JP2009086788A - Vehicle surrounding monitoring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem of conventional techniques having difficulty in alerting a driver of an obstacle that suddenly appears from a blind spot where a monitoring area is blocked by installations on a road, etc. <P>SOLUTION: This vehicle surrounding monitoring device includes: a warning unit; an obstacle detecting unit for detecting any obstacle around one's own vehicle; a one's own vehicle movement detecting unit for detecting the movement of one's own vehicle; a one's own vehicle course trajectory estimating unit for estimating the course trajectory of one's own vehicle based on the direction and speed of travel of one's own vehicle; a blind spot estimating unit for estimating a blind spot created by an obstacle existing around one's own vehicle; a course obstacle appearance determining unit for estimating a position where one's own vehicle will collide with a course obstacle on the assumption that a course obstacle blocking the course trajectory of the one's own vehicle will appear form the blind spot; and a warning determining unit for computing the danger level of the blind spot based on the distance between the colliding position and the one's own vehicle and the speed of travel of the vehicle, and determining whether or not the warning unit should be operated according to the danger level. The warning unit provides a warning to the driver based on the determination by the warning determining unit. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a vehicle periphery monitoring device that prevents a collision accident in advance.

  Conventionally, various techniques for avoiding a collision between vehicles have been developed.

  Here, by irradiating a driver's blind spot area generated behind the vehicle with a plurality of laser beams, measuring the relative distance with the vehicle traveling in the blind spot area, and when the relative distance is below a certain reference value, There is a vehicle periphery monitoring device that issues an alarm (see Patent Document 1). Further, there is a rear-end collision prevention device that detects obstacles around the vehicle and avoids vehicles approaching from behind by providing a plurality of radars on the vehicle (see Patent Document 2).

JP-A-8-24149 JP 2005-182198 A

  According to the above prior art, the driver can be alerted to obstacles detected by a monitoring device such as a radar mounted on the host vehicle, but monitoring is performed by the vehicle, the wall surface, the installation on the road, etc. For obstacles that suddenly appear from the blind spot area where the area is blocked, it is difficult to give a warning that takes into account the time for the driver to avoid or brake. For example, at a crossroad intersection with poor visibility that is blocked by a wall, the surveillance area is blocked by the wall, so the traffic situation of the intersection road cannot be detected until it approaches the vicinity of the center of the intersection, causing an encounter accident Sometimes. Moreover, a pedestrian may jump out of a blind spot area caused by a stopped vehicle.

  In order to prevent such accidents caused by blind spots, monitoring devices such as cameras and radars are installed on the road, and wireless technologies such as optical beacons and DSRC (Dedicated Short Range Communications) are used to detect obstacles sequentially. Road-to-vehicle communication technology for informing the vehicle of existence and vehicle-to-vehicle communication technology for sharing obstacle detection information between vehicles have been developed. However, there is a problem that it takes a lot of cost and time for the monitoring apparatus to be sufficiently provided on roads throughout the country and to be mounted on the vehicle side.

  Accordingly, an object of the present invention is to provide a vehicle periphery monitoring device that notifies a driver of the danger of a blind spot area caused by an obstacle present around the host vehicle.

  One desirable embodiment of the present invention is as follows.

  The vehicle periphery monitoring device detects an alarm unit that gives a warning to the driver, obstacles around the vehicle as a point sequence of a plurality of detection points, and the relative distance, azimuth angle between the vehicle and each detection point, and Based on the obstacle detection unit for detecting the relative speed, the own vehicle motion detection unit for detecting the traveling direction and the traveling speed of the own vehicle, and the traveling direction and the traveling speed of the own vehicle detected by the own vehicle motion detection unit. From the own vehicle course trajectory estimation unit for estimating the own vehicle course trajectory, the blind spot region estimation unit for estimating the blind spot region caused by the obstacle existing around the own vehicle, and the blind spot region estimated by the blind spot region estimation unit, A path obstruction appearance determination section that estimates a position where the own vehicle and the path obstruction collide, and a path obstruction appearance determination section assume that a path obstruction that obstructs the vehicle path locus estimated by the path locus estimation section appears. The distance between the estimated collision position and the vehicle and the vehicle motion detector A warning determination unit that calculates the degree of danger in the blind spot area based on the traveling speed of the own vehicle and determines the operation of the alarm unit according to the degree of danger, and the alarm unit operates based on the determination of the alarm determination unit Alert the person.

  ADVANTAGE OF THE INVENTION According to this invention, the vehicle periphery monitoring apparatus which alert | reports to a driver the danger of the blind spot area | region which arises with the obstruction which exists around the own vehicle can be provided.

  Hereinafter, an embodiment of a vehicle periphery monitoring device will be described with reference to FIGS.

  FIG. 1 is a diagram showing a vehicle periphery monitoring device 300.

Obstacle detection unit 100a, 100b, 100c, 100d for detecting obstacles around the own vehicle 20, own vehicle motion detection for detecting the position and orientation of the own vehicle (including travel direction) and motion (including travel speed) The GPS 110, the acceleration / yaw rate sensor 120, the direction indicator 130, the wheel speed sensors 140a, 140b, 140c, and 140d, the steering angle sensor 150, and the gaze detection sensor 160 as a gaze detection unit that detects the driver's gaze are mounted. ing. Further, a map information storage device 200, a road traffic information reception device 210, a traffic accident information storage device 240, a calculation device 220 that performs calculation processing necessary for vehicle periphery monitoring, and an alarm device 230 are mounted. With this configuration, the following effects are obtained.
● “Blind area danger judgment”: The risk of the dead area caused by obstacles around the vehicle is autonomously judged from various sensors installed in the vehicle and an alarm is issued.
● “Danger judgment according to the driving scene”: Calculates the degree of danger in the blind spot area according to the driving scene, such as the driving state of the vehicle, past and current traffic conditions, and road surface conditions.
● “Affinity with the driver”: By detecting the driver's line of sight, an alarm action is performed in consideration of the driver's attention to the blind spot area, and further, the location of the blind spot is notified by an alarm.

  Next, a description will be given of the own vehicle motion detection unit that detects the position and orientation and motion of the own vehicle.

  A GPS (Global Positioning System) 110 is a sensor that detects the current position of the vehicle on the earth, and is generally used in vehicles as one of navigation systems.

  The acceleration / yaw rate sensor 120 is a sensor that detects longitudinal acceleration / lateral acceleration and yaw rate in the traveling direction and the vehicle width direction of the host vehicle.

  The direction indicator 130 is a device that outputs a signal corresponding to the course change direction of the host vehicle instructed by the driver.

  The wheel speed sensors 140a, 140b, 140c, and 140d are sensors that detect the rotational speeds of front, rear, left and right wheels that are generally installed in four-wheeled vehicles such as ordinary passenger cars.

  The steering angle sensor 150 is a sensor that detects the steering angle of the steering.

  These pieces of information are periodically transmitted to the arithmetic unit 220 via signal lines.

  Next, an obstacle detection unit that detects obstacles around the host vehicle will be described.

  A plurality of obstacle detection units 100a, 100b, 100c, and 100d are provided on the front, rear, and sides, and relative distance, relative speed, and azimuth information of a plurality of detection points obtained by scanning the detection area. Is output. Typical obstacle detection units include a millimeter wave radar 103, a laser range finder 102, a camera 101, and the like.

  The millimeter wave radar 103 scans an electromagnetic wave having a frequency of 24 GHz band or 30 to 300 GHz band in a one-dimensional or two-dimensional direction within a predetermined region, and measures the round-trip time of the electromagnetic wave reflected from the obstacle and the phase difference of the electromagnetic wave. The measurement data of the two-dimensional and three-dimensional relative distances, azimuths, and relative velocities at a plurality of detection points can be detected. Since the reflectance of electromagnetic waves varies depending on the material of the obstacle, the material of the obstacle can be specified by detecting the reflection intensity of the reflected wave.

  The laser range finder 102 is a radar that uses laser light with high directivity, and, like the millimeter wave radar 103, scans in a one-dimensional or two-dimensional direction in a predetermined area and measures the round trip time of light reflected from an obstacle. Thus, the two-dimensional and three-dimensional relative distance / azimuth information of a plurality of detection points can be detected. In general, the laser range finder 102 cannot directly detect the relative speed, so the relative speed is obtained by performing a difference between the data scanned one step before and the current scanned data. Similarly to the millimeter wave radar 103, the laser range finder 102 also has a different reflectance depending on the material of the obstacle. Therefore, the material of the obstacle can be specified by detecting the reflection intensity.

  The detection method differs depending on how many cameras 101 are used for obstacle detection by the camera 101. In binocular stereoscopic vision using two typical cameras 101, it is possible to detect three-dimensional relative distance and azimuth information in an imaged area based on triangulation using parallax. Since the camera 101 can acquire the luminance information of the obstacle, a template representing features such as the shape and color of the target to be recognized is prepared, and the obstacle is recognized by performing a matching process between the template and the captured image. It can be carried out. Even when the camera 101 is used, the relative speed cannot be directly detected in the same manner as the laser range finder 102, so the relative speed is obtained by performing a difference between the data scanned one step before and the currently scanned data. I can do things. The detection results of these obstacle detection units are transmitted to the arithmetic unit 220 via signal lines.

  In this embodiment, for convenience of explanation, a case where an obstacle detection unit capable of acquiring two-dimensional detection data is used will be described. Even when a three-dimensional obstacle detection unit is used, generality is not lost because the same processing procedure as in the two-dimensional case is followed. Further, the sensors in the present embodiment do not limit the content of this paper, and other sensors and measurement methods may be used, and the installation location of the vehicle is not limited.

  The line-of-sight detection sensor 160 mounted as a line-of-sight detection unit that detects the driver's line of sight is used to determine which direction the driver is paying attention by detecting the line of sight. The configuration and method of the sensor for detecting the line of sight are not particularly limited, but a method of detecting the line of sight based on the image data by photographing the eyes of the driver with a camera fixed in the vehicle is often used.

  The map information storage device 200 is commonly used in navigation systems, and various geographical information such as altitude, landform, address, building shape, road shape, road classification, etc., relating to part or all of the earth's surface. Is stored, and necessary geographic information is read in response to an information request from the arithmetic device 220. Even if all the above-mentioned stored information is mounted on the vehicle, only a part of the information is mounted on the vehicle, and the storage information stored in the base station is sequentially transmitted to the vehicle and updated by using wireless technology. It does not matter, and new geographical information may be added.

  The road traffic information receiving device 210 is a device that receives road traffic information ahead of the traveling road such as traffic jams, accident / traffic regulation information, and weather information. For example, the “Road Traffic Information Communication System Center” provides a traffic jam information providing service. By obtaining the traffic jam, accident / traffic regulation / weather information, it becomes possible to take measures to avoid danger in advance. From traffic jams / accidents / traffic regulation information, there is a high possibility that there is an accident vehicle on the road ahead, or there is a vehicle running at a low speed or when there is a traffic jam. An obstacle is likely to exist. On the other hand, when no traffic jam is predicted, the possibility that an obstacle exists in front of the traveling road is low. From the weather information, it is possible to infer road surface conditions such as dryness, wetness and freezing of the road surface. If the road surface state can be estimated, the coefficient of friction between the tire road surfaces can be estimated, so that it is possible to estimate the distance required for braking according to the weather.

  The traffic accident information storage device 240 is a storage device that stores statistics on traffic accidents that have occurred on the road in the past, and stores statistics on the number of accidents that occur according to the season, time, place, and road shape. And From this information, the season and time information at which accidents frequently occur at a certain point and the contents of accidents such as whether there are many accidents between vehicles or between vehicles and people are known.

  FIG. 2 is a block diagram illustrating a processing flow of the vehicle periphery monitoring apparatus 300.

  The obstacle detection unit 1 is a block that collectively represents 100a, 100b, 100c, and 100d in FIG. The own vehicle motion detection unit 2 is a block that collectively represents the GPS 110, the acceleration / yaw rate sensor 120, the direction indicator 130, the wheel speed sensors 140a, 140b, 140c, 140d, and the steering angle sensor 150 of FIG. The line-of-sight detection unit 3 is a line-of-sight detection sensor 160.

  The arithmetic device 220 is a device that performs arithmetic processing for monitoring the vehicle periphery, and includes an obstacle recognition unit 4, a map matching unit 8, a blind spot region estimation unit 6, a traveling direction estimation unit 7, an own vehicle position estimation unit 5, The obstacle appearance determination unit 9 and the alarm determination unit 10 perform arithmetic processing. The arithmetic device 220 processes information transmitted from the obstacle detection unit 1, the own vehicle motion detection unit 2, the map information storage device 200, the traffic accident information storage device 240, and the road traffic information reception device 210, and outputs the processing result. In response, the alarm device 230 is operated.

  The own vehicle position estimation unit 5 is an element that estimates the absolute position and posture of the own vehicle based on the output of the own vehicle motion detection unit 2 and outputs these estimation results. The detection of the obstacle detection unit 1 by this element It is possible to determine at which position around the own vehicle the blind spot region generated in the region is generated.

  In order to estimate the absolute position and orientation of the vehicle, not only the GPS 110 but also the vehicle motion based on the vehicle dynamics based on a plurality of information obtained from the acceleration yaw rate sensor 120, the wheel speed sensors 140a, 140b, 140c, 140d, and the steering angle sensor 150 Estimate the position and orientation using the model. Since position / orientation estimation using a vehicle motion model performs a time integration operation, there is a possibility that measurement errors of various sensors accumulate and the estimation error becomes enlarged. In contrast, it is possible to reduce the estimation error by adding position information obtained from the GPS 110. Integration such as detecting the position and orientation of the vehicle relative to the lane by detecting the lane with a camera or radar, and detecting the position and orientation of the vehicle based on information obtained from sensors or beacons provided on the road It is possible to improve the estimation accuracy by combining with a method that can detect the position and orientation relationship between the traveling road and the own vehicle directly without any operation.

  The absolute coordinate system of the vehicle position and orientation may be a latitude / longitude coordinate system defined based on the Japanese geodetic system or a latitude / longitude coordinate system defined based on the international earth reference coordinate system. Alternatively, a planar rectangular coordinate system may be used. Since the reference point differs depending on the reference coordinate system, each reference point may be used. The position / orientation estimation method does not limit the contents of this paper, and other position / orientation estimation methods may be used, or other coordinate systems may be used.

  Based on the output of the own vehicle motion detection unit 2, the own vehicle course estimation unit 7 estimates the absolute position of the course trajectory that the host vehicle is expected to travel in the future from the absolute position estimated by the own vehicle position estimation unit 5, This is an element that outputs the absolute position of the estimated track. Specifically, the vehicle will pass in the future based on the vehicle speed calculated from the wheel speed and wheel radius output from the wheel speed sensors 140a, 140b, 140c and 140d, and the yaw rate output from the acceleration / yaw rate sensor 120. A course locus is estimated by calculating a plurality of points.

  The course trajectory may be estimated using the longitudinal acceleration and lateral acceleration output from the acceleration / yaw rate sensor 120, or the steering angle output from the steering angle sensor 150 may be used. Alternatively, the path trajectory may be estimated using a vehicle motion model that expresses vehicle dynamics used in the vehicle position estimation unit 5. The course of the host vehicle may be determined based on the direction indicator 130 that is one of the host vehicle motion detection units 2. However, since the course locus of the host vehicle cannot be estimated only by the direction indicator 130, it is preferable to use it in combination with the method for estimating the course locus. The course trajectory estimation method does not limit the contents of this paper, and other estimation methods may be used.

  The obstacle recognition unit 4 performs a grouping process of detection point sequences that are determined to detect the same obstacle from the plurality of detection point sequences output by the obstacle detection unit 1. This element outputs data with a group number added to the output data. Specifically, it is recognized as the same obstacle by sequentially grouping point sequences of detection points whose distance between adjacent detection points is equal to or less than a threshold value in the scan direction. If the distance between the detection points exceeds a predetermined value, it is determined that the adjacent detection points are different obstacles, a group number is added to the grouped detection point data, and the next detection point is grouped. Move on. The process is performed on the data of all detection points, and a plurality of detection points are grouped. The grouping method uses only spatial information such as distances, but points detected from the same obstacle are point sequences of detection points that are less than or equal to a predetermined value, including speed differences and reflection intensity differences between detection points. It may be possible to use a grouping method for determining as follows. Further, as a result of the grouping process, when the number of detection points belonging to the group is less than a predetermined number, it is determined that the detection points are noise, and data may be removed from the detection point sequence.

  In this paper, the grouping processing method is not limited, and means other than the above method may be used. The obstacle recognition unit 4 does not need to be processed by the arithmetic device 220, and may be performed by an arithmetic device incorporated in the obstacle detection unit 1. However, when the processing is performed by the arithmetic device built in the obstacle detection unit 1, the detection data of the detection point including the identification result is transmitted to the arithmetic device 220.

  FIG. 3 is a diagram illustrating a processing example of the obstacle recognition unit 4. Here, an example is shown in which the detection range is scanned by the laser range finder, which is the obstacle detection unit 1 mounted on the front portion of the host vehicle 20, and obstacles around the host vehicle are detected.

  The own vehicle 20 is traveling on a road having a wall surface 22 on the right side, and there is another vehicle 21 parked in front. Under this situation, an obstacle around the own vehicle is detected by the laser range finder. Assume that data on the relative positions, azimuths, and relative speeds between the plurality of detection points illustrated by the white circles 30 corresponding to the other vehicle 21 and the wall surface 22 and the host vehicle 20 are obtained. The obstacle recognition unit 4 receives the data of the plurality of detection points and performs the grouping process to form groups 31 and 32 surrounded by dotted lines, and each of the detection point sequence data in the groups 31 and 32 has an identifier. A1 and A2 are added.

  The blind spot area estimation unit 6 is based on the information on a plurality of detection points to which the group number output by the obstacle recognition unit 4 is added and the absolute position and posture information of the own vehicle estimated by the own vehicle position estimation unit 5. This is an element that estimates the absolute location of the blind spot area of the obstacle detection unit 1 generated in the detection area due to the presence of the obstacle and outputs the absolute location information of the estimated blind spot area. The blind spot area of the obstacle detection unit 1 caused by the obstacle is an area where the transmission wave and the reflected light of the sunlight are blocked and data such as relative position and relative velocity cannot be measured. If the measurement point, the boundary point between the measurement point on the obstacle and the blind spot area (hereinafter, blind spot generation point), and the installation position of the obstacle detection unit 1 can be known, the blind spot area can be estimated. In order to determine the blind spot generation point from the measurement points obtained by the obstacle detection unit 1, the distance between adjacent detection points is a predetermined value or more at a plurality of detection points obtained from the obstacle detection unit 1. Find a point. Since the process of searching for the blind spot occurrence point has the same processing contents as the grouping process performed by the obstacle recognition unit 4, the detection points at both ends of the detection point sequence to which the group number has already been added by the obstacle recognition unit 4 Are made to correspond to two blind spot generation points. Thereby, a blind spot area can be estimated.

  When there are a plurality of adjacent blind spot areas, they may be combined into one blind spot area. Specifically, in a triangle whose apex is the adjacent blind spot generation point in the adjacent blind spot area and the installation point of the obstacle detection means 4, if the angle of the obstacle detection means 4 is less than a predetermined value, The matching blind spot areas may be regarded as the same blind spot area.

  FIG. 4 is a diagram illustrating a process performed by the blind spot area estimation unit 6. Here, the thick lines 33 and 34 represent the boundaries of the blind spot areas 40 and 41, respectively, and the black circles 60, 61, 62, and 63 represent the blind spot generation points.

  FIG. 4 shows a processing result when the blind spot area estimation unit 6 receives the data processed by the obstacle recognition unit 4 in FIG. 3, and shows two results positioned at both ends of the detection point sequence in the order of group numbers. A blind spot area that is detected by hatching from two straight lines and detection point sequences that pass through the origin of the coordinate system of the obstacle detection unit 1 and the blind spot generation points 60, 61 by detecting the blind spot occurrence points 60, 61, 62, 63. 40 is similarly estimated for the blind spot generation points 62 and 63.

  The map matching unit 8 acquires geographic information of a predetermined position from the geographic information storage device 200 based on the absolute position information. From the absolute position of the own vehicle obtained from the own vehicle position estimation unit 5, the absolute location of the blind spot region obtained from the blind spot region estimation unit 6, and the absolute position of the course locus obtained from the own vehicle route estimation unit 7, Geographic information about the periphery, the blind spot region, and the course locus periphery is extracted from the map information storage device 200. Based on the types and shapes of buildings and roads obtained from geographic information, it is possible to determine what kind of roads exist and whether there are buildings. The signals output from the map matching unit 8 are output values of the vehicle position estimation unit 5, the vehicle route estimation unit 7, and the blind spot area estimation unit 40 including geographic information.

  The path obstruction appearance determination unit 9 is a boundary position of a blind spot area including the detection point data of obstacles around the vehicle that has been grouped by the obstacle recognition unit 4 and the geographical information output by the map matching unit 8. On the basis of the trajectory information of the vehicle and its own vehicle, and the appearance of the obstruction obstructing the vehicle from the blind spot area is determined, and if it is determined to appear, the collision position with the obstruction is predicted and predicted. This element outputs the position. The type and size of the route obstruction can be set arbitrarily.

The path obstruction appearance determination unit 9 performs a path obstruction appearance determination process on a boundary of a blind spot area (hereinafter referred to as a search boundary) that satisfies the following conditions.
(Condition 1-1) A blind spot area existing within a predetermined range from the own vehicle (Condition 1-2) A boundary close to the own vehicle traveling locus excluding a boundary on an obstacle detection point

  The purpose of limiting the search boundary of (Condition 1-1) within a predetermined distance is to reduce the processing amount of the obstacle appearance determination unit 9 by processing only a short-range blind spot region that is highly dangerous for the vehicle. .

  The purpose of limiting to the boundary of the blind spot area close to the traveling locus of (Condition 1-2) is to correspond to the obstacle detection unit 1 detecting the path obstruction appearing from the blind spot area for the first time when passing through the boundary. It is.

When the following two conditions are satisfied at the search boundary selected as described above, it is determined that the path obstruction appears from the blind spot area. It is assumed that the route obstruction basically enters perpendicularly to the own vehicle route trajectory unless the route is restricted by the road shape.
(Condition 2-1) Geographical conditions where a path obstruction exists in the blind spot area near the search boundary are aligned (Condition 2-2) No obstacles enter the search boundary or the path of the path obstruction.

  The geographical conditions described in (Condition 2-1) are (Condition A) whether there is an area where a path obstruction can exist in the blind spot area, or (Condition B) where the blind spot area may contain a path obstruction. This is a condition set on the basis of geographical information obtained from map information such as whether or not it is land and whether or not there is a three-dimensional object that obstructs the course in the blind spot area. The determination of (Condition A) is, for example, when the vehicle is classified as a candidate for a path obstruction and a predetermined size is determined, and when it is determined that the area where the vehicle can exist is not in the blind spot area, The vehicle will be removed from the candidate obstacles. In the determination of (Condition B), for example, when the blind spot area is on the water surface or in a building, there is no course obstruction. As for (Condition C), for example, when there is guardrail or curb setting information, it is also possible to enter the vehicle course trajectory even if it is judged that there is a course obstruction in the judgment of (Condition A) and (Condition B). It is determined that there is no vehicle that can be used. However, since the determination of (Condition B) and (Condition C) has a wide variety of geographic information, the presence / absence pattern of the presence of a path obstruction based on the geographic information is stored in advance in a storage device as a database. desirable.

  The obstacle that intrudes into the search boundary described in (Condition 2-2) or obstructs the path of the path obstruction is an obstacle detected by the obstacle detection unit 1, and is the path of the search boundary or path obstruction. If there is an obstacle, the collision between the obstacle obstructed to be present in (Condition 2-1) and the obstacle is expected. It is determined that the obstruction does not appear on the own vehicle track.

  The above determination processing of (Condition 2-1) and (Condition 2-2) is performed in order from a location close to the vehicle position at the search boundary, and from the position satisfying (Condition 2-1) (Condition 2-2). It is determined that a path obstruction appears, and the determination process is terminated. When there are a plurality of search boundaries, the process is moved to the next search boundary and the same process is repeated to perform the route obstacle appearance determination process for all the search boundaries.

  FIG. 5 is a diagram illustrating a determination example of the route obstruction appearance determination unit 9. In particular, FIG. 4 is a diagram illustrating a result of the path obstruction appearance determination unit 9 receiving data processed by the blind spot estimation unit 6 of FIG. 3 and determining that the path obstruction 23 appears from the blind spot area 41.

  In FIG. 5, it is assumed that there are two blind spot areas 40 and 41 caused by the obstacles of the group numbers A1 and A2, and the host vehicle 20 proceeds in the direction of the arrow of the host vehicle track 50. When a search boundary that is located within a predetermined range from the vehicle 20 and is close to the vehicle track 50 is selected based on (Condition 1-1) (Condition 1-2) at the boundary of the blind spot area indicated by the bold line 33, The boundaries between the two blind spots displayed with lines are selected as the search boundaries 42 and 43.

  Next, on the two search boundaries 42 and 43, the appearance of an obstacle is determined based on (Condition 2-1) (Condition 2-2). Since the search boundary 42 belonging to the blind spot area 40 generated by the wall surface 22 of the group number A1 exists on the wall surface, it is determined that there is no course obstruction because the condition (2-1) is not satisfied. In addition, the search boundary 43 belonging to the blind spot area 41 generated by the other vehicle 21 that stops at the group number A2 has an area where a path obstruction may exist, is on a road, and a three-dimensional object that obstructs the path is present. Since it does not exist, (Condition 2-1) is satisfied, and since there is no obstacle in the vicinity that obstructs the path of the path obstruction, (Condition 2-2) is also satisfied. Therefore, it can be determined that the route obstruction 23 is present at the location shown in the figure.

  The predicted collision position 44 between the path obstruction 23 determined to appear and the path locus 50 of the host vehicle is perpendicular to the path locus 50 because there is no road shape that restricts the road around the path obstruction 23. Based on the assumption that the vehicle is moving, the position is indicated by x in the figure. When the other vehicle 21 of the group number A2 is traveling in the forward direction, (Condition 2-1) is satisfied. However, since the other vehicle 21 travels and enters the search boundary 43, (Condition 2) -2) is not satisfied, and it is determined that no obstacles appear.

  The warning determination unit 10 operates the warning device 230 based on the predicted collision position output by the path obstruction appearance determination unit 9, the output value of the own vehicle motion detection unit 2, the output value of the map matching unit 8, and the line-of-sight detection unit 3. Make a decision. The motion determination is obtained from the deceleration D required to stop at the predicted collision position and the past traffic accident information around the predicted collision position obtained from the traffic accident information storage device 240 and the road traffic information receiving device 210. The alarm determination parameter A composed of two variables of the accident occurrence parameter P calculated based on the real-time traffic information and the line-of-sight information obtained from the line-of-sight detection unit are used. The deceleration D is calculated from the following equation.

Here, V [m / s] is the speed of the own vehicle, μ is a coefficient of friction between the tire and the road surface, L [m] is a distance between the current position of the own vehicle and the predicted collision position, and g [m / s ² ] is Represents gravitational acceleration. The unit is [G]. From (Equation 1), the deceleration D is a variable that increases in value when the vehicle speed V is large, the distance L is small, and the friction coefficient μ is small. That is, as the deceleration increases, it becomes difficult to stop at the predicted collision position. In general, it is said that the deceleration that a person feels uncomfortable is 0.4 to 0.5 [G], and a value greater than that corresponds to a sudden braking operation. In view of the above, the deceleration D is used as one index for making an alarm determination.

  The speed V and the distance L can be calculated using information obtained from the own vehicle motion detection unit 2 and the map matching unit 9. The friction coefficient μ is a coefficient that changes depending on the degree of wear of the tire tread, the road surface condition, and the degree of wetness of the road surface, and it is difficult to calculate the value with high accuracy. However, if the information on whether the road surface is asphalt or gravel road and the weather information are known, the value of the friction coefficient μ can be estimated roughly. For example, μ = 0.7 for dry asphalt or concrete, μ = 0.5 for wet concrete, μ = 0.45-0.6 for wet asphalt, and μ = 0.55 for gravel roads. Is commonly used. Therefore, the value of the friction coefficient μ can be estimated by determining the type of road surface from the geographical information around the predicted collision position and determining the wetness of the road surface from the weather information obtained from the road traffic information receiving device 210. From the above, the deceleration D can be calculated by (Equation 1) based on each variable whose value has been determined.

  The accident occurrence parameter P is a parameter determined according to the past and current traffic conditions around the predicted collision position. The number of past accidents around the predicted collision position is small, and accidents and traffic jams occur in the current traffic condition. If not, set the value to a small value. If the number of past accidents is large and the current traffic situation has an accident or traffic jam, the value is set to a large value. The accident occurrence parameter is a value normalized from 0 to 1. The accident occurrence parameter P is 10 according to the situation and number of accidents and the degree of road congestion with respect to the past accident occurrence number and current road traffic information obtained from the traffic accident information storage device 240 and the road traffic information receiving device 210. The degree of danger on the road in the past and the present is numerically expressed as P1 and P2 in a stage evaluation, etc., and normalized to a number from 0 to 1, and can be obtained from the following formula.

  Here, α and β are positive real numbers, and are weight parameters for determining at what ratio the past and present accident information are related to the accident occurrence parameter P that is a criterion for alarm action determination. For example, when it is determined that the current information is more important for alarm judgment than the past information, the accident occurrence parameter P in which the current information becomes dominant is calculated by setting the value of β larger than α. can do. From the two indexes of the calculated deceleration D and accident occurrence parameter P, an alarm determination parameter A for performing an alarm operation determination is calculated as follows.

  The operation of the alarm device 230 can be determined according to the value of the alarm determination parameter calculated by (Equation 3). There are various operation patterns of the alarm device 230, and in this paper, the operation pattern of the alarm device 230 is not limited. For example, only when the alarm determination parameter A is equal to or greater than a predetermined threshold, the blind spot area determined to be dangerous is clearly indicated on the display. It is desirable to give a warning suitable for the driver's sensibility using a lamp, display, speaker, or buzzer, such as changing the buzzer sound according to the value of the warning judgment parameter A.

  Further, as a method of issuing a warning and canceling the warning, it is preferable to use the driver's line-of-sight information obtained from the line-of-sight detection unit 3. By using the line-of-sight information, it is possible to reduce the driver's discomfort caused by the operation of the alarm device while the driver is viewing the blind spot area. If the alarm can be canceled using the line-of-sight information, the influence on the steering wheel operation can be reduced. In order to determine that the driver has visually recognized the vicinity of the blind spot area boundary, it is only necessary to detect that the line of sight for a predetermined time is directed to the vicinity of the boundary of the blind spot area. When it is determined that the user has visually confirmed, the value of the alarm determination parameter A is set to 0 for a predetermined time. After the predetermined time elapses, the value of the alarm determination parameter A is calculated again, and thereafter, the above process is repeated to continuously monitor the blind spot area.

  In addition, a plurality of warning devices are installed in the passenger compartment, and the driver is operated by operating the warning device according to the location of the blind spot area in which the route obstruction is determined to appear by the route obstruction appearance determining unit 9. Can inform the location of the blind spot area. For example, if speakers and buzzers are installed in the front, rear, left and right of the room, and it is determined that a path obstruction appears from the blind spot area on the front left side of the vehicle, the speaker and buzzer installed in the front left of the vehicle By generating an alarm, the driver's attention can be directed to the left front side of the vehicle.

  According to this article, road-to-vehicle communication is estimated by estimating the blind spot area around the vehicle and calculating the degree of danger in the blind spot area based on geographical information, past and present traffic information, and weather information. It is possible to prevent collision accidents caused by blind spots autonomously without using vehicle-to-vehicle communication. Further, by calculating the degree of danger based on a plurality of information, it is possible to give a warning of a blind spot area corresponding to the traveling environment and the traveling time. In addition, by detecting the driver's line of sight and giving an alarm according to the driver's attention trend with respect to the blind spot area, the driver's blind spot area is prevented from being overlooked, and the driver feels uncomfortable with the vehicle periphery monitoring device alarm. Can be reduced. In addition, a plurality of warning sections are provided in the vehicle, and the warning section is operated according to the location where the blind spot area determined to be dangerous is present, thereby notifying the driver of the location of the blind spot area determined to be dangerous. As a result, it is possible to promote the driver's risk avoidance judgment.

The figure which shows a vehicle periphery monitoring apparatus. The block diagram which showed the flow of the process of a vehicle periphery monitoring apparatus. The figure which shows the process example of an obstruction recognition part. The figure explaining the process which a blind spot area estimation part performs. The figure which shows the example of a judgment of the course obstruction appearance judgment part 9. FIG.

Explanation of symbols

100 Obstacle detection unit 110 GPS
120 Acceleration / Yaw Rate Sensor 130 Direction Indicator 140 Wheel Speed Sensor 150 Steering Angle Sensor 200 Map Information Storage Device 210 Road Traffic Information Receiving Device 220 Arithmetic Device 230 Alarm Device 240 Traffic Accident Information Storage Device 300 Vehicle Perimeter Monitoring Device

Claims (8)

An alarm unit for giving an alarm to the driver;
An obstacle detection unit that detects obstacles around the vehicle as a point sequence of a plurality of detection points, and detects a relative distance, an azimuth angle, and a relative speed between the vehicle and each detection point;
A host vehicle motion detection unit for detecting a travel direction and a travel speed of the host vehicle;
Based on the traveling direction and traveling speed of the host vehicle detected by the host vehicle motion detecting unit, the host vehicle track locus estimating unit that estimates the host vehicle track locus;
A blind spot area estimator for estimating a blind spot area caused by an obstacle existing around the vehicle;
From the blind spot area estimated by the blind spot area estimation unit, assuming that a path obstruction that obstructs the own vehicle path trajectory estimated by the own vehicle path trajectory estimation unit appears, the position where the own vehicle and the path obstruction collide An estimated path obstruction appearance determination unit;
Based on the distance between the collision position estimated by the path obstruction appearance determination unit and the own vehicle, and the traveling speed of the own vehicle detected by the own vehicle motion detection unit, the degree of danger in the blind spot area is calculated, An alarm determination unit that determines the operation of the alarm unit according to the degree of danger,
The vehicle periphery monitoring device, wherein the warning unit gives a warning to the driver based on the determination of the warning determination unit. When there is a detection point whose distance between adjacent detection points is within a predetermined threshold among the plurality of detection points acquired by the obstacle detection unit, the adjacent detection points are obtained from the same obstacle. An obstacle recognition unit that determines that the detected point sequence is
The blind spot area estimation unit includes the detection point sequence including two half lines that pass through detection points at both ends where the detection point sequence is discontinuous, with the viewpoint of the obstacle detection unit as an origin. The vehicle periphery monitoring device according to claim 1, wherein an area excluding an area surrounded by the detection point sequence is determined as a blind spot area. A map information storage device that stores geographical information about the surface of the earth, and a vehicle position estimation unit that estimates a travel position of the vehicle;
The course obstruction appearance determination unit sets a boundary of the blind spot area in the predetermined range from the own vehicle in the vicinity of the own vehicle course locus, and stores the map information based on the travel position of the own vehicle. Geographic information near the boundary of the blind spot area obtained from the apparatus satisfies the geographical condition where a path obstruction can exist, and an obstacle that obstructs the course of the path obstruction to the own vehicle course locus is a boundary of the blind spot area. The vehicle periphery monitoring device according to claim 1, wherein when there is no nearby object, the route obstruction is determined to appear. The distance between the position of the host vehicle and the collision position is calculated, and the deceleration required for the host vehicle to stop at the collision position is calculated as the calculated distance, the vehicle speed, the friction coefficient between the road surface and the tire. Further comprising a deceleration calculation unit for calculating based on
4. The vehicle periphery monitoring device according to claim 1, wherein the warning determination unit gives a warning to the driver by setting the calculated deceleration as a risk level of the blind spot area. 5. A road traffic information receiving device for receiving road surface information of the own vehicle traveling path obtained from the map information storage device and road traffic information such as accidents, traffic jams, traffic regulation information and weather information on the road around the current vehicle A friction coefficient estimator for estimating a value of a friction coefficient between the road surface and the tire based on the weather information around the vehicle obtained from the vehicle,
The vehicle periphery monitoring device according to claim 4, wherein the deceleration calculation unit calculates a deceleration based on the friction coefficient estimated by the friction coefficient estimation unit. Based on the past and present traffic accident information around the vehicle obtained from the traffic accident information storage device storing past traffic accident statistical information and the road traffic information receiving device, the number of past accident occurrences is small, And if there is no accident or traffic jam in the current traffic situation, it will be a small value, and if there are many past accidents and the current traffic situation has an accident or traffic jam, it will be a large value. Accident occurrence parameter calculation unit to calculate,
6. The vehicle periphery monitoring device according to claim 5, wherein the warning determination unit gives a warning to the driver by calculating a degree of risk using the calculated accident occurrence parameter and the deceleration as independent variables. A line-of-sight detection unit that detects the line of sight of the driver;
When the warning determination unit detects that the line of sight is directed in the direction of the blind spot area where it is determined that a path obstruction that blocks the own vehicle path appears, the driving is performed in a state in which the risk level of the blind spot area is reduced. The vehicle periphery monitoring device according to any one of claims 1 to 6, which gives a warning to a person. The vehicle has a plurality of warning units,
The warning determination unit is configured to select the warning unit that issues a warning from the plurality of warning units according to the location of the blind spot area where it is determined that the obstruction obstructing the own vehicle path appears. The vehicle periphery monitoring apparatus according to any one of claims 1 to 7, wherein a driver is informed of a location where the blind spot area exists.

Images