Action probability estimation device and method, and traffic safety assistance system

Abstract

The invention provides an action probability estimation device and method, and a traffic safety assistance system. The present invention appropriately estimates an action probability of a steering action that can be executed by a driver of a target vehicle that is a target of driving assistance in accordance with a traffic condition. This behavior probability estimation device is provided with: a risk calculation unit that calculates a risk index value that is a numerical value indicating the degree of risk of collision between traffic participants around a target vehicle and the target vehicle; and a probability estimation unit that estimates an action probability, which is a probability that the driver performs a prescribed maneuvering action in the target vehicle, on the basis of the risk indicator value and driver characteristics of the target vehicle, the driver characteristics including a confidence level and a perseverance level when a dangerous situation occurs. The self-confidence level is a numerical value indicating the level of self-confidence of the driver with respect to the own driving skill, and the perseverance performance level during the dangerous situation is a numerical value indicating the level of tendency of the driver to know the perseverance performance of the dangerous behavior.

Description

Technical Field

[0001] The invention relates to an action probability estimation device, a method and a traffic safety auxiliary system. Background Art

[0002] In recent years, efforts have been actively made to allow disadvantaged people among traffic participants to use sustainable transportation systems that take them into consideration. To achieve this, we are working on research and development related to driver assistance technology to further improve traffic safety and convenience.

[0003] Patent document 1 discloses a traffic safety assistance system that assists in avoiding the risk of more than three traffic participants becoming parties. The traffic safety assistance system identifies two-wheeled motor vehicles, four-wheeled vehicles, and pedestrians in a traffic area as traffic participants, and predicts the future behavior of the two-wheeled motor vehicle, the future behavior of the four-wheeled vehicle corresponding to the future behavior of the two-wheeled motor vehicle, and the future chain risk of the pedestrians corresponding to the future behavior of at least one of the two-wheeled motor vehicle and the four-wheeled vehicle based on information related to the driving ability of the drivers of the two-wheeled motor vehicle and the four-wheeled vehicle.

[0004] Prior art literature

[0005] Patent Literature

[0006] Patent Document 1: International Publication No. 2023 / 089823 A1 Summary of the invention

[0007] Problems to be solved by the invention

[0008] However, in driving assistance technology, it is a challenge to provide driving assistance before the risk of an accident becomes apparent.

[0009] The purpose of the present application is to appropriately estimate the action probability of the maneuvering action that the driver of the target vehicle can perform according to the traffic conditions in order to solve the above-mentioned problems, and to give appropriate warnings to the target vehicle and surrounding traffic participants based on the above-estimated action probability, thereby contributing to the development of a sustainable transportation system.

[0010] Means for solving problems

[0011] One embodiment of the present invention is an action probability estimation device, which comprises: a risk calculation unit, which calculates a risk index value, which is a numerical value indicating the degree of the risk of collision between traffic participants around a target vehicle and the target vehicle; and a probability estimation unit, which estimates the action probability based on the risk index value and the characteristics of the driver of the target vehicle, i.e., the driver characteristics, the action probability is the probability that the driver performs a predetermined specified operating action in the target vehicle, the driver characteristics including: a confidence level, which is a numerical value indicating the degree of confidence of the driver in his or her own driving skills; and a resolute execution level in dangerous situations, which is a numerical value indicating the degree of the tendency to resolutely execute an action knowing that the danger is dangerous.

[0012] According to another aspect of the present invention, the confidence level and the risk-taking level are calculated in advance for each driver and stored in a storage device as a part of the driver's characteristics.

[0013] According to another embodiment of the present invention, the action probability estimation device also includes an emotion estimation unit, which calculates an emotion level, wherein the emotion level is a numerical value indicating the degree of magnitude of unpleasant emotions currently possessed by the driver, and the probability estimation unit estimates the action probability based on the driver characteristics, the risk indicator value and the emotion level.

[0014] According to another aspect of the present invention, the emotion estimation unit calculates an emotion level based on the physiological data of the driver and / or the action schedule of the driver.

[0015] According to another aspect of the present invention, a logistic regression model is used to estimate the action probability.

[0016] According to another aspect of the present invention, the predetermined control action is a lane change action of changing a driving lane of the target vehicle.

[0017] According to another embodiment of the present invention, the risk index value is represented by a front collision margin time and a rear collision margin time, wherein the front collision margin time is a collision margin time between a front vehicle traveling in front of the target vehicle as the traffic participant and the target vehicle, and the rear collision margin time is a collision margin time between a subsequent vehicle traveling behind the target vehicle as the traffic participant and the target vehicle.

[0018] According to another aspect of the present invention, the target vehicle is a two-wheeled motor vehicle, and the following vehicle is a four-wheeled automobile.

[0019] Another embodiment of the present invention is an action probability estimation method, which is executed by a computer to estimate the probability that the driver of the target vehicle will perform a prescribed operating action, i.e., the action probability, and includes the following steps: a risk calculation step, calculating a risk index value, wherein the risk index value is a numerical value indicating the degree of the risk of collision between the traffic participants around the target vehicle and the target vehicle; and an estimation step, estimating the action probability based on the risk index value and the characteristics of the driver of the target vehicle, i.e., the driver characteristics, wherein the action probability is the probability that the driver will perform a predetermined prescribed operating action in the target vehicle, and the driver characteristics include: a confidence level, which is a numerical value indicating the degree of confidence of the driver in his or her own driving skills; and a resolute execution level in dangerous situations, which is a numerical value indicating the degree of the tendency to resolutely perform an action knowing the danger.

[0020] Another embodiment of the present invention is a traffic safety assistance system, comprising: an action probability estimation device; and a notification device, which, based on the action probability estimated by the action probability estimation device, provides an attention-calling notification to the driver of the target vehicle and / or the traffic participant, wherein, when the rear collision margin time is less than a predetermined first time threshold, the notification device performs the following processing: when the action probability of the target vehicle is less than a predetermined probability threshold, a first level of attention-calling notification is performed in the target vehicle and the subsequent vehicle; when the action probability of the target vehicle is greater than the probability threshold, a second level of attention-calling notification with a significance level higher than the first level of attention-calling notification is performed in the target vehicle and the subsequent vehicle.

[0021] In another embodiment of the present invention, when the rear collision margin time is greater than the first time threshold and less than a predetermined second time threshold that is greater than the first time threshold, the notification device performs the following processing: when the action probability of the object vehicle is less than the probability threshold, a notification of an information prompt indicating the existence of the object vehicle is performed in the subsequent vehicle; when the action probability of the object vehicle is greater than the probability threshold, the first level of attention-calling notification is performed in the subsequent vehicle.

[0022] Effects of the Invention

[0023] According to the present invention, the action probability that the driver of the target vehicle performs a predetermined steering action depending on the degree of collision risk with a traffic participant can be appropriately estimated in consideration of characteristics such as the personality of the driver of the target vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1This is an explanatory diagram for explaining an overview of a behavior probability estimation device and a traffic safety support system including the behavior probability estimation device according to an embodiment of the present invention.

[0025] Figure 2 This is a diagram showing an example of a specific traffic scene in which the action probability estimation device and the traffic safety assistance system according to the present embodiment operate.

[0026] Figure 3 The diagram shows the structure of a behavior probability estimation device mounted on a target vehicle, which is a two-wheeled motor vehicle, and the structure of a traffic safety support system including the behavior probability estimation device.

[0027] Figure 4 This is a diagram showing an example of a device configuration in a traffic participant.

[0028] Figure 5 This is a flowchart showing the operation sequence of the traffic safety assistance system.

[0029] Description of Reference Numerals

[0030] 1…action probability estimation device, 2…traffic safety assistance system, 3…target vehicle, 4…first notification device, 5…first HMI device, 6…first communication device, 7…traffic participant, 7a…front vehicle, 7b…following vehicle, 8…second notification device, 9…second HMI device, 10…second communication device, 11…road, 12a, 12b…lane, 20…external camera, 21…object detection device, 22…seat sensor, 23…helmet, 24…wearable device, 25…portable terminal, 26…microphone, 27…first speaker, 28…HUD, 30…first processor, 31…first memory, 32…first program, 33…driver characteristics, 34…probability estimation model, 35…risk calculation unit, 36…emotion estimation unit, 37…probability estimation unit, 40…second processor, 41…second memory, 42…second program, 43…first notification unit, 50…display device, 51…second speaker, 60…third processor, 61…third memory, 62…third program, 63…second notification unit, D…driver, H…person. DETAILED DESCRIPTION

[0031] Hereinafter, embodiments of the present invention will be described with reference to the drawings.

[0032] [1. Summary]

[0033] Figure 1 This is an explanatory diagram for explaining the outline of a behavior probability estimation device 1 and a traffic safety assistance system 2 including the behavior probability estimation device 1 according to one embodiment of the present invention.

[0034] The action probability estimation device 1 and the traffic safety assistance system 2 are installed in a target vehicle 3 .

[0035] The action probability estimation device 1 estimates the action probability that the driver D of the target vehicle 3 performs a predetermined maneuvering action according to the degree of collision risk with the traffic participants 7 around the target vehicle 3. In this embodiment, the action probability estimation device 1 estimates the action probability of the driver in consideration of the characteristics such as the personality of the driver.

[0036] The traffic safety assistance system 2 provides safety assistance by notifying the target vehicle 3 and / or the traffic participant 7 based on the action probability estimated by the action probability estimation device 1 .

[0037] The traffic safety assistance system 2 includes an action probability estimation device 1 and a first notification device 4 provided in a target vehicle 3. The first notification device 4 notifies a driver D of the target vehicle and / or a traffic participant 7. The traffic safety assistance system 2 makes various notifications for traffic safety assistance to the target vehicle 3 and / or the traffic participant 7 based on the action probability estimated by the action probability estimation device 1 through the first notification device 4.

[0038] The above notification in the target vehicle 3 is performed, for example, by the first notification device 4 outputting various notifications to the driver D using the first HMI device 5 (HMI: Human Machine Interface) provided in the target vehicle 3. In addition, the above notification to the traffic participant 7 is performed by the first notification device 4 provided in the target vehicle 3 sending a notification instruction indicating various notifications to the traffic participant 7 via the first communication device 6. The second notification device 8 of the traffic participant 7 receives the notification instruction via the second communication device 10, and outputs various notifications via the second HMI device 9 according to the received notification instruction.

[0039] The traffic participant 7 may be other vehicles (other vehicles) or pedestrians existing around the target vehicle 3. When the traffic participant is another vehicle, the second notification device 8, the second HMI device 9, and the second communication device 10 may be vehicle-mounted devices. The second HMI device 9 outputs a notification to the person H who is the driver of the other vehicle. In addition, when the traffic participant 7 is the pedestrian H himself, the second notification device 8, the second HMI device 9, and the second communication device 10 may be, for example, a portable terminal such as a smartphone owned by the person H. The second HMI device 9 outputs a notification to the person H who is a pedestrian.

[0040] In the present embodiment, for example, the target vehicle 3 is a two-wheeled motor vehicle and the traffic participant is a four-wheeled motor vehicle. In the case of this combination, in particular, by using the action probability estimation device 1, a two-wheeled motor vehicle, which is more difficult to control the movement of the vehicle body and more important to call the driver's attention than a four-wheeled motor vehicle, is used as the target vehicle 3, and the action probability of the driver D is appropriately estimated, so that more appropriate assistance for safe driving can be performed. In addition, in a traffic scene including a two-wheeled motor vehicle and a four-wheeled automobile, which are difficult to coordinate safety assistance due to the difference in the mobility of the vehicle body, the above-mentioned coordinated traffic assistance can be realized based on the action probability appropriately estimated for the two-wheeled motor vehicle.

[0041] Figure 2 FIG. 2 is a diagram showing an example of a specific traffic scene in which the action probability estimation device 1 and the traffic safety assistance system 2 of the present embodiment operate. Figure 2 In the traffic scene of FIG. 1 , a two-wheeled motor vehicle, namely a target vehicle 3, is traveling in a lane 12a of a road 11 having two lanes (lane lines) 12a and 12b. On the road 11, there are a front vehicle 7a traveling in front of the target vehicle 3 and a follower vehicle 7b traveling behind the target vehicle 3 as traffic participants 7 around the target vehicle 3. The front vehicle 7a is traveling in the same lane 12a as the target vehicle 3, and the follower vehicle 7b is traveling in the lane 12b.

[0042] The action probability estimation device 1 calculates the action probability, which is the probability that the driver D of the object vehicle 3 performs a predetermined maneuvering action on the object vehicle 3, based on the above-mentioned traffic conditions including the object vehicle 3 and the front vehicle 7a and the following vehicle 7b as surrounding traffic participants. In the present embodiment, the above-mentioned predetermined maneuvering action is, for example, a lane change action of changing the driving lane of the object vehicle 3 from the lane 12a to the lane 12b.

[0043] In addition, the traffic safety assistance system 2 notifies the target vehicle 3, the front vehicle 7a, and / or the following vehicle 7b for traffic safety assistance based on the action probability estimated by the action probability estimation device 1. In the present embodiment, the traffic safety assistance system 2 notifies the target vehicle 3 and / or the following vehicle 7b for traffic safety assistance based on the action probability of the lane change action of the target vehicle 3 estimated by the action probability estimation device 1.

[0044] also, Figure 2The traffic scene shown is an example, and the action probability estimation device 1 and the traffic safety assistance system 2 can operate in any traffic scene encountered by the object vehicle 3. For example, the road on which the object vehicle 3 travels can also be an urban road with a crosswalk, an intersection, a merging point or branching point of a highway, or the inside of a parking lot. In addition, the prescribed maneuvering action performed by the driver D of the object vehicle 3 is not limited to the lane change action, and can be any unsafe risk action that can induce contact with traffic participants. Such risk actions can be, for example, sudden acceleration, sudden deceleration, sudden lane change, queue cutting, shortening the distance between the vehicle in front or the following vehicle, continuous driving across lanes, snaking, driving in the wrong direction, running a red light, driving at a speed faster than the surrounding moving objects, driving at a speed slower than the surrounding moving objects, or hindering the movement of surrounding traffic participants.

[0045] [2. Structure of Action Probability Estimation Device]

[0046] Figure 3 The diagram shows the configuration of a behavior probability estimation device 1 mounted on a target vehicle 3 which is a two-wheeled motor vehicle, and the configuration of a traffic safety support system 2 including the behavior probability estimation device 1 .

[0047] The action probability estimation device 1 and the first notification device 4 operate by obtaining information from various sensors and / or devices provided in the target vehicle 3 and / or worn or held by the driver D such as a laser radar.

[0048] For example, the target vehicle 3 includes an exterior camera 20, an object detection device 21, and a seat sensor 22. The exterior camera 20 is, for example, disposed at the front and rear of the vehicle body of the target vehicle 3, and outputs images or videos in front of and behind the target vehicle 3. The object detection device 21 is, for example, a radar, a lidar, and / or a sonar disposed at the front and rear of the vehicle body of the target vehicle 3, and outputs information on the positions of objects existing in front of and behind the target vehicle 3. In addition, the seat sensor 22 is provided at the seat where the driver D sits, and outputs physiological data of the driver D, such as the pulse and respiratory rate.

[0049] The driver D of the target vehicle 3 can wear a helmet 23 and a wearable device 24 such as a smart watch, for example. A microphone 26, a first speaker 27, and a HUD 28 (head-up display) can be provided on the helmet 23. The first speaker 27 and the HUD 28 are an example of a first HMI device 5 used in the target vehicle 3. Depending on the type of the target vehicle 3, the first HMI device 5 may include any other HMI device. Such an HMI device may be, for example, a tactile HMI such as a vibrating steering wheel having a vibration actuator on the steering wheel and transmitting information to the driver D holding the steering wheel through vibration. Alternatively, the HMI device may be an electric seat belt that can change the tension of the seat belt to call attention when the target vehicle 3 has a seat with a seat belt. The same is true for the second HMI device 9 possessed by the traffic participant 7.

[0050] The wearable device 24 detects physiological data such as pulse, blood pressure, and skin potential of the driver D. The driver D of the target vehicle 3 may also carry a portable terminal 25. The portable terminal 25 stores a schedule showing the action schedule of the driver D according to the conventional technology, for example.

[0051] The portable terminal 25 , the wearable device 24 , the microphone 26 , the first speaker 27 , and the HUD 28 of the helmet 23 can be communicatively connected to each other or to other devices included in the target vehicle 3 via the first communication device 6 , for example, by short-range wireless such as Bluetooth (registered trademark).

[0052] The action probability estimation device 1 includes a first processor 30 and a first memory 31. The first memory 31 is composed of, for example, a volatile and / or nonvolatile semiconductor memory, etc. The first memory 31 as a storage device stores a first program 32, a driver characteristic 33, and a probability estimation model 34.

[0053] The first processor 30 is, for example, a computer including a CPU, etc. The first processor 30 may also be a structure including a ROM in which a program is written, a RAM for temporary storage of data, etc. Furthermore, the first processor 30 includes a risk calculation unit 35, a mood estimation unit 36, and a probability estimation unit 37 as functional elements or functional units.

[0054] These functional elements of the first processor 30 are realized, for example, by the first processor 30 as a computer executing the first program 32 stored in the first memory 31. In addition, the first program 32 can be pre-stored in any computer-readable storage medium. Alternatively, all or part of the above-mentioned functional elements of the first processor 30 may be constituted by hardware each including one or more electronic circuit components.

[0055] The risk calculation unit 35 calculates a risk index value which is a numerical value indicating the magnitude of the collision risk between the target vehicle 3 and the traffic participants around the target vehicle 3. As described above, in the present embodiment, the traffic participants 7 are the preceding vehicle 7a and the following vehicle 7b.

[0056] In the present embodiment, the risk index values ​​are, for example, the collision margin time between the front vehicle 7a and the target vehicle 3, that is, the front collision margin time, and the collision margin time between the following vehicle 7b and the target vehicle 3, that is, the rear collision margin time. Thus, the contact risk between the target vehicle 3 and other vehicles, that is, the traffic participants 7, existing around the target vehicle 3 can be appropriately grasped as an objective numerical value.

[0057] Hereinafter, the front collision margin time and the rear collision margin time are referred to as front TTC and rear TTC, respectively. Figure 2 ).

[0058] The emotion estimation unit 36 ​​calculates the emotion level, which is a numerical value indicating the magnitude of the unpleasant emotion currently held by the driver D. Specifically, the emotion estimation unit 36 ​​calculates the emotion level based on, for example, the physiological data of the driver D and / or the action schedule of the driver D. The physiological data of the driver D can be obtained from the seat sensor 22 and / or the wearable device 24. In addition, the action schedule of the driver D can be obtained from the portable terminal 25.

[0059] In the present embodiment, the emotion estimation unit 36 ​​calculates, for example, a value "1" indicating an unpleasant emotion and a value "0" indicating an emotion that is not unpleasant ("pleasant" or "neutral emotion that is neither pleasant nor unpleasant") as the emotion level of the driver D. Specifically, for example, as a numerical value indicating an excited state, the emotion estimation unit 36 ​​divides the range of values ​​of blood pressure and respiratory rate, which are physiological data of the driver D, into a range of five stages from a quiet level to an excited level, and assigns a value of 1 to 5 to each stage. In addition, the emotion estimation unit 36 ​​divides the length of the margin time until the next event shown in the action schedule into five stages from margin to urgency, and assigns a value of 1 to 5 to each stage. In addition, the emotion estimation unit 36 ​​calculates a weighted addition value of the numerical value indicating the excited state and the numerical value indicating the temporal urgency. If the calculated weighted addition value is greater than a predetermined threshold, the emotion estimation unit 36 ​​sets the emotion level to "1" (unpleasant), and if the weighted addition value is less than the threshold, the emotion estimation unit 36 ​​sets the emotion level to "0" (pleasant or neutral). The weight used in the weighted addition can be predetermined.

[0060] In addition to the above, when music is output from the first speaker 27 of the helmet 23, the emotion estimation unit 36 ​​can consider the output of the music as a positive element that the emotion of the driver D is toward the "happy" direction, and reduce the value of the emotion level. Information on whether music is being output from the first speaker 27 can be obtained from the portable terminal 25, for example, when the music is output from the portable terminal 25 to the first speaker 27.

[0061] The probability estimation unit 37 estimates the probability that the driver D of the subject vehicle 3 performs a predetermined control action in the subject vehicle 3, i.e., the action probability, based on the risk index value calculated by the risk calculation unit 35 and the driver characteristics 33 stored in the first memory 31. As described above, in the present embodiment, the predetermined control action is, for example, a lane change action for changing the driving lane of the subject vehicle 3 from the lane 12a to the lane 12b in order to avoid the front vehicle 7a traveling in the lane 12a.

[0062] The driver characteristics 33 are characteristics of the driver D of the target vehicle 3. In the present embodiment, the driver characteristics 33 include a confidence level, which is a numerical value indicating the degree of confidence of the driver D in his / her driving skills, and a dangerous situation resolute execution level, which is a numerical value indicating the degree of the tendency to resolutely execute an action despite knowing the danger. Thus, the action probability of the driver D of the target vehicle 3 executing a prescribed maneuvering action according to the collision risk with the traffic participant 7 can be appropriately estimated by taking into account the characteristics such as the personality of the driver D of the target vehicle 3.

[0063] The confidence level and the risk-taking level are calculated in advance for each driver D and stored in the first memory 31 as part of the driver characteristics 33. Thus, even if the traffic conditions are the same, the action probability that may differ according to the characteristics of each driver can be appropriately estimated according to the characteristics of each driver.

[0064] The confidence level can be predetermined, for example, based on the score of each driver D's answer (e.g., answer of 5-level evaluation) related to a driving style checklist (RSQ: Rider Skill Questationnaire or DSQ: Driver Skill Questationnaire) according to the prior art. The driving style checklist is composed of a plurality of questions related to predetermined driving skills, such as "Do you care about causing a car accident while driving?"

[0065] In addition, the level of resoluteness in dangerous situations can be predetermined, for example, based on the scores of answers of each driver D related to a risk propensity questionnaire (RPQ) according to the prior art. The risk propensity questionnaire is composed of a plurality of predetermined questions about traffic actions in traffic scenes that may include dangerous situations, such as "While walking, whether to cross the road even if the traffic light is red and no car is coming".

[0066] The probability estimation unit 37 may estimate the action probability based on the emotion level calculated by the emotion estimation unit 36 ​​in addition to the risk index value and the driver characteristics 33. Thus, even if the traffic conditions are the same, the action probability that may differ depending on the emotional state of each driver at that time can be appropriately estimated by taking into account the current emotional state of the driver.

[0067] As described above, the emotion level can be calculated as a value corresponding to the excitement state and / or the time urgency state of the driver D. Therefore, by estimating the action probability in consideration of the emotion level, it is also possible to more appropriately estimate the action probability in consideration of the driver D's excitement state and / or the sense of urgency caused by the action schedule.

[0068] In the present embodiment, the probability estimation unit 37 calculates the probability that the driver D will perform the lane change action of the target vehicle 3, that is, the action probability, based on the risk index value, the emotion level, and the confidence level and the resolute execution level in dangerous situations as the driver characteristics 33, using the probability estimation model 34 stored in the first memory 31. Thus, the action probability of the lane change action of the target vehicle 3, which is likely to affect the characteristics of the driver D such as the personality of the target vehicle 3 and has a great influence on the surrounding traffic participants 7, can be appropriately estimated in consideration of the characteristics of the driver D.

[0069] The probability estimation model 34 is a model that represents the relationship between the risk index value, the emotion level, the confidence level of the driver D as various individuals, and the resolute execution level in dangerous situations, and the action probability of the driver D performing a predetermined driving action (in this embodiment, a lane change action). The probability estimation model 34 is, for example, a logistic regression model, and the partial regression coefficient of the logistic function representing the action probability is given as a function of the risk index value, the emotion level, and the confidence level and resolute execution level of the driver D in dangerous situations. The above function is, for example, a weighted addition of the risk index value, the emotion level, the confidence level, and the resolute execution level in dangerous situations.

[0070] This makes it possible to objectively estimate the probability of an action that may occur as a result of multiple factors using a statistical method.

[0071] Specifically, the logistic regression model given by the probability estimation model 34 is given by the following equations (1) and (2).

[0072] [Formula 1]

[0073]

[0074] in,

[0075] Z=a1x1+a2x2+a3x3+a4x4+a5x5 (2)

[0076] Here, x1 and x2 are the front TTC and rear TTC, respectively, which are risk index values. In addition, x3 and x4 are the confidence level and the resolute execution level in dangerous situations, respectively, which are the driver characteristics 33. In addition, x5 is the emotional level. a1, a2, a3, a4, and a5 in formula (1) are weighting coefficients determined by logistic regression analysis of the relationship between the front TTC, the rear TTC, the confidence level, the resolute execution level in dangerous situations, and the emotional level and the action probability. In addition, when the emotional level is not used when estimating the action probability, the last term a5x5 shown in formula (2) is not used.

[0077] The probability estimation unit 37 outputs the calculated action probability to the first notification device 4 described later.

[0078] [3. Structure of traffic safety assistance system]

[0079] The traffic safety assistance system 2 includes an action probability estimation device 1 and a first notification device 4 .

[0080] The first notification device 4 notifies the driver D of the target vehicle 3 and / or the road user 7 .

[0081] The traffic safety assistance system 2 makes various notifications for traffic safety assistance to the driver D of the target vehicle 3 and / or the traffic participant 7 through the first notification device 4 based on the action probability estimated by the action probability estimation device 1 .

[0082] In the present embodiment, the traffic safety assistance system 2 makes a notification for traffic safety assistance to the target vehicle 3 and / or the following vehicle 7 b based on the action probability of the lane change action of the target vehicle 3 estimated by the action probability estimation device 1 .

[0083] The first notification device 4 includes a second processor 40 and a second memory 41. The second memory 41 is composed of, for example, a volatile and / or nonvolatile semiconductor memory, etc. A second program 42 is stored in the second memory 41.

[0084] The second processor 40 is, for example, a computer including a CPU, etc. The second processor 40 may also include a ROM in which a program is written, a RAM for temporarily storing data, etc. Furthermore, the second processor 40 includes a first notification unit 43 as a functional element or a functional unit.

[0085] The first notification unit 43 is implemented by the second processor 40 as a computer executing the second program 42 stored in the second memory 41. In addition, the second program 42 can be pre-stored in any computer-readable storage medium. Alternatively, all or part of the above-mentioned functional elements of the second processor 40 can be constituted by hardware each including one or more electronic circuit components.

[0086] When the rear TTC is less than a predetermined first time threshold (e.g., less than 6 seconds), the first notification unit 43 performs a first level attention calling notification to the target vehicle 3 and the following vehicle 7b when the action probability of the target vehicle 3 is less than a predetermined probability threshold. In addition, when the rear TTC is less than a predetermined first time threshold, the first notification unit 43 performs a second level attention calling notification having a higher significance level than the first level attention calling notification to the target vehicle 3 and the following vehicle 7b when the action probability of the target vehicle 3 is greater than the probability threshold.

[0087] Thus, based on the estimated action probability of the target vehicle 3 taking into account the characteristics of the driver D, the attention-calling notification is not only given to the target vehicle 3, but also to the following vehicle 7b which is the traffic participant 7, so that it is possible to appropriately and in advance prevent the driver D of the target vehicle 3 from being in a dangerous state that may occur according to his characteristics and situation. In addition, since the attention-calling notification based on the high-precision action probability estimation performed by the action probability estimation device 1 taking into account the characteristics of the driver D is provided to the driver D, the recognition of the notification by the driver D and the traffic participant 7 can be improved. As a result, the trust of the driver D and the traffic participant 7 in the traffic safety assistance system 2 can be improved, and the effect of the traffic safety assistance based on the attention-calling notification can be improved.

[0088] In the above, specifically, the first level of attention calling notification and the second level of attention calling notification in the following vehicle 7b can be respectively performed by the first notification unit 43 using the first communication device 6 to send a first notification instruction instructing to perform the first level of attention calling notification and a second notification instruction instructing to perform the second level of attention calling notification to the following vehicle 7b.

[0089] As described later, upon receiving the first notification instruction, the second notification device 8 of the following vehicle 7b, which is the traffic participant 7, uses the second HMI device 9, i.e., the display device 50 and / or the second speaker 51, mounted on the following vehicle 7b, to perform a first-level attention-calling notification in the following vehicle 7b. In addition, upon receiving the second notification instruction, the second notification device 8 of the following vehicle 7b uses the second HMI device 9, i.e., the display device 50 and / or the second speaker 51, mounted on the following vehicle 7b, to perform a second-level attention-calling notification in the following vehicle 7b.

[0090] Here, the significance level of the attention-calling notification refers to the intensity of the stimulation to the auditory or visual sense of a person output by the first HMI device 5 and the second HMI device 9 as the attention-calling notification to attract or draw the attention of the person.

[0091] For example, the significance level of the sound, i.e., the auditory stimulation, outputted from the first speaker 27 or the second speaker 51 corresponds to the intensity of the sound, the height of the frequency, the length of the repetition period, and / or the length of the change period of the intensity or frequency of the sound. The greater the intensity of the sound, the higher the frequency, the shorter the above-mentioned repetition period, and / or the shorter the change period of the intensity or frequency, the higher the significance level of the auditory stimulation given by the sound.

[0092] In addition, for example, when the visual information is output as graphic elements such as text and graphics displayed on these display devices, the significance level of the visual information displayed by the HUD 28 or the display device 50 can be determined based on the brightness, brightness change cycle, flicker cycle or color tone of the displayed graphic elements. For example, the higher the brightness, the shorter the brightness change or flicker cycle, or the closer the color tone is from cold color to warm color, the higher the significance level of the visual information.

[0093] Alternatively, in the case of notification by tactile stimulation as vibration from a steering wheel or the like, the significance level of the tactile stimulation corresponds to the intensity of the vibration, the height of the frequency, the length of the repetition period, and / or the length of the intensity or frequency variation period. The greater the intensity of the vibration, the higher the frequency, the shorter the above-mentioned repetition period, and / or the shorter the intensity or frequency variation period, the higher the significance level of the tactile stimulation imparted by the vibration.

[0094] Alternatively, when the tactile stimulation is output as the tension of the electric seat belt, the significance level of the tactile stimulation corresponds to the magnitude of the tension. The greater the tension, the higher the significance level of the tactile stimulation given by the tension.

[0095] In addition, in each of the first level attention calling notification and the second level attention calling notification, the details of the form of auditory stimulation and / or visual information output by the first HMI device 5 and the second HMI device 9 can be predetermined in the first notification device 4 and the second notification device 8. The same applies to the case where the first HMI device 5 and the second HMI device 9 perform notification by tactile stimulation.

[0096] In addition, when the rear TTC is greater than or equal to the first time threshold (for example, greater than or equal to 6 seconds) and less than or equal to a predetermined second time threshold (for example, less than or equal to 10 seconds) that is greater than the first time threshold, when the action probability of the target vehicle 3 is less than the probability threshold, the first notification unit 43 notifies the following vehicle 7b of an information prompt indicating the presence of the target vehicle 3. In addition, when the rear TTC is greater than or equal to the first time threshold and less than or equal to the second time threshold, when the action probability of the target vehicle 3 is greater than or equal to the probability threshold, the first notification unit 43 notifies the following vehicle 7b of a first-level attention call. The notification of the information prompt may be, for example, an information message such as "a two-wheeled motor vehicle is traveling in the adjacent lane ahead".

[0097] Therefore, when the following vehicle 7b may have enough time to take evasive action, only information prompts or notifications to draw attention are provided to the following vehicle 7b, without troubling the driver D of the target vehicle 3, and the risk of contact accidents can be effectively reduced.

[0098] In addition, specifically, the notification of the information prompt in the following vehicle 7b and the first level of attention call notification can be respectively performed by the first notification unit 43 sending an information notification instruction to instruct the notification of the information prompt and the above-mentioned first notification instruction to the following vehicle 7b using the first communication device 6. The information notification instruction can include, for example, the above-mentioned information message.

[0099] For example, when the first notification unit 43 sends the first notification instruction, the second notification instruction, and the information notification instruction to the second notification device 8 of a specific traffic participant 7 such as the following vehicle 7b, the first notification unit 43 uses the access information of the second communication device 10 of the traffic participant 7 (for example, the identification ID such as the communication address of the second communication device 10) to establish communication with the second communication device 10. When the traffic participant 7 is a vehicle, the access information of the second communication device 10 of the traffic participant 7 can be associated with the vehicle number (license plate) of the vehicle in advance, for example. Information on the correspondence between the vehicle number and the access information can be obtained, for example, from a server (not shown) that provides the information.

[0100] In the present embodiment, the first notification unit 43 sends the first notification instruction, the second notification instruction, and the information notification instruction to the following vehicle 7 b as the traffic participant 7 , but the destinations of these instructions may include other traffic participants 7 including the preceding vehicle 7 a .

[0101] [4. Device structure among traffic participants]

[0102] Figure 4 1 is a diagram showing an example of a device configuration in a traffic participant 7. The traffic participant 7 includes a second notification device 8, a display device 50, a second speaker 51, and a second communication device 10. The display device 50 and the second speaker 51 are examples of a second HMI device 9. In the present embodiment, the traffic participant 7 is a front vehicle 7a and a following vehicle 7b, and the second notification device 8, the display device 50, the second speaker 51, and the second communication device 10 are configured as, for example, vehicle-mounted devices. For example, the display device 50 and the second speaker 51 may be provided on a dashboard in a vehicle compartment.

[0103] The second notification device 8 includes a third processor 60 and a third memory 61. The third memory 61 is composed of, for example, a volatile and / or nonvolatile semiconductor memory, etc. A third program 62 is stored in the third memory 61.

[0104] The third processor 60 is, for example, a computer including a CPU, etc. The third processor 60 may also have a ROM in which a program is written, a RAM for temporarily storing data, etc. Furthermore, the third processor 60 includes a second notification unit 63 as a functional element or a functional unit.

[0105] The second notification unit 63 is implemented by the third processor 60 as a computer executing the third program 62 stored in the third memory 61. In addition, the third program 62 can be pre-stored in any computer-readable storage medium. Alternatively, all or part of the above-mentioned functional elements of the third processor 60 can be constituted by hardware each including one or more electronic circuit components.

[0106] When receiving the first notification instruction from the first notification device 4 of the target vehicle 3, the second notification unit 63 uses the display device 50 and / or the second speaker 51 as the second HMI device 9 to perform a first-level attention-calling notification to the traffic participant 7. In addition, when receiving the second notification instruction from the first notification device 4 of the target vehicle 3, the second notification unit 63 uses the display device 50 and / or the second speaker 51 to perform a second-level attention-calling notification having a higher significance level than the first-level attention-calling notification to the traffic participant 7. In addition, when receiving the information notification instruction from the first notification device 4 of the target vehicle 3, the second notification unit 63 uses the display device 50 and / or the second speaker 51 to perform a notification of information presentation indicating the presence of the target vehicle 3 to the traffic participant 7.

[0107] The forms of the first level attention calling notification and the second level attention calling notification output by the second HMI device 9 for the traffic participant 7 can be predetermined in the second notification device 8 .

[0108] [4. Operation of traffic safety assistance system]

[0109] Next, the operation procedure of the traffic safety assistance system 2 will be described.

[0110] Figure 5 This is a flowchart showing the operation procedure of the traffic safety assistance system 2 including the action probability estimation device 1 . Figure 5 The illustrated processing is executed by the first processor 30 and the second processor 40 which are computers included in the traffic safety assistance system 2 . Figure 5 The processing shown is started at least when the behavior probability estimation device 1 of the target vehicle 3 and the first notification device 4 are powered on, and is repeatedly executed.

[0111] When the process starts, first, the risk calculation unit 35 of the action probability estimation device 1 calculates a risk index value indicating the magnitude of the risk of collision between the traffic participants 7 around the target vehicle 3 and the target vehicle 3 (S100). In the present embodiment, the risk index value is the front TTC between the front vehicle 7a traveling in front of the target vehicle 3 and the target vehicle 3, and the rear TTC between the follow-up vehicle 7b traveling behind the target vehicle 3 and the target vehicle 3.

[0112] Next, the emotion estimation unit 36 ​​calculates the emotion level, which is a numerical value indicating the magnitude of the unpleasant emotion currently held by the driver D (S102). Next, the probability estimation unit 37 calculates the action probability, which is the probability that the driver D of the target vehicle 3 performs a predetermined specified operation action in the target vehicle 3 (S104). For example, the probability estimation unit 37 obtains the confidence level and the resolute execution level of the driver D in dangerous situations with reference to the driver characteristics 33 stored in the first memory 31. Then, the probability estimation unit 37 calculates the action probability of the target vehicle 3 performing the lane change action based on the front TTC and the rear TTC calculated in step S100, the emotion level calculated in step S102, and the confidence level and the resolute execution level in dangerous situations obtained based on the driver characteristics 33, using the probability estimation model 34 stored in the first memory 31 (for example, the above-mentioned formulas (1) and (2)).

[0113] The action probability estimation device 1 transmits the risk index value calculated by the risk calculation unit 35 and the action probability calculated by the probability estimation unit 37 to the first notification device 4 .

[0114] Next, the first notification unit 43 of the first notification device 4 determines whether the rearward TTC is less than the first time threshold Tth1 (S106). When the rearward TTC is less than the first time threshold Tth1 (S106, Yes), the first notification unit 43 determines whether the action probability is less than the probability threshold Pth (S110).

[0115] Then, when the action probability is less than the probability threshold value Pth in step S110 (S110, Yes), the first notification unit 43 uses the first HMI device 5 of the target vehicle 3 to perform a first level attention-calling notification to the driver D (S112). In the present embodiment, specifically, the first HMI device 5 is the first speaker 27 and HUD 28 of the helmet 23 worn by the driver D of the target vehicle 3, which is a motorcycle.

[0116] Furthermore, the first notification unit 43 transmits a first notification instruction to the following vehicle 7b to instruct the following vehicle 7b to perform a first-level attention calling notification (S114), and ends the present process.

[0117] On the other hand, when the action probability is greater than or equal to the probability threshold value Pth (S110, No), the first notification unit 43 uses the first HMI device 5 of the target vehicle 3 to perform a second-level attention-calling notification having a higher significance level than the first-level attention-calling notification (S116) to the driver D of the target vehicle 3. In addition, the first notification unit 43 transmits a second notification instruction to the following vehicle 7b to instruct the following vehicle 7b to perform the second-level attention-calling notification (S118), and ends the present process.

[0118] On the other hand, when the rearward TTC is greater than or equal to the first time threshold value Tth1 in step S106 (S106, No), the first notification unit 43 determines whether the rearward TTC is less than or equal to the second time threshold value Tth2 which is greater than the first time threshold value Tth1 (S108). When the rearward TTC is equal to or less than the second time threshold value Tth2 (S108, Yes), the first notification unit 43 determines whether the action probability is less than the probability threshold value Pth (S120).

[0119] Then, when the action probability is less than the probability threshold value Pth (S120, YES), the first notification unit 43 transmits an information notification instruction to the following vehicle 7b to instruct the following vehicle 7b to notify information presentation (S122), and ends the present process.

[0120] On the other hand, when the action probability is equal to or greater than the probability threshold value Pth (S120, No), the first notification unit 43 transmits a first notification instruction to instruct the following vehicle 7b to perform a first-level attention calling notification (S124), and ends the present process.

[0121] On the other hand, when the rear TTC exceeds the second time threshold value Tth2 in step S108 ( S108 , No), the first notification unit 43 ends the present process.

[0122] In addition, Figure 5 In the embodiment, steps S100 to S104 correspond to the action probability estimation method executed by the computer of the action probability estimation device 1, namely, the first processor 30. In addition, steps S100 and S104 correspond to the risk calculation step and the estimation step in the present disclosure, respectively.

[0123] [5. Other embodiments]

[0124] In the above-mentioned embodiment, the driver characteristics 33 and / or the probability estimation model 34 are pre-stored in the first memory 31 of the action probability estimation device 1, but the first processor 30 (eg, the probability estimation unit 37) can also download them from a server (not shown) via a communication network.

[0125] The action probability estimation device 1 and the traffic safety assistance system 2 may also be composed of a server device that is communicatively connected to the target vehicle 3 and the traffic participant 7 via a communication network. In this case, the risk calculation unit 35 may also calculate the risk index value based on an image or video acquired using a traffic infrastructure such as a street light camera. In addition, the emotion estimation unit 36 ​​can receive the sensor output of the seat sensor 22 and the like from the target vehicle 3 and calculate the emotion level.

[0126] The probability estimation model 34 is not limited to a model expressed by an analytical formula such as a logistic regression model, and may be a model generated by machine learning using a neural network.

[0127] In addition, the present invention is not limited to the configurations of the above-described embodiments, and can be implemented in various forms within the scope not departing from the gist of the present invention.

[0128] [6. Structure supported by the above-mentioned embodiment]

[0129] The above-mentioned embodiment supports the following structure.

[0130] (Structure 1) An action probability estimation device, comprising: a risk calculation unit, which calculates a risk index value, which is a numerical value indicating the degree of the risk of collision between traffic participants around a target vehicle and the target vehicle; and a probability estimation unit, which estimates an action probability based on the risk index value and the characteristics of the driver of the target vehicle, i.e., the driver characteristics, wherein the action probability is the probability that the driver performs a predetermined specified operating action in the target vehicle, wherein the driver characteristics include: a confidence level, which is a numerical value indicating the degree of confidence of the driver in his or her own driving skills; and a resolute execution level in dangerous situations, which is a numerical value indicating the degree of the tendency to resolutely execute an action knowing that the danger is dangerous.

[0131] According to the action probability estimation device of configuration 1, it is possible to appropriately estimate the action probability that the driver of the target vehicle will perform a predetermined steering action according to the collision risk with a traffic participant, taking into account characteristics such as the personality of the driver of the target vehicle.

[0132] (Structure 2) The action probability estimation device described in Structure 1, wherein the confidence level and the resolute execution level in the dangerous situation are calculated in advance for each person who becomes the driver and stored in a storage device as part of the driver's characteristics.

[0133] According to the action probability estimation device of Configuration 2, even if the traffic conditions are the same, it is possible to appropriately estimate the action probability that may differ depending on the characteristics of each driver.

[0134] (Structure 3) An action probability estimation device as described in Structure 1 or 2, wherein the action probability estimation device further comprises an emotion estimation unit, which calculates an emotion level, wherein the emotion level is a numerical value indicating the degree of magnitude of unpleasant emotions currently possessed by the driver, and the probability estimation unit estimates the action probability based on the driver characteristics, the risk indicator value, and the emotion level.

[0135] According to the action probability estimation device of Configuration 3, even if the traffic conditions are the same, it is possible to appropriately estimate the action probability that may differ depending on the emotional state of each driver at that time, taking into account the current emotional state of the driver.

[0136] (Structure 4) The action probability estimation device according to Structure 3, wherein the emotion estimation unit calculates the emotion level based on the physiological data of the driver and / or the action schedule of the driver.

[0137] According to the action probability estimation device of configuration 4, it is also possible to appropriately estimate the action probability by taking into account the driver's mental state such as the excitement state and / or the sense of urgency caused by the action schedule.

[0138] (Structure 5) The action probability estimation device described in any one of Structures 1 to 4, wherein the action probability is estimated using a logistic regression model, and the logistic regression model is a logistic function.

[0139] According to the action probability estimation device of structure 5, it is possible to objectively estimate the action probability that may occur as a result of multiple main causes using a statistical method.

[0140] (Configuration 6) The action probability estimation device according to any one of Configurations 1 to 5, wherein the predetermined control action is a lane change action of changing a driving lane of the target vehicle.

[0141] According to the action probability estimation device of configuration 6, it is possible to appropriately estimate the action probability of the lane change action of the target vehicle which is likely to affect the characteristics such as the personality of the driver of the target vehicle and has a great impact on surrounding traffic participants, taking into account the characteristics of the driver.

[0142] (Structure 7) An action probability estimation device as described in any one of Structures 1 to 6, wherein the risk index value is represented by a front collision margin time and a rear collision margin time, the front collision margin time being a collision margin time between a front vehicle traveling in front of the target vehicle as the traffic participant and the target vehicle, and the rear collision margin time being a collision margin time between a subsequent vehicle traveling behind the target vehicle as the traffic participant and the target vehicle.

[0143] According to the behavior probability estimation device of Configuration 7, when the traffic participant existing around the target vehicle is another vehicle, the contact risk between the target vehicle and the traffic participant can be appropriately grasped as an objective numerical value.

[0144] (Structure 8) The action probability estimation device described in Structure 7, wherein the target vehicle is a two-wheeled motor vehicle and the following vehicle is a four-wheeled automobile.

[0145] According to the action probability estimation device of structure 8, a motor two-wheeled vehicle, which is more difficult to control and intervene in the vehicle body movement than a four-wheeled vehicle and is more important for calling the driver's attention, is taken as the target vehicle, and the action probability of the driver is appropriately estimated, thereby enabling more appropriate assistance for safe driving.

[0146] In addition, according to the action probability estimation device of structure 8, in a traffic scene including motor two-wheeled vehicles and four-wheeled automobiles where coordinated safety assistance is difficult to perform due to differences in the maneuverability of the vehicle bodies, the above-mentioned coordinated traffic assistance can be implemented based on the action probability appropriately estimated for the motor two-wheeled vehicles.

[0147] (Structure 9) An action probability estimation method, which is executed by a computer and estimates the probability that the driver of a target vehicle will perform a prescribed manipulation action, i.e., the action probability, wherein the method comprises the following steps: a risk calculation step, calculating a risk index value, wherein the risk index value is a numerical value indicating the degree of the risk of collision between traffic participants around the target vehicle and the target vehicle; and an estimation step, estimating the action probability based on the risk index value and the characteristics of the driver of the target vehicle, i.e., the driver characteristics, wherein the action probability is the probability that the driver will perform a predetermined prescribed manipulation action in the target vehicle, wherein the driver characteristics include: a confidence level, which is a numerical value indicating the degree of confidence of the driver in his or her own driving skills; and a resolute execution level in dangerous situations, which is a numerical value indicating the degree of the tendency to resolutely execute an action knowing that the danger is dangerous.

[0148] According to the action probability estimation method of configuration 9, the action probability that the driver of the target vehicle performs a predetermined steering action according to the collision risk with a traffic participant can be appropriately estimated by taking into account the characteristics of the driver of the target vehicle, such as the personality.

[0149] (Structure 10) A traffic safety assistance system, comprising: an action probability estimation device as described in Structure 7 or 8; and a notification device, which, based on the action probability estimated by the action probability estimation device, issues an attention-calling notification to the driver of the target vehicle and / or the traffic participant, wherein the notification device performs the following processing when the rear collision margin time is less than a predetermined first time threshold: when the action probability of the target vehicle is less than a predetermined probability threshold, a first level of attention-calling notification is issued to the target vehicle and the subsequent vehicle; when the action probability of the target vehicle is greater than the probability threshold, a second level of attention-calling notification having a higher significance level than the first level of attention-calling notification is issued to the target vehicle and the subsequent vehicle.

[0150] According to the traffic safety assistance system of structure 10, based on the estimated probability of the action of the target vehicle taking into account the characteristics of the driver, an attention-calling notification is not only given to the target vehicle, but also to the following vehicles as traffic participants, thereby being able to appropriately and in advance prevent dangerous conditions that may occur to the driver of the target vehicle based on his characteristics and conditions.

[0151] (Structure 11) A traffic safety assistance system described in Structure 10, wherein the notification device performs the following processing when the rear collision margin time is greater than the first time threshold and less than a predetermined second time threshold that is greater than the first time threshold: when the action probability of the object vehicle is less than the probability threshold, a notification of an information prompt indicating the existence of the object vehicle is performed in the subsequent vehicle; and when the action probability of the object vehicle is greater than the probability threshold, the first level of attention-calling notification is performed in the subsequent vehicle.

[0152] According to the traffic safety assistance system of structure 11, when there may be time margin for the following vehicle to take evasive action, only information prompts or attention-calling notifications are provided to the following vehicle, which will not trouble the driver of the target vehicle and can effectively reduce the risk of contact accidents.

Claims

1. An action probability estimation device, comprising: a risk calculation unit that calculates a risk index value, the risk index value being a numerical value indicating a degree of magnitude of a collision risk between a traffic participant around the target vehicle and the target vehicle; as well as a probability estimation unit that estimates an action probability, which is a probability that the driver performs a predetermined control action in the target vehicle, based on the risk index value and a characteristic of the driver of the target vehicle, namely, a driver characteristic; The driver characteristics include: a confidence level, which is a numerical value indicating the degree of confidence of the driver in his or her own driving skills; The risk-taking level is a numerical value indicating the degree of the tendency to act resolutely despite knowing the danger.

2. The action probability estimation device according to claim 1, wherein: The confidence level and the risk-taking level are calculated in advance for each driver and stored in a storage device as part of the driver's characteristics.

3. The action probability estimation device according to claim 1, wherein: The action probability estimation device further includes an emotion estimation unit that calculates an emotion level, the emotion level being a numerical value indicating the magnitude of an unpleasant emotion currently possessed by the driver. The probability estimation unit estimates the action probability based on the driver characteristics, the risk index value, and the emotion level.

4. The action probability estimation device according to claim 3, wherein: The emotion estimation unit calculates an emotion level based on the physiological data of the driver and / or the action schedule of the driver.

5. The action probability estimation device according to claim 1, wherein: The action probabilities are estimated using a logistic regression model, which is a logistic function.

6. The action probability estimation device according to claim 1, wherein: The predetermined maneuvering action is a lane changing action for changing a driving lane of the target vehicle.

7. The action probability estimation device according to claim 1, wherein: The risk index value is represented by a front collision margin time and a rear collision margin time, wherein the front collision margin time is a collision margin time between the front vehicle, which is the traffic participant, and the target vehicle, and the rear collision margin time is a collision margin time between the follow-up vehicle, which is the traffic participant, and the target vehicle.

8. The action probability estimation device according to claim 7, wherein: The target vehicle is a two-wheeled motor vehicle, and the following vehicle is a four-wheeled automobile.

9. An action probability estimation method, which is executed by a computer and estimates the probability that a driver of a target vehicle will perform a prescribed control action, that is, the action probability, wherein: The following steps are involved: a risk calculation step of calculating a risk index value, the risk index value being a numerical value indicating the magnitude of a risk of collision between the target vehicle and traffic participants around the target vehicle; as well as an estimating step of estimating an action probability based on the risk index value and the characteristics of the driver of the target vehicle, namely, the driver characteristics, wherein the action probability is the probability that the driver performs a predetermined specified operating action in the target vehicle, The driver characteristics include: a confidence level, which is a numerical value indicating the degree of confidence of the driver in his or her own driving skills; The risk-taking level is a numerical value indicating the degree of the tendency to act resolutely despite knowing the danger.

10. A traffic safety assistance system comprising: The action probability estimation device according to claim 7 or 8; and a notification device for providing a notification to call attention to the driver of the target vehicle and / or the traffic participant based on the action probability estimated by the action probability estimation device, wherein: When the rear collision margin time is less than a predetermined first time threshold, the notification device performs the following processing: When the action probability of the target vehicle is less than a predetermined probability threshold, performing a first level of attention calling notification in the target vehicle and the following vehicle; When the action probability of the target vehicle is equal to or greater than the probability threshold, a second-level attention calling notification having a higher significance level than the first-level attention calling notification is performed in the target vehicle and the following vehicle.

11. The traffic safety assistance system according to claim 10, wherein: When the rear collision margin time is equal to or greater than the first time threshold and equal to or less than a predetermined second time threshold that is greater than the first time threshold, the notification device performs the following processing: When the action probability of the target vehicle is less than the probability threshold, notifying the following vehicle of information indicating the presence of the target vehicle; When the action probability of the target vehicle is equal to or greater than the probability threshold, the first level of attention calling notification is performed in the following vehicle.

Images