JP3638980B2 - Vehicle alarm device - Google Patents

Description

[0001]
[Industrial application fields]
The present invention relates to an alarm device that assists a driver's safe driving while traveling in a vehicle such as an automobile, and more particularly to prevention of a driver's sleep, collision, and lane departure.
[0002]
[Prior art]
In recent years, a comprehensive driving assistance system (ADA, Active Drive Assist system) that actively assists driving operations has been developed in order to dramatically improve the safety of vehicles in response to an increasing trend of traffic accidents. In this system, it is indispensable to recognize the external environment of the vehicle, but it processes the image information of the scenery and objects in front of the vehicle captured by multiple cameras, so that the road and traffic environment have sufficient accuracy and time for practical use. It has become possible to recognize three-dimensionally. Therefore, the ADA system uses the image data of the road and traffic environment, and if the driver makes an operation mistake, looks aside, or falls asleep during monotonous driving, there is a possibility of a collision or lane departure. The aim is to assist from various angles in order to predict the presence or absence and lead to the safe side.
[0003]
Here, as a way of driving support, first, preventive safety (not falling into a dangerous state from the beginning) is aimed at based on the idea that a vehicle is operated by a human. That is, when the possibility of a collision or lane departure is predicted, an alarm is sounded to assist the driver in safe driving. If the driver does not perform the appropriate avoidance operation even when the alarm is sounded, the driving operation system of the brake, throttle or steering is automatically controlled to the safe side to avoid a collision, lane departure, etc. It has been proposed to assist directly to act as a substitute.
[0004]
Therefore, alarms in the ADA system are very important in predicting various possibilities and improving preventive safety. For this reason, it is possible to determine what is the possibility in what situation, effectively predict the possibility by effectively utilizing the information of the image data, and sound the alarm properly in a state where the avoidance operation by the driver is possible Etc. are required.
[0005]
Here, as various possibilities, a drowsy driving due to a decrease in the driver's arousal level can be considered. The driver's arousal level generally decreases when a monotonous operation with a small amount of driving operation of the driver continues for a long time. However, even if the driver's driving volume is low due to the continuous straight road, the driver has psychological stimuli, such as psychological burdens such as passing cars and interrupting cars, narrow road lanes, and visual stimuli. If there is even one, the driver's arousal level is kept high. Therefore, it is desirable that the degree of driver's arousal level is detected according to a psychological stimulus reflected in the driver's eyes, and a doze alarm is based on this.
[0006]
Further, it is known that in the case of drowsy driving, the possibility of encountering a collision with a preceding vehicle or the like, a lane departure, or the like is very high. Therefore, when the possibility of a collision or lane departure is predicted in a state where the driver's arousal level is low, the timing of the warning is advanced earlier than usual to prevent a drowsiness and a collision or lane departure due to the above. It is desirable to make a preventive safety alarm.
[0007]
Conventionally, with regard to the preventive safety alarm, for example, there is a prior art disclosed in Japanese Patent Publication No. 59-16968. When the number of driving operations in a certain section is less than a set value, it is determined that the section is a monotone section, and the accumulation of this monotonous section is performed. It is shown that a wake-up alarm is issued when the number exceeds the reference number. In the prior art disclosed in Japanese Patent Publication No. 59-16969, it is shown that a monotonous running state is judged and an alarm is issued when the frequency at which the steering wheel operation amount or the accelerator operation amount does not reach a set value. In the prior art disclosed in Japanese Patent Publication No. 61-53250, an environment in which the driver's level of consciousness may be lowered depending on the state of steering wheel operation is judged, and in this environment, the steering angle value is cumulatively calculated and the road condition is monotonous. It is shown that an alarm will be given if it is determined.
[0008]
Regarding the alarm timing, for example, there is a prior art disclosed in JP-A-5-162596, and the safe inter-vehicle distance is calculated based on the vehicle speed. If the current inter-vehicle distance is less than the safe inter-vehicle distance, it is determined that there is a possibility of a rear-end collision. Alarm. In this case, it is shown that the driver's awareness level is evaluated from the inter-vehicle distance during brake operation, the average inter-vehicle distance during normal driving, and the standard deviation, and the alarm output timing is adjusted according to the evaluation result. .
[0009]
[Problems to be solved by the invention]
By the way, in all of the above prior arts, the monotonous driving is judged by the operation pattern in which the steering wheel operation amount and the accelerator operation amount of the driving operation are reduced, and further, the road condition is judged and an alarm is given. Therefore, since the degree of arousal level due to the psychological stimulus of the driver is not detected, an alarm is often issued unnecessarily. Also, when adjusting the alarm timing, the driver's level of consciousness is evaluated based on the distance between the vehicles at the time of brake operation. However, since the brake operation state changes depending on the road and traffic environment, the level of consciousness is accurately evaluated. It is difficult.
[0010]
In view of these points, the present invention has an object to directly and accurately detect the degree of arousal by a driver's psychological stimulus, and to reliably prevent falling asleep and causing a collision or lane departure.
[0011]
[Means for Solving the Problems]
  In order to achieve this object, the vehicle alarm device according to claim 1 of the present invention, as shown in FIG. 1, is an image recognition means 27 for obtaining image data obtained by three-dimensionally recognizing the road and traffic environment ahead of the vehicle. Whether there are roads ahead of the vehicle based on the image data, moving objects, obstacles, road shapes, etc. that are visible to the driver's eyes depending on the traffic environment and vehicle speed, and there is a psychological stimulus that is a driver's psychological burden and visual stimulus The road lane width is less than the set value, the radius of curvature is greater than the set value, there is a possibility of a rear-end collision with the preceding vehicle, there is a parked vehicle, and there are other moving obstacles However, if there is at least one psychological stimulus, if there is at least one psychological stimulus, the psychological stimulus determination means 43 determines that there is no psychological stimulus.Things and psychologyAwakening degree detection means 44 for detecting that the driver's awakening degree has dropped when no state stimulus is present for a set time, and a collision prediction means for predicting the possibility of a collision with an obstacle based on image data and vehicle speed 46, lane departure prediction means 47 for predicting the possibility of lane departure from image data, and when detecting a decrease in wakefulness, when predicting a collision or lane departure when the wakefulness decreases, the warning timing is changed respectively. The alarm timing control means 48 which issues an alarm is provided.
[0012]
  The vehicle alarm device according to claim 2 is:The warning timing control means 48 is characterized in that when a collision or lane departure is predicted when the arousal level is lowered, the warning is corrected and corrected so that the collision or lane departure is predicted earlier, and the warning is issued early.
[0013]
  According to a third aspect of the present invention, there is provided an alarm device for a vehicle according to a third aspect, wherein image recognition means for obtaining image data obtained by three-dimensionally recognizing a road and traffic environment ahead of the vehicle; Whether there are moving objects, obstacles, road shapes, etc., and whether there is a psychological stimulus that is a driver's psychological burden or visual stimulus,When the width of the road lane is less than or equal to the set value, when the change in the radius of curvature of the road is greater than or equal to the set value, there are obstacles or moving objects, and various driving operations are required, or visual stimulation If there isThe psychological stimulus determination means for determining that there is a psychological stimulus when any one of them is satisfied, and the driver's arousal level is high if there is at least one psychological stimulusThings and psychologyWakefulness detection means for detecting that the driver's wakefulness has been lowered when no state stimulus continues for a set time, and collision prediction means for predicting the possibility of a collision with an obstacle based on image data and vehicle speed; Lane departure prediction means for predicting the possibility of lane departure from image data, and when detecting a decrease in wakefulness, when a collision or lane departure is predicted when the wakefulness decreases, an alarm is generated by changing the alarm timing respectively. And an alarm timing control means. According to a fourth aspect of the present invention, there is provided a vehicle alarm device comprising: an image recognition means for obtaining image data obtained by three-dimensionally recognizing a road ahead of the vehicle and a traffic environment; and the width of the lane, the road ahead of the vehicle, the traffic environment A psychological stimulus judging means for judging whether there is a psychological stimulus which is a moving object, an obstacle, a road shape, etc., which is visible to the driver's eyes according to the vehicle speed, and which is a psychological stimulus or visual stimulus for the driver; Arousal level detection means for detecting that the driver's arousal level is high when there is even one psychological stimulus and that the driver's arousal level has decreased when a state in which there is no psychological stimulus continues for a set time, image data, Collision prediction means that predicts the possibility of collision with an obstacle based on vehicle speed, lane departure prediction means that predicts the possibility of lane departure based on image data, and if a decrease in arousal level is detected, a collision or lane when the arousal level decreases When predicting the de characterized by comprising a warning timing control means for issuing an alarm by changing the alarm time by decreasing correcting the width of the lane.
[0014]
[Action]
  Accordingly, in claim 1 of the present invention, the road and traffic environment ahead of the vehicle are always recognized by the image recognition means 27 when the vehicle is traveling, and image data relating to the position, shape, speed, etc. of the white line and various obstacles is obtained. can get. And psychological stimulus determination means 43As a psychological stimulus reflected in the driver's eyes, the width of the road lane is less than the set value, the change in the radius of curvature is greater than the set value, there is a preceding vehicle, and there is a possibility of a rear-end collision with the preceding vehicle. If any one of the various elements is present, there is a psychological stimulus and the driver's arousal level is appropriately determined, and one of them is not satisfied It is appropriately determined that there is no psychological stimulus and the driver's arousal level may be reduced.For this reason, when the driving state is monotonous due to the road shape and there is no psychological stimulus at all for a set time, the arousal level detecting means 44 directly and accurately detects that the driver's arousal level has decreased. The
[0015]
The collision prediction means 46 predicts the possibility of a collision, and the lane departure prediction means 47 predicts the possibility of a lane departure. Then, when the alarm timing control means 48 detects a decrease in the arousal level, the driver is prevented from falling asleep. Also, if a collision or lane departure is predicted when the arousal level is lowered, the alarm timing is changed and a warning is issued, so that the driver's doze and the resulting collision or lane departure can be reliably prevented.
[0016]
  In claim 2,If a collision or lane departure is predicted when the degree of arousal is reduced, the warning timing control means 48 predicts them early and issues an early warning, thereby reliably preventing the driver from falling asleep and causing a collision or lane departure. .
[0017]
  According to claim 3, the psychological stimulus determination means at least determines whether or not there is a psychological stimulus,When the width of the road lane is less than or equal to the set value, when the change in the radius of curvature of the road is greater than or equal to the set value, there are obstacles or moving objects, and various driving operations are required, or visual stimulation If there isBy determining that there is a psychological stimulus when any one of them is satisfied, it is possible to accurately determine the presence or absence of a psychological stimulus that is reflected in the eyes of the driver and becomes a psychological burden or a visual stimulus. According to the fourth aspect of the present invention, the warning timing control means corrects the lane width to be reduced and changes the warning timing to issue an alarm so that the lane departure is predicted early. The lane departure can be reliably prevented.
[0018]
【Example】
Embodiments of the present invention will be described below with reference to the drawings. In FIG. 2, the outline of the entire vehicle and the ADA system will be described. First, in the vehicle 1, an engine 2 is connected to a rear wheel 9 via a clutch 4, a transmission 5, a propeller shaft 6, a rear differential device 7, a rear axle 8, and the like, and a center differential device 17 and a front differential device are connected from the propeller shaft 6. 18 is connected to the front wheel 10 via the front axle 19 and the like, and is configured to be able to travel by driving the front and rear wheels 9 and 10. As a driving operation system of the vehicle 1, an accelerator pedal 11 is provided so as to change the engine output by opening and closing the throttle valve 3 of the engine 2. Further, a brake device 12 is provided so as to introduce a brake pressure to the wheel cylinder side of the front and rear wheels 9 and 10 through a brake line 14 by an operation of the brake pedal 13 and brake the brake pressure. Further, a steering device 15 is provided to steer the front wheel 10 by operating the handle 16.
[0019]
The ADA system 20 is provided with a throttle actuator 21 in the throttle valve 3 as various actuators for vehicle control so that the engine output is forcibly reduced by a throttle signal. Further, a brake actuator 22 is provided in the brake pipe line 14 of the brake device 12 so as to automatically brake by increasing and decreasing the brake pressure by a brake signal. Further, a steering actuator 23 is provided in the steering device 15 so as to be automatically steered by a steering signal.
[0020]
The electronic control system will be described. As the image recognition means, for example, two CCD cameras 25 arranged on the left and right front of the vehicle 1 and an image recognition unit 26 of a stereo image processor are provided. The two CCD cameras 25 capture and capture a landscape and an object in front of the vehicle in a stereo manner. The image recognition unit 26 calculates the distance by processing the image signals picked up by the two CCD cameras 25 by the triangulation method by the stereo method, and creates a distance image having a three-dimensional distance distribution on the entire screen. Then, a lane, a preceding vehicle, an obstacle, and the like are separated and detected from the distance image, and the left and right white lines and the three-dimensional shape of the road are recognized from the lane. In addition, it recognizes what the object ahead is, the relative distance and speed with respect to the preceding vehicle and obstacles, and thus obtains image data of the road and traffic environment.
[0021]
The ADA control unit 40 includes an alarm control system that predicts various possibilities and gives an alarm, and a vehicle control system that is used when the driver does not perform an avoidance operation despite the alarm. The vehicle control system calculates the acceleration / deceleration based on the image data and other various sensor signals so as to maintain a safe distance from, for example, a preceding vehicle or an obstacle on the road, and an appropriate throttle opening based on this acceleration / deceleration. The throttle signal is output to the throttle actuator 21 to control the engine output. In addition, a brake signal of an appropriate brake pressure based on acceleration / deceleration is output to the brake actuator 22 to perform automatic brake control, thereby keeping a safe inter-vehicle distance or preventing a collision when a driver's operation mistake or avoidance operation is not performed. It is possible. On the other hand, a target trajectory at a predetermined distance position on the image is set, and a predicted trajectory is calculated when the vehicle has traveled to a predetermined distance position in the current traveling state, and according to a deviation between the target trajectory and the predicted trajectory A steering signal is output to the steering actuator 23 for automatic steering control, so that it is possible to avoid collision and prevent lane departure in the event of a driver's operation error or the like.
[0022]
As an alarm control system, an alarm for a driver's arousal level reduction, collision, and lane departure will be described. First, there are moving objects, obstacles, road shapes, etc. that are visible to the driver's eyes depending on the road and traffic environment in front of the vehicle based on the image data, and the vehicle speed of the vehicle speed sensor 30, which may be a psychological burden on the driver or a visual stimulus The psychological stimulus to be received is determined, and it is detected whether or not the driver's arousal level has decreased due to the presence or absence of the psychological stimulus. At the same time, the possibility of collision with obstacles and lane departure is predicted. When the awakening level is reduced and a collision or lane departure is predicted, the alarm 31 is changed and the alarm 31 is issued.
[0023]
Next, alarm control will be described. First, image signals captured by the two CCD cameras 25 while the vehicle is running are processed by the image recognition unit 26 to obtain image data of roads and traffic environments (step S1). With this image data, for example, in the case of a two-lane road, as shown in FIG. 6 (a), in the XY coordinates based on the position and direction of the host vehicle 1, the left and right and center white lines H1 to H3 in front of the vehicle are displayed. Data on position and shape coordinates, position coordinates of a moving obstacle M3 such as a preceding vehicle M1, a parked vehicle M2, a crossing vehicle or an oncoming vehicle in an oncoming lane of the host vehicle 1, and a speed vector are obtained.
[0024]
Therefore, the determination control of the psychological stimulus of the driver will be described with reference to the functional block of FIG. 3 and the flowchart of FIG. The image data is input to the road shape calculation means 41 of the ADA control unit 40 to calculate the lane width B and the lane curvature radius change ΔR that are necessary for proper steering operation as seen by the driver's eyes (step S2). ). As shown in FIG. 6B, the lane width B is determined by measuring positions YL and YR from the center of the vehicle width of the host vehicle 1 to the left and right white lines H1 and H2 of the travel lane. Calculated by YR-YL.
[0025]
When calculating the change ΔR in the curvature radius of the lane, first, as shown in FIG. 6C, the coordinates (x1, y1) and (x) of the center line O with respect to the three dividing lines N1, N2, N3 in front of the vehicle. x2, y2), (x3, y3) are obtained, and the curvature radius R of the lane is determined from these data as R = {(x−a)² + (Y−b)² }^1/2Calculated by Then, the change ΔR of the previous curvature radius R (n−1) and the current curvature radius R (n) calculated at the set time pitch is calculated by subtracting both. The lane width B and the curvature radius change ΔR are input to the psychological stimulus determination means 43, the lane width B is compared with the set value Bs, and the change ΔR is also compared with the set value Rs (step S3). When the lane width B is narrower than the set value Bs and the change ΔR is greater than the set value Rs, it is determined that there is a psychological stimulus that requires an appropriate steering operation by the driver (step S8). .
[0026]
Further, the image data is input to the obstacle detection means 42, and is a moving object and an obstacle reflected in the eyes of the driver. Various driving operations are required, or a preceding vehicle M1, a parked vehicle M2, which receives a visual stimulus, The other moving obstacle M3 is detected. The preceding vehicle M1 recognizes whether or not the obstacle in the traveling lane has a speed in the forward direction. Further, the possibility of a rear-end collision is determined based on the relationship between the host vehicle 1 and the preceding vehicle M1. That is, the inter-vehicle distance L1 between the host vehicle 1 and the preceding vehicle M1, the host vehicle speed V, the preceding vehicle speed V1, the set acceleration / deceleration α of the host vehicle 1, the set acceleration / deceleration α1 of the preceding vehicle M1, as shown in FIG. The possibility of a rear-end collision is determined by the following equation based on the set time T1 for taking a margin.
L1 ≦ (−V1² / 2α1) + {(V² / 2α) + V · T1}
[0027]
The parked vehicle M2 recognizes a vehicle having a speed vector of zero in the travel lane as the parked vehicle M2. For other moving obstacles M3, a predetermined range A is set near the left and right sides of the range B of the white lines H1 and H2 of the traveling lane as shown in FIG. 6E, and these ranges A and B are added. The range C for identifying the obstacle is determined. A moving obstacle M3 such as an oncoming vehicle, a passing vehicle, a crossing vehicle, a two-wheeled vehicle, or a pedestrian having a velocity vector in the entire range C is detected. These detection signals are also input to the psychological stimulus determination means 43 to determine the presence / absence of a preceding vehicle M1 (step S4), to determine the possibility of a rear-end collision (step S5), and to determine the presence / absence of a parked vehicle M2 ( In step S6), it is determined whether or not there is another moving obstacle M3 (step S7). If there is a possibility of various obstacles and rear-end collisions, and a psychological burden or visual stimulus that requires various operations of the driver is detected, the presence of a psychological stimulus is determined.
[0028]
As described above, the width B of the road lane is not more than the set value Bs, the change in curvature radius ΔR is not less than the set value Rs, there is a preceding vehicle M1, there is a possibility of a rear-end collision with the preceding vehicle M1, there is a parked vehicle M2, etc. If any one of the moving obstacles M3 is satisfied and there is a driver's psychological burden or visual stimulus, the presence of a psychological stimulus is determined. On the other hand, if none of them is satisfied and there is no psychological burden or visual stimulus of the driver, it is determined that there is no psychological stimulus (step S9).
[0029]
Next, arousal level determination and alarm control will be described with reference to the functional block of FIG. 3 and the flowchart of FIG. First, a psychological stimulus signal is input to the arousal level detection means 44 to determine whether or not there is a psychological stimulus (step S11). If there is even one psychological stimulus, the driver's arousal level is high. Detect that there is no risk of falling asleep. If there is no psychological stimulus, the duration 45 of the monotonous operation is counted by the timer 45 (step S12), the duration T is compared with the set time Ts (step S13), and the set time Ts is exceeded. It is directly detected that the driver's arousal level has decreased (step S14).
[0030]
As a result, the driver is psychologically stimulated in road conditions where the lane width B is narrow or the curvature radius change ΔR is large as in general roads, and there are some obstacles M1 to M3 ahead of the vehicle. It is properly detected that the degree of arousal is high. On the other hand, in a monotonous driving state of a dedicated road such as an expressway where the width B of the lane is wide, the change ΔR of the radius of curvature is small, and the amount of driving operation of the driver is small, a moving obstacle M3 such as a preceding vehicle M1, an oncoming vehicle, etc. If there is no driver's psychological burden and no visual stimulus continues for a long time, the driver's arousal level tends to decrease. In this case, the decrease in the arousal level is reliably detected.
[0031]
On the other hand, the image data and the vehicle speed V are input to the collision prediction means 46, and the possibility of collision with the obstacle M4 in the traveling lane of the host vehicle 1 is predicted. 7A, the relative distance L2 between the host vehicle 1 and the obstacle M4, the host vehicle speed V, the forward speed V1 of the obstacle M4, the set acceleration / deceleration β of the host vehicle 1, and the setting of the obstacle M4. The possibility of a collision is predicted by the following equation based on the acceleration / deceleration β1 and the set time T2 for taking a margin.
L2 ≦ (−V1² / 2β1) + {(V² / 2β) + V · T2}
[0032]
Further, the image data is input to the lane departure prediction means 47 to predict the possibility of departure from the traveling lane of the host vehicle 1. That is, as shown in FIG. 7B, the width B of the lane and the positions YL and YR of the left and right white lines H1 and H2 are measured, and the position Y of the host vehicle 1 with respect to the lane centerline O is determined as Y = (B / 2) Calculate by -YR = (YL-YR) / 2. Based on the tread W of the host vehicle 1, the position Y, and the width B of the lane, when Y ≦ (−B / 2) − (W / 2), a deviation from the left lane is predicted, and Y ≧ (B / 2) − ( When W / 2), the right lane departure is predicted.
[0033]
Signals for detecting a decrease in the driver's arousal level, predicting a collision with an obstacle or a lane departure are input to the alarm timing control means 48 to perform alarm timing control. That is, when the arousal level is lowered, the possibility of a collision is predicted (step S15). When there is no possibility, the possibility of lane departure is predicted (step S16). When there is no possibility of lane departure, An alarm signal is output to the alarm 31 at the time of detection of a decrease in wakefulness level (step S17). For this reason, an alarm sounds at the time when the driver's arousal level decreases, and the driver's sleep is reliably prevented.
[0034]
On the other hand, when the possibility of a collision is further predicted in a state where the driver's arousal level is detected, the set time T2 for taking a margin in the above-described determination formula is set by the correction amount ΔT provided for the low arousal level. , T2 + ΔT is corrected to increase and alarmed (step S18). For this reason, the above-mentioned determination condition is satisfied at a time earlier by ΔT than the time when the possibility of a collision is actually predicted, and an alarm is sounded early. Thus, the alarm prevents a doze like the above, and the driver can sufficiently perform the brake avoidance operation with an early alarm, thereby reliably preventing the collision.
[0035]
When the possibility of lane departure is predicted, the lane width B is corrected to decrease to B−ΔB by the correction amount ΔB provided for the arousal level reduction in the above-described determination formula, and an alarm is given (step S19). Therefore, the determination is made in relation to the width of the lane narrowed by ΔB, and the possibility of lane departure is predicted early, and an alarm is issued early. Therefore, the alarm prevents the dozing in the same manner as described above, and the driver can sufficiently perform the steering wheel avoidance operation with the early warning, thereby reliably preventing the departure from the lane.
[0036]
Although the embodiment of the present invention has been described above, the reference value and the measured value of the operation pattern and the vehicle movement pattern can be calculated by a method other than the embodiment. In the embodiments of the present invention, the alarm means is described using the alarm 34. However, as long as the driver's arousal level can be increased, the alarm means is used in a display device, a seat cushion, or a seat back provided in the driver's field of view. An internal vibration device that emits vibration may be used.
[0037]
【The invention's effect】
  As described above, in the vehicle alarm device according to claim 1 of the present invention, the image recognition means for obtaining image data obtained by three-dimensionally identifying the road and traffic environment ahead of the vehicle, and the vehicle Whether there are moving objects, obstacles, road shapes, etc. that are visible to the driver's eyes depending on the road, traffic environment and vehicle speed, and whether there is a psychological stimulus that is a psychological burden or visual stimulus for the driver, There is psychological stimulation when the width is less than the set value, the change in the radius of curvature is greater than the set value, there is a possibility of a rear-end collision with the preceding vehicle, there is a parked vehicle, or there are other moving obstacles. If there is at least one psychological stimulus, the driver's arousal level is high.Things and psychologyAnd arousal level detection means for detecting that the driver's arousal level has decreased when no state stimulus is present for a set time, the degree of arousal level is determined based on the driver's psychological stimulus. It can be detected directly and accurately. For this reason, the accuracy of the preventive safety doze alarm is greatly improved.
[0038]
Collision prediction means that predicts the possibility of collision of obstacles based on image data and vehicle speed, lane departure prediction means that predicts the possibility of lane departure based on image data, and when a decrease in arousal level is detected, a collision occurs when the arousal level decreases When a collision or lane departure is predicted when the degree of arousal is reduced, the warning timing changes and the driver changes the warning timing. Can be reliably prevented from falling asleep and causing collisions and lane departures.
[0039]
  In the vehicle alarm device according to claim 2,When the alarm timing control means predicts a collision or a lane departure when the arousal level is reduced, the alarm timing control means corrects the collision or the lane departure so as to be predicted earlier, so that an alarm can be issued early. For this reason, collision and lane departure can be reliably prevented by the dozing alarm.
[0040]
  In the vehicle alarm device according to claim 3, the psychological stimulus determination means at least whether or not there is a psychological stimulus,When the width of the road lane is less than the set value, when the change in the radius of curvature of the road is greater than the set value, there are obstacles or moving objects and various driving operations are required, or there is a visual stimulus If presentBy determining that there is a psychological stimulus when any one of them is satisfied, it is possible to accurately determine whether there is a psychological stimulus that is reflected in the eyes of the driver and becomes a psychological burden or a visual stimulus. Further, in the vehicle alarm device according to claim 4, the alarm timing control means corrects the lane width to be corrected by decreasing the lane width and changing the alarm timing to issue an alarm, so that the lane departure is predicted early. A warning can be issued early, and lane departure can be reliably prevented.
[Brief description of the drawings]
FIG. 1 is a claim correspondence diagram showing a configuration of a vehicle alarm device according to the present invention.
FIG. 2 is an explanatory diagram showing an outline of the entire vehicle and an ADA system.
FIG. 3 is a functional block diagram of an embodiment of the present invention.
FIG. 4 is a flowchart of psychological stimulus determination control.
FIG. 5 is a flowchart of arousal level detection and alarm control.
FIG. 6 is an explanatory diagram for determining image data and psychological stimulation.
FIG. 7 is an explanatory diagram for predicting a collision and a lane departure.
[Explanation of symbols]
27 Image recognition means
43 Psychological stimulus determination means
44 Arousal level detection means
46 Collision prediction means
47 Lane departure prediction means
48 Alarm timing control means

Claims (4)

Image recognition means for obtaining image data obtained by three-dimensionally recognizing a road in front of the vehicle and a traffic environment;
Whether there is a road in front of the vehicle based on the image data, a moving object, an obstacle, a road shape, etc. that are visible to the driver's eyes depending on the traffic environment and vehicle speed, and whether there is a psychological stimulus that is a psychological burden or visual stimulus for the driver The road lane width is less than the set value, the change in the radius of curvature is greater than the set value, there is a possibility of rear-end collision with the preceding vehicle, there is a parked vehicle, and there are other moving obstacles Psychological stimulation determination means for determining presence of psychological stimulation when satisfied, and determining that there is no psychological stimulation when none of them is satisfied;
Arousal level detection means for detecting that the driver's arousal level is high if there is even one psychological stimulus and that the driver's awakening level has decreased when a state in which there is no psychological stimulus continues for a set time;
A collision prediction means for predicting the possibility of a collision with an obstacle based on the image data and the vehicle speed;
Lane departure prediction means for predicting the possibility of lane departure from the image data;
An alarm device for a vehicle, comprising: an alarm timing control unit that issues an alarm by changing an alarm timing when a collision or a lane departure is predicted when an arousal level is reduced.   The warning timing control means, when predicting a collision or a lane departure when the arousal level is reduced, corrects the collision or the lane departure so as to predict it earlier, and issues a warning early. Vehicle warning device. Image recognition means for obtaining image data obtained by three-dimensionally recognizing a road in front of the vehicle and a traffic environment;
Whether there is a road in front of the vehicle based on the image data, a moving object, an obstacle, a road shape, etc. that are visible to the driver's eyes depending on the traffic environment and vehicle speed, and whether there is a psychological stimulus that is a psychological burden or visual stimulus for the driver If at least the width of the lane of the road is less than or equal to a set value, if the change in the radius of curvature of the road is greater than or equal to the set value, there are obstacles or moving objects and various driving operations are required. A psychological stimulus determination means for determining that there is a psychological stimulus when any one of the case or the case where the visual stimulus exists is satisfied,
Arousal level detection means for detecting that the driver's arousal level is high if there is even one psychological stimulus and that the driver's awakening level has decreased when a state in which there is no psychological stimulus continues for a set time;
A collision prediction means for predicting the possibility of a collision with an obstacle based on the image data and the vehicle speed;
Lane departure prediction means for predicting the possibility of lane departure from the image data;
An alarm device for a vehicle, comprising: an alarm timing control unit that issues an alarm by changing an alarm timing when a collision or a lane departure is predicted when an arousal level is reduced. Image recognition means for obtaining image data obtained by three-dimensionally recognizing a road in front of the vehicle and a traffic environment;
The image data includes a lane width, a road ahead of the vehicle, a moving object, an obstacle, a road shape, etc. that are visible to the driver according to the traffic environment and vehicle speed, and a psychological stimulus that is a psychological burden or visual stimulus for the driver. Psychological stimulus determination means for determining whether or not there is,
Arousal level detection means for detecting that the driver's arousal level is high if there is even one psychological stimulus and that the driver's awakening level has decreased when a state in which there is no psychological stimulus continues for a set time;
A collision prediction means for predicting the possibility of a collision with an obstacle based on the image data and the vehicle speed;
Lane departure prediction means for predicting the possibility of lane departure from the image data;
An alarm timing control means for generating an alarm by changing the alarm timing by correcting the decrease in the width of the lane when a collision or a lane departure is predicted when the arousal level is decreased. Vehicle warning device.

Images