Driving assistance method and driving assistance device

Abstract

To provide driver assistance methods that can further improve driving safety. [Solution] The support unit 12 assists the driver in driving vehicle C. The discrimination unit 11 acquires brain activity data from the brain activity measurement unit 9, which measures the driver's brain activity, and determines, based on the acquired brain activity data, what state a predetermined specific driving behavior characteristic is in for the driver. The support unit 12 then determines the content of the driving assistance based on the determination result of the driving behavior characteristic state.

Description

Technical Field

[0001] The present disclosure relates to a driving assistance method and a driving assistance device.

Background Art

[0002] Conventionally, a device has been proposed that calculates a group to which a driver belongs based on the driver's image data, estimates the driver's reliability based on the calculated group, and provides an ADAS function based on the estimated reliability (see, for example, Patent Document 1). Patent Document 1 lists a group of Uber (registered trademark) drivers as an example of a group to which a driver belongs.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, for example, even among Uber drivers, there are those who are eager to drive and those who avoid driving. Therefore, when the device described in Patent Document 1 is applied to uniformly provide the same ADAS function to Uber drivers, the assistance of vehicle driving by the ADAS function may become inappropriate. As a result, it may be difficult to improve driving safety. An object of the present disclosure is to provide a driving assistance method and a driving assistance device that can further improve driving safety.

Means for Solving the Problems

[0005] One aspect of the present disclosure is a driving assistance method for assisting a driver in driving a vehicle, comprising: acquiring brain activity data from a measurement unit that measures the driver's brain activity; determining the state of a predetermined specific driving behavior characteristic for the driver based on the acquired brain activity data; and determining the content of the driving assistance based on the determination result of the state of the driving behavior characteristic.

[0006] Furthermore, a driver assistance device according to one aspect of this disclosure includes a support unit that assists the driver in driving a vehicle, a measurement unit that measures the driver's brain activity and generates brain activity data, and a discrimination unit that determines, based on the brain activity data generated by the measurement unit, what state a predetermined specific driving behavior characteristic is in for the driver, and the support unit determines the content of the driving assistance based on the discrimination result of the driving behavior characteristic state obtained by the discrimination unit. [Effects of the Invention]

[0007] This disclosure provides a driver assistance method and a driver assistance device that can further improve driving safety. [Brief explanation of the drawing]

[0008] [Figure 1] This figure shows the overall configuration of the driver assistance device according to the embodiment. [Figure 2] This diagram shows the various functions that an information processing device can perform. [Figure 3] This diagram shows the learning unit that trains the model. [Figure 4] This flowchart shows the overall flow of the driving assistance method according to the embodiment. [Figure 5] This flowchart shows the overall flow of the driving assistance method for variation (2). [Figure 6] This figure shows the relationship between response time and the total Hb level in the subjects' brains. [Figure 7] This flowchart shows the overall flow of the driving assistance method for variation (3). [Figure 8] This is a flowchart showing the overall flow of the driving assistance method for variation (4). [Modes for carrying out the invention]

[0009] The embodiments of this invention will be described in detail below with reference to the drawings. Note that the drawings are schematic and may differ from actual ones. Furthermore, the embodiments of the present invention described below are illustrative examples of devices and methods for realizing the technical concept of the present invention, and the technical concept of the present invention is not limited to the structure, arrangement, etc., of the components described below. The technical concept of the present invention can be modified in various ways within the technical scope defined by the claims described in the patent claims.

[0010] (Embodiment) In this embodiment, as shown in Figure 1, an example is given in which the driving assistance method and driving assistance device of this disclosure are applied to a driving assistance device 1 that is mounted on a vehicle C and assists the driver in driving the vehicle C. Figure 1 is a diagram showing the overall configuration of the driving assistance device 1 according to this embodiment. As shown in Figure 1, the driver assistance system 1 includes an external sensor 2, a vehicle sensor 3, an actuator 4, a human-machine interface (HMI) 5, a navigation device 6, a brain activity detection device 7, and an information processing device 8. The external sensor 2 is a sensor that detects various information (external environment information) about the external environment surrounding vehicle C. Examples of external sensor 2 include cameras, laser radar, millimeter-wave radar, and LiDAR. The detection results (external environment information) are output to the information processing device 8.

[0011] Vehicle sensor 3 is a sensor that detects various vehicle status information (vehicle information) obtained from vehicle C. Examples of vehicle sensors 3 include a vehicle speed sensor that detects the vehicle speed of vehicle C, a wheel speed sensor that detects the rotational speed of the vehicle C's tires, a three-axis acceleration sensor that detects the acceleration and deceleration of vehicle C in three axes, a steering angle sensor that detects the steering angle of the steering wheels, a gyro sensor that detects the angular velocity of vehicle C, a yaw rate sensor that detects the yaw rate of vehicle C, an accelerator sensor that detects the accelerator opening of vehicle C, and a brake sensor that detects the amount of brake operation of vehicle C. The detection results (vehicle information) are output to the information processing device 8. Actuator 4 comprises a steering actuator, an accelerator opening actuator, and a brake control actuator. The steering actuator controls the steering direction and amount of the steering wheel of vehicle C. The accelerator opening actuator controls the accelerator opening of vehicle C. The brake control actuator controls the brake system of vehicle C. Actuator 4 (including the steering actuator) controls the steering wheel, accelerator opening, and brake system of vehicle C in accordance with the control signals output from the information processing device 8, thereby generating predetermined vehicle behavior in vehicle C.

[0012] HMI5 is an interface device that exchanges information between the information processing device 8 and the driver of vehicle C. HMI5 can include, for example, a display that shows image information visible to the driver of vehicle C (e.g., a rear-view monitor, a back-view monitor, or an around-view monitor), a speaker that outputs audio information audible to the driver, and controls that the driver can operate (e.g., buttons, levers, or a touch panel). Examples of rear-view monitors, back-view monitors, and around-view monitors include monitors that can switch between displaying images of the vehicle C's surroundings captured by the camera of the external sensor 2 in two dimensions and three dimensions. For three-dimensional display, for example, glasses-free stereoscopic display (three-dimensional display without glasses) can be used. Navigation device 6 is a device that detects the current position of vehicle C using a positioning device, acquires map data for the detected current position, sets a target driving route to the destination entered by the driver, and provides route guidance to the driver according to the target driving route. As the target driving route, for example, the shortest route from the current position of vehicle C to the destination can be adopted. In route guidance, navigation device 6 provides display guidance and voice guidance of information on the route that vehicle C should take (hereinafter also referred to as "route information"). Route information includes, for example, guidance on the direction of travel (for example, guidance on turning right or left at intersections, and guidance on lane selection). In addition, in display guidance, route information is displayed as an image on the display of HMI 5. In addition, in voice guidance, route information is output as voice from the speaker of HMI 5. Furthermore, navigation device 6 can control (change) the target driving route according to the control signal output from information processing device 8.

[0013] The brain activity detection device 7 is a device for detecting the brain activity of a driver. The brain activity detection device 7 comprises a brain activity measurement unit 9 (or "measurement unit" in a broader sense) and a brain activity analysis unit 10. The brain activity measurement unit 9 is attached to the driver's head and measures the driver's brain activity. For example, it detects the activity of multiple regions of interest (ROIs) in the driver's brain and generates brain activity data showing the activity levels of the multiple ROIs. The brain activity measurement unit 9 can be, for example, an fNIRS (functional near-infrared spectroscopy) sensor, an electroencephalograph (EEG sensor), or a magnetoencephalograph (MEG sensor). In this embodiment, the case in which an fNIRS sensor is used as the brain activity detection device 7 is described, which irradiates near-infrared light and measures the concentration of hemoglobin in the blood of the brain (oxy-Hb, deoxy-Hb, total-Hb) from the intensity of the near-infrared light detected at a detection point at a predetermined distance from the irradiation point. Brain activity data is continuously measured while the brain activity measurement unit 9 is attached to the driver's head. The measurement results (brain activity data) are output to the brain activity analysis unit 10. The brain activity analysis unit 10 performs preprocessing (for example, removal of noise components) on the brain activity data generated by the brain activity measurement unit 9. The brain activity data after preprocessing is output to the information processing device 8.

[0014] The information processing device 8 includes a processor 8a and peripheral components such as a storage device 8b that stores computer programs and the like. As the processor 8a, for example, a CPU (Central Processing Unit) or an MPU can be adopted. Also, as the storage device 8b, for example, a semiconductor storage device, a magnetic storage device, or an optical storage device can be adopted. The storage device 8b may include memories such as registers, cache memories, a ROM and a RAM used as a main storage device. Each function of the information processing device 8 described below is realized, for example, when the processor 8a executes a computer program stored in the storage device 8b.

[0015] Next, each function of the information processing device 8 will be described in detail. As shown in FIG. 2, the information processing device 8 realizes the functions of a discrimination unit 11 and a support unit 12. FIG. 2 is a diagram showing each function realized by the information processing device 8. The discrimination unit 11 acquires brain activity data from the brain activity detection device 7 and, based on the acquired brain activity data, determines the state of a predetermined specific driving behavior characteristic for the driver. Specifically, the information processing device 8 first extracts feature quantities from the brain activity data used to determine the state of the driver's driving behavior characteristics. In extracting feature quantities from brain activity data, for example, first, brain activity data generated by the brain activity measurement unit 9 is acquired for a certain period of time T1 minutes (for example, 10 seconds) for each region of interest ROI. This obtains time-series data of brain activity data for each region of interest ROI. Subsequently, a feature extraction algorithm is applied to each acquired time-series data to obtain multiple types of features (for example, maximum value, minimum value, mean, median, mean, standard deviation, variance, trend) from a single time-series data. As a feature extraction algorithm, for example, an algorithm that extracts multiple types of features from time-series data can be used. For example, the FRESH (feature extraction based on scalable hypothesis tests) algorithm can be used. One example of a FRESH algorithm is tsfresh, a Python library. The total number of features extracted is "number of ROIs of interest" × "number of features obtained from one time series data".

[0016] Subsequently, the discrimination unit 11 uses the learned model 13 to discriminate the state of the driving behavior characteristics of the driver based on the feature amounts of the extracted brain activity data of the driver. As the model 13, as shown in FIG. 3, for example, a model that has been learned using the feature amounts of the brain activity data of a plurality of subjects as input data and the discrimination result of the state of the driving behavior characteristics of the subject as output data is used. FIG. 3 is a diagram showing a learning unit 14 that performs learning of the model 13. As the brain activity data of the subject, for example, brain activity data when the subject is made to perform a predetermined task by showing captured images obtained by a camera mounted on the vehicle can be adopted. Further, as the feature amounts used for learning the model 13, a plurality of types of feature amounts obtained by a feature amount extraction algorithm (such as the FRESH algorithm) that extracts a plurality of types of feature amounts from the time-series data of the brain activity data of the subject can be adopted. That is, the feature amounts of the brain activity data used for learning are obtained by acquiring the brain activity data for a fixed time T1 minutes for each region of interest ROI of the subject's brain, and applying the feature amount extraction algorithm to each of the acquired time-series data to obtain a plurality of types of feature amounts from one time-series data.

[0017] Furthermore, the results of determining the state of driving behavior characteristics used in learning Model 13 can be those obtained by existing driving behavior classification methods. Existing driving behavior classification methods include, for example, driving behavior classification methods that determine the state of driving behavior characteristics based on the subject's self-report. Examples include DBQ (Driver Behavior Questionnaire) and DSQ (Driving Style Questionnaire). When using DSQ, driver characteristics can include, for example, (1) confidence in driving skills, (2) passivity towards driving, (3) impatient driving tendency, (4) meticulous driving tendency, (5) preparatory driving in response to traffic signals, (6) car as a status symbol, (7) unstable driving tendency, and (8) anxious tendency. In addition, the state of driving behavior characteristics can be, for example, "high" or "low". That is, in determining the state of driving behavior characteristics, a driving behavior classification method such as DSQ is administered to the subject, and the determination result of the state of driving behavior characteristics (e.g., high or low confidence in driving skills) is obtained based on the subject's self-report. In DSQ, the determination of the state of driving behavior characteristics takes a numerical value from 1 to 4. Therefore, when using DSQ, if the determination of the state of driving characteristics by DSQ is above the threshold, it is judged as "high," and if it is below the threshold, it is judged as "low." Furthermore, the training of the trained model 13 is performed using, for example, a machine learning algorithm. One example is the SVM (Support Vector Machine) algorithm.

[0018] Here, we investigated the validity of the method for discriminating the state of driving behavior characteristics using the pre-trained model 13. Specifically, model 13 was trained using "confidence in driving skills" as the driving behavior characteristic, the "FRESH algorithm" as the feature extraction algorithm, and the "SVM algorithm" as the machine learning algorithm. When the state of "confidence in driving skills" was determined using the pre-trained model 13, the accuracy rate was 72.7%. Furthermore, when the driving behavior characteristic was changed to "unstable driving tendency" and model 13 was trained again, the accuracy rate for discriminating the state of motor behavior characteristics using the pre-trained model 13 was 63.4%. When the driving behavior characteristic was changed to "preparatory driving in response to traffic signals" and model 13 was trained again, the accuracy rate for discriminating the state of motor behavior characteristics using the pre-trained model 13 was 63.0%. Furthermore, when the driving behavior characteristic was changed to "passivity towards driving" and model 13 was trained again, the accuracy rate for discriminating the state of motor behavior characteristics using the pre-trained model 13 was 91.7%.

[0019] The support unit 12 assists the driver in operating vehicle C. For example, when the driving mode is switched from manual driving mode to automatic driving mode, the support unit 12 makes vehicle C autonomously drive according to the target driving route set by the navigation device 6. In autonomous driving mode, the support unit 12 uses the external sensor 2, vehicle sensor 3, and navigation device 6 to generate control signals for the actuator 4 so that vehicle C autonomously drives according to the target driving route. The support unit 12 also determines the content of the driving support based on the determination result of the driving behavior characteristics obtained by the discrimination unit 11. Once the content of the driving support is determined, the support unit 12 switches to the determined content of the driving support, or proposes to the driver that they switch to the determined content of the driving support. For example, if the discrimination unit 11 is configured to determine the level of the driver's "confidence in driving skills," and it is determined that the "confidence in driving skills" is low when in manual driving mode, the support unit 12 will determine the content of the driving support to be "automatic driving" and switch from "manual driving" to "automatic driving," or propose to the driver that they switch from "manual driving" to "automatic driving." Suggestions to the driver are made, for example, via the HMI5's display or speaker. On the other hand, if the system determines that the driver has high confidence in their driving skills, it will decide to provide driving assistance in the form of "manual driving" and continue with "manual driving."

[0020] (operation) Next, we will explain the process for determining the details of the driving assistance for vehicle C. First, when the driving mode of vehicle C is set to manual driving mode, the driver starts driving vehicle C (S101 in Figure 4). Next, measurement of the driver's brain activity begins (S102 in Figure 4). Brain activity is measured by the brain activity detection device 7 and is continuously measured while the brain activity measurement unit 9 is attached to the driver's head. Figure 4 is a flowchart showing the overall flow of the driving assistance method of this embodiment. Furthermore, the discrimination unit 11 of the information processing device 8 acquires brain activity data for a certain period of time T1 minutes (for example, 10 seconds) from the brain activity data obtained by the brain activity detection device 7 and extracts characteristic quantities of the acquired brain activity data (S103 in Figure 4).

[0021] Furthermore, the discrimination unit 11 uses a trained model 13 to determine the state of the driver's driving behavior characteristics based on the features of the extracted brain activity data (S104 in Figure 4). That is, it determines what state (e.g., high, low) a predetermined specific driving behavior characteristic (e.g., characteristics (1) to (8) above) is in for the driver. Next, the support unit 12 determines whether the driver's "confidence in driving skills" is low based on the discrimination result obtained by the discrimination unit 11 (S105 in Figure 4). If it determines that the "confidence in driving skills" is high (S105 "No" in Figure 4), it maintains manual driving (S106 in Figure 4), determines that driving has not been completed (S107 "No" in Figure 4), returns to S102, and repeats the flow from S102 to S107, starting with measuring brain activity for a certain period of time T1 minutes and extracting features of the brain activity data, thereby repeatedly determining the state of the driver's driving behavior characteristics at predetermined intervals (e.g., 10 seconds). On the other hand, if the discrimination unit 11 determines that the driver has low confidence in their driving skills (S105 "Yes" in Figure 4), it decides on "autonomous driving" as the content of the driving support and proposes to the driver that they switch from "manual driving" to "autonomous driving" (autonomous driving mode), or switch from "manual driving" to "autonomous driving" (autonomous driving mode) (S108 in Figure 4). Subsequently, if it determines that the driving has not been completed (S107 "No" in Figure 4), it returns to S102 and repeats the flow from S102 to S105, S108 and S107, starting with measuring brain activity for a certain period of time T1 minutes and extracting features from the brain activity data, thereby repeatedly determining the state of the driving behavior characteristics.

[0022] (Effects of this embodiment) (1) In this embodiment, the support unit 12 assists the driver in driving vehicle C. The discrimination unit 11 acquires brain activity data from the brain activity measurement unit 9, which measures the driver's brain activity, and determines, based on the acquired brain activity data, what state a predetermined specific driving behavior characteristic is in for the driver. The support unit 12 then determines the content of the driving assistance based on the determination result of the state of the driving behavior characteristic. Therefore, in this embodiment, driving assistance can be provided according to the driver's driving behavior characteristics. As a result, more appropriate driving assistance can be provided to the driver. This can further improve driving safety.

[0023] (2) In this embodiment, the brain activity measurement unit 9 is an fNIRS sensor, an electroencephalograph, or a magnetoencephalograph. This reduces the burden on the driver compared to other measurement devices, for example.

[0024] (3) In this embodiment, in determining the state of driving behavior characteristics, a model 13 that has been trained with the feature quantities of brain activity data of multiple subjects as input data and the determination result of the state of the subject's driving behavior characteristics as output data is used to determine the state of the driver's driving behavior characteristics based on the feature quantities of the driver's brain activity data. This improves the accuracy of determining driving behavior characteristics.

[0025] (4) In this embodiment, the discrimination result used for learning Model 13 is the discrimination result of the state of driving behavior characteristics obtained by a driving behavior classification method that discriminates the state of driving behavior characteristics based on the subject's self-report. This allows for more appropriate learning.

[0026] (5) In this embodiment, the features used for training Model 13 are multiple types of features obtained by a feature extraction algorithm that extracts multiple types of features from the time-series data of the subject's brain activity. This allows for more appropriate training.

[0027] (6) In this embodiment, the training of model 13 is performed using a machine learning algorithm. This makes it possible to train model 13 more appropriately and easily.

[0028] (7) In addition, in this embodiment, once the content of the driving assistance is determined, the system switches to the determined driving assistance, or suggests to the driver that the system switch to the determined driving assistance. This allows the content of the driving assistance to be changed appropriately, either automatically or through the driver's operation.

[0029] (modified version) (1) In this embodiment, an example was shown in which the support unit 12 performs autonomous driving as driving assistance, but other configurations can also be adopted. For example, a configuration that provides driving assistance with a lower degree of automation than autonomous driving may be used. Examples of driving assistance with a lower degree of automation include automatic braking, lane keeping assist control (LKAS), and adaptive cruise control (ACC).

[0030] (2) In this embodiment, we have shown an example in which "confidence in driving skills" is used as the driver's driving behavior characteristic, and when it is determined that the "confidence in driving skills" is low, autonomous driving is started (driving assistance is started). However, other configurations can also be adopted. For example, a configuration may be used in which the degree of driving assistance is adjusted based on the determination result of the state of the driver's driving behavior characteristics. For example, "unstable driving tendency" may be used as the driver's driving behavior characteristic, the navigation device 6 may be used to display a map of the target driving route as driving assistance for vehicle C, and the configuration may be used to decide whether to switch the surrounding image of vehicle C displayed on the HMI 5 display (e.g., rear view monitor, back view monitor, around view monitor) from a 2D display to a 3D display. As for the 3D display, for example, naked-eye stereoscopic display, that is, 3D display without glasses can be used. The relationship between the degree of driving assistance is 2D display < 3D display. In this case, for example, when the surrounding image displayed on the display is a two-dimensional display, as shown in Figure 5, it is determined whether the "unstable driving tendency" is low (S201 in Figure 5). If it is determined that the "unstable driving tendency" is high (S201 "No" in Figure 5), the surrounding image is kept in a two-dimensional display (S202 in Figure 5). Figure 5 is a flowchart showing the overall flow of the driving assistance method of the modified example (2). Figure 5 illustrates the case where S105, S106, and S108 in Figure 4 are replaced with S201, S202, and S203. On the other hand, if it is determined that the "unstable driving tendency" is low (S201 "Yes" in Figure 5), the content of the driving assistance is decided to be "three-dimensional display of the surrounding image," and a command is output to the HMI5 to switch the surrounding image from a two-dimensional display to a three-dimensional display (the determined content), or the driver is suggested to perform an operation to switch to a three-dimensional display (S203 in Figure 5). Suggestions to the driver are made, for example, through the HMI5's display and speakers.

[0031] In this modified version (2), as described above, the degree of driving assistance is adjusted based on the results of the determination of the state of driving behavior characteristics. Therefore, more appropriate driving assistance can be provided. Here, for example, a test was conducted in which subjects were shown 2D and 3D displays of captured images containing multiple objects at different distances from the camera, and were asked to identify the objects closest to the camera in order. According to this test, as shown in Figure 6, it was confirmed that drivers with a low "unstable driving tendency" experienced significantly lower cognitive load when using the 3D display compared to the 2D display. Therefore, in this modified version (2), when it is determined that a driver has a low "unstable driving tendency," the surrounding image is switched from a 2D display to a 3D display. Figure 6 shows the relationship between response time and the total-Hb (total hemoglobin) level in the subject's brain. Note that for drivers with a high "unstable driving tendency," the data variability was large, and it was not possible to determine a significant difference between the 2D and 3D displays.

[0032] (3) Alternatively, for example, "preparatory driving in response to traffic signals" may be used as a driver behavior characteristic, and guidance of target driving route information by the navigation device 6 (such as map display guidance or voice guidance) may be used as support for driving vehicle C, and the degree of driving support may be adjusted by deciding whether to switch the target driving route from the shortest route (i.e., the target driving route used in normal route guidance) to a route with fewer traffic signals. The relationship between the degree of driving support is shortest route < route with fewer traffic signals. In this case, for example, when the target driving route displayed by the navigation device 6 is the shortest route, it is determined whether "preparatory driving in response to traffic signals" is low (S301 in Figure 7), and if it is determined that "preparatory driving in response to traffic signals" is high (S301 "No" in Figure 7), the target driving route is maintained as the shortest route (S302 in Figure 7). Figure 7 is a flowchart showing the overall flow of the driving support method of modification (3). Figure 7 illustrates the case where S105, S106, and S108 in Figure 4 are replaced with S301, S302, and S303. On the other hand, if it is determined that the "preparatory driving for signals" is low (S301 "Yes" in Figure 7), the driving assistance is determined to be route guidance to a "route with fewer signals," and a command is output to the navigation device 6 to switch the target driving route from the shortest route to a route with fewer signals (determined content), or the driver is suggested to perform an operation to switch to a route with fewer signals (S303 in Figure 7). The suggestion to the driver is made, for example, via the display or speaker of the HMI 5.

[0033] (4) Alternatively, for example, "passivity towards driving" may be used as a driver behavior characteristic, and guidance of target driving route information by the navigation device 6 (such as map display guidance or voice guidance) may be used as support for driving vehicle C, and the degree of driving support may be adjusted by deciding whether to switch the target driving route from the shortest route (i.e., the target driving route used in normal route guidance) to a route that goes through well-maintained, wide roads. The relationship between the degree of driving support is shortest route < route that goes through well-maintained, wide roads. In this case, for example, when the target driving route displayed by the navigation device 6 is the shortest route, as shown in Figure 8, it is determined whether the "passivity towards driving" is high (S401 in Figure 8), and if it is determined that the "passivity towards driving" is low (S401 "No" in Figure 8), the target driving route is maintained as the shortest route (S402 in Figure 8). Figure 8 is a flowchart showing the overall flow of the driving support method of modification (4). Figure 8 illustrates the case where S105, S106, and S108 in Figure 4 are replaced with S401, S402, and S403. On the other hand, if it is determined that the driver has a high level of "passivity towards driving" (S401 "Yes" in Figure 8), the driver assistance is determined to be route guidance to a "route with fewer traffic lights," and a command is output to the navigation device 6 to switch the target driving route from the shortest route to a route that uses well-maintained, wide roads (determined content), or the driver is suggested to perform the operation to switch to a route that uses well-maintained, wide roads (S403 in Figure 8). The suggestion to the driver is made, for example, via the display or speaker of the HMI 5.

[0034] (5) In this embodiment, an example was shown in which the model 13 is trained using feature quantities of brain activity data from multiple subjects as input data and the results of the discrimination of the state of the subject's driving behavior characteristics as output data. However, other configurations can also be adopted. For example, the model 13 may be further trained using the driver's brain activity data. In this case, for example, the feature quantities of the driver's brain activity data are stored in the memory device 8b during driving, and after driving is completed, the results of the discrimination of the driver's characteristics state are obtained using DSQ or the like (a driving behavior classification method based on the subject's self-report) and stored in the memory device 8b, and these are used for further training of the model 13. [Explanation of Symbols]

[0035] 1…Driving assistance system, 2…External environment sensor, 3…Vehicle sensor, 4…Actuator, 6…Navigation system, 7…Brain activity detection device, 8…Information processing device, 8a…Processor, 8b…Memory device, 9…Brain activity measurement unit, 10…Brain activity analysis unit, 11…Discrimination unit, 12…Support unit, 13…Model, 14…Learning unit

Claims

1. A driver assistance method that assists the driver in operating a vehicle, Brain activity data is acquired from the measurement unit that measures the driver's brain activity. Based on the acquired brain activity data, it is determined what state the driver is in regarding predetermined specific driving behavior characteristics. Based on the results of determining the state of the aforementioned driving behavior characteristics, the content of the driving support is determined. Driving assistance methods.

2. The measurement unit is an fNIRS sensor, an electroencephalograph, or a magnetoencephalograph. The driving assistance method according to claim 1.

3. In determining the state of the driving behavior characteristics, a model trained with the feature quantities of the brain activity data of multiple subjects as input data and the determination results of the state of the driving behavior characteristics of the subjects as output data is used to determine the state of the driver's driving behavior characteristics based on the feature quantities of the driver's brain activity data. The driving assistance method according to claim 1.

4. The discrimination result used in training the aforementioned model is the discrimination result of the state of the driving behavior characteristics obtained by a driving behavior classification method that discriminates the state of the driving behavior characteristics based on the subject's self-report. The driving assistance method according to claim 3.

5. The features used to train the aforementioned model are the multiple types of features obtained by a feature extraction algorithm that extracts multiple types of features from the time-series data of the subject's brain activity. The driving assistance method according to claim 3.

6. The aforementioned model is trained using machine learning algorithms. The driving assistance method according to claim 3.

7. Once the content of the aforementioned driving assistance is determined, the system switches to the determined driving assistance, or proposes to the driver that they switch to the determined driving assistance. The driving assistance method according to claim 1.

8. A support unit that assists the driver in operating the vehicle, A measurement unit that measures the driver's brain activity and generates brain activity data, The system includes a discrimination unit that determines, based on the brain activity data generated by the measurement unit, what state a predetermined specific driving behavior characteristic of the driver is in, The support unit determines the content of the driving support based on the determination result of the state of the driving behavior characteristics obtained by the determination unit. Driving assistance system.

Images