Fatigue level determination method

The method uses helmet-mounted sensors to detect and determine fatigue in two-wheeled vehicle drivers by analyzing helmet movements, effectively addressing the instability of existing methods and providing reliable fatigue detection.

JP2024162925APending Publication Date: 2024-11-21F2F GRP CO LTD

Patent Information

Application Number
JP2023078939
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-11
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing methods for detecting driver fatigue in two-wheeled vehicles, such as motorcycles and bicycles, are unreliable due to the dynamic nature of the driver's posture and the use of helmets that can obstruct facial recognition, making it difficult to stabilize fatigue level determination.

Method used

A method that detects temporal changes in the movement of a helmet worn by the driver using an acceleration sensor, determining fatigue level based on metrics such as line-of-sight angle and helmet inclination angle relative to the vehicle's motion, without requiring additional installation on the vehicle.

Benefits of technology

Stably determines the fatigue level of two-wheeled vehicle drivers by accurately tracking helmet movements, providing early warnings to prevent accidents, and is cost-effective without needing additional vehicle-mounted equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024162925000001_ABST
    Figure 2024162925000001_ABST
Patent Text Reader

Abstract

To provide a method for stably determining the fatigue level of a driver who operates a two-wheeled vehicle.SOLUTION: A fatigue level determination method for determining a fatigue level of a driver who operates a two-wheeled vehicle, comprises: a detection step of detecting a temporal change in motion of a helmet worn by the operator; and a determination step of determining the fatigue level from the temporal change in the motion of the helmet extracted in the detection step. In the detection step, the detection of the motion of the helmet may be performed by an acceleration sensor disposed on the helmet.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a fatigue level determination method. [Background technology]

[0002] As the number of elderly drivers increases, there is an urgent need to reduce car accidents caused by elderly drivers. One of the causes of car accidents is impaired judgment due to driver fatigue. Therefore, if driver fatigue can be detected, preventive measures can be taken, such as warning the driver or forcibly stopping the car.

[0003] For example, a device for preventing drowsiness in a driver is known (see JP 2019-154613 A). The device detects the driver's fatigue based on drowsiness. Specifically, the device captures moving images of the driver's face using a video camera mounted on the vehicle and can determine drowsiness from changes in facial expression. In this way, images captured by a camera or the like are generally used to detect the driver's fatigue in a vehicle. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2019-154613 A Summary of the Invention [Problem to be solved by the invention]

[0005] In a car, the driver sits in a seat and is somewhat fixed in position by a seat belt, so the range of facial movement is limited, making it relatively easy to capture facial expressions using a camera mounted in the car.

[0006] On the other hand, it is desirable to reduce motorcycle (motorcycle and bicycle) accidents, just as with automobiles. However, it is not easy to use the above-mentioned method of determining drowsiness from changes in facial expressions to detect fatigue in motorcycle drivers. Motorcycle drivers change their posture depending on the condition of the motorcycle. For example, when approaching a curve, they will also tilt their body position according to the curvature of the curve. For this reason, there is a high possibility that the face will be outside the field of view of the camera in fixed-point observation like that of automobiles, and it is also not easy to make the camera's field of view follow the driver's posture. For this reason, it is difficult to stably determine the fatigue level of motorcycle drivers.

[0007] In addition, wearing a helmet is mandatory (recommended for cyclists), and if a full-face helmet is used, for example, it may not be possible to obtain facial expressions in the first place.

[0008] The present invention has been made in light of the above circumstances, and has an object to provide a method for stably determining the fatigue level of a rider who operates a motorcycle. [Means for solving the problem]

[0009] A fatigue level assessment method according to one embodiment of the present invention is a fatigue level assessment method for assessing the fatigue level of a driver of a two-wheeled vehicle, and includes a detection step of detecting changes over time in the movement of a helmet worn by the driver, and a assessment step of assessing the fatigue level from the changes over time in the helmet movement extracted in the detection step.

[0010] The inventors focused on helmet movement in order to detect impaired judgment due to fatigue in motorcycle drivers. That is, the inventors determined that it is possible to detect the fatigue of a driver as the slow movement of the helmet itself worn on the head, since it is believed that head movement becomes slower as the driver's fatigue increases. Therefore, in this fatigue level determination method, the fatigue level is determined from the change in helmet movement over time. Furthermore, because the helmet movement can be detected regardless of the type of helmet or the driver's posture, it is possible to stably determine the fatigue level of a motorcycle driver.

[0011] In the detection step, the detection of the movement of the helmet may be performed by an acceleration sensor disposed on the helmet. By detecting the movement of the helmet by the acceleration sensor in this manner, the movement of the helmet can be detected easily and at low cost.

[0012] The helmet movement detected in the detection step may include a line of sight angle between the forward direction of the helmet and the traveling direction of the motorcycle. The line of sight angle indicates the movement of the driver's line of sight, and as the driver becomes fatigued, the line of sight movement decreases. Therefore, the fatigue of the driver can be determined from the decrease in the line of sight movement.

[0013] The helmet movement detected in the detection step may include the helmet tilt angle relative to the vertical direction. When a motorcycle leans to turn a corner, the rider tilts his / her head, i.e., helmet, in the opposite direction to keep his / her line of sight horizontal. As the rider becomes fatigued, the amount of head tilt decreases, and the helmet tilt angle increases. Therefore, the rider's fatigue can be determined from this increase in the helmet tilt angle.

[0014] The detection step includes a first detection step of detecting a movement of the helmet, a second detection step of detecting a movement of the motorcycle corresponding to the movement of the helmet detected in the first detection step, and an extraction step of extracting a time change in a relative relationship of the movement of the helmet in the first detection step to the movement of the motorcycle in the second detection step, and the time change in the relative relationship is preferably used as the time change in the helmet movement. The rider's posture often changes in conjunction with the movement of the motorcycle. Also, since the change in the rider's posture is also reflected in the movement of the helmet, the rider's fatigue level can be determined more accurately by capturing the movement of the helmet relative to the movement of the motorcycle.

[0015] The method may include an information acquisition step of acquiring time information or position information of the motorcycle, and in the determination step, a history of changes in helmet movement over time corresponding to the time or position corresponding to the change in helmet movement over time is referenced, and if the history exists, the degree of fatigue is determined by comparing with the history. Drivers often drive in a similar manner at the same time or in the same place. Therefore, if a history of changes in helmet movement over time corresponding to the corresponding time or position exists, the degree of fatigue can be determined more accurately by comparing with the history. Effect of the Invention

[0016] The fatigue level determination method of the present invention can stably determine the fatigue level of a rider who operates a motorcycle. [Brief description of the drawings]

[0017] [Figure 1] FIG. 1 is a flow chart showing a flow of a fatigue level determination method according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic diagram showing an example of the configuration of a helmet used in the fatigue level determination method of FIG. [Diagram 3] FIG. 3 is an explanatory diagram for explaining the line of sight angle on a straight road. [Figure 4] FIG. 4 is an explanatory diagram for explaining the line of sight angle on a curved road. [Diagram 5] FIG. 5 is an explanatory diagram for explaining the helmet inclination angle. [Figure 6] FIG. 6 is a flow diagram showing a flow of a detection process according to the third embodiment of the present invention. [Figure 7] FIG. 7 is a flow chart showing the flow of a fatigue level determination method according to the fourth embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] Hereinafter, a fatigue level determination method according to each embodiment of the present invention will be described with reference to the drawings as appropriate.

[0019] [First embodiment] The fatigue level determination method according to the first embodiment of the present invention is a fatigue level determination method for determining the fatigue level of a rider who drives a motorcycle. As shown in Fig. 1, the fatigue level determination method includes a detection step S1 and a determination step S2.

[0020] In the present invention, the term "two-wheeled vehicle" refers to both two-wheeled motor vehicles, such as motorcycles, and non-self-propelled bicycles. In the following, the explanation will be given using a two-wheeled motor vehicle as an example, but the present invention can also be used for bicycles, and the same effects can be obtained.

[0021] <Detection process> In the detection step S1, a change over time in the movement of the helmet worn by the driver is detected.

[0022] 2 shows an example of the configuration of a helmet 1 used in the fatigue level determination method. The helmet 1 includes a helmet body 10, a visor 20, an acceleration sensor 30, a display unit 40, a vibration oscillator unit 50, and a control unit 60.

[0023] The helmet body 10 is a part that is worn so as to cover at least the top of the head of a driver. The helmet body 10 shown in FIG. 1 is a full-face type for motorcycles. The helmet body 10 in FIG. 1 does not include a face shield, but may include a face shield. The helmet body 10 is not limited to a full-face type, and may be an off-road helmet, a street jet helmet, or a bicycle helmet.

[0024] The visor 20 is provided so as to protrude forward from above the driver's eye line of the helmet body 10. The visor 20 is plate-shaped and may be curved or bent.

[0025] The visor 20 is preferably configured to be detachable. By replacing the visor 20 with a helmet equipped with an existing detachable visor 20, the fatigue level determination method can be used. Furthermore, the helmet 1 is generally stored under the seat or in a rear box of a motorcycle when not in use, but by making the visor 20 detachable, the storage volume can be reduced, making storage easier.

[0026] A known acceleration sensor, for example a six-axis acceleration sensor, can be used as the acceleration sensor 30. In the fatigue level determination method, in the detection step S1, the detection of the movement of the helmet 1 is performed by the acceleration sensor 30 disposed on the helmet 1.

[0027] The detection of the movement of the helmet 1 can also be performed, for example, by a camera attached to the helmet 1. When a camera is attached, it is limited to a small and lightweight one in terms of the functionality and design of the helmet 1. In addition, detecting the movement of the helmet 1 by image analysis from the obtained video requires advanced and large-scale calculations, and a high-speed, large-capacity computing device is required. For this reason, the necessary equipment tends to be expensive, and the operating time is limited due to the increase in the required power. In contrast, when the acceleration sensor 30 is used, the movement of the helmet 1 can be easily traced based on the direction and magnitude of acceleration, and the acceleration sensor 30 is relatively inexpensive. Therefore, by detecting the movement of the helmet 1 by the acceleration sensor 30 arranged on the helmet 1, the movement of the helmet 1 can be detected easily and at low cost.

[0028] The acceleration sensor 30 may be disposed inside the helmet body 10, but is preferably disposed at the tip of the visor 20. The tip of the visor 20 is a position where displacement increases with respect to the movement of the helmet 1, and it is easy to detect the direction and magnitude of acceleration. Furthermore, if the acceleration sensor 30 is disposed inside the helmet body 10, there is a risk that it will become difficult to use the fatigue level determination method for a helmet equipped with an existing detachable visor 20 due to replacement of the visor 20.

[0029] The display unit 40 is a means for visually notifying the driver that fatigue has been detected when fatigue of the driver is detected by the fatigue level determination method.

[0030] The display unit 40 transmits information to the driver, for example, by emitting light. Fig. 2 shows a configuration in which information is transmitted by dot-shaped light-emitting elements. The light-emitting elements may be elements capable of emitting different colors. The light-emitting elements can be realized, for example, by LEDs.

[0031] The display unit 40 may be a light-emitting element group in which a plurality of light-emitting elements are arranged in succession. By arranging a plurality of light-emitting elements in succession in this manner, the degree of the light can be expressed by the number and positions of the light-emitting elements that are emitting light. In addition, for example, by combining three light-emitting elements with different colors such as green, yellow, and red, a configuration can be formed in which, for example, the degree of danger can be accurately conveyed to the driver.

[0032] The vibration oscillator 50 notifies the driver that fatigue has been detected by vibration. By providing a mechanism for transmitting information by vibration in this manner, information can be transmitted to the driver more reliably.

[0033] The vibration oscillator 50 may be, for example, a bone conduction speaker capable of emitting voice or a specific frequency sound. Alternatively, the vibration oscillator 50 may be one that emits vibrations that rock the helmet body 10. By vibrating the entire helmet body 10 in this way, it is expected that the driver will be awakened.

[0034] When the vibration oscillator 50 emits sound, it is preferable that the vibration oscillator 50 is arranged inside the helmet body 10 so as to be located near both ears of a driver wearing the helmet body 10, as shown in FIG. 2.

[0035] The control unit 60 detects the movement of the helmet 1 from the measurement values ​​of the acceleration sensor 30, and executes a judgment process S2 described later based on the change over time.If fatigue of the driver is detected, the control unit 60 controls the display unit 40 and the vibration oscillation unit 50.

[0036] The control unit 60 can be configured with, for example, a microcontroller, a battery, a memory, etc. Also, it may have a built-in antenna that can communicate with the outside. In this case, the detection and determination of the movement of the helmet 1 can be calculated by an external computer such as a cloud server connected to a network via the antenna.

[0037] In this fatigue level determination method, the movement of the helmet detected in the detection step S1 is the line of sight angle (angle α in FIG. 3, angle β in FIG. 4) made by the front direction of the helmet 1 relative to the traveling direction of the motorcycle 2. Here, the front direction of the helmet 1 is the direction that the front of the helmet 1 faces when the rider is wearing it and faces the traveling direction of the motorcycle 2. This line of sight angle represents the movement of the rider's line of sight.

[0038] These gaze angles α and β make it possible to identify the displacement of helmet 1 from the direction of acceleration and its magnitude, so a new gaze angle can be obtained by adding the above displacement to the previous gaze angle of helmet 1.

[0039] The gaze angle may be calculated analytically, or the gaze angle may be calculated inferentially using artificial intelligence (AI). In the fatigue level determination method, as described later, the driver's fatigue level is determined from the amount of change in the gaze angle for checking the surrounding traffic environment while driving on a curved road or in an urban area. However, the change in the gaze angle is not necessarily caused by checking the surrounding traffic environment, and may change, for example, when checking the surrounding scenery. However, since the change pattern of the gaze angle is considered to be different depending on the cause, the target gaze angle change can be easily extracted by using AI. That is, a method can be adopted in which the time change of the known gaze angle and the time range during which the surrounding traffic environment is checked are used as input data, and the gaze angle when the surrounding traffic environment is checked is used as teacher data to determine the time and the gaze angle during the check of the surrounding traffic environment from the time change of the gaze angle using a prediction model machine-learned. In addition, it is also possible to use AI, adopt an individual learning model, or adopt a common learning model from the data of many people. The former is easy to reflect the unique situation of each individual driver, and the latter is easy to use as an introduction model when personal information is insufficient.

[0040] In the above-mentioned method using the acceleration sensor 30, it is preferable that the line of sight angle (reference angle) at a reference time (e.g., time 0) is known. By setting the reference angle, it becomes possible to specify in which direction the driver is looking. This reference angle can be determined from the time change in the detected movement of the helmet 1, in addition to a method of performing an initialization operation to specify the traveling direction when the driver gets on the motorcycle (e.g., by aligning the front direction of the helmet 1 with the traveling direction and pressing a reset button), for example. In general, since the driver often keeps his / her line of sight forward, the direction in which the movement of the helmet 1 is stable can be recognized as the traveling direction. It is also possible to detect the time change in the movement of the helmet 1 as the relative movement of the line of sight angle without setting a reference angle. In other words, the time change in the movement of the helmet 1 may be determined using the line of sight angle at the start of detection as the reference angle. Although it is not possible to specify the direction in which the driver is facing, the degree of fatigue can still be determined in this case.

[0041] <Judgment process> In the determination step S2, the degree of fatigue is determined from the change over time in the movement of the helmet 1 extracted in the detection step S1.

[0042] First, it will be described how the driver's fatigue can be determined based on the line of sight angle when passing through a curved road.

[0043] When driving on a straight road, the driver generally keeps his / her gaze P in the direction of travel or slightly to the right of the direction of travel, as shown in Figure 3. In other words, the gaze angle α is relatively close to 0°. In contrast, when approaching a curved road, the driver shifts his / her gaze P in the direction of the turn, so that the gaze angle β becomes significantly larger than the gaze angle α, as shown in Figure 4.

[0044] On the other hand, when the driver becomes tired, the movement of turning the head to shift the line of sight when turning becomes slower. In other words, the difference between the line of sight angle β and the line of sight angle α (the amount of line of sight movement) becomes smaller than when the driver is not tired. Therefore, the driver's fatigue can be determined from the decrease in the amount of line of sight movement P.

[0045] The same applies when driving in urban areas. When driving in urban areas, the line of sight angle β changes from the direction of travel in order to check for the presence of cars or people entering from around the road, check traffic lights and signs, and look left and right at intersections. When the driver becomes tired, the amount of movement of the line of sight P decreases, the movement time becomes shorter, or the act of moving the line of sight P is omitted altogether (the number of times the line of sight P moves decreases). Therefore, the driver's fatigue can be determined from the changes in these amounts.

[0046] Taking the case of passing through a curved road as an example, a method of determining the occurrence of fatigue will be described. For example, when it is possible to extract that the driver is passing through a curved road by using AI or the like, it is determined whether the change in the gaze angles α and β has decreased. It is considered that the number of cases in which the change in the gaze angles α and β is small gradually increases as the driver becomes fatigued, but on the other hand, the change in the gaze angles α and β also tends to become small when turning on a curved road with low curvature, and the change in the gaze angles α and β also depends on the type of curved road (e.g. curvature). For this reason, it is preferable to make the determination based on the average value of the change in the gaze angles α and β over a certain period of time. This is because it is considered possible to capture the trend in the change in the gaze angles α and β by looking at the average value over a certain period of time, even if there is a localized fluctuation.

[0047] The lower limit of the certain period is preferably 1 minute, more preferably 3 minutes, and even more preferably 10 minutes. On the other hand, the upper limit of the certain period is preferably 60 minutes, and more preferably 30 minutes. If the certain period is less than the lower limit, for example, the driver may be strongly influenced by small changes in the line of sight angles α and β due to the factors described above, and the accuracy of the determination of the driver's fatigue level may decrease. Conversely, if the certain period exceeds the upper limit, the interval between determinations may become too long. In other words, even if the driver is determined not to be fatigued at a certain determination time, the time until the next determination time is long, and as a result, the time between the onset of fatigue and the determination may be long, which may result in insufficient prevention of accidents.

[0048] When passing through urban areas, the driver's fatigue can be judged in the same manner as when passing through curved roads, except that the criteria for judgment are a decrease in the amount of movement of the line of sight P, a decrease in travel time, a decrease in the number of movements, or a combination of these.

[0049] Although the methods for determining whether the motorcycle is passing through a curved road or an urban area have been described, either one of them may be used or they may be used together. In this case, if the factor of the change in the gaze angles α and β in the time change of the gaze angles α and β is not specified, it may be difficult to make the determination. In the fatigue level determination method, the acceleration sensor 30 is used to detect the movement of the helmet 1. For example, when the acceleration sensor 30 observes acceleration in the forward and backward directions, it is usually linked to the acceleration and deceleration of the motorcycle. In this way, the movement of the motorcycle itself may be estimated from the measurement results of the acceleration sensor 30. Then, the factor of the change in the gaze angles α and β can be estimated based on the estimated movement. For example, when the motorcycle is frequently accelerated and decelerated, it is estimated that the motorcycle is passing through an urban area. Then, the movement of the gaze P when the motorcycle is stopped is estimated to be checking the traffic light or checking the left and right at an intersection, and further, checking the traffic light involves the movement of the gaze P in the up and down direction, whereas checking the left and right at an intersection is mainly considered to be a horizontal movement. In addition, the movement of the helmet 1 when passing through a curved road is considered to be such that first, the helmet 1 itself rotates as the line of sight P is shifted in the direction of the turn, and then a lateral acceleration (towards the center of curvature of the road) is added due to turning the curved road. The rotational movement of the helmet 1 itself and the lateral acceleration due to turning can be distinguished from each other because the directions of acceleration are different. Furthermore, it is considered possible to estimate the curvature of the road from the above-mentioned lateral acceleration. In this way, by estimating the cause of the change in the line of sight angles α and β from the measurement results of the acceleration sensor 30, the fatigue level of the rider can be determined with higher accuracy. The cause of the change in the line of sight angles α and β may be estimated algorithmically, but it is effective to use AI. In other words, the movement of the rider and the motorcycle can be obtained from the acceleration observed by the acceleration sensor 30 using a prediction model that is machine-learned using the acceleration observed by the acceleration sensor 30 as input data and the actual movement of the rider and the motorcycle as teacher data.

[0050] On the other hand, when the cause of the change in the gaze angles α and β is not specified in the time change of the gaze angles α and β, the cause of the change may include not only passing through a curved road or an urban area as described above, but also checking the surrounding scenery. In this case, a pattern may occur in which the driver does not check the surrounding scenery as he or she becomes tired, so it is preferable to consider the number of times the gaze angles α and β change in addition to or instead of the decrease in the amount of change in the gaze angles α and β. In this way, by considering the number of times the gaze angles α and β change, the driver's fatigue can be determined without estimating the cause of the change.

[0051] Also, instead of or in addition to the number of times the gaze angles α and β change, the deviation of the gaze angles α and β within the above-mentioned certain period can be obtained and judged. This is because it is considered that as the driver becomes fatigued, the deviation tends to decrease regardless of the cause of the change.

[0052] In either method, when a decrease in the value of the predetermined amount or more is recognized and fatigue of the driver is detected, the control unit 60 controls the display unit 40 and the vibration oscillator 50 to issue a warning to the driver.

[0053] <Advantages> Focusing on the fact that the fatigue of a rider can be detected as a slow movement of the helmet 1 itself worn on the head, this fatigue level determination method determines the fatigue level from the change over time in the movement of the helmet 1. And because the movement of the helmet 1 can be detected regardless of the type of helmet 1 or the rider's posture, the fatigue level of a rider operating a motorcycle 2 can be determined stably.

[0054] Furthermore, since the fatigue level determination method determines the fatigue level using only the time change in the movement of the helmet 1, it is not necessary to mount a measuring device or the like on the motorcycle 2. Therefore, the method can be implemented inexpensively and easily.

[0055] [Second embodiment] The fatigue level determination method according to the second embodiment of the present invention is a fatigue level determination method for determining the fatigue level of a rider who drives a two-wheeled vehicle 2, and similarly to the first embodiment, includes a detection process S1 for detecting changes over time in the movement of a helmet 1 worn by the rider, and a determination process S2 for determining the fatigue level from the changes over time in the movement of the helmet 1 extracted in the detection process S1 (see Figures 1 and 2).

[0056] This fatigue level determination method is similar to the fatigue level determination method of the first embodiment, except that the movement of the helmet 1 detected in the detection step S1 is the helmet tilt angle with respect to the vertical direction. Therefore, hereinafter, only the helmet tilt angle will be described, and other descriptions will be omitted.

[0057] As shown in Figure 5, helmet inclination angle θ1 refers to the angle between the vertical direction and the up-down direction of helmet 1 (the direction from the head length to the chin when rider X is wearing it). For example, when turning a corner, motorcycle 2 generally turns at an incline, as shown in Figure 5. Meanwhile, rider X tries to keep his line of sight horizontal even when turning a corner. For this reason, he tilts his head, i.e., helmet 1, in the opposite direction to the inclination of motorcycle 2. Therefore, with respect to motorcycle inclination angle θ2, which is the angle between the vertical direction and the vertical direction of motorcycle 2, the relationship θ1<θ2 holds.

[0058] Here, the helmet inclination angle θ1 is usually 0° or a value close to 0°. If a value far from 0° (e.g., 30° or more) is observed for a certain period of time or more, it is presumed that the driver X is experiencing a physical condition abnormality, and therefore it is assumed that the driver X is fatigued and some kind of warning should be issued. The motorcycle inclination angle θ2 changes depending on the curvature of the corner, etc., but is often between 5° and 10°, and is thought to be at most about 30°.

[0059] On the other hand, when the rider X becomes tired, he or she will no longer bend his or her head, and the helmet inclination angle θ1 will shift in a direction that follows the inclination of the motorcycle 2. In other words, the helmet inclination angle θ1 will become larger than when the rider X is not tired. Therefore, the fatigue of the rider X can be determined from the increase in the helmet inclination angle θ1 or the increase in the deviation of the helmet inclination angle θ1.

[0060] <Advantages> Even in the case where the helmet inclination angle θ1 is used as in this fatigue level determination method, the same effects as those of the first embodiment can be achieved.

[0061] [Third embodiment] A fatigue level determination method according to a third embodiment of the present invention is a fatigue level determination method for determining the fatigue level of a driver X who drives a two-wheeled vehicle 2, and similarly to the first embodiment, includes a detection step S1 for detecting a change over time in the movement of a helmet 1 worn by the driver X, and a determination step S2 for determining the fatigue level from the change over time in the movement of the helmet 1 extracted in the detection step S1 (see Figs. 1 and 2). The helmet 1 used in this fatigue level determination method is similar to that in the first embodiment, and therefore a detailed description thereof will be omitted.

[0062] <Detection process> In the fatigue level determination method, the detection step S1 includes a first detection step S11, a second detection step S12, and an extraction step S13, as shown in FIG.

[0063] (First detection step) In a first detection step S11, the movement of the helmet 1 is detected.

[0064] The first detection step S11 is similar to the detection step S1 in the first or second embodiment, and therefore a detailed description thereof will be omitted.

[0065] (Second detection step) In the second detection step S12, the movement of the two-wheeled vehicle 2 corresponding to the movement of the helmet 1 detected in the first detection step S11 is detected.

[0066] In the first detection step S11, as in the first embodiment, if the movement of the helmet 1 is defined as the line of sight angles α, β formed by the front direction of the helmet 1 relative to the traveling direction of the motorcycle 2, the movement of the motorcycle 2 detected in the second detection step S12 can be defined as, for example, the cornering curvature corresponding to the change in the line of sight angles α, β. Since it is believed that the line of sight angles α, β change significantly when the curvature is large, it is believed that there is a certain correlation between the two.

[0067] In this case, it is considered that the two-wheeled vehicle 2 turns with a certain time delay after the line-of-sight angles α, β change, so the detection of the movement of the two-wheeled vehicle 2 in the second detection process S12 has a certain delay with respect to the first detection process S11. This delay may be a fixed value, but may also be changed according to the speed of the two-wheeled vehicle 2. The higher the speed of the two-wheeled vehicle 2, the smaller the delay.

[0068] Furthermore, if the two-wheeled vehicle 2 does not turn despite changes in the gaze angles α and β, or turns in the opposite direction to the expected direction, it is deemed that there is no correlation between the changes in the gaze angles α and β and the movement of the two-wheeled vehicle 2, and the two-wheeled vehicle 2 should not be subject to the extraction process S13 described below.

[0069] For example, when the motorcycle 2 is stopped by a traffic light while driving in an urban area, or when it is known that a sign is present near the motorcycle 2, the angle of the motorcycle 2 may be determined based on this object. It is considered that the line of sight angle of the rider X changes in correlation with the angle between the object and the motorcycle 2. Therefore, this angle can be used instead of the above-mentioned curvature. The positional relationship between the motorcycle 2 and the positions of the traffic lights and signs can be obtained, for example, by combining a GPS and a map.

[0070] In the first detection process S11, as in the second embodiment, when the movement of the helmet 1 is determined to be a helmet inclination angle θ1 with respect to the vertical direction, the movement of the motorcycle 2 detected in the second detection process S12 can be determined to be a motorcycle inclination angle θ2 with respect to the vertical direction corresponding to this movement of the helmet 1. In this case, as the rider X becomes fatigued, the helmet inclination angle θ1 tends to increase, but the amount of increase is considered to be greater as the motorcycle inclination angle θ2 increases, and therefore there is considered to be a certain correlation between the two.

[0071] In this case, since the change in the helmet inclination angle θ1 and the motorcycle inclination angle θ2 are considered to occur almost simultaneously, the detection of the movement of the motorcycle 2 in the second detection step S12 can be performed simultaneously with the first detection step S11.

[0072] Furthermore, if the motorcycle tilt angle θ2 does not change despite a change in the helmet tilt angle θ1, or changes in the opposite direction to the expected direction, it is deemed that there is no correlation between the change in the helmet tilt angle θ1 and the movement of the motorcycle 2, and the motorcycle 2 is not subjected to the extraction process S13 described later.

[0073] (Detection process) In the detection step S13, a change over time in the relative relationship of the movement of the helmet 1 in the first detection step S11 with respect to the movement of the two-wheeled vehicle 2 in the second detection step S12 is extracted.

[0074] In the first detection step S11, when the movement of the helmet 1 is determined as the gaze angle formed by the front direction of the helmet 1 relative to the traveling direction of the motorcycle 2, the above-mentioned relative relationship may be the ratio of the amount of change in the gaze angle to the curvature (the angle between the object and the motorcycle 2 in case of driving in urban areas). Even when the driver X becomes fatigued, it is believed that the tendency for the change in the gaze angle to be greater in corners with larger curvatures remains. For this reason, by using the ratio of the amount of change in the gaze angle to the curvature as the relative relationship, the above ratio decreases regardless of the curvature when the driver X becomes fatigued, so that the degree of fatigue of the driver X can be determined more reliably.

[0075] In the first detection step S11, when the movement of the helmet 1 is the helmet tilt angle θ1 relative to the vertical direction, the above relative relationship may be the ratio of the amount of change in the helmet tilt angle θ1 to the motorcycle tilt angle θ2. When the motorcycle tilt angle θ2 is small, the inclination of the head may be small to maintain the helmet tilt angle θ1. In this case, even if the rider X becomes fatigued, it is considered that the fluctuation of the helmet tilt angle θ1 tends to be smaller than when the rider X is not fatigued. Therefore, by using the relative relationship normalized by the motorcycle tilt angle θ2, when the rider X becomes fatigued, the above ratio increases regardless of the motorcycle tilt angle θ2, so that the degree of fatigue of the rider X can be determined more reliably.

[0076] <Judgment process> In the fatigue level determination method, in the determination step S2, the change over time of the relative relationship is used as the change over time of the movement of the helmet 1.

[0077] Except for using the time change in the relative relationship, the process can be similar to the determination step S2 in the first or second embodiment, so a detailed description will be omitted.

[0078] <Advantages> As described above, the posture of the rider X often changes in conjunction with the movement of the motorcycle 2. In addition, since the change in the posture of the rider X is also reflected in the movement of the helmet 1, the fatigue level of the rider X can be determined more accurately by capturing the movement of the helmet 1 relatively to the movement of the motorcycle 2. In addition, data in which the movement of the helmet 1 is not correlated with the movement of the motorcycle 2 can be excluded from the determination target, so that the fatigue level of the rider X can be determined more accurately in this respect as well.

[0079] [Fourth embodiment] The fatigue level determination method according to the fourth embodiment of the present invention is a fatigue level determination method for determining the fatigue level of a driver X who drives a two-wheeled vehicle 2, and includes a detection process S1, an information acquisition process S3, and a determination process S2, as shown in FIG. 7.

[0080] The helmet 1 used in this fatigue level determination method is similar to that in the first embodiment, and therefore a detailed description thereof will be omitted.

[0081] <Detection process> In the detection step S1, a change over time in the movement of the helmet 1 worn by the driver X is detected.

[0082] The detection step S1 can be the same as the detection step S1 in the first embodiment or the detection step S1 in the second embodiment. In the following, the description will be continued taking as an example a case where the detection step S1 is the same as the detection step S1 in the first embodiment.

[0083] <Information acquisition process> In the information acquisition step S3, time information or position information of the two-wheeled vehicle 2 is acquired. Either the time information or the position information may be acquired, or both may be acquired. In addition, the information acquisition step S3 and the detection step S1 may be performed in any order. The information acquisition step S3 may be performed prior to the detection step S1. It is preferable that the two steps are performed consecutively.

[0084] The time information means the time when the detection step S1 is performed. The method of acquiring the time information is not particularly limited, and may be performed by a timer mounted on the control unit 60, information may be acquired from a device mounted on the two-wheeled vehicle 2, or information may be acquired from a mobile terminal or the like owned by the rider X.

[0085] The time information may include the hour and minute of a day, the day of a week, the day of a month, and the date of a year, but a combination of the day of the week and the hour or the day of the week and the hour and minute is preferable. Furthermore, the weather may be taken into consideration. Weather information may be obtained, for example, from an external server.

[0086] The location information means the location where the two-wheeled vehicle 2 is located when the detection step S1 is performed. The method of acquiring the location information is not particularly limited, and the information can be acquired from a GPS function installed in the two-wheeled vehicle 2 or a mobile terminal owned by the rider X, for example.

[0087] <Judgment process> The determination step S3 determines the degree of fatigue from the change over time in the movement of the helmet 1 extracted in the detection step S1. In this fatigue level determination method, in this determination step S3, a history of the change over time in the movement of the helmet 1 corresponding to the time or position corresponding to the change over time in the movement of the helmet 1 to be determined is referenced, and if such history exists, the degree of fatigue is determined by comparing with the history. The determination of the degree of fatigue by comparing with the history will be described in detail below. Note that apart from this point, the determination configuration can be configured in the same way as the determination configuration S3 of the first embodiment, and therefore a detailed description will be omitted.

[0088] In the information acquisition step S3, for example, when time information of the hour and minute is acquired, the history for the past time information is searched and the history with the same hour and minute is extracted. The time change in the movement of the helmet 1 detected in the detection step S1 is judged based on the time change in the movement of the helmet 1 in this history. Specifically, if the gaze angle detected in the detection step S1 is found to be lower than the gaze angle at the same time in the history by a predetermined amount or more, it is judged that the driver X is fatigued.

[0089] Driver X often drives in a specific location at a specific time. This can be easily imagined, for example, when Driver X uses the motorcycle 2 to commute to work. In this case, Driver X will leave his / her home at a specific time and travel to his / her workplace via a specific route. For this reason, the driving conditions at the same time are often similar. Therefore, by comparing with past history, it becomes easier to extract abnormal cases.

[0090] This also applies when location information is acquired in the information acquisition step S3. In the same location, the road conditions are often the same, and driver X often drives in a similar manner. Even in this case, it becomes easier to extract abnormal cases by comparing with past history.

[0091] In addition, the position information can be used to determine whether the two-wheeled vehicle 2 is traveling in an urban area, and to identify the presence or absence of an intersection and a sign, so that it is also possible to determine the line of sight angle of the driver X, which is the direction that the driver X should be facing, as described in the third embodiment.

[0092] If there is no past history, the determination can be made in the same manner as in the first embodiment. Alternatively, even if there is a past history, the determination can be made by using the same determination method as in the first embodiment.

[0093] The result of the determination in the determination step S3 may be registered as a new history and used for subsequent determinations. In order to achieve this configuration, it is preferable that the history be managed by a cloud server.

[0094] <Advantages> In this fatigue level determination method, if there is a history of the change over time in the movement of the helmet 1 corresponding to the corresponding time or position, the fatigue level can be determined more accurately by comparing with that history.

[0095] [Other embodiments] The above-mentioned embodiment does not limit the configuration of the present invention. Therefore, the above-mentioned embodiment may omit, replace or add components of each part of the above-mentioned embodiment based on the description in this specification and common general technical knowledge, and it should be understood that all of these are within the scope of the present invention.

[0096] In the above first embodiment, a case has been described in which the helmet movement detected in the detection process is the line of sight angle made by the front direction of the helmet relative to the traveling direction of the motorcycle, and in the above second embodiment, a case has been described in which the helmet movement detected in the detection process is the helmet inclination angle relative to the vertical direction, but both the line of sight angle and the helmet inclination angle may be used. When both are used, the rider's fatigue may be determined to be fatigued overall when both are determined to be fatigued, or may be determined to be fatigued overall when either one is determined to be fatigued.

[0097] Moreover, the helmet movement detected in the detection step may be other helmet movement linked to the fatigue of the driver, or may be a combination of one or both of the line of sight angle and helmet tilt angle and other movement. In the third embodiment, the helmet movement detected may be both the line of sight angle and helmet tilt angle, or may be other helmet movement or a combination of this with one or both of the line of sight angle and helmet tilt angle.

[0098] In the above third embodiment, when the helmet movement is the helmet tilt angle relative to the vertical direction, the detection of the motorcycle movement (tilt angle) in the second detection step is performed simultaneously with the first detection step, but the time response between the helmet tilt angle and the motorcycle tilt angle may be detected and the amount of delay may be obtained. As the rider becomes fatigued, the response speed to the motorcycle tilt is thought to decrease, so it is also possible to determine the rider's fatigue from the amount of delay.

[0099] The present invention also contemplates a fatigue level determination method that includes both the detection step of the third embodiment and the information acquisition step of the fourth embodiment.

[0100] In the above embodiment, a case has been described in which the detection of fatigue is communicated to the driver by the display unit and the vibration oscillator unit, but the communication to the driver may be by either one of them, or by other means. [Industrial Applicability]

[0101] As described above, the fatigue level determination method can stably determine the fatigue level of a rider who operates a two-wheeled vehicle. [Explanation of symbols]

[0102] 1. Helmet 10 Helmet body 20 Visor 30 Acceleration Sensor 40 Display section 50 Vibration Oscillator 60 Control section 2 Two-wheeled vehicles α, β sight angle P Sight θ1ヘルメットtilt angle θ2 Bicycle tilt angle X-Movers

Claims

1. A fatigue level determination method for determining a fatigue level of a driver who drives a motorcycle, comprising: A detection step of detecting a change over time in the movement of the helmet worn by the driver; a determination step of determining a fatigue level from a time change in the helmet movement extracted in the detection step; A fatigue level determination method comprising:

2. 2. The method for determining a degree of fatigue according to claim 1, wherein in said detection step, the detection of the movement of said helmet is performed by an acceleration sensor disposed in said helmet.

3. 3. The fatigue level determination method according to claim 1, wherein the movement of the helmet detected in the detection step includes a line of sight angle formed by the front direction of the helmet with respect to the traveling direction of the motorcycle.

4. 3. The method for determining a degree of fatigue according to claim 1, wherein the movement of the helmet detected in the detection step includes an inclination angle of the helmet relative to a vertical direction.

5. The detection step comprises: a first detection step of detecting a movement of the helmet; a second detection step of detecting a movement of the motorcycle corresponding to the movement of the helmet detected in the first detection step; an extraction step of extracting a time change in a relative relationship of a movement of the helmet in the first detection step with respect to a movement of the motorcycle in the second detection step; having 3. The method for determining a degree of fatigue according to claim 1, wherein a change in the relative relationship over time is used as the change in the movement of the helmet over time.

6. An information acquisition step of acquiring time information or position information of the motorcycle, A fatigue level determination method as described in claim 1 or claim 2, wherein in the determination step, a history of the change in helmet movement over time corresponding to the time or position corresponding to the change in helmet movement over time to be determined is referenced, and if the history exists, the fatigue level is determined by comparing with the history.

Citation Information

Patent Citations

  • Drowsiness detection system, drowsiness detection data generation system, drowsiness detection method, computer program, and detection data

    JP2019154613A

Cited By

  • A computer-implemented method for training of a machine-learnng model in vehicle driver behavior assessment for a two-wheeled or a three-wheeled vehicle

    EP4764904A1

  • A computer-implemented method for training of a machine-learning model in vehicle driver behavior assessment for a two-wheeled or a three-wheeled vehicle

    WO2026131944A1