a first obstacle is detected in an image of the road taken by a traffic information acquisition means positioned facing the road, object information is generated that includes information on the attributes of the first obstacle, information on the lane in which the first obstacle was detected, and position information in the direction of travel where the first obstacle was detected. Risk area information is generated by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the driving lane and the position information. The route information is updated by selecting another lane to bypass the portion included in the risk area information in the aforementioned route information. A driving support method in which, when the vehicle is traveling at a position ahead of the detection range of the first obstacle of the traffic information acquisition means and a second obstacle is detected, the information of the second obstacle, including information of the driving lane in which the second obstacle was detected and position information in the direction of travel in which the second obstacle was detected, is added to the object information, and when the second obstacle and the first obstacle are detected in the same driving lane, a predetermined risk value is associated with the position information from the position of the second obstacle to the position of the first obstacle in the object information to generate the risk area information.

3. To generate route information that represents the driving route of the vehicle from its current location to its destination, and includes information on multiple roads connecting the current location to the destination in series, and information on one of the driving lanes selected from among multiple driving lanes arranged in parallel on each road based on the driving route, When a first obstacle is detected in an image of the road taken by a traffic information acquisition means positioned facing the road, object information is generated that includes information on the attributes of the first obstacle, information on the lane in which the first obstacle was detected, and position information in the direction of travel where the first obstacle was detected. Risk area information is generated by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the driving lane and the position information. The route information is updated by selecting another lane to bypass the portion included in the risk area information in the aforementioned route information. A driving support method that has information on a third obstacle detected on the aforementioned driving path at a position outside the detection range of the first obstacle of the traffic information acquisition means, and when the vehicle is driving at a position ahead of the detection range of the first obstacle of the traffic information acquisition means and a second obstacle is detected, the method adds information on the second obstacle, which includes information on the driving lane where the second obstacle was detected and position information in the direction of travel where the second obstacle was detected, and the information on the third obstacle to the object information, and when the second obstacle and the third obstacle are detected in the same driving lane, the method generates risk area information by associating a predetermined risk value with the position information from the position of the second obstacle to the position of the third obstacle in the object information.

4. To generate route information that represents the driving route of the vehicle from its current location to its destination, and includes information on multiple roads connecting the current location to the destination in series, and information on one of the driving lanes selected from among multiple driving lanes arranged in parallel on each road based on the driving route, When a first obstacle is detected in an image of the road taken by a traffic information acquisition means positioned facing the road, object information is generated that includes information on the attributes of the first obstacle, information on the lane in which the first obstacle was detected, and position information in the direction of travel where the first obstacle was detected. Risk area information is generated by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the driving lane and the position information. The route information is updated by selecting another lane to bypass the portion included in the risk area information in the aforementioned route information. A driving support method that generates risk area information by associating a predetermined risk value with the position information from the position of the second obstacle to the position of the fourth obstacle in the object information when the second obstacle and the fourth obstacle are detected in the same driving lane.

5. The driving assistance method according to claim 4, wherein when the fourth obstacle and the first obstacle are detected in the same driving lane, a predetermined risk value is associated with the position information from the position of the fourth obstacle to the position of the first obstacle in the object information to generate the risk area information.

6. A driving assistance method according to any one of claims 1 to 5, which generates the risk area information with respect to the road that the vehicle will enter after the road on which it is currently traveling.

7. To generate route information that represents the driving route of the vehicle from its current location to its destination, and includes information on multiple roads connecting the current location to the destination in series, and information on one of the driving lanes selected from among multiple driving lanes arranged in parallel on each road based on the driving route, When a first obstacle is detected in an image of the road taken by a traffic information acquisition means positioned facing the road, object information is generated that includes information on the attributes of the first obstacle, information on the lane in which the first obstacle was detected, and position information in the direction of travel where the first obstacle was detected. Risk area information is generated by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the driving lane and the position information. The route information is updated by selecting another lane to bypass the portion included in the risk area information in the aforementioned route information. A driving support method that sets an estimated stay time for the first obstacle at a predetermined location based on the attributes and location of the first obstacle, and sets to zero a predetermined risk value associated with the information of the first obstacle such that the time until the vehicle encounters the first obstacle is longer than the estimated stay time.

8. A driving support method that performs automatic driving of the vehicle according to the driving path set by the driving support method described in any one of claims 1 to 7.

9. A process for generating route information that includes a representation of the vehicle's current location to its destination, information on multiple roads connecting the current location to the destination in series, and information on one of the multiple parallel driving lanes within each road selected based on the aforementioned driving route. When a first obstacle is detected in an image of the road taken by a traffic information acquisition means positioned facing the road, a step is made to generate object information including information on the attributes of the first obstacle, information on the lane in which the first obstacle was detected, and position information in the direction of travel where the first obstacle was detected. A step of generating risk area information by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the driving lane and the position information, The steps include updating the route information by selecting another lane to bypass the portion included in the risk area information in the route information, A driving support program that, when two of the first obstacles are detected side by side in the same driving lane along the direction of travel, generates risk area information by associating a predetermined risk value with the position information in the object information, from the position of the preceding first obstacle to the position of the succeeding first obstacle in the direction of travel.

10. Route information generation means that generates route information including the driving route of the vehicle from its current location to its destination, information on multiple roads connecting the current location to the destination in series, and information on one of the driving lanes selected from multiple driving lanes arranged in parallel on each road based on the driving route, A means for acquiring traffic information, which is positioned facing the road and takes images of the road, When a first obstacle is detected in the aforementioned image, the following information is obtained: information on the attributes of the first obstacle, information on the lane in which the first obstacle was detected, and information on the direction of travel in which the first obstacle was detected. Object information generation means that generates object information including location information, Risk area information generation means generates risk area information by associating a predetermined risk value corresponding to the attributes of the first obstacle with the information of the driving lane and the position information, The route information update means includes updating the route information by selecting other driving lanes to bypass the portion included in the risk area information in the route information, The risk area information generation means is a driving support system that generates risk area information by associating a predetermined risk value with the position information in the object information, from the position of the preceding first obstacle to the position of the succeeding first obstacle in the direction of travel, when two first obstacles are detected side by side in the same driving lane along the direction of travel.

Images