JP2008003789A - Driver state determination device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a driver state determination device capable of highly precisely detecting an abnormal state of a driver. <P>SOLUTION: The driver state detector includes a driving state determination ECU 2 for obtaining an evaluation of a driving state and a right/left turn trajectory operation ECU 3 for obtaining an evaluation during right/left turn. The driving state determination ECU 2 and the right/left turn trajectory ECU 3 output obtained evaluations to a vacant state determination ECU 5. Criterion running states being criteria for these evaluations are stored in a normal driving data recording part 4, and these criterion running states are outputted to the vacant state determination ECU 5. The vacant state determination ECU 5 compares evaluations outputted from the driving state determination ECU 2 and the right/left turn trajectory ECU 3 with criterion running state outputted from the normal driving data recording part 4 to determine a vacant driving state of the driver. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a driver state determination device that determines an abnormality that occurs in a driver driving a vehicle.

2. Description of the Related Art Conventionally, there is an apparatus that determines an abnormality such as a snoozing that occurs in a driver by steering of the driver (driver). As such a device, for example, there is an abnormal steering determination device disclosed in Japanese Patent Laid-Open No. 5-85221 (Patent Document 1). The abnormal steering determination device detects a steering amount and a vehicle speed of a vehicle and a lane change, detects a response delay of a steering operation based on the steering amount, and travels the vehicle position and the vehicle based on the steering amount and the vehicle speed. Calculate the deviation amount of the lane you want. Then, by comparing the response delay and deviation amount in the normal state of the driver with consideration of the vehicle speed and lane change, and the calculated response delay and deviation amount, the driver is in a normal state or an abnormal state. It is to detect whether there is.
JP-A-5-85221

  However, when the vehicle is traveling along the lane, the vehicle can easily travel along the lane even when the driver does not intentionally steer due to the effects of the vehicle self-straightness and the road lateral gradient. Here, in the abnormal steering determination device disclosed in Patent Document 1, since the detection is performed in a state where the vehicle is traveling along the lane, the steering wheel is operated between the abnormal state and the normal state of the driver. There is a problem that the difference in the response delay of the operation and the deviation amount of the lane is small, and the accuracy of determining the abnormal state of the driver is low.

  Then, the subject of this invention is providing the driver state determination apparatus which can detect a driver | operator's abnormal state accurately.

  The driver state determination apparatus according to the present invention that has solved the above problems includes a scene detection unit that detects that the vehicle is in a course change scene at an intersection, and the vehicle traveling when the course change scene is detected by the scene detection unit. A driving state detecting unit for detecting a state, a driving state storing unit for storing a reference driving state serving as a reference for a driving state when the vehicle is in a course change scene, a driving state detected by the driving state detecting unit, and a driving Determination means for determining the state of the driver based on the deviation from the reference running state stored in the state storage means.

  In the driver state determination apparatus according to the present invention, the driver state is determined when the vehicle is in a course change scene at an intersection. When the vehicle is in a course change scene at an intersection, there is a greater difference in traveling state between when the driver is in a normal state and in an abnormal state than when the vehicle is traveling along a lane. For this reason, when the vehicle is in a course change scene, the abnormal state of the driver can be accurately detected by detecting the abnormal state based on the deviation between the detected traveling state and the reference traveling state.

  Here, the driving state may be a mode in which the turning radius of the vehicle is set, and the driving state may be a mode in which the position of the vehicle with respect to the reference position set at the intersection is set. Can do. When the driver's concentration is reduced and the vehicle enters an abnormal state, the vehicle is likely to make a large turn. Therefore, the driver's abnormality can be detected based on the turning radius of the vehicle or the position of the vehicle with respect to the reference position set at the intersection. . Examples of the reference position here include a curb at an intersection, a center line, and the center of the intersection.

  Or it can be set as the aspect in which the turning start point of the vehicle is set as a running state, and it can also be set as the aspect in which the magnitude | size of the deceleration before steering operation is set as a running state. When the driver's concentration is reduced, the turning start point is deviated or the amount of braking operation varies. For this reason, the driver's abnormality can be detected from the turning point of the vehicle and the magnitude of the deceleration before the start of the steering operation as the running state.

  Moreover, it can be set as the aspect by which the vehicle speed at the time of steering operation start is set as a driving state. The vehicle speed at the start of the steering operation is greatly different when the driver is in an abnormal state compared to when the driver is in a normal state. For this reason, since the vehicle speed at the start of the steering operation is set as the traveling state, it is possible to detect an abnormality occurring in the driver with higher accuracy.

  Furthermore, it can be set as the aspect in which the driving | running | working state according to the right / left turn of a vehicle is set as a driving | running state. There is a difference in traveling state between when the vehicle turns right and when it turns left. By detecting and comparing the traveling state according to this difference, it is possible to detect an abnormality occurring in the driver with higher accuracy.

  The determining means determines that the driver's condition is abnormal when the deviation between the traveling state detected by the traveling state detecting means and the reference traveling state stored in the traveling state storage means exceeds a predetermined threshold. It can be set as the aspect determined to exist. In this way, when the deviation between the driving state and the reference driving state exceeds a predetermined threshold value, for example, by determining that there is an abnormality in the driver's state, for example, a determination criterion for issuing an alarm when in an abnormal state And so on.

  According to the driver state determination device of the present invention, an abnormal state of the driver can be detected with high accuracy.

  Hereinafter, an embodiment of a driver state determination device according to the present invention will be described with reference to the drawings. Note that description of the same members and elements may be omitted.

  FIG. 1 is a block configuration diagram of a driver state determination device according to the present embodiment.

  As shown in FIG. 1, the driver state determination device includes a vehicle information recognition ECU 1, a driving state determination ECU 2, a right / left turn trajectory calculation ECU 3, a normal driving data recording unit 4, a blurred state determination ECU 5, and an alarm device 6. Yes. Further, the vehicle information recognition ECU 1 includes various sensors, specifically, an accelerator stroke sensor 11, a brake stroke sensor 12, a steering rudder angle sensor 13, a vehicle speed sensor 14, a front / rear / left / right acceleration sensor 15, an angular acceleration sensor 16, a right side of the vehicle. An information camera 17 and a vehicle left side information camera 18 are connected to each other.

  The accelerator stroke sensor 11 is provided, for example, in an accelerator pedal of the vehicle, and detects the amount of depression of the accelerator pedal. The accelerator stroke sensor 11 outputs the detected depression amount of the accelerator pedal to the vehicle information recognition ECU 1 as an accelerator pedal stroke.

  The brake stroke sensor 12 is provided, for example, in a vehicle brake pedal, and detects the amount of depression of the brake pedal. The brake stroke sensor 12 outputs the detected depression amount of the brake pedal to the vehicle information recognition ECU 1 as a brake pedal stroke.

  The steering rudder angle sensor 13 is attached to, for example, a steering rod, and detects the steering rudder angle by operating the steering wheel. The steering rudder angle sensor 13 outputs the detected steering rudder angle to the vehicle information recognition ECU 1.

  The vehicle speed sensor 14 is provided, for example, on a vehicle wheel, and detects the vehicle speed by measuring the number of rotations of the wheel. The vehicle speed sensor 14 outputs the detected vehicle speed to the vehicle information recognition ECU 1. The front / rear / left / right acceleration sensor 15 is attached to the body of the vehicle, for example, and detects acceleration generated in the front / rear direction and the left / right direction of the vehicle. The front / rear / left / right acceleration sensor 15 outputs the detected acceleration to the vehicle information recognition ECU 1.

  The angular acceleration sensor 16 is attached to a vehicle body, for example, and detects angular accelerations of the yaw rate, roll rate, and pitch rate of the vehicle. The angular acceleration sensor 16 outputs the detected angular acceleration to the vehicle information recognition ECU 1.

  The vehicle right side information camera 17 is attached to, for example, a door mirror on the right side of the vehicle, and images the periphery of the right side of the vehicle. The vehicle right side information camera 17 outputs the captured image to the vehicle information recognition ECU 1. The vehicle information recognition ECU 1 performs image processing such as edge detection on the output image, such as white lines and curbs around the right side of the vehicle. Get information.

  The vehicle left side information camera 18 is attached to, for example, a door mirror on the left side of the vehicle, and images the periphery of the left side of the vehicle. The vehicle right information camera 17 outputs the captured image to the vehicle information recognition ECU 1. The vehicle information recognition ECU 1 performs image processing such as edge detection on the output image, and acquires information such as white lines and curbs around the left side of the vehicle.

  An instruction signal is output from the winker device to the vehicle information recognition ECU 1 when the winker instructs to turn right or left. The vehicle information recognition ECU 1 detects that the vehicle is in a course change scene at an intersection when an instruction signal is output from the blinker device, and performs the following calculation, output, and the like. Further, when an instruction signal is output from the winker device, right / left turn information corresponding to the instruction signal is output to the right / left turn trajectory calculation ECU 3 and the instruction signal is sent to the blur state determination ECU 5 via the right / left turn trajectory calculation ECU 3. The left / right turn information corresponding to is output. The vehicle information recognition ECU 1 constitutes a scene detection unit. Moreover, vehicle information recognition ECU1 comprises the driving state detection means which detects the driving state of a vehicle when a course change scene is detected.

  The vehicle information recognition ECU 1 includes an accelerator pedal depression amount output from the accelerator stroke sensor 11, a brake pedal depression amount output from the brake stroke sensor 12, a steering angle output from the steering rudder angle sensor 13, and a vehicle speed sensor 14. Is output to the driving state determination ECU 2. Further, the vehicle information recognition ECU 1 detects the vehicle speed before the brake pedal is depressed, the magnitude of the deceleration, and whether or not the steering operation is started, and outputs it to the driving state determination ECU 2.

  The vehicle information recognition ECU 1 is based on the vehicle acceleration output from the longitudinal / lateral acceleration sensor 15 and the angular acceleration such as the yaw rate output from the angular acceleration sensor 16. Calculate the turning radius. The vehicle information recognition ECU 1 outputs the calculated turning radius to the left / right turn trajectory calculation ECU 3.

  Further, the vehicle information recognition ECU 1 determines the distance from the left curb, the distance from the right center line, and the center of the right intersection based on information obtained by image processing the images output from the vehicle information cameras 17 and 18. The distance and the distance from the right center line are calculated. The vehicle information recognition ECU 1 outputs the calculated distance to the left curb, the distance to the right center line, the distance to the center of the right intersection, and the distance to the right center line to the right / left turn trajectory calculation ECU 3.

  The driving state determination ECU 2 outputs the accelerator pedal depression amount, the brake pedal depression amount, the steering angle, and the vehicle speed output from the vehicle information recognition ECU 1, and the vehicle speed and deceleration before the brake pedal depression is started. Based on the above, it is determined whether the “brake action evaluation” corresponds to “less than weak”, “weak”, “center”, “strong”, or “strong super”. The driving state determination ECU 2 outputs this “brake action evaluation” to the normal driving data recording unit 4 and the blurred state determination ECU 5.

  The driving state determination ECU 2 determines that the “steering wheel steering evaluation” is “faster”, “faster”, “center”, “slow”, “slower” based on the steering angle and the images taken by the vehicle information cameras 17, 18. Find which one corresponds to it. The driving state determination ECU 2 outputs this “steering wheel steering evaluation” to the normal driving data recording unit 4 and the blurred state determination ECU 5.

  Based on the right / left turn information output from the vehicle information recognition ECU 1 and the turning radius when the vehicle starts turning, the “turning radius evaluation” is “less than minimum”, “minimum”, “center”. Determine which corresponds to “maximum” or “greater than”. The left / right turn trajectory calculation ECU 3 outputs this “turn radius evaluation” to the normal operation data recording unit 4 and the blurred state determination ECU 5.

  Further, the right / left turn trajectory calculation ECU 3 determines that “distance evaluation with the left curb” is “riding”, “zero”, “minimum”, “minimum” based on the right / left turn information output from the vehicle information recognition ECU 1 and the distance to the left curb. Determine which corresponds to “center” or “maximum”. The right / left turn trajectory calculation ECU 3 outputs this “distance evaluation with the left curb” to the normal operation data recording unit 4 and the blurred state determination ECU 5.

  Further, the right / left turn trajectory calculation ECU 3 determines that “distance evaluation with the center line” is “override”, “zero”, “minimum” based on the right / left turn information output from the vehicle information recognition ECU 1 and the distance to the right center line. It is determined whether it corresponds to “center” or “maximum”. The left / right turn trajectory calculation ECU 3 outputs this “distance evaluation with the center line” to the normal operation data recording unit 4 and the blurred state determination ECU 5.

  Further, the right / left turn trajectory calculation ECU 3 determines that the “distance evaluation with respect to the intersection center” is “less than minimum”, “minimum”, “center” based on the right / left turn information output from the vehicle information recognition ECU 1 and the distance to the right intersection center. ”“ Maximum ”or“ Exceeding the maximum ”is determined. The right / left turn trajectory calculation ECU 3 outputs the distance to the “distance evaluation with respect to the intersection center” to the normal operation data recording unit 4 and the blurred state determination ECU 5.

  The normal operation data recording unit 4 is a running state storage unit, and stores a left turn determination index map shown in FIG. 2 and a right turn determination index map shown in FIG. 3 as reference running states. Here, the left turn determination index map shown in FIG. 2 includes a deceleration determination index (a) corresponding to “brake action evaluation”, a timing determination index (b) corresponding to “steering wheel steering evaluation”, and “turning radius evaluation”. The small turn determination index (c) corresponding to, the ride determination index (d) corresponding to “distance evaluation with the left curb”, and the bulge determination index (e) corresponding to “distance evaluation with the center line” are recorded. Yes.

  The deceleration determination index (a) is the smallest when the magnitude of the brake is a median value, and becomes a larger value as the deceleration of the vehicle becomes stronger or weaker. However, the amount of increase in the deceleration determination index becomes larger when the deceleration becomes weaker than when the deceleration becomes stronger. The timing determination index (b) is smallest when the steering start timing is a median value, and increases as the steering steering start timing is delayed or accelerated. However, the amount of increase in the timing determination index is larger when the start timing is earlier than when the start timing is later.

  Further, the small turning determination index (c) is smaller when the turning radius is smaller, and becomes larger as the turning radius is larger. The riding determination index (d) is the smallest value when the distance to the left curb is the allowable minimum value, and is a large value when it is less than the minimum and becomes zero or riding. When the value is larger than the minimum value, the larger the value, the larger the value. However, the value in the case of zero or riding is a larger value. The bulge determination index (e) decreases as the distance from the center line increases, and increases as the distance from the center line decreases.

  The right turn determination index map shown in FIG. 3 includes a deceleration determination index (A) corresponding to “brake action evaluation”, a timing determination index (B) corresponding to “steering wheel steering evaluation”, and “turning radius evaluation”. Corresponding small turn determination index (C), bulge determination index (D) corresponding to “distance evaluation with left curb”, determination index (E) from the right corresponding to “distance evaluation with center line”, and “intersection center” The in-cut determination index (F) corresponding to “distance evaluation” is recorded.

  The deceleration determination index (A) is the smallest when the magnitude of the brake is a median value, and becomes a larger value as the deceleration of the vehicle becomes stronger or weaker. However, the amount of increase in the deceleration determination index becomes larger when the deceleration becomes weaker than when the deceleration becomes stronger. The timing determination index (B) is smallest when the steering start timing is a median value, and becomes larger as the steering start timing is delayed or earlier. However, the amount of increase in the timing determination index is larger when the start timing is earlier than when the start timing is later. Further, the small turning determination index (C) is smaller when the turning radius is smaller and becomes larger as the turning radius is larger.

  The bulge determination index (D) is the smallest value when the distance to the left curb is minimum, and is a large value when the distance is less than the minimum or zero. When the value is larger than the minimum value, the larger the value, the larger the value. However, the value in the case of zero or riding is a larger value. The rightward determination index (E) is smaller as the distance from the center line is larger, and becomes larger as the distance from the center line is smaller. The in-cut determination index (F) is smaller as the distance from the intersection center is smaller, and increases as the distance from the intersection center increases.

  In the normal driving data recording unit 4, “brake action evaluation” and “steering wheel steering evaluation” output from the driving state determination ECU 2, “turning radius evaluation” output from the right / left turn trajectory calculation ECU 3, “distance to the left curb” Based on "Evaluation", "Distance evaluation with center line", and "Distance evaluation with intersection center", the left turn determination index map and the right turn determination index map are updated. For example, the normal driving data recording unit 4 stores a history of “braking action evaluation” output from the driving state determination ECU 2 and outputs a result close to “weak” as “braking action evaluation”. In the case of increasing, the deceleration determination index is adjusted to 1 when “braking action evaluation” is “weak”. In this way, determination according to the driver's habit can be performed.

  Further, the normal operation data recording unit 4 stores the blur determination values x (for example, 15), y (for example, 25), and z (for example, 35) that serve as threshold values for determining that the vehicle is driving in a blurred manner when turning right or left. ing. When the blurred state characteristic value described later exceeds the first blurred determination value x, it is determined as “slightly blurred operation”. Further, when the blurred state characteristic value exceeds the second blurred determination value y, it is determined as “blurred driving”, and when it exceeds the third blurred determination value, it is determined as “substantially blurred driving”. The normal operation data recording unit 4 outputs the left turn determination index map shown in FIG. 2, the right turn determination index map shown in FIG. 3, and the blurred determination values x, y, and z to the blurred state determination ECU 5.

  When the left turn information is output from the vehicle information recognition ECU 1, the blurred state determination ECU 5 receives the deceleration determination index (a), the timing determination index (b), the small turn determination index (c), the riding determination index (d), and the bulge A determination index (e) is obtained. The deceleration determination index (a) is obtained by referring to the “brake action evaluation” output from the driving state determination ECU 2 in the left turn determination index map output from FIG. 3. The timing index (b) is “steering wheel steering evaluation”, the small turning determination index (c) is “turning radius evaluation”, the riding determination index (d) is “distance evaluation with the left curb”, and the bulge determination index ( e) is obtained by referring to the “distance evaluation with the center line” in the left turn determination index map.

  In addition, when the right turn information is output from the vehicle information recognition ECU 1, the blurred state determination ECU 5 receives the deceleration determination index (A), the timing determination index (B), the small turn determination index (C), the bulge determination index (D), A rightward determination index (E) and an incut determination index (F) are obtained. The deceleration determination index (A) is obtained by referring to the “brake action evaluation” output from the driving state determination ECU 2 in the left turn determination index map output from FIG. 3. Further, the timing index (B) is “steering wheel steering evaluation”, the small turning determination index (C) is “turning radius evaluation”, the bulge determination index (D) is “distance evaluation with the left curb”, and the right side determination index ( E) is obtained by referring to the “distance evaluation with the center line” and the in-cut determination index is referred to by referring to the “distance evaluation with the center of the intersection” in the determination index map on the right turn.

  Further, the blurred state determination ECU 5 uses the deceleration determination index (a), the timing determination index (b), the small turn determination index (c), the riding determination index (d), and the bulge determination index (e), and the following (1 ) To obtain the dim state characteristic value KL when turning left.

KL = a + b + c + (d + e) / 2 (1)
In the above equation (1), the ride determination index (d) and the bulge determination index (e) appear in opposite directions, so that by using the average, weighting is performed between the other index. Is going.

  Further, the blurred state determination ECU 5 includes a deceleration determination index (A), a timing determination index (B), a small turn determination index (C), a bulge determination index (D), a rightward determination index (E), and an incut determination index (F). ) To obtain the blurred state characteristic value KR when turning right according to the following equation (2).

KR = A + B + C + (D + E + F) / 3 (2)
In the above formula (2), the bulge determination index (D), the right-side determination index (E), and the incut determination index (F) appear in directions opposite to each other. Weighting is performed between other indices.

  Further, the blurred state determination ECU 5 compares the left turn blurred state characteristic value KL or the right turn blurred state characteristic value KR with the blurred determination value x, y, z output from the normal operation data recording unit 4. As a result, if the left turn blurred state characteristic value KL or the right turn blurred state characteristic value KR exceeds the blurred determination values x, y, and z, the blurred information corresponding to the mode is output to the alarm device 6. Specifically, when the left turn blurred state characteristic value KL or the right turn blurred state characteristic value KR exceeds the first blurred determination value x, “slightly blurred information” is displayed. If the “blurring information” exceeds the third blurring determination value z, “substantially faint information” is output to the alarm device 6.

  The alarm device 6 is provided in a vehicle, for example, and includes a speaker that emits a voice to the driver. The warning device 6 emits a sound corresponding to the blurred information output from the blurred state determination ECU 5 to alert the driver.

  A control processing procedure of the driver state determination device according to the present embodiment having the above configuration will be described. FIG. 4 is a flowchart showing the procedure of the driver state determination device according to the present embodiment. At the intersection, when the vehicle makes a right or left turn, the correspondence between the key preft principle and the inner wheel difference of the vehicle requires fast steering operation at a large steering angle as well as vehicle speed control by deceleration. And tension will be forced. For this reason, when a low state of arousal, a drowsiness, or a blurred state where attention is paid to other than driving due to thoughts, a change in the behavior of the vehicle is more likely to occur than in normal times.

  For example, as shown in FIG. 5, when a vehicle enters an intersection and makes a left turn, in a normal case, the vehicle turns left after passing through positions P11 to P14, as indicated by a phantom line P15. , Too close to the inner curb X1, and more likely to get on. Alternatively, a behavior such as crossing the center line is likely to occur as in the position P16.

  On the other hand, when the vehicle enters the intersection and makes a right turn, in a normal case, the vehicle turns right after passing through the positions P21 to P24, and rides on the outer curb X2, or after turning right like the position P25. It becomes easy to step on the center line. Therefore, the driver state determination device according to the present embodiment detects the blurred state from the behavior of the vehicle when entering the intersection.

  As shown in FIG. 4, in the driver state determination device according to the present embodiment, first, based on the presence / absence of an instruction signal output from the winker device, it is determined whether to make a right turn or a left turn (S1). . As a result, when it is determined that neither a right turn nor a left turn is performed, the process returns to step S1 and the determination is repeated until a right turn or a left turn is performed.

  In step S 1, when a left turn signal is output from the winker device and it is determined that a left turn is to be made, the vehicle information recognition ECU 1 outputs the accelerator pedal depression amount output from the accelerator stroke sensor 11 and the brake stroke sensor 12. Until the steering wheel is cut out based on the brake pedal depression amount, the steering angle output from the steering angle sensor 13, the vehicle speed output from the vehicle speed sensor 14, the acceleration output from the front / rear / left / right acceleration sensor 15, etc. The vehicle speed when the vehicle is decelerated and the steering wheel is cut out is detected (S2). Based on the vehicle deceleration and the vehicle speed when the steering wheel is cut out, the driving state determination ECU 2 obtains “brake action evaluation”.

  Subsequently, the curb angle on the left side of the vehicle and the handle cut-out position are calculated based on the image photographed by the vehicle left side information camera 18 (S3). Here, the curb angle means an angle formed by the traveling direction of the vehicle and the curb side edge. Based on the curb angle and the handle cut-out position, “steering wheel steering evaluation” is obtained. Further, based on the vehicle speed output from the vehicle speed sensor 14 and the yaw rate output from the angular acceleration sensor 16, a turning radius during a left turn is calculated (S4). Based on the turning radius, the driving state determination ECU 2 calculates “turning radius evaluation”.

  Then, the distance between the left curb and the left rear wheel is obtained based on the image taken by the vehicle left information camera 18 (S5). Based on the distance between the left curb and the left rear wheel, the right / left turn trajectory calculation ECU 3 obtains “distance evaluation with the left curb”. Thereafter, based on the image taken by the vehicle right side information camera 17, the distance between the road center line (lane right side lane) and the right front wheel after the left turn is obtained (S6). Based on the distance between the center line and the right front wheel, the right / left turn trajectory calculation ECU 3 calculates “distance evaluation with the center line”.

  Further, the driving state determination ECU 2 outputs “brake action evaluation” and “handle operation evaluation” to the normal driving data recording unit 4 and the blurred state determination ECU 5. The right / left turn trajectory calculation ECU 3 outputs “turn radius evaluation”, “distance evaluation with the left curb”, and “distance evaluation with the center line” to the normal operation data recording unit 4 and the blurred state determination ECU 5. Further, the normal operation data recording unit 4 updates the data and outputs the left turn determination index map and the blurred determination values x, y, and z to the blurred state determination ECU 5.

  Thereafter, the vague state determination ECU 5 performs “brake action evaluation”, “steering wheel operation evaluation”, “turning radius evaluation”, “distance evaluation with the left curb”, and “distance evaluation with the center line” respectively. With reference to the map, a deceleration determination index (a), a timing determination index (b), a small turn determination index (c), a riding determination index (d), and a bulge determination index (e) are obtained (S7). Then, using the equation (1), the left turn blurred state characteristic value KL is calculated (S8).

  Subsequently, the left turn blurred state characteristic value KL and the blurred determination values x, y, z are compared (S20). As a result, if the left turn blurred state characteristic value KL exceeds the blurred determination values x, y, and z, it is determined that each is blurred (S21), and the blurred information corresponding to the blurred state is output to the alarm device 6. .

  When entering the intersection and making a left turn, if the steering speed of the steering wheel is slow while the vehicle speed is high, the trajectory of the vehicle at the time of the left turn swells and crosses the center line of the road at the left turn destination. On the other hand, if the steering wheel is cut out at an early timing so that the behavior of the vehicle does not swell, it will ride on the curb this time. When these behaviors occur, an attempt is made to determine the blurred state of the driver.

  In the driver state determination device according to the present embodiment, "braking action evaluation", "steering wheel steering evaluation", "turning radius evaluation", "distance evaluation with left curb", and "distance evaluation with center line" when turning left These five evaluation items are set, and the blurred state characteristic value is calculated using an index corresponding to these evaluation items. In general, when the driver's degree of blur is high, the braking action tends to be weaker than usual or, conversely, too strong. However, in many cases, the brakes tend to be weakened. In addition, the steering timing becomes slower than usual or, on the contrary, too early. However, many cases of blurring tend to be slow.

  Furthermore, if the driver is obscured, there is a tendency to make a large turn when turning, but conversely, the turning radius is too small and the steering wheel is cut out quickly, so the rider rides on the curb on the left side or the curb. It can get very close. Also, when turning left, the center line is usually not stepped on, but when the degree of blur is high, the center line may be stepped over.

  Based on the characteristics of the behavior when these drivers are blurred, the blurred state characteristic value KL at the time of left turn is obtained by the above (1). The blurring state characteristic value KL at the time of left turn includes a deceleration determination index (a) corresponding to “brake action evaluation”, a timing determination index (b) corresponding to “steering wheel steering evaluation”, and a small turning determination index corresponding to “turning radius evaluation”. (C) Substituting the ride determination index (d) corresponding to “distance evaluation with the left curb” and the bulge determination index (e) corresponding to “distance evaluation with the center line” into the above equation (1) Ask. Since the blurred state characteristic value KL at the time of the left turn is compared with the blurred determination values x, y, z, the blurred state of the driver is determined, so that the blurred state (abnormal state) of the driver can be detected with high accuracy. .

  Here, when the driver is not blurred, a value near the median of past normal driving data is expected. For example, if the left turn blurred state characteristic value KL calculated by the equation (1) is in the vicinity of 14, it can be said that the driver is not in a blurred state and is in a driving state with no problem. In addition, three levels are set as the dimness determination values x, y, and z, and a warning according to the degree of dimness is given to the driver by determining “slightly sloppy driving”, “sloppy driving”, and “slightly sloppy driving”. You can make a call.

  In step S1, if a left turn signal is output from the winker device and it is determined that a right turn is to be made, the vehicle deceleration until the steering wheel is cut out and the vehicle speed when the steering wheel is cut out are determined as in the case of a left turn. It detects (S11). Based on the vehicle deceleration and the vehicle speed when the steering wheel is cut out, the driving state determination ECU 2 obtains “brake action evaluation”.

  Subsequently, a steering wheel cutout position is calculated (S12), and "steering wheel steering evaluation" is obtained based on the curb angle and the steering wheel cutout position. Further, the turning radius at the time of the left turn is calculated (S13), and “turning radius evaluation” is obtained in the driving state determination ECU 2 based on the turning radius. Subsequently, the distance between the left curb and the left rear wheel is obtained based on the image taken by the vehicle left information camera 18 (S14), and the right / left turn turning trajectory based on the distance between the left curb and the left rear wheel. In the calculation ECU 3, “distance evaluation with the left curb” is obtained.

  Then, the right / left turn trajectory calculation ECU 3 extrapolates each center line of the road before and after the right turn and calculates the intersection, that is, the position of the intersection center (S15). The intersection of the center line W1 before the right turn imaged by the vehicle right information camera 17 and the center line W2 after the right turn is detected by image processing, and this intersection point is obtained as the center of the intersection. Subsequently, the positional relationship (distance) between the intersection center and the right rear wheel is calculated (S16). The positional relationship (distance) between the intersection center and the right rear wheel is the shortest distance from the intersection center to the turning trajectory of the right rear wheel. Based on the positional relationship (distance) with the intersection center, “distance evaluation with the intersection center” is obtained.

  In calculating this positional relationship, the inner direction is positive. The distance between the center of this intersection and the center line (white line on the right side of the lane) W2 after the right turn and the right front wheel is calculated (S17). Based on the distance between the center line of the road after the right turn and the right rear wheel, “distance evaluation with the center line” is obtained.

  Then, the driving state determination ECU 2 outputs “brake action evaluation” and “steering wheel operation evaluation” to the normal driving data recording unit 4 and the blurred state determination ECU 5. Further, the right / left turn trajectory calculation ECU 3 performs “turning radius evaluation”, “distance evaluation with the left curb”, “distance evaluation with the center line”, and “distance evaluation with the intersection center” in the normal operation data recording unit 4. And output to the blurred state determination ECU 5. Further, the normal operation data recording unit 4 updates the data and outputs the left turn determination index map and the blurred determination values x, y, and z to the blurred state determination ECU 5.

  Thereafter, the vague state determination ECU 5 performs “brake action evaluation”, “steering wheel operation evaluation”, “turning radius evaluation”, “distance evaluation with the left curb”, “distance evaluation with the center line”, and “intersection center”. “Evaluation of distance” is referred to the right turn determination index map, the deceleration determination index (A), the timing determination index (B), the small turn determination index (C), the bulge determination index (D), the right shift determination index (E), and An incut determination index (F) is obtained (S18). Then, the blurred state characteristic value KR is calculated using the above equation (2) (S19).

  Subsequently, the blurring state characteristic value KR at the time of right turn is compared with the blurring determination values x, y, z (S20). As a result, when the right turn blurred state characteristic value KR exceeds the blurred determination values x, y, and z, it is determined as a blurred state (S21), and the blurred information corresponding to the blurred state is output to the alarm device 6. .

  When entering the intersection and making a right turn, if the steering speed of the steering wheel is slow while the vehicle speed is high, the trajectory of the vehicle at the time of the left turn swells and rides on the curb at the tip of the right turn. On the other hand, if the steering wheel is cut out at an early timing so that the behavior of the vehicle does not swell, this time, the vehicle will step on the center line of the road where it turns right or pass through a position that is too close to the center of the intersection. When these behaviors occur, an attempt is made to determine the blurred state of the driver.

  In the driver state determination device according to the present embodiment, when making a right turn, `` brake action evaluation '', `` steering wheel steering evaluation '', `` turning radius evaluation '', `` distance evaluation with left curb '', `` distance evaluation with center line '', In addition, six evaluation items for distance evaluation with respect to the intersection center are set, and the blurred state characteristic value is calculated using an index corresponding to these evaluation items. In general, when the driver's degree of blur is high, the braking action tends to be weaker than usual or, conversely, too strong. However, in many cases, the brakes tend to be weakened. In addition, the steering timing becomes slower than usual or, on the contrary, too early. However, many cases of blurring tend to be slow.

  Furthermore, if the driver is obscured, there is a tendency to make a large turn when turning, but conversely, the turning radius is too small and the steering wheel is cut out quickly, so the rider rides on the curb on the left side or the curb. It can get very close. Also, when turning left, the center line is usually not stepped on, but when the degree of blur is high, the center line may be stepped over.

  Based on the characteristics of the behavior when these drivers are blurred, the blurred state characteristic value KR at the time of right turn is obtained by the above (2). The blurring state characteristic value KR when turning right is a deceleration determination index (A) corresponding to “brake action evaluation”, a timing determination index (B) corresponding to “steering wheel steering evaluation”, and a small turning determination index corresponding to “turning radius evaluation”. (C), bulge determination index (D) corresponding to “distance evaluation with left curb”, right-side determination index (E) corresponding to “distance evaluation with center line”, and “distance evaluation with intersection center” The corresponding incut determination index (F) is obtained by substituting it into the above equation (2). Since the driver's blurred state is determined by comparing the blurred state characteristic value KR at the right turn with the blurred determination values x, y, and z, the driver's blurred state (abnormal state) can be accurately detected. .

  Here, when the driver is not blurred, a value near the median of past normal driving data is expected. For example, if the right turn blurred state characteristic value KR calculated by the equation (2) is in the vicinity of 14, it can be said that the driver is not in a blurred state and is in a driving state with no problem. In addition, three levels are set as the dimness determination values x, y, and z, and a warning according to the degree of dimness is given to the driver by determining “slightly sloppy driving”, “sloppy driving”, and “slightly sloppy driving”. You can make a call.

  The preferred embodiment of the present invention has been described above, but the present invention is not limited to the above embodiment. For example, in the above-described embodiment, evaluation items of 5 items and 6 items are set for turning right and left, respectively, but any of these items can be appropriately selected and evaluated. In the above embodiment, an alarm is issued when the blurred state characteristic values KL and KR exceed the blurred determination values x, y, and z. For example, the cumulative value exceeding the blurred determination value has a predetermined value. An alarm may be issued when the number of times exceeds the limit, or an alarm may be generated when the number of times exceeding the blur determination value exceeds a predetermined number of times within a predetermined time.

It is a block block diagram of a driver | operator state determination apparatus. It is a figure which shows the determination index map at the time of a left turn. It is a figure which shows the determination index map at the time of a right turn. It is a flowchart which shows the treatment procedure of the driver | operator state determination apparatus which concerns on this embodiment. It is a top view which shows the behavior of the vehicle in an intersection.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Vehicle information recognition ECU, 2 ... Driving state determination ECU, 3 ... Right / left turn trajectory calculation ECU, 4 ... Normal driving data recording part, 5 ... Blurred state determination ECU, 6 ... Alarm device, 11 ... Accelerator stroke sensor, 12 DESCRIPTION OF SYMBOLS ... Brake stroke sensor, 13 ... Steering steering angle sensor, 14 ... Vehicle speed sensor, 15 ... Front / rear / left / right acceleration sensor, 16 ... Angular acceleration sensor, 17 ... Vehicle right side information camera, 18 ... Vehicle left side information camera.

Claims (8)

A scene detection means for detecting that the vehicle is in a course change scene at an intersection;
Traveling state detection means for detecting a traveling state of the vehicle when a course change scene is detected by the scene detection means;
Traveling state storage means for storing a reference traveling state serving as a reference of a traveling state when the vehicle is in a course change scene;
A determination unit that determines a driver's state based on a deviation between the traveling state detected by the traveling state detection unit and a reference traveling state stored in the traveling state storage unit;
A driver state determination device comprising:   The driver state determination device according to claim 1, wherein a turning radius of the vehicle is set as the traveling state.   The driver state determination device according to claim 1, wherein a position of the vehicle with respect to a reference position set at an intersection is set as the traveling state.   The driver state determination device according to claim 1, wherein a turning start point of the vehicle is set as the traveling state.   The driver state determination device according to claim 1, wherein a magnitude of deceleration before the start of a steering operation is set as the traveling state.   The driver state determination device according to claim 1, wherein a vehicle speed at the start of a steering operation is set as the traveling state.   The driver state determination device according to claim 1, wherein a traveling state corresponding to a right or left turn of the vehicle is set as the traveling state. The determination means is abnormal in a driver's condition when a deviation between the running condition detected by the running condition detection means and a reference running condition stored in the running condition storage means exceeds a predetermined threshold value. The driver state determination device according to any one of claims 1 to 7, wherein the driver state determination device determines that there is a drive.

Images