Ontology update device, vehicle, and ontology update method

Abstract

An ontology update device according to an embodiment of the present disclosure comprises a control unit that is capable of updating an ontology. The control unit can execute the following two steps. (1) The degree of similarity between a surrounding situation for a first vehicle and a dangerous situation that is described in an ontology is determined, and, when it is determined as a result that the degree of similarity is low, it is determined that the surrounding situation is not described in the ontology. (2) On the basis of data obtained from a device that is installed in the first vehicle when the surrounding situation is acquired, a determination is made as to whether a dangerous event is occurring around the first vehicle, and, when it is determined as a result that a dangerous event is occurring around the first vehicle, the surrounding situation and the dangerous event are associated with one another and added to the ontology, thereby updating the ontology.

Description

Ontology update device, vehicle, and ontology update method

[0001] The present disclosure relates to an ontology update device, a vehicle, and an ontology update method.

[0002] There are known technologies that support vehicle driving while taking into consideration dangers that exist around the vehicle. For example, Patent Literature 1 discloses a technology that issues a warning to drivers who do not comply with traffic rules described in an ontology. Also, Patent Literature 2 discloses a technology that predicts whether a vehicle will suddenly jump out in front of the driver based on traffic rules and inference rules described in an ontology.

[0003] Patent No. 5932984 Patent No. 6978313

[0004] An ontology update device according to a first aspect of the present disclosure includes a control unit capable of updating an ontology. The control unit is capable of performing the following two operations: (A1) determining the similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology, and if the similarity is determined to be low as a result, determining that the surrounding situation is not described in the ontology; and (A2) determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired, and if the similarity is determined to be low as a result, updating the ontology by linking the surrounding situation and the dangerous event and adding them to the ontology.

[0005] A vehicle according to a second aspect of the present disclosure includes a storage unit storing an ontology and a control unit capable of updating the ontology. The control unit is capable of performing the following two operations: (B1) determining the similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology, and if the similarity is determined to be low as a result, determining that the surrounding situation is not described in the ontology; and (B2) determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired, and if the similarity is determined to be low as a result, updating the ontology by linking the surrounding situation and the dangerous event and adding them to the ontology.

[0006] A driving assistance method according to a third aspect of the present disclosure includes the following (C1) and (C2): (C1) determining a similarity between a surrounding situation of a first vehicle and a dangerous situation described in an ontology, and if it is determined that the similarity is low as a result, determining that the surrounding situation is not described in the ontology; (C2) determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired, and if it is determined that there is a dangerous event around the first vehicle as a result, updating the ontology by linking the surrounding situation and the dangerous event and adding them to the ontology.

[0007] The accompanying drawings are included to provide a further understanding of the disclosure, and are incorporated in and constitute a part of this specification. The drawings illustrate one embodiment and, together with the description, serve to explain the principles of the disclosure.

[0008] FIG. 1 is a diagram illustrating an example of a schematic configuration of a cruise control system according to an embodiment of the present disclosure. FIG. 2 is a diagram illustrating an example of inference rules described in the ontology DB of FIG. 1. FIG. 3 is a diagram illustrating an example of traffic rules described in the ontology DB of FIG. 1. FIG. 4 is a diagram illustrating an example of road concepts described in the ontology DB of FIG. 1. FIG. 5 is a diagram illustrating an example of traffic conditions and scenarios for dangerous situation A. FIG. 6 is a diagram illustrating a state in which the AEB is activated under the traffic conditions of FIG. 5. FIG. 7 is a diagram illustrating a state in which the ABS is activated under the traffic conditions of FIG. 5. FIG. 8 is a diagram illustrating a state in which the AES is activated under the traffic conditions of FIG. 5. FIG. 9 is a diagram illustrating a state in which a VDC warning is activated under the traffic conditions of FIG. 5. FIG. 10 is a diagram illustrating a state in which an airbag is activated under the traffic conditions of FIG. 5. FIG. 11 is a diagram illustrating an example of a driving assistance procedure for the traffic conditions of FIG. 5. FIG. 12 is a diagram illustrating an example of an ontology update procedure. FIG. 13 is a diagram illustrating a modified example of the ontology update procedure. FIG. 14 is a diagram showing a modified example of the ontology update procedure.

[0009] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings.

[0010] Some exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. Note that the following description illustrates one specific example of the present disclosure and should not be construed as limiting the present disclosure. For example, each element, including numerical values, shapes, materials, parts, the position of each part, and the connection method of each part, is merely an example and should not be construed as limiting the present disclosure. Furthermore, in the following exemplary embodiments, components not described in independent claims based on the highest concept of the present disclosure are optional and may be provided as needed. The drawings are schematic and are not intended to be drawn to scale. Throughout this specification and the drawings, components having substantially the same function and configuration are designated by the same reference numerals, and redundant description will be omitted. Furthermore, components not directly related to one embodiment of the present disclosure are not shown in the drawings.

[0011] <1. Background> In recent years, development of automatic driving control technologies for vehicles such as automobiles, which allow the vehicle to travel automatically without the need for driver operation, has been progressing. Furthermore, various driving assistance devices that utilize this type of automatic driving control technology to perform various controls to assist the driver in driving operations have been proposed and are becoming generally put into practical use. Technologies related to such driving assistance devices are disclosed, for example, in Patent Documents 1 and 2.

[0012] Patent Document 1 discloses a technology for issuing a warning to drivers who do not comply with traffic rules described in an ontology. Patent Document 2 discloses a technology for predicting a vehicle's sudden jump in front of a vehicle based on traffic rules and inference rules described in an ontology. However, the inventions described in Patent Documents 1 and 2 have a problem in that inference does not hold for rules not described in the ontology.

[0013] Therefore, after extensive research, the inventors of the present invention have come up with a technology that enables inference to be established even for inference rules that are not described in the ontology. The following describes in detail a cruise control system that realizes this technology.

[0014] 2. Embodiments> [Configuration Example] Fig. 1 illustrates a schematic configuration example of a cruise control system 1 according to an embodiment of the present disclosure. As shown in Fig. 1, the cruise control system 1 includes cruise control devices 10 mounted on a plurality of vehicles, and a control device 200 provided in a network environment NW to which the plurality of cruise control devices 10 are connected via wireless communication. The cruise control device 10 corresponds to a specific example of an "ontology update device" according to an embodiment of the present disclosure.

[0015] The control device 200 sequentially integrates and updates the road map information transmitted from the cruise control devices 10 of the vehicles, and transmits the updated road map information to the vehicles. The control device 200 includes, for example, a road map information integration ECU 201 and a transceiver 202.

[0016] The road map boundary information integration ECU 201 integrates road map information collected from multiple vehicles via the transceiver 202, and sequentially updates road map information surrounding the vehicle on the road. The road map information may be, for example, a dynamic map, and includes static information and quasi-static information that mainly constitute road information, and quasi-dynamic information and dynamic information that mainly constitute traffic information.

[0017] The static information that makes up road information is composed of information that requires updates within one month, such as roads, structures on roads, structures around roads, lane information, road surface information, and permanent regulation information. "Roads" include, for example, road locations and shapes, intersections, and road attributes (e.g., national roads, prefectural roads, city roads, private roads, priority roads, non-priority roads, general roads, and expressways). "Structures on roads" include, for example, traffic signs, traffic lights, convex mirrors, and pedestrian bridges. "Structures around roads" include, for example, various buildings and parks.

[0018] The quasi-static information that constitutes the road information is made up of information that needs to be updated every hour, such as traffic regulation information due to road construction or events, wide-area weather information, and congestion forecasts.

[0019] The semi-dynamic information that constitutes traffic information is composed of information that requires updating within one minute, such as the actual traffic congestion situation at the time of observation, driving restrictions, temporary driving obstructions such as fallen objects and obstacles, actual accident conditions, and narrow-area weather information.

[0020] The dynamic information that makes up the traffic information is made up of information that needs to be updated every second, such as information sent and exchanged between mobile bodies, information on currently displayed traffic signals, information on pedestrians and bicycles at intersections, information on vehicles traveling on roads, etc. Such road map information is maintained and updated periodically until the next information is received from each vehicle, and the updated road map information is transmitted to each vehicle as appropriate via the transceiver 202.

[0021] The cruise control device 10 has a driving environment recognition unit 11 and a locator unit 12 as units for recognizing the driving environment around the vehicle. The cruise control device 10 also has a driving control unit (hereinafter referred to as "driving_ECU") 21, an engine control unit (hereinafter referred to as "E / G_ECU") 22, a power steering control unit (hereinafter referred to as "PS_ECU") 23, a brake control unit (hereinafter referred to as "BK_ECU") 24, and an ontology control unit (hereinafter referred to as "ontology_ECU") 25. These control units 21 to 25 are connected to the driving environment recognition unit 11 and the locator unit 12 via an in-vehicle communication line such as a Controller Area Network (CAN).

[0022] The travel_ECU 21 controls the vehicle according to, for example, a driving mode. Examples of the driving modes include a manual driving mode and a driving control mode. The manual driving mode requires the driver to maintain steering, and the host vehicle is driven by the driver's driving operations, such as steering, accelerating, and braking. The driving control mode supports the driver in driving operations to increase the safety of pedestrians and other vehicles around the vehicle (host vehicle). In the driving control mode, for example, when the vehicle (host vehicle) approaches an intersection, the travel_ECU 21 predicts the behavior of a traveling or stopped vehicle (hereinafter referred to as a "target vehicle") on a road intersecting the intersection. If the prediction indicates that the target vehicle is likely to enter the intersection, the travel_ECU 21 can, for example, alert or warn the driver, and even perform risk avoidance control such as braking. Detailed processing in the driving control mode will be described later.

[0023] A throttle actuator 26 is connected to the output side of the E / G_ECU 22. This throttle actuator 26 opens and closes a throttle valve of an electronically controlled throttle provided in a throttle body of the engine. The E / G_ECU 22 controls the operation of the throttle actuator 26 by outputting a drive signal to the throttle actuator 26. The throttle actuator 26 opens and closes the throttle valve based on the drive signal from the E / G_ECU 22 to adjust the intake air flow rate, thereby generating a desired engine output.

[0024] An electric power steering motor 27 is connected to the output side of the PS_ECU 23. This electric power steering motor 27 applies steering torque to the steering mechanism by the rotational force of the motor. The PS_ECU 23 controls the operation of the electric power steering motor 27 by outputting a drive signal to the electric power steering motor 27. During autonomous driving, the electric power steering motor 27 performs lane keeping control, which keeps the vehicle traveling in the current lane, and lane change control, which moves the vehicle to an adjacent lane (lane change control for overtaking control, etc.), based on the drive signal from the PS_ECU 23.

[0025] A brake actuator 28 is connected to the output side of the BK_ECU 24. This brake actuator 28 adjusts the brake hydraulic pressure supplied to the brake wheel cylinders provided on each wheel. The BK_ECU 24 controls the operation of the brake actuator 28 by outputting a drive signal to the brake actuator 28. Based on the drive signal from the BK_ECU 24, the brake actuator 28 generates a braking force on each wheel using the brake wheel cylinders, forcibly decelerating the wheel.

[0026] An ontology database 29 is connected to the Ontology_ECU 25. The ontology database 29 is a database written in OWL (Web Ontology Language) and stored in a large-capacity storage medium such as a HDD. The Ontology_ECU 25 reads data from the ontology database 29 and updates the data in the ontology database 29 under the control of the Travel_ECU 21.

[0027] The ontology database 29 has an ontology data structure in which inference rules 28A, traffic rules 28B, and traffic information 28C, which embody concepts such as those shown in FIGS. 2 to 4, are implemented. In the inference rules 28A, multiple dangerous situations (A, B, ..., S) that may exist around the vehicle are described in the form of a scenario. In the scenario, an identifier is assigned to each vehicle around the vehicle, and the position and speed of each vehicle are associated with each vehicle. The scenario also associates the lane and lane type of each vehicle with each vehicle, and includes traffic rules and traffic information. The traffic rules refer to rules that traffic participants must follow to participate in traffic while complying with the rules, and refer, for example, to rules regarding road information collected by the road map boundary information integration ECU 201. The traffic information refers, for example, to traffic information collected by the road map boundary information integration ECU 201. In the inference rules 28A, each dangerous situation is associated with a dangerous event that may occur in that dangerous situation. An example of a dangerous event is a traffic participant suddenly appearing from a blind spot. In the inference rule 28A, a dangerous situation (scenario) is described as a condition term, and a dangerous event is described as a result term. In the traffic rule 28B, rules regarding road information collected by the road map boundary information integration_ECU 201 are described. In the traffic information 28C, traffic information collected by the road map boundary information integration_ECU 201 is described. The traffic rule 28B and the traffic information 28C serve as background knowledge necessary for inferring a dangerous event based on a scenario.

[0028] The driving environment recognition unit 11 is fixed, for example, to the center of the upper part of the front interior of the vehicle. The driving environment recognition unit 11 has an in-vehicle camera (stereo camera) consisting of a main camera 11a and a sub-camera 11b, an image processing unit (IPU) 11c, and a driving environment detection unit 11d.

[0029] The main camera 11a and the sub-camera 11b are autonomous sensors that sense the real space around the vehicle. The main camera 11a and the sub-camera 11b are, for example, arranged at symmetrical positions on either side of the center of the vehicle in the width direction, and are capable of capturing stereo images of the area in front of the vehicle from different viewpoints.

[0030] The IPU 11c is capable of generating a distance image calculated from the amount of deviation in the positions of corresponding objects based on a pair of stereo images of the area in front of the vehicle obtained by capturing images with the main camera 11a and the sub-camera 11b.

[0031] The driving environment detection unit 11d can, for example, determine lane markings that divide the road around the vehicle based on the distance image received from the IPU 11c. The driving environment detection unit 11d can also, for example, determine the road curvature [1 / m] of the markings that divide the left and right sides of the road (driving lane) on which the vehicle is traveling, and the width between the left and right markings (vehicle width). The driving environment detection unit 11d can also, for example, perform predetermined pattern matching on the distance image to detect lanes and three-dimensional objects such as structures that exist around the vehicle.

[0032] Here, the detection of a three-dimensional object by the driving environment detection unit 11d includes, for example, detecting the type of the three-dimensional object, the distance to the three-dimensional object, the speed of the three-dimensional object, and the relative speed between the three-dimensional object and the vehicle (host vehicle). Examples of three-dimensional objects to be detected include traffic lights, intersections, road signs, stop lines, other vehicles, pedestrians, and various buildings. The driving environment detection unit 11d is capable of outputting information about the detected three-dimensional objects to the driving_ECU 21, for example.

[0033] The locator unit 12 estimates the vehicle's position (host vehicle position) on a road map and includes a locator calculation unit 13 that estimates the host vehicle position. Sensors required for estimating the vehicle's position (host vehicle position) are connected to the input side of the locator calculation unit 13. Examples of such sensors include an acceleration sensor 14, a vehicle speed sensor 15, a gyro sensor 16, and a GNSS receiver 17. The acceleration sensor 14 is capable of detecting the longitudinal acceleration of the vehicle. The vehicle speed sensor 15 is capable of detecting the vehicle's speed. The gyro sensor 16 is capable of detecting the vehicle's angular velocity or angular acceleration. The GNSS receiver 17 is capable of receiving positioning signals transmitted from multiple positioning satellites. The locator calculation unit 13 is also connected to a transceiver 18 that transmits and receives information to and from the control device 200 and other vehicles.

[0034] A high-precision road map database 19 is also connected to the locator calculation unit 13. The high-precision road map database 19 is a large-capacity storage medium such as an HDD, and stores high-precision road map information (dynamic map). This high-precision road map information, like the road map information included in the road map information integration_ECU 201, mainly includes static information and quasi-static information constituting road information, and quasi-dynamic information and dynamic information constituting traffic information.

[0035] The locator calculation unit 13 includes, for example, a map information acquisition unit 13a, a vehicle position estimation unit 13b, and a driving environment recognition unit 13c.

[0036] The vehicle position estimation unit 13b is capable of acquiring the position coordinates of the vehicle (host vehicle) based on the positioning signal received by the GNSS receiver 17. The vehicle position estimation unit 13b is also capable of estimating the host vehicle's position on a road map by map-matching the acquired position coordinates with route map information. The map information acquisition unit 13a is capable of acquiring map information of a predetermined range including the vehicle (host vehicle) from map information stored in the high-precision road map database 19, based on the position coordinates of the vehicle (host vehicle) acquired by the vehicle position estimation unit 13b.

[0037] In an environment where valid positioning signals from positioning satellites cannot be received due to reduced sensitivity of the GNSS receiver 17, such as when driving inside a tunnel, the vehicle position estimation unit 13b can switch to autonomous navigation, which estimates the vehicle's position based on the vehicle speed detected by the vehicle speed sensor 15, the angular velocity detected by the gyro sensor 16, and the longitudinal acceleration detected by the acceleration sensor 14, and estimate the vehicle's position on a road map.

[0038] As described above, the vehicle position estimation unit 13b estimates the position of the vehicle (host vehicle position) on a road map based on the positioning signal received by the GNSS receiver 17 or information detected by the gyro sensor 16, etc., and is then able to determine the road type, etc. of the road on which the vehicle (host vehicle) is traveling based on the estimated host vehicle position on the road map.

[0039] The driving environment recognition unit 13c is capable of updating the road map information stored in the high-precision road map database 19 to the latest state by using road map information acquired through external communication (roadside-to-vehicle communication and vehicle-to-vehicle communication) via the transceiver 18. This information update is performed not only for static information but also for quasi-static information, quasi-dynamic information, and dynamic information. As a result, the road map information includes road information and traffic information acquired through communication with the outside of the vehicle, and information on moving bodies such as vehicles traveling on roads is updated in real time.

[0040] The driving environment recognition unit 13c verifies road map information based on the driving environment information recognized by the driving environment recognition unit 11, and is capable of updating the road map information stored in the high-precision road map database 19 to the latest state. This information update is performed not only on static information, but also on quasi-static information, quasi-dynamic information, and dynamic information. As a result, information on moving objects such as vehicles traveling on roads recognized by the driving environment recognition unit 11 is updated in real time.

[0041] The road map information thus updated is then transmitted to the control device 200 and vehicles around the vehicle (host vehicle) by road-to-vehicle communication and vehicle-to-vehicle communication via the transceiver 18. Furthermore, the driving environment recognition unit 13c is capable of outputting, from the updated road map information, map information of a predetermined range including the host vehicle position estimated by the vehicle position estimation unit 13b, together with the host vehicle position (vehicle position information), to the driving_ECU 21.

[0042] The driving control device 10 further includes a driving assistance system 31. The driving assistance system 31 includes, for example, an Autonomous Emergency Braking (AEB) system, an Automatic Emergency Steering (AES) system, an Anti-lock Brake System (ABS), a Vehicle Dynamics Control (VDC) system, and an airbag system.

[0043] The AEB system detects a preceding vehicle or an obstacle ahead, and when it determines that a collision with the detected preceding vehicle or obstacle is unavoidable, performs brake control on behalf of the driver. The AEB system outputs an AEB primary brake operation flag when the AEB primary brake is performed, and outputs an AEB secondary brake operation flag when the AEB secondary brake is performed. When the AEB system is activated, the AEB system outputs an AEB warning flag.

[0044] The AES system detects preceding vehicles and obstacles ahead and performs steering control to avoid collision with the detected preceding vehicles and obstacles. The AES system outputs an AES operation flag when steering control is executed.

[0045] The ABS is a system that performs brake control to prevent tires from locking when the driver suddenly brakes, and outputs an ABS operation flag when brake control is executed.

[0046] The VDC system is a system that controls the traction of a vehicle and outputs a VDC warning flag when the vehicle is reaching its traction limit.

[0047] An airbag system is a system that reduces the impact on the driver's head when the driver collides with the steering wheel, instrument panel, windshield, etc. when the vehicle collides with a preceding vehicle or an obstacle ahead. When the airbag is activated, the airbag system outputs an airbag activation flag.

[0048] Next, the inference rule 28A will be described in detail. Figure 5 shows an example of traffic conditions and scenarios of dangerous situation A included in the inference rule 28A.

[0049] In danger situation A, it is assumed that a first vehicle (host vehicle) 100a is traveling on a road with one lane in each direction. The first vehicle 100a corresponds to a specific example of a "first vehicle" according to an embodiment of the present disclosure. This one-lane road is composed of a driving lane L1 in which the first vehicle 100a is traveling and an oncoming lane L2 that is provided along the driving lane L1 via a center line. An unsignalized intersection CL is provided ahead of the first vehicle 100a on this one-lane road. This one-lane road is a priority road Lm in relation to the road that intersects with this one-lane road at the unsignalized intersection CL. In other words, the first vehicle 100a is traveling on the priority road Lm.

[0050] On the other hand, the road that intersects with the priority road Lm at the unsignalized intersection CL is a non-priority road Ls in relation to the priority road Lm. A second vehicle (risky vehicle) 100b is traveling on the non-priority road Ls toward the unsignalized intersection CL. There are no traffic lights at the unsignalized intersection CL.

[0051] The driver of the first vehicle 100a recognizes that the first vehicle 100a is traveling on the priority road Lm. Therefore, the first vehicle 100a is about to enter the unsignalized intersection CL without slowing down. At this time, a second vehicle 100b is traveling on the non-priority road Ls toward the unsignalized intersection CL. However, the second vehicle 100b is in the blind spot of a fifth vehicle 100e traveling in the oncoming lane L2 to the driver of the first vehicle 100a, and the driver of the first vehicle 100a is unaware of the presence of the second vehicle 100b. A fourth vehicle 100d is traveling ahead of the first vehicle 100a. The fourth vehicle 100d is traveling while slowing down toward the unsignalized intersection CL. In addition to the fifth vehicle 100e, a third vehicle 100c is also present in the oncoming lane L2. The third vehicle 100c is stopped just before the unsignalized intersection CL. The driver of the second vehicle 100b is aware of the presence of the decelerating fourth vehicle 100d and the stopped third vehicle 100c. However, the driver of the second vehicle 100b is unaware of the presence of the first vehicle 100a because the first vehicle 100a is in the blind spot of the fifth vehicle 100e. Therefore, the driver of the second vehicle 100b intends to pass through the unsignalized intersection CL immediately after the fourth vehicle 100d has passed through the unsignalized intersection CL. Under such traffic conditions, there is a high possibility that the first vehicle 100a and the second vehicle 100b will collide head-on at the unsignalized intersection CL.

[0052] Such a scenario of dangerous situation A is stored in the inference rules 28A. In the inference rules 28A, the scenario of dangerous situation A describes, for example, the following content: - Scenario of dangerous situation A - A priority road Lm and a non-priority road Ls intersect at an unsignalized intersection CL. - A first vehicle 100a is traveling on the priority road Lm and heading toward the unsignalized intersection CL. - A second vehicle 100b is traveling on the non-priority road Ls and heading toward the unsignalized intersection CL. - The second vehicle 100b is hidden in a blind spot as seen from the first vehicle 100a. - A third vehicle 100c is stopped in the oncoming lane L2 of the first vehicle 100a just before the unsignalized intersection CL. - A fourth vehicle 100d is traveling ahead of the first vehicle 100a and decelerating.

[0053] In dangerous situation A, when the driver of the first vehicle 100a performs a braking operation, for example, as shown in Fig. 6, the AEB system of the first vehicle 100a operates and an AEB operation flag (for example, an AEB primary brake flag, an AEB secondary brake flag, or an AEB warning flag) is output from the AEB system. Alternatively, in dangerous situation A, when the driver of the first vehicle 100a performs a braking operation, for example, as shown in Fig. 7, the ABS system of the first vehicle 100a operates and an ABS operation flag is output from the ABS system.

[0054] Alternatively, in dangerous situation A, when the driver of the first vehicle 100a is performing driving operations, the AES system of the first vehicle 100a is activated, and an AES activation flag is output from the AES system, as shown in Fig. 8. Alternatively, in dangerous situation A, when the driver of the first vehicle 100a is performing driving operations, the VDC system of the first vehicle 100a is activated, and a VDC warning flag is output from the VDC system, as shown in Fig. 9. Alternatively, in dangerous situation A, for example, as shown in Fig. 10, the first vehicle 100a collides with the second vehicle 100b, causing the airbag system to activate, and an airbag activation flag is output from the airbag system.

[0055] In this way, under the circumstances where the AEB activation flag, the ABG activation flag, the AES activation flag, the VDC warning flag, or the airbag activation flag is output, there is a high possibility that the first vehicle 100a and the second vehicle 100b will collide head-on at the unsignalized intersection CL. In this specification, an event that is likely to occur in the dangerous situation A is referred to as a dangerous event. In the inference rule 28A, for example, the scenario of the dangerous situation A is associated with the dangerous event that is likely to occur in the dangerous situation A.

[0056] (Driving Assistance Procedure) Next, a driving assistance procedure in the cruise control system 1 will be described with reference to Fig. 11. Fig. 11 shows an example of a driving assistance procedure in the cruise control system 1.

[0057] First, a stereo camera mounted on the first vehicle 100a captures images of the area ahead of the first vehicle 100a and outputs the resulting stereo images to the IPU 11c. The IPU 11c generates a distance image based on the stereo images captured by the stereo camera and outputs the image to the driving environment detection unit 11d. The driving environment detection unit 11d performs predetermined pattern matching on the distance image generated by the IPU 11c to detect the priority road Lm, the driving lane L1, the oncoming lane L2, the non-priority road Ls, the unsignalized intersection CL, vehicles on the priority road Lm (e.g., 100a, 100c-100e), and vehicles on the non-priority road Ls (e.g., 100b).

[0058] Next, the driving environment recognition unit 13c uses road map information acquired via external communication to detect the priority road Lm, driving lane L1, oncoming lane L2, non-priority road Ls, unsignalized intersections CL, vehicles on the priority road Lm (e.g., 100a, 100c to 100e), and vehicles on the non-priority road Ls (e.g., 100b). Here, it is assumed that the road map information acquired via external communication includes information on vehicles on the priority road Lm (e.g., 100a, 100c to 100e) and vehicles on the non-priority road Ls (e.g., 100b). In this case, the driving environment recognition unit 13c can detect vehicles on the priority road Lm (e.g., 100a, 100c to 100e) and vehicles on the non-priority road Ls (e.g., 100b) using the road map information acquired via external communication.

[0059] The vehicle position estimation unit 13b acquires the position coordinates of the first vehicle 100a based on the positioning signal received by the GNSS receiver 17. The vehicle position estimation unit 13b further acquires the vehicle speed (speed of the first vehicle 100a) detected by the vehicle speed sensor 15.

[0060] Next, the travel_ECU 21 acquires road information Da and vehicle information Db based on various information obtained from the travel environment detection unit 11d, the vehicle position estimation unit 13b, and the travel environment recognition unit 13c (step S101). Here, the road information Da includes information about the priority road Lm, the driving lane L1, the oncoming lane L2, the non-priority road Ls, and the unsignalized intersection CL detected by the travel environment detection unit 11d or the travel environment recognition unit 13c. The vehicle information Db includes information about the speed (vehicle speed) of the first vehicle 100a acquired from the vehicle position estimation unit 13b, and information about vehicles on the priority road Lm (e.g., 100a, 100c to 100f) and vehicles on the non-priority road Ls (e.g., 100b) acquired from the travel environment detection unit 11d or the travel environment recognition unit 13c.

[0061] Next, the traveling_ECU 21 outputs the road information Da and the vehicle information Db to the Ontology_ECU 25. Upon acquiring the road information Da and the vehicle information Db from the traveling_ECU 21, the Ontology_ECU 25 creates a scenario of the surrounding conditions of the first vehicle 100a based on the acquired road information Da and vehicle information Db (step S102). The Ontology_ECU 25 creates a scenario of the surrounding conditions of the first vehicle 100a at predetermined intervals (e.g., 0.5 seconds) from the time when the time (margin time) until the first vehicle 100a reaches the unsignalized intersection CL falls below a predetermined threshold.

[0062] The Ontology_ECU 25 determines the degree of similarity between the surrounding situation (scenario) of the first vehicle 100a and each dangerous situation (scenarios) described in the ontology database 29 (step S103). At this time, the Ontology_ECU 25 determines the degree of similarity based on, for example, the instance generation ratio between the surrounding situation (scenario) of the first vehicle 100a and each dangerous situation (scenarios) described in the ontology database 29.

[0063] For example, if the instance generation rate is equal to or greater than a predetermined threshold, the Ontology_ECU 25 determines that the similarity is high (step S104; Y). At this time, the Ontology_ECU 25 determines that the surrounding situation (scenario) of the first vehicle 100a is described in the ontology database 29. On the other hand, if the instance generation rate is less than a predetermined threshold, the Ontology_ECU 25 determines that the similarity is low (step S104; N). At this time, the Ontology_ECU 25 determines that the surrounding situation (scenario) of the first vehicle 100a is not described in the ontology database 29.

[0064] When the Ontology_ECU 25 determines that the similarity is high, it notifies the driver of the occurrence of a dangerous event corresponding to the dangerous situation (scenario) determined to have a high similarity (step S105). The Ontology_ECU 25, for example, generates warning sound data and outputs it to the speaker of the first vehicle 100a, and the speaker of the first vehicle 100a outputs a sound based on the input warning sound data.

[0065] When the Ontology_ECU 25 determines that the similarity is high, it further outputs a control signal indicating that there is a high possibility that a dangerous event corresponding to the dangerous situation (scenario) determined to have a high similarity will occur to the traveling_ECU 21. When the traveling_ECU 21 receives such a control signal from the Ontology_ECU 25, it performs traveling control to avoid a dangerous event corresponding to the dangerous situation (scenario) determined to have a high similarity (step S106).

[0066] (Ontology Update Procedure) Next, an ontology update procedure in the cruise control system 1 will be described with reference to Fig. 12. Fig. 11 shows an example of a driving assistance procedure in the cruise control system 1.

[0067] When the Ontology_ECU 25 determines that the similarity is low, it determines whether a predetermined flag of the driving assistance system 31 has been output (step S107). The Ontology_ECU 25 determines whether, for example, an AEB primary brake operation flag, an AEB secondary brake operation flag, an AES operation flag, an ABS operation flag, a VDC warning flag, or an airbag operation flag has been output. The Ontology_ECU 25 determines whether or not there is a high possibility of a dangerous event occurring corresponding to the surrounding situation (scenario) of the first vehicle 100a using the predetermined flag of the driving assistance system 31 (step S108). The Ontology_ECU 25 determines whether or not there is a high possibility of a dangerous event occurring corresponding to the surrounding situation (scenario) of the first vehicle 100a based on whether, for example, an AEB primary brake operation flag, an AEB secondary brake operation flag, an AES operation flag, an ABS operation flag, a VDC warning flag, or an airbag operation flag has been output. If the predetermined flag of the driving assistance system 31 is not output, this means that the driving assistance system 31 is not operating.

[0068] The Ontology_ECU 25 infers a dangerous event corresponding to the surrounding situation (scenario) of the first vehicle 100a based on the surrounding situation (scenario) of the first vehicle 100a and the traffic rules 28B and traffic information 28C in the ontology database 29. If a predetermined flag of the driving assistance system 31 has been output, the Ontology_ECU 25 determines that there is a high possibility that the dangerous event corresponding to the surrounding situation (scenario) of the first vehicle 100a will occur (step S108; Y). On the other hand, if the predetermined flag of the driving assistance system 31 has not been output, the Ontology_ECU 25 determines that there is a low possibility that the dangerous event corresponding to the surrounding situation (scenario) of the first vehicle 100a will occur (step S108; N).

[0069] When the Ontology_ECU 25 determines that there is a high possibility that a dangerous event corresponding to the surrounding circumstances (scenario) of the first vehicle 100a will occur, the Ontology_ECU 25 links (associates) the surrounding circumstances (scenario) of the first vehicle 100a with the dangerous event obtained by inference and adds them to the ontology database 29. In this way, the Ontology_ECU 25 updates the ontology database 29 by adding a set of new inference rules 28A to the ontology database 29, the set of new inference rules 28A having the surrounding circumstances (scenario) of the first vehicle 100a as a condition term and the dangerous event obtained by inference as a result term (step S109). The Ontology_ECU 25 starts updating the ontology database 29 when it determines that there is a dangerous event around the first vehicle 100a. On the other hand, when the Ontology_ECU 25 determines that there is a low possibility that a dangerous event corresponding to the surrounding circumstances (scenario) of the first vehicle 100a will occur, the Ontology_ECU 25 does not update the ontology database 29.

[0070] [Effects] Next, effects of the cruise control system 1 according to one embodiment of the present disclosure will be described.

[0071] In this embodiment, the similarity between the surrounding situation (scenario) of the first vehicle 100a and each dangerous situation (scenarios) described in the ontology database 29 is determined. If the similarity is determined to be low, it is determined that the surrounding situation (scenario) of the first vehicle 100a is not described in the ontology database 29. If the similarity is determined to be low, it is determined whether or not a dangerous event exists around the first vehicle 100a based on a predetermined flag of the driving assistance system 31. If the similarity is determined to be low, the surrounding situation (scenario) of the first vehicle 100a is associated with the dangerous event obtained by inference and added to the ontology database 29, thereby updating the ontology database 29.

[0072] As described above, in this embodiment, when the first vehicle 100a encounters a dangerous situation (scenario) that does not exist in the existing ontology database 29 and a dangerous event occurs around the first vehicle 100a, the ontology database 29 is updated by associating the dangerous situation (scenario) with the dangerous event at that time and adding the associations to the ontology database 29. Therefore, after the ontology database 29 is updated, when the first vehicle 100a encounters a similar dangerous situation, it becomes possible to perform control that predicts the occurrence of the dangerous event read from the ontology database 29.

[0073] In this embodiment, a determination is made as to whether or not a dangerous event is occurring around the first vehicle 100a based on a signal (for example, a predetermined activation flag) output from the driving assistance system 31. This makes it possible to avoid adding situations (scenarios) that are not particularly dangerous to the ontology database 29. As a result, the first vehicle 100a can perform control that predicts the occurrence of a dangerous event read from the ontology database 29 only when the first vehicle 100a is in an actually dangerous situation.

[0074] In this embodiment, when it is determined that a dangerous event exists around the first vehicle 100a, the ontology database 29 starts to be updated, which makes it possible to always keep the ontology database 29 up to date.

[0075] In this embodiment, a set of new inference rules 28A, each of which has the surrounding situation (scenario) of the first vehicle 100a as a condition term and the dangerous event obtained by inference as a result term, is added to the ontology database 29, thereby updating the ontology database 29. After the ontology database 29 is updated, when the first vehicle 100a encounters a similar dangerous situation, it becomes possible to perform control that predicts the occurrence of the dangerous event read from the ontology database 29.

[0076] In this embodiment, the similarity is determined based on the instance generation ratio between the surrounding situation (scenario) of the first vehicle 100a and each dangerous situation (scenarios) described in the ontology database 29. As a result, for example, if the instance generation ratio is equal to or greater than a predetermined threshold, it is possible to determine that the similarity is high and that the surrounding situation (scenario) of the first vehicle 100a is described in the ontology database 29. On the other hand, for example, if the instance generation ratio is less than a predetermined threshold, it is possible to determine that the similarity is low and that the surrounding situation (scenario) of the first vehicle 100a is not described in the ontology database 29. As a result, it is possible to easily determine whether or not the ontology database 29 needs to be updated based on the instance generation ratio.

[0077] 3. Modifications The present disclosure has been described above by giving embodiments, but the present disclosure is not limited to these embodiments and various modifications are possible.

[0078] [Modification 3-1] In the above embodiment, the Ontology_ECU 25 determines whether there is a high possibility that a dangerous event corresponding to the surrounding situation (scenario) of the first vehicle 100a will occur, based on whether a predetermined flag has been output from the driving assistance system 31. However, in the above embodiment, the Ontology_ECU 25 may determine whether there is a high possibility that a dangerous event corresponding to the surrounding situation (scenario) of the first vehicle 100a will occur, taking into account not only the predetermined flag of the driving assistance system 31 but also other factors.

[0079] 13 shows a modified example of the procedure for updating the ontology database 29. After executing step S107, the ontology_ECU 25 detects, for example, one or more risk factors contained in information (stereo images) obtained from the driving environment detection unit 11d. After executing step S107, the ontology_ECU 25 may also detect, for example, one or more risk factors contained in data obtained by external communication (road-to-vehicle communication and vehicle-to-vehicle communication) via the transceiver 18.

[0080] The Ontology_ECU 25 determines whether a portion of another traffic participant (e.g., the second vehicle 100b) appears from a blind spot with low accuracy (whether there is a risk factor) based on, for example, information obtained from the driving environment detection unit 11d (information about three-dimensional objects such as structures present around the first vehicle 100a) (step S110). The Ontology_ECU 25 may determine whether a portion of another traffic participant (e.g., the second vehicle 100b) appears from a blind spot with low accuracy (whether there is a risk factor) based on, for example, data obtained through external communication (road-to-vehicle communication and vehicle-to-vehicle communication) via the transceiver 18. Here, an example of a "blind spot" is a vehicle parked on the road. Note that if a portion of another traffic participant (e.g., the second vehicle 100b) appears from a blind spot with high accuracy, the Ontology_ECU 25 can determine the similarity using information about the other traffic participant in the above-mentioned step S103.

[0081] The Ontology_ECU 25 further determines the driving behavior of the driver of the first vehicle 100a based on data obtained from an imaging device (sensor) or the like installed in the first vehicle 100a. The Ontology_ECU 25 determines, for example, whether the driver of the first vehicle 100a is gazing ahead based on data obtained from an imaging device (sensor) or the like installed in the first vehicle 100a. The Ontology_ECU 25 determines, for example, whether the driver of the first vehicle 100a is gazing at the other traffic participants based on data obtained from the imaging device or the like installed in the first vehicle 100a (step S111).

[0082] The Ontology_ECU 25 further determines whether the driver of the first vehicle 100a is taking evasive action (step S112). The Ontology_ECU 25 may determine whether the driver of the first vehicle 100a is taking evasive action from the following three types of data combinations based on data obtained from various sensors installed in the first vehicle 100a. Note that in the following (2) and (3), the fact that the path sign of the first vehicle 100a is opposite to the direction of other traffic participants suggests that the driver of the first vehicle 100a is taking evasive action: (1) Brake pressure change rate + deceleration of the first vehicle 100a (2) Steering angle change rate + path sign of the first vehicle 100a (3) Steering wheel torque change rate + path sign of the first vehicle 100a

[0083] In this way, the Ontology_ECU 25 determines whether there is a high possibility of the occurrence of a dangerous event corresponding to the surrounding circumstances (scenario) of the first vehicle 100a, taking into account not only the predetermined flag of the driving assistance system 31 but also other factors (e.g., the presence of other traffic participants, the driving behavior of the driver of the first vehicle 100a, and the biometric indicators of the driver of the first vehicle 100a) (step S108). In this way, it is possible to estimate with higher accuracy the possibility of the occurrence of a dangerous event corresponding to the surrounding circumstances (scenario) of the first vehicle 100a. This makes it possible to avoid adding situations (scenarios) that are not particularly dangerous to the ontology database 29. As a result, the first vehicle 100a can perform control that predicts the occurrence of a dangerous event read from the ontology database 29 only under actually dangerous situations.

[0084] In addition, in this modified example, even in a situation where the driving assistance system 31 is not operating, by using other factors (e.g., the presence of other traffic participants, the driving attitude of the driver of the first vehicle 100a, and the biometric indicators of the driver of the first vehicle 100a) as judgment factors, it is possible to estimate the possibility of a dangerous event occurring corresponding to the surrounding circumstances (scenario) of the first vehicle 100a.

[0085] In this modification, it is determined with low accuracy whether a part of another traffic participant (for example, the second vehicle 100b) is appearing from a blind spot based on information obtained from the driving environment detection unit 11d (information about three-dimensional objects such as structures present around the first vehicle 100a) and data acquired through external communication (road-to-vehicle communication and vehicle-to-vehicle communication) via the transceiver 18. This makes it possible to perform a reliable estimation, even if it is with low accuracy.

[0086] [Variation 3-2] In the above-described variation 3-1, instead of step S112, the Ontology_ECU 25 may determine whether data acquired from a sensor capable of detecting biometric indicators of the driver of the first vehicle 100a exceeds a predetermined threshold (step S113), for example, as shown in FIG. 14 . By using the biometric indicators, the Ontology_ECU 25 can estimate the psychological state of the driver of the first vehicle 100a. This allows the Ontology_ECU 25 to determine whether the driver of the first vehicle 100a is taking evasive action even in a situation where the driving assistance system 31 is not operating. As a result, the Ontology_ECU 25 can determine whether a dangerous event exists around the first vehicle 100a.

[0087] [Variation 3-3] In the above-described embodiment and its variations, the present disclosure has been applied to driving assistance at an intersection CL where a priority road Lm and a non-priority road Ls intersect. However, in the above-described embodiment and its variations, the present disclosure may also be applied to driving assistance at a junction where a non-priority road Ls merges with a priority road Lm, for example. Even in such a case, it is possible to obtain the same effects as those of the above-described embodiment and its variations.

[0088] [Variation 3-4] In the above-described embodiment and its variations, if it is difficult for the first vehicle 100a to communicate with the network environment NW, the travel_ECU 21 may acquire road information Da and vehicle information Db, for example, based on various data obtained from various sensors mounted on the first vehicle 100a. Here, the road information Da includes information about the priority road Lm, the driving lane L1, the oncoming lane L2, the non-priority road Ls, and the unsignalized intersection CL detected by the driving environment recognition unit 13c. The vehicle information Db includes information about the speed (vehicle speed) of the first vehicle 100a acquired from the vehicle position estimation unit 13b, as well as information about vehicles on the priority road Lm (e.g., 100a, 100c to 100e) and vehicles on the non-priority road Ls (e.g., 100b). In this way, even if a determination is made as to whether or not there is a dangerous incident around the first vehicle 100a based on the traffic conditions ahead of the first vehicle 100a obtained from various sensors mounted on the first vehicle 100a, it is possible to obtain the same effects as those of the above-mentioned embodiment and its modified examples.

[0089] Note that the effects described in this specification are merely examples. The effects of the present disclosure are not limited to the effects described in this specification. The present disclosure may have effects other than the effects described in this specification.

[0090] Furthermore, for example, the present disclosure may be configured as follows: (1) An ontology updating device including a control unit capable of updating an ontology, wherein the control unit is capable of: determining a similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology, and, if it is determined that the similarity is low as a result, determining that the surrounding situation is not described in the ontology; and determining whether or not a dangerous event exists around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired, and, if it is determined that a dangerous event exists around the first vehicle as a result, updating the ontology by linking the surrounding situation with the dangerous event and adding the link to the ontology. (2) The ontology updating device according to (1), wherein the device is a driving assistance system mounted on the first vehicle, and the control unit is capable of determining whether or not a dangerous event exists around the first vehicle based on a signal output from the driving assistance system. (3) The ontology update device according to (2), wherein the device is capable of determining whether or not a hazardous event exists around the first vehicle by utilizing an operation flag of the driving assistance system. (4) The ontology update device according to any one of (1) to (3), wherein the device includes a first sensor capable of detecting traffic conditions ahead of the first vehicle, and the control unit is capable of determining whether or not a hazardous event exists around the first vehicle based on data obtained from the first sensor in a situation where a driving assistance system mounted on the first vehicle is not operating. (5) The ontology update device according to (4), wherein the first sensor is a binocular camera capable of acquiring stereo images ahead of the first vehicle, and the control unit is capable of detecting one or more risk factors included in the stereo images obtained by the binocular camera and determining whether or not a hazardous event exists around the first vehicle based on the detection result.(6) The ontology update device described in (4), wherein the device includes the first sensor and a second sensor capable of detecting a driving attitude, evasive behavior, or biological indicators of a driver of the first vehicle, and the control unit is capable of determining whether or not a hazardous event exists around the first vehicle based on data obtained from both the first sensor and the second sensor in a situation where the driving assistance system is not operating. (7) The ontology update device described in (4), wherein the control unit is capable of determining whether or not a hazardous event exists around the first vehicle by acquiring the event from at least one of a third sensor mounted on the first vehicle and capable of detecting an event ahead of the first vehicle, and a communication unit capable of acquiring the event ahead of the first vehicle from an external device. (8) The ontology update device described in any one of (1) to (7), wherein the control unit is capable of starting updating the ontology at a timing when it is determined that a hazardous event exists around the first vehicle. (9) The ontology update device according to any one of (1) to (8), wherein the control unit is capable of updating the ontology by adding to the ontology a set of new inference rules having the surrounding situation as a condition term and the hazardous event as a result term. (10) The ontology update device according to any one of (1) to (9), wherein the control unit is capable of determining the similarity based on an instance generation rate between the surrounding situation and a hazardous situation described in the ontology.(11) A vehicle comprising: a storage unit storing an ontology; and a control unit capable of updating the ontology, wherein the control unit is capable of: determining a similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology; and, if it is determined that the similarity is low as a result, determining that the surrounding situation is not described in the ontology; and determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired; and, if it is determined that there is a dangerous event around the first vehicle as a result, updating the ontology by linking the surrounding situation and the dangerous event and adding them to the ontology. (12) An ontology updating method including: determining a similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology; and, if it is determined that the similarity is low as a result, determining that the surrounding situation is not described in the ontology; and determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired; and, if it is determined that there is a dangerous event around the first vehicle as a result, updating the ontology by linking the surroundings with the dangerous event and adding them to the ontology.

[0091] The cruise control device 10 shown in FIG. 1 can be implemented by circuitry including at least one semiconductor integrated circuit, such as at least one processor (e.g., a central processing unit (CPU)), at least one application-specific integrated circuit (ASIC), and / or at least one field-programmable gate array (FPGA). The at least one processor can be configured to perform all or a portion of the various functions of the cruise control device 10 shown in FIG. 1 by reading instructions from at least one non-transitory, tangible computer-readable medium. Such medium can take various forms, including, but not limited to, various magnetic media such as hard disks, various optical media such as CDs or DVDs, and various semiconductor memories (i.e., semiconductor circuits) such as volatile or non-volatile memories. Volatile memory can include DRAM and SRAM. Non-volatile memory can include ROM and NVRAM. An ASIC is an integrated circuit (IC) specialized to perform all or a portion of the various functions of the cruise control device 10 shown in FIG. 1. An FPGA is an integrated circuit that is designed to be configurable after manufacture so as to perform all or part of the various functions of the cruise control device 10 shown in FIG.

Claims

1. An ontology updating device comprising a control unit capable of updating an ontology, wherein the control unit is capable of executing the following: determining a similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology, and when it is determined that the similarity is low as a result, determining that the surrounding situation is not described in the ontology; determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired, and when it is determined that there is a dangerous event around the first vehicle as a result, updating the ontology by linking the surrounding situation with the dangerous event and adding them to the ontology.

2. The ontology update device according to claim 1, wherein the device is a driving assistance system mounted on the first vehicle, and the control unit is capable of determining whether or not there is a dangerous event around the first vehicle based on a signal output from the driving assistance system.

3. The ontology update apparatus according to claim 2, wherein the device is capable of determining whether or not there is a dangerous event around the first vehicle by using an operation flag of the driving assistance system.

4. The ontology update device described in claim 1, wherein the device includes a first sensor capable of detecting traffic conditions ahead of the first vehicle, and the control unit is capable of determining whether or not there is a dangerous event around the first vehicle based on data obtained from the first sensor under conditions in which a driving assistance system installed in the first vehicle is not operating.

5. The ontology update device described in claim 4, wherein the first sensor is a binocular camera capable of acquiring stereo images of the area in front of the first vehicle, and the control unit is capable of detecting one or more risk factors contained in the stereo images acquired by the binocular camera and determining whether or not a dangerous event is present around the first vehicle based on the detection results.

6. The ontology update device described in claim 4, wherein the device includes the first sensor and a second sensor capable of detecting the driving attitude, evasive behavior, or biometric indicators of the driver of the first vehicle, and the control unit is capable of determining whether or not there is a dangerous event around the first vehicle based on data obtained from both the first sensor and the second sensor under a situation where the driving assistance system is not operating.

7. The ontology update device described in claim 4, wherein the control unit is capable of determining whether or not there is a dangerous event around the first vehicle by acquiring the event from at least one of a third sensor mounted on the first vehicle and capable of detecting an event ahead of the first vehicle, and a communication unit capable of acquiring the event ahead of the first vehicle from an external device.

8. The ontology update device according to claim 1, wherein the control unit is capable of starting updating of the ontology at a timing when it is determined that a dangerous event exists around the first vehicle.

9. The ontology updating device according to claim 1, wherein the control unit is capable of updating the ontology by adding to the ontology a set of new inference rules in which the surrounding circumstances are used as a condition term and the hazardous event is used as a result term.

10. The ontology update device according to claim 1, wherein the control unit is capable of determining the similarity based on an instance generation ratio between the surrounding situation and a dangerous situation described in the ontology.

11. A vehicle comprising: a memory unit in which an ontology is stored; and a control unit capable of updating the ontology, wherein the control unit is capable of: determining a similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology, and when it is determined that the similarity is low as a result, determining that the surrounding situation is not described in the ontology; determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired, and when it is determined that there is a dangerous event around the first vehicle as a result, updating the ontology by linking the surrounding situation with the dangerous event and adding them to the ontology.

12. An ontology updating method comprising: determining a similarity between a surrounding situation of a first vehicle and a dangerous situation described in the ontology, and, when it is determined that the similarity is low as a result, determining that the surrounding situation is not described in the ontology; determining whether or not there is a dangerous event around the first vehicle based on data obtained from a device mounted on the first vehicle when the surrounding situation is acquired, and, when it is determined that there is a dangerous event around the first vehicle as a result, updating the ontology by linking the surroundings with the dangerous event and adding them to the ontology.

Images