JP2006343904A - Driving support device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a technique for accurately reporting to a driver a problem of safety in driving. <P>SOLUTION: Safety drive or not is determined based on a difference between distribution of behavior information of the driver and a reference distribution. The behavior information of the driver is stored as history for each traveling state. When behavior information in the history in a similar state tends to perform no safety drive, the driver's attention is drawn. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a vehicle safe driving support method for determining whether a driver is performing safe driving and notifying the driver to perform safe driving when it is determined that the driver is not performing safe driving.

It is related to safe driving support that determines whether the driver is driving safely based on driving operations or related information, and notifies the driver and controls driving when it is determined that driving is not safe. There is technology.
As a specific example, a steering rotation angle distribution is obtained at a predetermined time interval, and a case where a parameter representing the distribution exceeds a predetermined standard is determined as a state of reduced consciousness such as falling asleep (for example, a patent) Reference 1).
When it is determined that the inter-vehicle distance is shortened, it is checked whether or not an accelerator pedal, a brake pedal, or the like is operated. If not, a warning is issued by an alarm sound or an alarm lamp (for example, Patent Document 2) ).
Obtaining a driver's gaze direction distribution at a predetermined time interval, and determining that there is a safety problem when the difference between the distribution and the reference distribution is large (for example, Patent Document 3) ).

JP 6-107029 A

JP-A-10-188200 JP 2003-327003 A

The above-described conventional technology is a technology for inputting information related to driving operation and traveling conditions and determining whether the driver is driving safely based on the input information.
In Patent Literature 1, safety is determined from information related to driving operation acquired at predetermined time intervals, and there is a possibility that determination as to whether there is a problem in safe driving is immediately before a point where attention is to be paid. There is a problem that there is. In order to prevent traffic accidents in advance, it is necessary to predict in advance that there is a safety problem and to prompt the driver with sufficient attention in advance.
Further, Patent Document 2 is intended only for preventing a rear-end collision with a preceding vehicle, but it is not sufficient to determine whether or not a driving operation has been performed in order to determine driving safety in various situations. It is necessary to make a determination including whether or not the driving operation is performed. For example, it is necessary to perform sufficient deceleration operations before intersections and curves, and to avoid sudden steering operations. In this case, as the vehicle approaches the intersection, the brake operation time becomes longer, and it is necessary to determine the driving safety depending on whether the steering wheel is changing at a change rate within a certain range.

  On the other hand, in Patent Document 3, safety is determined based on the driver's gaze direction distribution. In this respect, it is considered possible to give attention to the driver before falling into danger. However, it only determines whether or not the driver is paying attention to the surroundings, and does not determine what kind of driving is being performed. Paying attention to the surroundings is a driver's behavior that is important for safety, but it is not possible to determine whether or not sufficient deceleration has been performed immediately before in the situation such as the above intersection. For example, the speed of the car can be used, but the optimal deceleration varies depending on the individual, so there is a possibility that it can interfere with comfortable driving with one report that can ensure safety and improve it.

In order to achieve the above object, in the present invention, information related to driving operations for parts such as a steering, an accelerator, a brake, and a blinker is detected, and the distribution of the detected information is calculated as driver behavior information. In addition, information on roads such as the type and width of roads around the vehicle, the shape, the position of intersections, and the position and type of signals and road signs, roads such as pedestrian crossings and temporary stop lines, etc. The mark etc. drawn on the top is extracted as a running situation. Based on the extracted driving conditions, a reference distribution for determining whether or not the driver is driving safely is selected. Based on the difference between the behavior information and the reference distribution, the driver Determine if you are driving.
In addition, the driver's behavior information is stored as a history for each driving situation, the behavior information in a situation similar to the driving situation in the traveling direction is searched from the history, and the searched behavior information performs safe driving. If it is determined that there is not, or if the frequency of safe driving is low, the driver is notified to perform safe driving.

According to the present invention, by determining whether or not safe driving is performed based on the difference between the distribution of information related to driving operation and the reference distribution, driving that may be dangerous based on criteria suitable for the driver. It is possible to detect the state accurately and notify the driver.
In addition, the driver's behavior information is stored as a history for each driving situation, the behavior information in a situation similar to the driving situation in the traveling direction is searched from the history, and the searched behavior information performs safe driving. By notifying the driver when it is determined that there is no such thing or when it is determined that the safe driving is performed less frequently, it is possible to effectively perform a notification with a margin.

A first embodiment of the present invention will be described with reference to FIGS.
FIG. 1 is a conceptual block diagram of a first embodiment to which a vehicle safe driving support method according to the present invention is applied. The driving operation detection means 101 in FIG. 1 is a means for detecting information related to driving operations on a vehicle part such as a driver's steering, accelerator, brake, blinker, etc. as driver behavior information. Steering operation can be detected by using a well-known rotation angle sensor. Further, the operation of the accelerator, the brake, and the winker can be detected by installing a switch on each operation tool. The behavior information extraction unit 102 obtains how information related to the driving operation detected from the driving operation detection unit 101 is distributed within a predetermined time.

  The vehicle position detection means 103 detects longitude and latitude representing the vehicle position as an absolute position where the vehicle is currently traveling. As this method, a well-known technique, GPS (Global Positioning System), can be used. In addition, the map data 104 includes information on roads such as the type, width, shape, and intersection position of roads around the location where the vehicle is traveling, as well as the presence or absence of signals, the position and type of road signs, pedestrian crossings, Stores information to be careful of when driving safely, such as marks drawn on roads such as temporary stop lines. Information that should be noted when performing safe driving can be easily searched for information around the position where the vehicle is traveling, for example, by storing it together with the latitude and longitude values representing the respective positions. It becomes like this. In addition, in order to distinguish from information such as features that are not particularly important for safe driving, a flag indicating whether or not it is important for safe driving may be added. As the vehicle position detection means 103 and the map data 104, a car navigation system that is well known as an integrated system can be used as it is.

1 detects, based on the vehicle position detected by the vehicle position detection means 103, that the vehicle is located at a location where danger notification is required. Specifically, the map data includes the type, width, and shape of the road that is running, the distance to points of interest such as intersections and curves in the direction of travel, pedestrian crossings, and temporary stop lines, and the types of surrounding road signs. Search from 104. Further, the behavior information extracting means 103 is notified of the timing for extracting the behavior information of the driver. Specifically, the distance to the points to be careful of when driving safely, such as the type, width, and shape of the road on which it is traveling, intersections and curves in the direction of travel, pedestrian crossings, temporary stop lines, etc., were determined in advance. If the value is equal to or less than the threshold value, the behavior information extraction unit 103 is notified that the behavior information should be extracted. In response to the notification, the behavior information extraction means 103 divides the current vehicle position and the points to be careful of when performing safe driving into a plurality of regions, and sums up information on driving operations for each region, Find the distribution. For example, it is assumed that a point to be noted in performing safe driving is an intersection, and the current vehicle position and the intersection position are 201 and 202 in FIG. 2, respectively. In this case, the area between the vehicle and the intersection is divided into areas such as A (203), B (204), C (205), and D (206) in FIG. Find the distribution of information about.
When the information related to the driving operation is steering, the frequency at which the steering angle changes, the cumulative change angle of the steering, the average change angle of the steering, etc. can be used as the information related to the driving operation. For the accelerator, the brake, the winker, etc., the frequency of operation and the total operation time can be used as information related to the driving operation. As a result, in the case of the positional relationship shown in FIG. 2, for the brake operation time, for example, a frequency distribution as shown in FIG. 3 can be obtained. In FIG. 3, the brake operation times 301, 302, 303, and 304 in the respective regions A, B, C, and D have a distribution that represents an operation state that gradually increases as the vehicle approaches the intersection.

  The reference behavior data 106 in FIG. 1 is data serving as a reference for determining whether or not the driver is driving safely, and a driving operation in which the distribution of face and line-of-sight directions is created by the behavior information extraction unit 102. The distribution of information related to driving operations during safe driving is stored in the same format as the distribution of information related to. However, since the distribution of information related to driving operations varies depending on the surrounding situation and the situation in the direction of travel, prepare distribution data of information related to driving operations as a reference for each situation, and each distribution data is in a format with information on the situation attached. It may be stored. Information on the situation includes the type, width, and shape of the road that is being searched for by the driving situation extraction means 105 as a location where the driver is required to perform a safety check or the like, an intersection or curve in the direction of travel, a pedestrian crossing, It is possible to use the distance to a point to be careful such as a temporary stop line, the type of surrounding road signs, and the like. Moreover, since the situation of the driving operation having a safety problem varies depending on the driver, the reference behavior data may be stored for each driver.

  The driving state determination unit 107 determines whether or not the driver is driving safely based on the information extracted by the behavior information extraction unit 102 and the traveling state extraction unit 105 and the data in the reference behavior data 106. FIG. 4 shows the flow of processing in the operating state determination means 107. In step 401 in FIG. 4, distribution data of information related to the driving operation that matches the condition is selected from the reference behavior data 106 on the condition that the information extracted by the traveling state extraction unit 105 is used. As conditions, for example, a combination of road type, width, shape, intersection, curve, pedestrian crossing, temporary stop line, type of surrounding road signs, etc., indicating that the location is required for danger notification may be used. . As a selection method, it is only necessary to select data having the largest amount of matching information among the information set as conditions. Since the number of matching information may be the same, the type of information may be weighted, the number of matching conditions may be obtained by weighting, and the data with the largest number of matching may be selected. The weight for each information may be determined in advance as a value unique to each information. In step 402, the difference between the distribution of information relating to the driving operation extracted by the behavior information extraction means 102 and the reference behavior data selected in step 401 is obtained. Now, it is assumed that the distribution of information related to the driving operation is as shown in FIG. 5, and the reference behavior data is as shown in FIG. In this case, the difference between the two is as indicated by broken arrows 601 to 604 in FIG. In step 402, the sum of the absolute values of the differences indicated by reference numerals 601 to 604 is obtained as the difference between the distribution of information relating to the driving operation and the reference behavior data. Alternatively, the sum of the squares of the differences may be obtained. Furthermore, a weight may be set for each region, and the absolute value of the difference or the weighted sum of the square of the difference may be obtained. Further, when the difference is obtained, even if the trend of the distribution is the same, the frequency number may be greatly different. Therefore, the difference may be obtained after normalizing the distribution. Normalization can be performed, for example, by dividing the value of each frequency by the maximum frequency or average frequency in the distribution. In step 403, it is determined whether or not the difference obtained in step 402 is larger than a predetermined threshold value. If larger, the process proceeds to step 404 to notify the driver that safe driving is not being performed. If the difference is smaller than the threshold value, the process ends. In the above example, it is only determined whether the driver is driving safely based on the difference between the distribution of information related to driving operation and the reference behavior data. It is also possible to notify the driver of why there is not. For example, in the difference between the distribution of information related to the driving operation shown in FIG. 6 and the reference behavior data, the brake operation time is significantly lower in C and D than the reference behavior data, as indicated by 603 and 604. This indicates that sufficient deceleration is not performed before the intersection from the region shown in FIG. Therefore, it is possible to notify the driver that there is a problem with the driving operation when entering the intersection.

In order to make a notification as described above, a threshold is set for the difference for each area, and if the magnitude of the difference in each area is larger than a predetermined value, it is determined that there is a problem with the driving operation in that area. , Just notify the driver. As a method for notifying the driver, the content of notification can be changed based on the difference between the distribution of information related to driving operation and the reference behavior data. For example, when notification is performed by voice or sound, the volume for notification is increased if the difference is large, and the volume is decreased if the difference is small. Further, after the notification is once performed, the comparison of the distribution of information related to the driving operation and the reference behavior data is continued, and when it is determined that the driver's attention is not improved, the notification may be performed again. When the notification is performed again, the notification may be performed with a content different from the content of the notification performed previously.
The output unit 108 in FIG. 1 presents the notification to the driver determined by the driving state determination unit 107 by voice or sound, or displays it by characters or symbols. As the output means, a generally used device such as a speaker or a monitor, or a head-up display for displaying on a windshield can be used.

  As described above, according to the first embodiment, it is determined whether or not safe driving is performed based on the difference between the distribution of information related to driving operation and the reference distribution, thereby obtaining a reference suitable for the driver. Therefore, it is possible to accurately detect a driving state that may be dangerous and notify the driver.

A second embodiment will be described with reference to FIG.
The driving operation history 701 in FIG. 7 shows the distribution of information regarding driving operation, the difference between the distribution of information regarding driving operation and the reference behavior data, and the result of determining whether or not safe driving is performed based on the difference. It is a means for storing each as a history. As shown in FIG. 7, the driving operation history 701 is provided between the driving state determination unit 702 and the output unit 703, and the output unit 703 determines the determination result based on the distribution of the information related to the acquired driving operation and the information in the history. To decide whether to pay attention to the driver. For example, when the ratio determined as not safe driving in the information retrieved from the driving operation history 701 for the determination result in the driving situation similar to the driving situation at that time is higher than a predetermined ratio, There is a high possibility of neglecting danger avoidance. Therefore, the determination can be performed by combining this information and the determination result based on the reference behavior data. Moreover, the magnitude of the difference between the distribution of information related to the driving operation and the reference behavior data can be used instead of the determination result of whether or not the driving is safe. As a result, even if the difference between the distribution and the reference behavior data is within the range where it is determined that the vehicle is safe driving, even if a relatively large difference continues, it is determined that the driver should be careful. Can be notified. Furthermore, instead of the difference between the distribution of all judgment results and information on driving operations in the history and the reference behavior data, only the information immediately before or just before the predetermined number of times may be used. .

  As described above, according to the second embodiment, the determination result of the behavior information of the driver is stored as a history for each driving situation, and the determination result of the behavior information in a situation similar to the driving situation in the traveling direction is stored. If the vehicle is searched for from the history and it is determined that safe driving is not being performed, or if it is determined that safe driving is infrequent, the driver is notified so that there is sufficient margin. This can be done effectively.

A third embodiment will be described with reference to FIGS.
FIG. 8 is a conceptual block diagram of the third embodiment. In FIG. 8, a face / eye detection means 801 is added to the first embodiment shown in FIG.
The face / eye direction detection means 801 detects either or both of the position and direction of the driver's face and / or the direction of the eyes as the behavior information of the driver at predetermined time intervals. . As a method of detecting the position and direction of the face, for example, the technique in “Vehicle Driver Monitoring Device (Japanese Patent Laid-Open No. 11-161798)” is used. As the method of detecting the position of the eyes and the direction of the line of sight, for example, “ This can be realized by using the technology in “Front Gaze Degree Detection Device (Japanese Patent Laid-Open No. 7-96768)”.

In the behavior information extraction unit 802 in FIG. 8, in addition to the distribution of information related to the driving operation, the face direction or the direction of the line of sight detected by the face / eye direction detection unit 801 is within the predetermined time. Find out how it is distributed in the region. A method for realizing this will be described below.
For the sake of explanation, it is assumed that the equipment in the vehicle in the driver's field of view is as shown in FIG. In FIG. 9, 901 is a front window, 902 is a rear view mirror, 903 is a steering, 904 is an instrument such as a speedometer, 905 is a screen or operation panel of a car navigation system or an audio system, and 906 is a right side. 907 denotes a left side rearview mirror. In order to obtain the distribution of the face direction and the line-of-sight direction, first, the driver's visual field area shown in FIG. 9 is divided into appropriate areas. As a division method, for example, as shown in FIG. 10, the visual field region may be divided into rectangular regions equally. Alternatively, the division method can be changed according to the importance of the visual field area and the way of gazing. For example, since it is important whether or not the rearview mirror and the left and right rearview mirrors are being watched, each can be made into one area. In addition, since confirmation of the instruments is also necessary for safe driving, they can be considered as one area. In addition, safety issues arise wherever other indoor equipment is watched, so the area corresponding to the lower indoor equipment other than the instruments may be a single area. In the example of the division method shown in FIG. 10, each region has a shape that does not overlap with other regions, but may be a region that has a portion overlapping with another region. Further, the division may be performed not by a square or a straight line but by an arbitrary shape such as a circle or a curve. The behavior information extraction unit 802 obtains the distribution of the face and line-of-sight directions based on the divided visual field regions. Since the equipment in the vehicle is unique for each vehicle, the position of the divided area is also unique for each vehicle. Therefore, it is possible to determine which region the direction of the face or line of sight is facing from the relationship between the position of the face or eyes and the position of each region. Based on the determination result of the area in which the direction of the face or line of sight is directed, the frequency with which the face or line of sight is directed within a predetermined time is obtained for each area. Thereby, for example, a frequency distribution as shown in FIG. 11 can be obtained for each region in the row indicated by “D” in FIG. FIG. 11 shows a state in which driving is performed while paying attention to the front (1101, 1102 and 1103) and attention to the left and right (1104, 1105, 1106 and 1107).

  The method described in the first embodiment can be used as it is as a method for determining whether or not safe driving is performed from the behavior information detected by the face / eye detection means 801 and the driving operation detection means 803. In addition, the method described in the first embodiment can be used as the method for notifying the driver when it is determined that safe driving is not performed. Furthermore, both the difference between the distribution of the behavior information generated from the face / eye detection unit 801 and the reference behavior data and the difference between the distribution of the behavior information generated from the driving operation detection unit 803 and the reference behavior data are determined in advance. If the threshold value is larger than the threshold value, it may be determined that the driver is not driving safely and a notification for alerting the driver may be issued. Further, as a method for selecting the reference behavior data, it is also possible to select the information related to the driving operation in addition to the traveling state.

  As described above, according to the third embodiment, in addition to the distribution of information related to driving operations, only information related to driving operations is determined by determining whether safe driving is performed based on the distribution of face and eye movements. The driver can be notified with higher accuracy than when using the.

The conceptual block diagram which shows the 1st Example of this invention. An example of area division when obtaining a distribution of information related to driving operations. An example of distribution of information about driving operation. The flowchart which shows the process which determines whether a driver | operator is performing safe driving | operation. The 2nd example of distribution of information about driving operation. The figure which shows the difference of the distribution shown in FIG. 3, and the distribution shown in FIG. The conceptual block diagram which shows the 2nd Example of this invention. The conceptual block diagram which shows the 3rd Example of this invention. The figure which shows an example of the equipment in a vehicle. The example which divided | segmented the driver | operator's visual field area | region in a vehicle into equal intervals. An example of the distribution of face and gaze direction.

Explanation of symbols

101, 803 Driving operation detection means 102, 802 Behavior information extraction means 103 Vehicle position detection means 104 Map data 105 Travel condition extraction means 106 Reference behavior data 107, 702 Driving state determination means 108, 703 Output means 701 History database 801 Face / eye Detection means.

Claims (8)

Driving operation detecting means for detecting information related to driving operation in the vehicle;
Vehicle position detecting means for detecting the position of the vehicle;
A storage unit for storing map data;
Behavior information extracting means for extracting, as driver behavior information, a distribution of information relating to the driving operation in a plurality of consecutive sections on the map based on the vehicle position and the map data;
Reference behavior data for storing reference data for determining whether the driver is driving safely;
Driving state determination means for determining whether safe driving is performed from the reference behavior data and the driver behavior information;
A driving support device comprising: an output means for notifying the driver when the driving operation detecting means determines that the driver is not driving safely. Based on the vehicle position and the map data, a traveling condition extracting means for detecting that the vehicle is located at a location where danger notification is required in the map data;
2. The driving support apparatus according to claim 1, wherein the behavior information extracting unit extracts behavior information when a danger notification location is detected by the traveling state extracting unit.   The driving support apparatus according to claim 1 or 2, wherein the behavior information extracting means acquires a distribution of information related to driving operation over a predetermined section.   The driving state determination means compares the behavior information with reference data, and determines that the driver is not performing safe driving when the difference between them is larger than a predetermined value. The driving assistance apparatus in any one of thru | or 3.   5. The driving support apparatus according to claim 4, wherein the output means changes a method of notifying the driver that safe driving is not being performed based on the magnitude of the difference.   6. The reference data according to claim 2, wherein the reference data is stored for each kind of location where the danger notification is requested, and the reference data is selected based on information of the traveling state extracting means. The driving support device according to 1. It has a history database that accumulates past behavior information,
7. The driving support according to claim 2, wherein whether or not the safe driving is performed is determined using history data associated with the driving condition detected by the driving condition means. apparatus. The driving operation detection means also detects the position and direction of the driver's face and eyes,
The behavior information extracting means extracts the driver's face and eye position and direction distribution as the driver's behavior information, and uses the information on the face and eye position and direction to perform the driving safely. The safe driving support apparatus according to any one of claims 2 to 7, wherein it is determined whether or not there is any.

Images